Вы находитесь на странице: 1из 475

Chertezh-gibkikh-podatlivikh-svyazey-frilktsi-unikon-ankernie-bolti-friktsionno-podvizhnie-

soedineniya-pall-friction-damper-spbgasu

Чертежи гибких податливых связей фрикционных Фирмы Уникон анкерных болтов на фрикционно
подвижных энергопоглощающих соединениях с рассеивающими и энергопоглощающими
устройствами для снижения взрывной и сейсмической энергии

Демонстрация балочного энергопоглотителя с моделированием взаимодействия сооружения с


особыми воздействиями (ударной волной) и геологической средой , в том числе численным
нелинейным методом расчета в ПК SCAD в механике деформируемых сред и конструкций
методом оптимизации и идентификации статических и динамических задач теории устойчивости с
целью повышения надежности здания, путем обеспечения многокаскадного демпфирования при
динамических нагрузках и увеличения демпфирующих свойств, преимущественно при импульсных
растягивающих нагрузках от ударной взрывной волны

Chertezh-gibkikh-podatlivikh-svyazey-frilktsi-unikon-ankernie-bolti-friktsionno-podvizhnie-
soedineniya-pall-friction-damper-spbgasu

Аналогичная демонстрация важности разложения и поглощение сейсмической энергии,


использующего
изобретения проф ПГУПС дтн А.М.Уздина №№ 1143895, 1168755, 1174616, 2010136746 «Способ
защиты зданий и сооружений при взрыве с использованием сдвигоустойчивых легко
сбрасываемых соединений , использующие систему демпфирования фрикционности и
сейсмоизоляцию для поглощения взрывной энергии», полезная модель № 165076 «Опора
сейсмостойкая», № 154506 «Панель противовзрывная» для поглощение энергии трением
(вынуждает ограничитель) и быстро удалять энергию входа.
http://www.palldynamics.com/FrictionDampers.htm

http://www.palldynamics.com/Pall_Damper_Publications.htm

Опасности резонанса в структуре (рамке), особенно без разложения энергии также


иллюстрируются.

Вход энергии в различных частотах запасен структурой(рамкой). Когда не имеется никакого


разложения энергии, структура(рамка) может легко войти в резонанс, и это стимулирует
большие деформации.

С установленным сухого трения, энергия преобразована (рассеивается ) в высокую


температуру, что означает, что это больше не запасено как потенциальная энергия в системе,
результат - меньшие деформации, силы и полная лучшая стабильность.
https://www.iitk.ac.in/nicee/wcee/article/13_3446.pdf

Эти принципы применяются в полном масштабе в структурах, поскольку землетрясение будет


иметь различные преобладающие частоты.

Объединяя структуры энергопоглощение трением , может остаться упругим и может быть


разработан (предназначен) для меньших сил, заканчивающихся более низкой стоимостью,
лучше защищенные здания и сооружений https://www.quaketek.com/seismic-friction-dampers/

1
Pall Friction Dampers Video
https://www.quaketek.com/seismic-design/
https://www.youtube.com/watch?v=Pkhfy8AsRxo
Brief overview of Pall Friction Dampers. Posted by Pall Dynamics. For more information please visit
www.PallDynamics.com
https://www.youtube.com/channel/UC2G9tzlAYSwg3gS_BkAoR-Q
https://www.youtube.com/watch?v=-yLDBk9RNw4
http://www.palldynamics.com/DesignWithFrictionDampers.htm
http://www.palldynamics.com/SeismicDampers_ViscousDampers_EnergyDissipatingDe
vices.htm
https://www.google.ru/search?newwindow=1&sxsrf=ALeKk00E6Ilwk-
7M6bPLHvRU_JHJUPSjoA:1598185883944&source=univ&tbm=isch&q=Pall+Friction+D
ampers+Video&client=opera&sa=X&ved=2ahUKEwjK15avqrHrAhWns4sKHUnKDsYQs
AR6BAgJEAE&biw=1360&bih=640
https://www.youtube.com/watch?v=ZRJcowT24I8
A demonstration of the importance of energy dissipation using a Quaketek friction damper (force
limiter) to quickly remove the input energy. The dangers of resonance in a frame, particularly without
energy dissipation is also illustrated. The energy input at various frequencies is stored by the frame.
When there is no energy dissipation the frame can easily go into resonance and this induces large
deformations. With the friction damper installed, the energy is converted into heat which means it is
no longer stored as potential energy in the system, the result is smaller deformations, forces and
overall better stability. These principles apply at full scale in structures as an earthquake will have
various predominant frequencies. By integrating friction dampers structures can remain elastic and
can be designed for smaller forces resulting in lower cost, better protected buildings.
http://www.palldynamics.com/Pall_Damper_Publications.htm
http://palldynamics.com/pdf/Paper001955.pdf
https://www.iitk.ac.in/nicee/wcee/article/13_1955.pdf
Авторы исследуют Использование сдвиговых трубчатых балочных кольцевых конвертных
квадратных струнных упругопластических шарниров для обеспечения устойчивости сооружения
на особые воздействия от ударной волны, для рассеивания и поглощения взрывной энергии, в
том числе с использованием численного моделирования в ПК SCAD на основе изобретений проф
ПГУПС дтн А.М.Уздина №№ 1143895, 1168755, 1174616, 2010136746 «Способ защиты зданий и
сооружений при взрыве с использованием сдвигоустойчивых легко сбрасываемых соединений ,
использующие систему демпфирования фрикционности и сейсмоизоляцию для исключения
прогрессирующего обрушения на особые воздействия для обеспечения устойчивости сооружений
от ударной волны, за счет использования сдвиговых упругопластических шарниров и балочных
энергопоглотителей, в том числе нелинейным методом расчета в ПК SCAD, существующих зданий
в Бейруте, за счет рассеивания энергии и использования упругопластических шарниров.
http://www.palldynamics.com/pdf/Paper00901.pdf

2
Ключевые слова: прогрессирующее разрушение, аварии, особые воздействия, пластические
шарниры , сдвиговые связи, энергопоглощающие устройства , рассеивание энергии,
обеспечение устойчивости существующих сооружений, зданий от особых воздействиях за счет
рассеивания энергии и использования упругопластических шарниров и поглощение энергии
шарнирными балочными, трубчатыми, конвертными, квадратными, крестовидными, тросовыми,
струнными энергопоглотителями

The basic tasks of mathematical modeling in ПК SCAD

The authors investigate USE shift tubular балочных ring конвертных square string упругопластических
of hinges for maintenance of stability of a structure on the special influences from a shock wave, for
dispersion and absorption of explosive energy, including with use of numerical modeling in PK SCAD on
the basis of the inventions prof PGUPS dtn A.М.Uzdinf №№ 1143895, 1168755, 1174616, 2010136746 «
the Way of protection of buildings and structures at explosion with use sdvigoustoychivikh is easy
сбрасываемых of connections using system demfirovaniya, friktsionnosti and seismoisolation for
exception progressing обрушения on the special influences for maintenance of stability of structures from
a shock wave, at the expense of use shift упругопластических of hinges and балочных
энергопоглотителей, including nonlinear method of account in ПК SCAD, existing buildings in Beirut, at
the expense of dispersion of energy and use uprugoplsticheskikh of hinges.

Key words: progressing destruction, failures, special influences, plastic hinges, shift
communications(connections), энергопоглощающие of the device, dispersion of energy, maintenance of
stability of existing structures, buildings from the special influences at the expense of dispersion of energy
both use упругопластических of hinges and absorption of energy шарнирными балочными, tubular,
конвертными, square, крестовидными, тросовыми, string энергопоглотителями

Предложена методология научно-технического обоснования эффективности повышения


надежности лестниц , счет устройство энергопоглощающих устройств на фрикционно –подвижных
соединениях и фрикционно-демпфирующих опорах. На конкретных примерах произведены
нелинейные расчеты в ПК SCAD, систем устройства энергопоглощающих устройств .
http://www.palldynamics.com/pdf/Paper001946.pdf

Литература

Работа выполнена при финансовой поддержке организация «Сейсмофонд» при СПб ГАСУ
ОГРН:1022000000824, ИНН 2014000780

Президента организации «Сейсмофонд» Хасан Нахоевича Мажиева, проф. дфмн СПб ГУ Олег
Алексеевича Малафеев, проф дтн ПГУПС Александр Михайловича Уздина , ктн доцент СПб ГАСУ
Ирина Утарбаевна Аубакарова , ученый секретарь СПб ГАСУ Елен Ивановна Андреева

Президент организации «Сейсмофонд», Мажиев Хасан Нажоевич Президент организации


«Сейсмофонд»ИНН 2014000780 t3487810@interzet.ru, зам президента организации
«Сейсмофонд» ОГРН 1022000000824
Зам президента организации «Сейсмофонд» ОГРН 1022000000824 O.A.Малафеев доктор
физико-математических наук, профессор кафедры моделирования социально-экономических
систем, заведующий кафедрой Санкт-Петербургский государственный университет
produsertifikatsiya@yahoo.com
Зам президента организации «Сейсмофонд» ОГРН 1022000000824 , ассистент –стажер ученый
секретарь кафедры ТСМиМ СПб ГАСУ, инж –мех ЛПИ им Калинина Е.И.Андреева ,
t89995354729@bk.ru тел 999-535 47 29

Зам президента организации «Сейсмофонд» ИНН 2014000780 ученый секретарь кафедры


ТСМиМ СПб ГАСУ ктн, доцент Аубакирова Ирина Утарабаевна с9995354729@yandex.ru

3
зам президента организации «Сейсмофонд» ОГРН 1022000000824, научный консультант проф
.дтн ПГУПС Уздин Александр Михайлович sertifikatsiyaprodutsii@yandex.ru

Х.Н.Мажиев, О.А.Малафеев, А.М.Уздин, И.У.Аутарабаевна, Е.И.Андреева


Санкт-Петербургский государственный Архитектурно -Строительный Университет , 190005, СПб,
2-я Красноармейская ул. д 4 , организация «Сейсмофонд» ОГРН:1022000000824, ИНН 2014000780

Математическое и компьютерное моделирование взаимодействия сооружения с особыми


воздействиями и геологической средой , в том числе численным нелинейным методом расчета в
ПК SCAD в механике деформируемых сред и конструкций методом оптимизации и идентификации
статических и динамических задач теории устойчивости с целью повышения надежности здания,
путем обеспечения многокаскадного демпфирования при динамических нагрузках и увеличения
демпфирующих свойств, преимущественно при импульсных растягивающих нагрузках от ударной
взрывной волны выполнена организацией СПб ГАСУ совместно с общественной организацией
«Сейсмофонд» по адрес: 190005, СПб , 2-я Красноармейская ул д 4 СПб ГАСУ на строительном
факультете кафедры технологии строительных материалов и метрологии, ТСМиМ СПб ГАСУ

СПб ГАСУ и организацией «Сейсмофонд» проведены лабораторные испытания узлов и


фрагментов энергопоглотителй и демпфирующей сейсмоизоляции по использованию
сдвиговых, трубчатых, балочных, кольцевых, конвертных, квадратных, струнных и тросовых
энергопоглощающих упругопластических шарниров, для обеспечения устойчивости сооружения
на особые воздействия от ударной взрывной волны, для рассеивания и поглощения взрывной
энергии, в том числе с использованием численного моделирования в ПК SCAD на основе
изобретений проф ПГУПС дтн А.М.Уздина №№ 1143895, 1168755, 1174616, 2010136746 «Способ
защиты зданий и сооружений при взрыве с использованием сдвигоустойчивых легко
сбрасываемых соединений , использующие систему демпфирования фрикционности и
сейсмоизоляцию для поглощения взрывной энергии», полезная модель № 165076 «Опора
сейсмостойкая», № 154506 «Панель противовзрывная»

Проведены лабораторные испытания с применением численного моделирования методом


оптимизации и идентификации разрушенных (например ) от взрыва зданий в Бейруте (Ливан) в
августе 2020 от ударной взрывной волны для повышения надежности и использовании новых
техническое решение предназначено для защиты существующих зданий в Бейруте (Ливан),
сооружений, мостов, магистральных трубопроводов, линий электропередач, от
террористических актов (взрывов) на основе использования фрикци –болта ( см заявку на
изобретение № 2018105803/20 (008844) от 11.05.2018 «Антисейсмическое фланцевое
фрикционно-подвижное соединение для трубопроводов» F 16L 23/02) и фрикционно-подвижных
соединениях (ФПС) при знакопеременных нагрузках и многокаскадном демпфировании и
динамических нагрузках на протяжных фрикционное- податливых соединений проф. ПГУПС дтн
Уздина А М "Болтовое соединение" №№ 1143895 , 1168755 , 1174616 "Болтовое соединение
плоских деталей" для лестничных маршей повышенной надежности.

Известны фрикционные соединения для защиты от особых воздействий , железнодорожных


мостов , от динамических взрывных воздействий. Известно, например, болтовое соединение
плоских деталей встык, патент RU №1174616, F15B5/02 с пр. от 11.11.1983, RU 2249557 D 66C
7/00 " Узел упругого соединения трех главного рельса с подкрановой балкой ", RU № 2148 805
G 01 L 5/24 "Способ определения коэффициента закручивания резьбового соединения "
Известна Японо-Американская фирма RUBBER BEARING FRICTION DAMPER (RBFD)
HTTPS://WWW.DAMPTECH.COM/-RUBBER-BEARING-FRICTION-DAMPER-RBFD
HTTPS://WWW.DAMPTECH.COM/-RUBBER-BEARING-FRICTION-DAMPER-RBFD
https://www.damptech.com/for-buildings-cover https://www.youtube.com/watch?v=r7q5D6516qg
Авторы исследуют системы повышение взрывостойкости существующих зданий, сооружений
повышенной надежности , взрывозащиты современных зданий и сооружений в г Бейруте (Ливан)
.

4
Предложена методология научно-технического обоснования повышение взрывостойоксти зданий
и сооружений повышенной надежности на фрикционно-демпфирующих опорах с
энергопоглощающими узлами. На конкретных примерах произведены нелинейные расчеты систем
повышение взрывостойоксти существующих зданий и сооружений повышенной надежности .

Отмечается так же важность пересмотра действующих нормативных документов и методов


расчета зданий и сооружений повещенной взрывостойкости на воздействия воздушной волны ,
расчет на взрывную ударную волну , зданий и сооружений, согласно изобретениям СПб ГАСУ и
организации «Сейсмофонд».

Введение. Взаимодействия зданий в Бейруте (Ливане ) на особые воздействия для обеспечения


устойчивости сооружений , от ударной волны, за счет использования сдвиговых
упругопластических шарниров и балочных энергопоглотителей, в том числе нелинейным
методом расчета в ПК SCAD, повышенной надежности разработана для примененные на
объектах повышенного риска, вокзалы, аэропорт , зрелищные объекты с использованием
фрикционо-подвижных соединениях, не имеют аналогов в мировой практике взрывозащиты
жилых зданий с аркой , где имеется лестница , само слабое место в здании ,на особые
воздействия для обеспечения устойчивости сооружений , от ударной волны, за счет
использования сдвиговых балочных, трубчатых, квадратных, струнных, тросовых, колцевых
упругопластических энергопоглощающих шарниров , балочных энергопоглотителей, в том числе
нелинейным методом расчета в ПК SCAD

Их высокие защитные качества обеспечиваются как при укреплении и усилении задний,


сооружений, мостов, трубопроводов , линий электропередач, АЭС, так и при максимальных
расчетных особых воздействий или землетрясениях. Эта система взрывозащиты с
энергопоглощающими узлами крепления с пластическим шарниром позволяет прогнозировать
характер, возможного накопления повреждений в конструкции, от воздействия ударной волны,
сохранить здание, сооружение, лестничный о марш, мост в ремонтопригодном состоянии в случае
разрушительного особого воздействия или землетрясения, а также обеспечивает нормальную
эксплуатацию . частично разрушенного здания, лестничного марша , не приводя к
прогрессирующему обрушению зданий сооружения, железнодорожного моста, магитарльного
трубопровода пути при динамических особых воздействиях или при многокасадном
демпфировании сооружения, здания, моста, трубопровода

C заявками на изобретение демпфирующих сдвиговых энернопоглотителей для обеспечения


устойчивости сооружений , от ударной волны, за счет использования сдвиговых
упругопластических шарниров и балочных энергопоглотителей, от особых воздействий,
(интеллектуальная собственность передается с альбомом специальные технические условия
(СТУ) можно ознакомится по ссылкам:

Описание изобретения на полезную модель Сейсмостойкая фрикционно 18


стр https://yadi.sk/i/JZ0YxoW0_V6FCQ

Заявка на изобретение полезную модель Энергопоглощающие дорожное барьерное


ограждение 23 стр https://yadi.sk/d/dWKraP12fvXAlA

Описание изобретения на полезную модель Взрывостойкая лестница 10


стр https://yadi.sk/i/EDoOs4AFUWKYEg

Заявка на изобретение полезная модель Опора сейсмоизолирующая гармошка 20


стр https://yadi.sk/i/JOuUB_oy2sPfog

Заявка на полезную модель Опора сейсмоизолирующая маятниковая 32


стр https://yadi.sk/i/Ba6U0Txx-flcsg

5
Виброизолирующая опора Е04Н 9 02 РЕФЕРАТ изобретения полезная 17
стр https://yadi.sk/i/dZRdudxwOald2w

Обеспечение взрывостойкости существующих железнодорожных мостов на основе 15


стр https://yadi.sk/i/en6RGTLgfhrg_A
Доклад в СПб ГАСУ усиление опор Крымского моста https://yadi.sk/i/RpW2sh5lMdx35A

Скачать научную статью Сейсмофонд при СПб ГАСУ( опубликованную в США, Японии и др
странах ), можно по ссылке : Использование лего сбрасываемых конструкций для повышения
сейсмостойкости сооружений http://scienceph.ru/f/science_and_world_no_3_43_march_vol_i.pdf

Изобретения с демпфирующей сейсмоизоляций «Сейсмофонд» широк используются


американской фирмой RUBBER BEARING FRIKTION DAMPER (RBFD) в Японии, Новой Зеландии,
США, Китае, Тайване и др странах https://www.damptech.com/-rubber-bearing-friction-damper-
rbfd https://www.damptech.com/for-buildings-cover
http://downloads.hindawi.com/journals/sv/2018/5630746.pdf
https://www.youtube.com/watch?v=r7q5D6516qg

Теория сейсмостойкости находится в кризисе, а жизнь миллионов граждан проживающих в ЖБ


гробах не относится к государственной безопасности http://www.myshared.ru/slide/971578/
https://yadi.sk/i/JfXt8hs_aXcKRQ https://yadi.sk/i/p5IgwFurPlgp1w

Оценка возможности инициирования сейсмического геофизического и техногенного оружия с


применением существующих технических средств и
технологий https://yadi.sk/i/3VmQxa78RhhBBA
ГОСТ 6249-52 «Шкала для определения силы землетрясения в пределах от 6 до 9 баллов»
http://scaleofintensityofearthquakes.narod.ru http://scaleofintensityofearthquakes2.narod.ru
http://scaleofintensityofearthquakes3.narod.ru http://peasantsinformagency1.narod.ru
http://s-a-m-a-r-a-citi.narod.ru http://sergeyshoygu.narod.ru/pdf1.pdf

Обеспечение взрывостойкости существующих железнодорожных мостов на основе 15


стр https://yadi.sk/i/en6RGTLgfhrg_A

Патенты изобретения взрывозащите противовзрывная https://yadi.sk/i/-PwJxeHVvI_eoQ

Научный доклад на 67 конференции СПб ГАСУ 4 стр https://yadi.sk/i/sMuk8V-J0Ui_lw

Научная статья в журнале СПб ГАСУ https://yadi.sk/i/Vf_86hLPmeYIsw


Доклад на конференции изобретателей Попов ЛПИ Политех 5 стр https://yadi.sk/i/c1D-6wvsIeJWnA

Антисейсмическое фланцевое фрикционн 4 стр https://yadi.sk/i/pXaZGW6GNm4YrA

Обеспечение взрывостойкости существующих лестничных маршей 8 стр https://yadi.sk/i/ZJNyX-


y0gsfEyQ

Доклад сообщение научное Испытание математических моделей ФПС 60 стр + выводы


https://yadi.sk/d/6lNXCB4lw-HgpA

Научная статья доклад сообщения конференции с 5 по 7 февраля 2014 19


стрhttps://yadi.sk/i/CnFN36oKLYPpzQ

Научное сообщение доклад на 67 конференции проходившей в начале 3 5 февраля 2010 г в


СПб ГАСУ стр 208 стр 211 2 страницы https://yadi.sk/i/MaKtKmd5GP9ecw

6
Доклад сообщение Маживеа Уздина Испытание математических моделей на сейсмостойкость 137
стр https://yadi.sk/d/MDvdSPojHUpe3w

ЛИСИ Научные статьи изобретателя СПбГАСУ научной конференции 9 стр


https://yadi.sk/i/uLbA_SwO5GHO2w
http://www.palldynamics.com/pdf/Paper001946.pdf
Приложение: изобретение СПОСОБ ЗАЩИТЫ ЗДАНИЯ И СООРУЖЕНИЯ ПРИ ВЗРЫВЕ С
ИСПОЛЬЗОВАНИЕМ СДВИГОУСТОЙЧИВЫХ И ЛЕГКОСБРАСЫВАЕМЫХ СОЕДИНЕНИЙ,
ИСПОЛЬЗУЮЩИЕ СИСТЕМУ ДЕМПФИРОВАНИЯ ФРИКЦИОННОСТИ И СЕЙСМОИЗОЛЯЦИЮ
ДЛЯ ПОГЛОЩЕНИЯ ВЗРЫВНОЙ И СЕЙСМИЧЕСКОЙ ЭНЕРГИИ РОССИЙСКАЯ ФЕДЕРАЦИЯ

ФЕДЕРАЛЬНАЯ СЛУЖБА ПО ИНТЕЛЛЕКТУАЛЬНОЙ СОБСТВЕННОСТИ


(19)
RU
(11)

2010 136 746


(13)

(51) МПК
* E04C 2/00 (2006.01)

(12) ЗАЯВКА НА ИЗОБРЕТЕНИЕ


Состояние делопроизводства:
Экспертиза завершена (последнее изменение статуса: 02.10.2013) (21)(22) Заявка: 2010136746/03,
01.09.2010
Приоритет(ы):
(22) Дата подачи заявки: 01.09.2010
(43) Дата публикации заявки: 20.01.2013 Бюл. № 2
Адрес для переписки:
443004, г.Самара, ул.Заводская, 5, ОАО "Теплант"
(71) Заявитель(и):
Открытое акционерное общество "Теплант" (RU)
(72) Автор(ы):
Подгорный Олег Александрович (RU),
Акифьев Александр Анатольевич (RU),
Тихонов Вячеслав Юрьевич (RU),
Родионов Владимир Викторович (RU),
Гусев Михаил Владимирович (RU),
Коваленко Александр Иванович (RU)

(54) СПОСОБ ЗАЩИТЫ ЗДАНИЯ И СООРУЖЕНИЯ ПРИ ВЗРЫВЕ С ИСПОЛЬЗОВАНИЕМ


СДВИГОУСТОЙЧИВЫХ И ЛЕГКОСБРАСЫВАЕМЫХ СОЕДИНЕНИЙ, ИСПОЛЬЗУЮЩИЕ СИСТЕМУ
ДЕМПФИРОВАНИЯ ФРИКЦИОННОСТИ И СЕЙСМОИЗОЛЯЦИЮ ДЛЯ ПОГЛОЩЕНИЯ ВЗРЫВНОЙ
И СЕЙСМИЧЕСКОЙ ЭНЕРГИИ
(57) Формула изобретения
1. Способ защиты здания от разрушений при взрыве или землетрясении, включающий
выполнение проема/проемов рассчитанной площади для снижения до допустимой величины
взрывного давления, возникающего во взрывоопасных помещениях при аварийных внутренних
взрывах, отличающийся тем, что в объеме каждого проема организуют зону, представленную в
виде одной или нескольких полостей, ограниченных эластичным огнестойким материалом и
установленных на легкосбрасываемых фрикционных соединениях при избыточном давлении

7
воздухом и землетрясении, при этом обеспечивают плотную посадку полости/полостей во всем
объеме проема, а в момент взрыва и землетрясения под действием взрывного давления
обеспечивают изгибающий момент полости/полостей и осуществляют их выброс из проема и
соскальзывают с болтового соединения за счет ослабленной подпиленной гайки.
2. Способ по п.1, отличающийся тем, что «сэндвич»-панели, щитовые панели смонтированы на
высокоподатливых с высокой степенью подвижности фрикционных, скользящих соединениях с
сухим трением с включением в работу фрикционных гибких стальных затяжек диафрагм жесткости,
состоящих из стальных регулируемых натяжений затяжек сухим трением и повышенной
подвижности, позволяющие перемещаться перекрытиям и «сэндвич»-панелям в горизонтали в
районе перекрытия 115 мм, т.е. до 12 см, по максимальному отклонению от вертикали 65 мм, т.е.
до 7 см (подъем пятки на уровне фундамента), не подвергая разрушению и обрушению
конструкции при аварийных взрывах и сильных землетрясениях.
3. Способ по п.2, отличающийся тем, что каждая «сэндвич»-панель крепится на сдвигоустойчивых
соединениях со свинцовой, медной или зубчатой шайбой, которая распределяет одинаковое
напряжение на все четыре-восемь гаек и способствует одновременному поглощению сейсмической
и взрывной энергии, не позволяя разрушиться основным несущим конструкциям здания, уменьшая
вес здания и амплитуду колебания здания.
4. Способ по п.3, отличающийся тем, что за счет новой конструкции сдвигоустойчивого
податливого соединения на шарнирных узлах и гибких диафрагмах «сэндвич»-панели могут
монтироваться как самонесущие без стального каркаса для малоэтажных зданий и сооружений.
5. Способ по п.4, отличающийся тем, что система демпфирования и фрикционности и поглощения
сейсмической энергии может определить величину горизонтального и вертикального перемещения
«сэндвич»-панели и определить ее несущую способность при землетрясении или взрыве прямо на
строительной площадке, пригрузив «сэндвич»-панель и создавая расчетное перемещение по
вертикали лебедкой с испытанием на сдвиг и перемещение до землетрясения и аварийного
взрыва прямо при монтаже здания и сооружения.

6. Способ по п.5, отличающийся тем, что расчетные опасные перемещения определяются,


проверяются и затем испытываются на программном комплексе ВК SCAD 7/31 r5, ABAQUS 6.9,
MONOMAX 4.2, ANSYS, PLAKSIS, STARK ES 2006, SoliddWorks 2008, Ing+2006, FondationPL 3d,
SivilFem 10, STAAD.Pro, а затем на испытательном при объектном строительном полигоне прямо
на строительной площадке испытываются фрагменты и узлы, и проверяются экспериментальным
путем допустимые расчетные перемещения строительных конструкций (стеновых «сэндвич»-
панелей, щитовых деревянных панелей, колонн, перекрытий, перегородок) на возможные при
аварийном взрыве и при землетрясении более 9 баллов перемещение по методике разработанной
испытательным центром ОО «Сейсмофонд» - «Защита и безопасность городов».
http://www.palldynamics.com/pdf/Paper001949.pdf
https://www.iitk.ac.in/nicee/wcee/article/13_3446.pdf
Панель противовзрывная полезная модель изобретение
РОССИЙСКАЯ ФЕДЕРАЦИЯ

ФЕДЕРАЛЬНАЯ СЛУЖБА
ПО ИНТЕЛЛЕКТУАЛЬНОЙ СОБСТВЕННОСТИ
(19)
RU (11)

154 506 (13)

U1 (51) МПК
* E04B 1/92 (2006.01)

(12) ОПИСАНИЕ ПОЛЕЗНОЙ МОДЕЛИ К ПАТЕНТУ


Статус:

8
не действует (последнее изменение статуса: 07.08.2018) (21)(22) Заявка: 2014131653/03,
30.07.2014
(24) Дата начала отсчета срока действия патента:
30.07.2014
Приоритет(ы):
(22) Дата подачи заявки: 30.07.2014
(45) Опубликовано: 27.08.2015 Бюл. № 24
Адрес для переписки:
197371, Санкт-Петербург, пр. Королева, 30, корп. 1, кв. 135, Коваленко Александр Иванович
(72) Автор(ы):
Андреев Борис Александрович (RU),
Коваленко Александр Иванович (RU)
(73) Патентообладатель(и):
Андреев Борис Александрович (RU),
Коваленко Александр Иванович (RU)
(54) ПАНЕЛЬ ПРОТИВОВЗРЫВНАЯ
(57) Реферат:
Техническое решение относится к области строительства и предназначено для защиты
помещений от возможных взрывов. Конструкция позволяет обеспечить надежный и быстрый сброс
легкосбрасываемой панели, сброс давления при взрыве и зависание панели на опорной плите,
Конструкция представляет собой опорную плиту с расчетным проемом, которая жестко крепится на
каркасе защищаемого сооружения. На опорной плите крепежными элементами, имеющими
ослабленное резьбовое поперечное сечение, закреплена панель легкосбрасываемая.
Ослабленное резьбовое соединение каждого крепежного элемента образовано лысками
выполненными с двух сторон резьбовой части. Кроме того опорная плита и легкосбрасываемая
панель соединены тросом один конец которого жестко закреплен на опорной плите, а другой конец
соединен с крепежным элементом через планку, с возможностью перемещения. 4 ил.

Техническое решение относится к области строительства и предназначено для защиты


помещений содержащих взрывоопасные среды.
Известна панель для легкосбрасываемой кровли взрывоопасных помещений по Авт.св. 617552,
М.Кл. 2 E04B 1/98 с пр. от 21.11.75. Панель включает ограждающий элемент с шарнирно
закрепленными на нем поворотными скобами, взаимодействующими через опоры своими
наружными полками с несущими элементами. С целью защиты от воздействия ветровой нагрузки,
панель снабжена подвижной плитой, шарнирно соединенной с помощью тяг с внутренними
концами поворотных скоб, которые выполнены Т-образными. Недостатком предлагаемой
конструкции является низкая надежность шарнирных соединений при переменных внешних и
внутренних нагрузках. Известна также легкосбрасываемая ограждающая конструкция
взрывоопасных помещений по Патенту SU 1756523, МПК5 E06B 5/12 с пр. от 05.10.1990.

Указанная конструкция содержит поворотную стеновую панель, состоящую из нижней и верхней


секций и соединенную с каркасом временной связью. Нижняя секция в нижней части шарнирно
связана с каркасом здания, а в верхней части - шарнирно соединена с верхней секцией панели.
Верхняя секция снабжена роликами, установленными в направляющих каркаса здания.
Недостатком указанной конструкции является низкая надежность вызванная большим количеством
шарнирных соединений, требующих высокой точности изготовления в условиях строительства.
Известна также противовзрывная панель по Патенту RU 2458212, E04B 1/92 с пр. от 13.04.2011,
которую выбираем за прототип. Изобретение относится к защитным устройствам применяемым во
взрывоопасных объектах.

Противопожарная панель содержит металлический каркас с бронированной обшивкой и


наполнителем-свинцом. Панель имеет четыре неподвижных патрубка-опоры, а в покрытии
взрывоопасного объекта жестко заделаны четыре опорных стержня, которые телескопически
вставлены в неподвижные патрубки-опоры панели. Наполнитель выполнен в виде дисперсной

9
системы воздух-свинец, а опорные стержни выполнены упругими. Недостатком вышеуказанной
панели является низкая надежность срабатывания телескопических сопряжений при воздействии
переменных внешних и внутренних нагрузок.

Задачей заявляемого устройства является обеспечение надежности открывания проема при


взрыве (сбрасывания легкосбрасываемой панели) за минимальное время и обеспечение
зависания панели после сброса.

Сущность заявляемого решения состоит в том, что для защиты стен, оборудования и персонала
от возможного взрыва, помещение снабжено панелью противовзрывной, обеспечивающей
надежное и быстрое открытие проема при взрыве и сброс избыточного давления, а также
зависание панели на плите опорной. Панель противовзрывная содержит плиту опорную которая
жестко закреплена на стене защищаемого помещения и имеет проем соответствующий проему в
стене, а с другой стороны плиты опорной винтами с резьбой, ослабленной по сечению, закреплена
панель легкосбрасываемая.

Площадь проема плиты опорной и проема помещения определяется в зависимости от объема


помещения, от взрывоопасной среды, температуры горения, давления, скорости распространения
фронта пламени и др. параметров. Винты имеют резьбовую часть, ослабленную по сечению с двух
сторон лысками до размера <Z> и т. о. образуется ослабленное резьбовое сопряжение,
разрушаемое под воздействием взрывной волны.
Сущность предлагаемого решения поясняется чертежами где:
на фиг. 1 изображен разрез Б-Б (фиг. 2) панели противовзрывной;
на фиг. 2 изображен разрез ?-A (фиг. 1);
на фиг. 3 изображен вид по стрелке В (фиг. 1) в увеличенном масштабе;
на фиг. 4 изображен разрез Г-Г (фиг. 2), узел крепления троса в увеличенном масштабе.

Панель противовзрывная состоит из опорной плиты 1, которая жестко крепится к каркасу


защищаемого помещения (на чертеже не показано). В каркасе помещения и в опорной плите
выполнен проем 2, имеющий расчетную площадь S=b*h, которая зависит от объема защищаемого
помещения, температуры горения, давления, скорости распространения фронта пламени и др.
параметров. На опорной плите 1, резьбовыми крепежными элементами, например саморежущими
шурупами 3, имеющими ослабленное поперечное резьбовое сечение, закреплена
легкосбрасываемая панель 4. Кроме того, легкосбрасываемая панель соединена с опорной плитой
гибким узлом, состоящим из планки 5, закрепленной с одной стороны на тросе 6, а с др. стороны
сопряженной с крепежным элементом 3. Ослабленное поперечное сечение резьбовой части
образовано лысками, выполненными с двух сторон по всей длине резьбы до размера <Z>.
Ослабленная резьбовая часть в совокупности с обычным резьбовым отверстием в опорной плите
1, образуют ослабленное резьбовое сопряжение, разрушаемое под действием взрывной волны.
Разрушение (вырыв) в ослабленном резьбовом соединении возможно или за счет разрушения
резьбы в опорной плите, или за счет среза резьбы крепежного элемента-самореза 3, в
зависимости от геометрии резьбы и от соотношения пределов прочности материалов самореза и
плиты опорной. Рассмотрим пример. На опорной плите 1 толщиной 5 мм, изготовленной из стали
3, самосверлящими шурупами 3 размером 5,5/6,3?105, изготовленными из стали У7А, закреплена
легкосбрасываемая панель 4, изготовленная из
стали 20. Усилие вырыва при стандартной резьбе для одного шурупа составляет 1500 кгс.
Опытным путем установлено, что после доработки шурупа путем стачивания резьбы с двух сторон
до размера Z=3 мм, величина усилия вырыва составляет 700 кгс. Соответственно, при креплении
плиты четырьмя шурупами, усилие вырыва составит 2800 кгс. При условии, что площадь проема
S=10000 см2, распределенная нагрузка для вырыва должна быть не менее 0,28 кгс/см2. Таким
образом, зная параметры взрывоопасной среды, объем и компоновку защищаемого помещения,
выбираем конструкцию крепежных элементов после чего, в зависимости от заданного усилия
вырыва, можно определить величину <Z> - толщину ослабленной части резьбы.
Панель противовзрывная работает следующим образом. При возникновении взрывной нагрузки,
взрывная волна через проем 2 в опорной плите 1 воздействует по площади легкосбрасываемой

10
панели 4, закрепленной на опорной плите 1 четырьмя саморежущими шурупами 3, имеющими
ослабленное резьбовое сечение. При превышении взрывным усилием предела прочности
резьбового соединения, резьбовое соединение разрушается по ослабленному сечению,
легкосбрасываемая панель освобождается от механического крепления, после чего сбрасывается,
сечение проема открывается и давление сбрасывается до атмосферного. После сбрасывания
панель легкосбрасываемая зависает на тросе 6, один конец которого закреплен на опорной плите,
а другой, через планку 5 сопряжен с крепежным элементом 3.
Формула полезной модели

1. Панель противовзрывная, содержащая опорную плиту, на которой резьбовыми крепежными


элементами закреплена панель легкосбрасываемая, отличающаяся тем, что в опорной плите
выполнен проем, а панель легкосбрасываемая выполнена сплошной, при этом крепежные
элементы, скрепляющие панель легкосбрасываемую с опорной плитой, имеют ослабленное
поперечное сечение резьбовой части, образованное лысками, выполненными с двух сторон по
всей длине резьбы и, кроме того, панель легкосбрасываемая соединена с опорной плитой тросом,
один конец которого жестко закреплен в опорной плите, а другой конец соединен с панелью
легкосбрасываемой.

2. Панель противовзрывная по п.1, отличающаяся тем, что трос соединен с панелью


легкосбрасываемой через планку, сопряженную с крепежным элементом.

Опора сейсмостойкая полезная модель изобретение Коваленко


РОССИЙСКАЯ ФЕДЕРАЦИЯ

ФЕДЕРАЛЬНАЯ СЛУЖБА
ПО ИНТЕЛЛЕКТУАЛЬНОЙ СОБСТВЕННОСТИ
(19)

RU

(11)

165 076
(13)

U1

(51) МПК
* E04H 9/02 (2006.01)

(12) ОПИСАНИЕ ПОЛЕЗНОЙ МОДЕЛИ К ПАТЕНТУ


Статус:
не действует (последнее изменение статуса: 26.09.2019) (21)(22) Заявка: 2016102130/03,
22.01.2016
(24) Дата начала отсчета срока действия патента:
22.01.2016
Приоритет(ы):
(22) Дата подачи заявки: 22.01.2016
(45) Опубликовано: 10.10.2016 Бюл. № 28
Адрес для переписки:
197371, Санкт-Петербург, пр. Королева, 30, корп. 1, кв. 135, Коваленко Александр Иванович
(72) Автор(ы):
Андреев Борис Александрович (RU),
Коваленко Александр Иванович (RU)
(73) Патентообладатель(и):

11
Андреев Борис Александрович (RU),
Коваленко Александр Иванович (RU)
(54) ОПОРА СЕЙСМОСТОЙКАЯ
(57) Реферат:

Опора сейсмостойкая предназначена для защиты объектов от сейсмических воздействий за счет


использования фрикцион но податливых соединений. Опора состоит из корпуса в котором
выполнено вертикальное отверстие охватывающее цилиндрическую поверхность щтока. В
корпусе, перпендикулярно вертикальной оси, выполнены отверстия в которых установлен
запирающий калиброванный болт. Вдоль оси корпуса выполнены два паза шириной <Z> и длиной
<I> которая превышает длину <Н> от торца корпуса до нижней точки паза, выполненного в штоке.
Ширина паза в штоке соответствует диаметру калиброванного болта. Для сборки опоры шток
сопрягают с отверстием корпуса при этом паз штока совмещают с поперечными отверстиями
корпуса и соединяют болтом, после чего одевают гайку и затягивают до заданного усилия.
Увеличение усилия затяжки приводит к уменьшению зазора<Z>корпуса, увеличению сил трения в
сопряжении корпус-шток и к увеличению усилия сдвига при внешнем воздействии. 4 ил.

Предлагаемое техническое решение предназначено для защиты сооружений, объектов и


оборудования от сейсмических воздействий за счет использования фрикционно податливых
соединений. Известны фрикционные соединения для защиты объектов от динамических
воздействий. Известно, например Болтовое соединение плоских деталей встык по Патенту RU
1174616, F15B 5/02 с пр. от 11.11.1983. Соединение содержит металлические листы, накладки и
прокладки. В листах, накладках и прокладках выполнены овальные отверстия через которые
пропущены болты, объединяющие листы, прокладки и накладки в пакет. При малых
горизонтальных нагрузках силы трения между листами пакета и болтами не преодолеваются. С
увеличением нагрузки происходит взаимное проскальзывание листов или прокладок относительно
накладок контакта листов с меньшей шероховатостью.

Взаимное смещение листов происходит до упора болтов в края овальных отверстий после чего
соединения работают упруго. После того как все болты соединения дойдут до упора в края
овальных отверстий, соединение начинает работать упруго, а затем происходит разрушение
соединения за счет смятия листов и среза болтов. Недостатками известного являются:
ограничение демпфирования по направлению воздействия только по горизонтали и вдоль
овальных отверстий; а также неопределенности при расчетах из-за разброса по трению. Известно
также Устройство для фрикционного демпфирования антиветровых и антисейсмических
воздействий по Патенту TW 201400676 (A)-2014-01-01. Restraint anti-wind and anti-seismic friction
damping device, E04B 1/98, F16F 15/10. Устройство содержит базовое основание, поддерживающее
защищаемый объект, нескольких сегментов (крыльев) и несколько внешних пластин. В сегментах
выполнены продольные пазы.

Трение демпфирования создается между пластинами и наружными поверхностями сегментов.


Перпендикулярно вертикальной поверхности сегментов, через пазы, проходят запирающие
элементы - болты, которые фиксируют сегменты и пластины друг относительно друга. Кроме того,
запирающие элементы проходят через блок поддержки, две пластины, через паз сегмента и
фиксируют конструкцию в заданном положении. Таким образом получаем конструкцию опоры,
которая выдерживает ветровые нагрузки но, при возникновении сейсмических нагрузок,
превышающих расчетные силы трения в сопряжениях, смещается от своего начального
положения, при этом сохраняет конструкцию без разрушения.

Недостатками указанной конструкции являются: сложность конструкции и сложность расчетов из-


за наличия большого количества сопрягаемых трущихся поверхностей.
Целью предлагаемого решения является упрощение конструкции, уменьшение количества
сопрягаемых трущихся поверхностей до одного сопряжения отверстие корпуса - цилиндр штока, а
также повышение точности расчета.

12
Сущность предлагаемого решения заключается в том, что опора сейсмостойкая выполнена из
двух частей: нижней - корпуса, закрепленного на фундаменте и верхней - штока, установленного с
возможностью перемещения вдоль общей оси и с возможностью ограничения перемещения за
счет деформации корпуса под действием запорного элемента. В корпусе выполнено центральное
отверстие, сопрягаемое с цилиндрической поверхностью штока, и поперечные отверстия
(перпендикулярные к центральной оси) в которые устанавливают запирающий элемент-болт.
Кроме того в корпусе, параллельно центральной оси, выполнены два открытых паза, которые
обеспечивают корпусу возможность деформироваться в радиальном направлении. В теле штока,
вдоль центральной оси, выполнен паз ширина которого соответствует диаметру запирающего
элемента (болта), а длина соответствует заданному перемещению штока.

Запирающий элемент создает нагрузку в сопряжении шток-отверстие корпуса, а продольные пазы


обеспечивают возможность деформации корпуса и «переход» сопряжения из состояния
возможного перемещения в состояние «запирания» с возможностью перемещения только под
сейсмической нагрузкой. Длина пазов корпуса превышает расстояние от торца корпуса до нижней
точки паза в штоке.

Сущность предлагаемой конструкции поясняется чертежами, где на фиг. 1 изображен разрез А-А
(фиг. 2); на фиг. 2 изображен поперечный разрез Б-Б (фиг. 1); на фиг. 3 изображен разрез В-В (фиг.
1); на фиг. 4 изображен выносной элемент 1 (фиг. 2) в увеличенном масштабе.
Опора сейсмостойкая состоит из корпуса 1 в котором выполнено вертикальное отверстие
диаметром «D», которое охватывает цилиндрическую поверхность штока 2 например по
подвижной посадке H7/f7. В стенке корпуса перпендикулярно его оси, выполнено два отверстия в
которых установлен запирающий элемент - калиброванный болт 3. Кроме того, вдоль оси
отверстия корпуса, выполнены два паза шириной «Z» и длиной «I». В теле штока вдоль оси
выполнен продольный глухой паз длиной «h» (допустмый ход штока) соответствующий по ширине
диаметру калиброванного болта, проходящего через этот паз.

При этом длина пазов «I» всегда больше расстояния от торца корпуса до нижней точки паза «Н». В
нижней части корпуса 1 выполнен фланец с отверстиями для крепления на фундаменте, а в
верхней части штока 2 выполнен фланец для сопряжения с защищаемым объектом. Сборка опоры
заключается в том, что шток 2 сопрягается с отверстием «D» корпуса по подвижной посадке.

Паз штока совмещают с поперечными отверстиями корпуса и соединяют калиброванным болтом 3,


с шайбами 4, с предварительным усилием (вручную) навинчивают гайку 5, скрепляя шток и корпус
в положении при котором нижняя поверхность паза штока контактирует с поверхностью болта
(высота опоры максимальна). После этого гайку 5 затягивают тарировочным ключом до заданного
усилия. Увеличение усилия затяжки гайки (болта) приводит к деформации корпуса и уменьшению
зазоров от «Z» до «Z1» в корпусе, что в свою очередь приводит к увеличению допустимого усилия
сдвига (усилия трения) в сопряжении отверстие корпуса - цилиндр штока.

Величина усилия трения в сопряжении корпус-шток зависит от величины усилия затяжки гайки
(болта) и для каждой конкретной конструкции (компоновки, габаритов, материалов, шероховатости
поверхностей, направления нагрузок и др.) определяется экспериментально. При воздействии
сейсмических нагрузок превышающих силы трения в сопряжении корпус-шток, происходит сдвиг
штока, в пределах длины паза выполненного в теле штока, без разрушения конструкции.
Формула полезной модели

Опора сейсмостойкая, содержащая корпус и сопряженный с ним подвижный узел, закрепленный


запорным элементом, отличающаяся тем, что в корпусе выполнено центральное вертикальное
отверстие, сопряженное с цилиндрической поверхностью штока, при этом шток зафиксирован
запорным элементом, выполненным в виде калиброванного болта, проходящего через поперечные
отверстия корпуса и через вертикальный паз, выполненный в теле штока и закрепленный гайкой с
заданным усилием, кроме того в корпусе, параллельно центральной оси, выполнено два открытых
паза, длина которых, от торца корпуса, больше расстояния до нижней точки паза штока.

13
Прилагаем текст положительного отзывы ГОССТРОЯ РФ МИНИСТЕРСТВА СТРОИТЕЛЬСТВА
РОССИЙСКОЙ ФЕДЕРАЦИИ МИНСТРОЙ РОССИИ 117987 ГСП 1 Москва ул. Строителей, 8,
корп. 2 24- номер 9У номер 3-3-1-33 "О рассмотрении проектной документации" Директору
крестьянского (фермерского) хозяйства "Крестьянская усадьба" 197371, Санкт-Петербург,
Директору ГП ЦПП В.Н.КАЛИНИНУ
Главное управление проектирования и инженерных изысканий рассмотрело проектную
документацию шифр 1010-2с.94 "Фундаменты сейсмостойкие с использованием
сейсмоизолирующего скользящего пояса для строительства малоэтажных зданий в районах
сейсмичностью 7, 8 и 9 баллов. "Выпуск 0-1". Фундаменты для существующих зданий.
Материалы для проектирования", выполненные КФХ "Крестьянская усадьба" по договору с
Минстроем России от 26 апреля 1994 г. N 4.2-09-133/94 (этап 2 "Разработка конструкторской
документации сейсмостойкого фундамента с использованием сейсмоизолирующего скользящего
пояса для существующих зданий").
Разработанная документация была направлена на экспертизу в Центр проектной продукции
массового применения (ГП ЦПП; экспертное заключение N 260/94), Камчатский Научно-
Технический Центр по сейсмостойкому строительству и инженерной защите от стихийных
бедствий (КамЦентр; экспертное заключение N 10-57/94), работа рассмотрена на заседании
секции "Сейсмостойкость сооружений" НТС ЦНИИСКа им.Кучеренко, а также заслушана на НТС
Минстроя России.
Результаты экспертиз и рассмотрений показали, что без проведения разработчиком
документации, экспериментальной проверки предлагаемых решений и последующего
рассмотрения результатов этой проверки в установленном порядке использование работы в
массовом строительстве пока нецелесообразно. ( Госстроем РФ рекомендовано проверить на
индивидуальных объектах, а изучив опыт, в дальнейшем широко использовать в РФ)
В связи с изложенным Главпроект считает работу по договору N 4.2-09-133/94 законченной и, с
целью осуществления авторами контроля за распространением документации, во изменение
письма от 21 сентября 1994 г. N 9-3-1/130, поручает ГП ЦПП вернуть КФХ "Крестьянская
усадьба" кальки чертежей шифр 1010-2С.94, выпуск 0-2. Главпроект обращает внимание
руководства КФХ "Крестьянская усадьба" и разработчиков документации на ответственность за
результаты применения в практике проектирования и строительства сейсмоизолирующего
скользящего пояса по чертежам шифр 1010-2С.94, выпуски 0-1 и 0-2, Приложение: экспертное
заключение КамЦентра на 6 л. Зам.начальника Главпроекта А.Сергеев. исполнитель Барсуков
(495) 930 54 87
МИНИСТЕРСТВО СТРОИТЕЛЬСТВА РОССИЙСКОЙ ФЕДЕРАЦИИ МИНСТРОЙ РОССИИ
117987, ГСП-1, Москва, ул. Строителей, 8, корп. 2 номер письма 9-3-1/199 "О рассмотрении
проектной документации"

Директору крестьянского (фермерского) хозяйства "Крестьянская усадьба" 197371, Санкт-


Петербург,
Директору ГП ЦПП В.Н.КАЛИНИНУ
Главное управление проектирования и инженерных изысканий рассмотрело проектную
документацию шифр 1010-2с. 94 "Фундаменты сейсмостойкие с использованием
сеисмоизолирующего скользящего пояса для строительства малоэтажных зданий в районах
сейсмичностью 7, 8 и 9 баллов. Выпуск 0-1. Фундаменты для существующих зданий.
Материалы для проектирования", выполненную КФЯ "Крестьянская усадьба" по договору с
Минстроем России от 26 апреля 1994 г. N 4.2-09-133/94 (этап 2 "Разработка конструкторской
документации сейсмостойкого фундамента с использованием сеисмоизолирующего
скользящего пояса для существующих зданий").

Разработанная документация была направлена на экспертизу в Центр проектной продукции


массового применения (ГП ЦПП; экспертное заключение N 260/94), Камчатский Научно-
Технический Центр по сейсмостойкому строительству и инженерной защите от стихийных
бедствий (КамЦентр; экспертное заключение N 10-57/94), работа рассмотрена на заседании

14
секции "Сейсмостойкость сооружений" НТС ЦНИИСКа им.Кучеренко, а также заслушана на НТС
Минстроя России.

Результаты экспертиз и рассмотрений показали, что без проведения разработчиком


документации экспериментальной проверки предлагаемых решений и последующего
рассмотрения результатов этой проверки в установленном порядке использование работы в
массовом строительстве нецелесообразно. В связи с изложенным Главпроект считает работу
по договору N 4.2-09-133/94 законченной и, с целью осуществления авторами контроля за
распространением документации, во изменение письма от 21 сентября 1994 г. N 9-3-1/130,
поручает ГП ЦПП вернуть КФХ "Крестьянская усадьба" кальки чертежей шифр 1010-2с.94,
выпуск 0-2.

Главпроект обращает внимание руководства КФХ "Крестьянская усадьба" и разработчиков


документации на ответственность за результаты применения в практике проектирования и
строительства сеисмоизолирующего скользящего пояса по чертежам шифр 1010-2С.94, выпуски
0-1 и 0-2. Приложение: экспертное заключение КамЦентра на 6 л. Зам.начальника Главпроекта
А.Сергеев.

Исполнитель Барсуков телефон (495) 930 54 87

Прилагаем положительную выписку отзыва из НТС Госстроя РОССИИ МИНИСТЕРСТВО


СТРОИТЕЛЬСТВА РОССИЙСКОЙ ФЕДЕРАЦИИ НАУЧНО ТЕХНИЧЕСКИЙ СОВЕТ ВЫПИСКА
ИЗ ПРОТОКОЛА заседания Секции научно-исследовательских и проектно изыскательских
работ, стандартизации и технического нормирования Научно-технического совета Минстроя
России г. Москва номер 23-13/3 15 ноября 1994 т. Присутствовали: от Минстроя России
: Вострокнутов Ю Г. , Абарыков В. П. , Гофман Г. Н. , Сергеев Д. А. , Гринберг И. Е. , Денисов Б.
И. , Ширяез Б. А. , Бобров Ф. В. , Казарян Ю. А. Задарено к А. Б. , Барсуков В. П. , Родина И.
В. , Головакцев Е. М. , Сорокин А. И , Сенина В. С. от ЦНИСК им. Кучеренко : - Айзенберг Я.
М Алексеенков Д. А. , Кулыгин Ю. С. , Смирнов В. И. , Чигрин С. И. , Ойзерман В. И. ,
Дорофеев В. М. , Сухов Ю. П. , Дашевский М. А. от ЦНИИпромзданий -Гиндоян А. П. ,
Иванова В. И. , Болтухов А. А. , Нейман А. И. , Малин И. С. , Севастьянов В.В, от ПНИИС-
Севастьянов В.В, от КФХ "Крестьянская усадьба" - Коваленко А.И, от НИИОСП им. Герсенова -
Ставницер М.Р АО ЦНИИС - Шестоперов Г.С. от КБ по железобетону им. Якушева- Афанасьев
П.Г . от Объединенного института физики земли РАН - Уломов В.И., Штейнберг В В от
ПромтрансНИИпроекта - Федотов В Г. от Научно-инженерного и координационного
сейсмологического центра РАН - Фролова Н.И . от ЦНИИпроектстальконструкция - Болодин Ю.И,
ИМЦ "Стройизыскания" - Ваулин Ю.И, Ассоциация "Югстройпроект"- Малик А.Н. от УКС
Минобороны России (г. Санкт-Петербург) - Беляев В.С 2. " О сейсмоизоляции существующих
жилых домов, как способ повышения сейсмостойкости малоэтажных жилых зданий" .

Рабочие чертежи серии • 1.010.-2с-94с. "Фундаменты сейсмостойкие с использованием


сейсмоизолирущего скользящего пояса для строительства малоэтажных зданий в районах
сейсмичностью 7, 8, 9 баллов" 1. Заслушав сообщение А. И. Коваленко, отметить, что по
договору N 4.2-09-133/94 с Минстроем России КФК "Крестьянская усадьба" выполняет работу
"Фундаменты сейсмостойкие с использованием сейсмоизолируюшего пояса для строительства
малоэтажных зданий в районах сейсмичностью 7, 8 и 9 баллов".

В основу работы положен принцип создания в цокольной части здания сейсмоизолируюшего


пояса, поглощающего энергию как горизонтальных, так и-вертикальных нагрузок от
сейсмических воздействий при помощи резино -щебеночных амортизаторов и ограничителей
перемещений. К настоящему времени завершен первый этап работы - подготовлены материалы
для проектирования фундаментов для вновь строящихся зданий. Второй этап работы,
направленный на повышение сейсмостойкости существующих зданий, не завершен.

15
Материалы работы по второму этапу предложены к промежуточному рассмотрению на
заседании Секции. Представленные материалы рассмотрены НТС ЦНИИСК им. Кучеренко (
Головной научно-исследовательской организацией министерства по проблеме
сейсмостойкости зданий и сооружений ).

Решили:

1. Принять к сведению сообщение А.И.Коваленко по указанному вопросу. 2. Рекомендовать


Главпроекту при принятии законченной разработки "проектно-сметной документации
сейсмостойкого Фундамента с использованием скользящего пояса (Типовые проектные
решения) учесть сообщение А. И. Коваленко и заключение НТС ЦНИИСК, на
котором были рассмотрены предложения сейсмоустойчивости инженерных систем
жизнеобеспечения ( водоснабжения, теплоснабжения, канализации и газораспределения). Зам.
председателя Секции научно-исследовательских и проектно-изыскательских работ,
стандартизации и технического нормирования Ю. Г. Вострокнутов В. С. Сенина
Ученый секретарь Секции научно-исследовательских и проектно-изыскательских работ,
стандартизации и технического нормирования. Прилагаем еще один положительный отзыв (
полный текст ): МИНИСТЕРСТВО СТРОИТЕЛЬСТВА РОССИЙСКОЙ ФЕДЕРАЦИИ МИНСТРОЙ
РОССИИ 117937 ГСП 1 Москва ул. Строителей 8 корп. 2 № 3-3-1 "О рассмотрении
проектной документации"
Директору крестьянского (фермерского) хозяйства "Крестьянская усадьба" А.И
КОВАЛЕНКО 197371, Санкт-Петербург,

Директору ГП ЦПП В.Н.КАЛИНИНУ. Главное управление проектирования и инженерных


изысканий рассмотрело проектную документацию (шифр 1010-2с.94 )"Фундаменты
сейсмостойкие с использованием сейсмоизолирующего скользящего пояса для строительства
малоэтажных зданий а районах сейсмичностью 7, 8 и 9 баллов. Выпуск 0-

2. Фундаменты для существующих зданий. Материалы для проектирования", выполненные КФХ


"Крестьянская усадьба" по договору с Минстроем России от 26 апреля 1994 г. N 4.2-09-133/94
(этап 2 "Разработка конструкторской документации сейсмостойкого фундамента с.
использованием сейсмоизолирующего скользящего пояса для существующих зданий").

Разработанная документация была направлена на экспертизу в Центр проектной продукции


массового применения (ГП ЦПП; экспертное заключение N 260/94), Камчатский Научно-
технический Центр по сейсмостойкому строительству и инженерной защите от стихийных
бедствий (КамЦентр; экспертное заключение N 10-57/94), работа рассмотрена на заседании
секции "Сейсмостойкость сооружений" НТС ЦНИИСКа им.Кучеренко, а также заслушана на НТС
Минстроя России.

Результаты экспертиз и рассмотрений показали, что без проведения разработчиком


документации экспериментальной проверки предлагаемых решений и последующего
рассмотрения результатов этой проверки в установленном порядке использование работы в
массовом строительстве нецелесообразно. В связи с изложенным Главпроект считает работу
по договору N 4.2-09-133/94 законченной и, с целью осуществления авторами контроля за
распространением документации, во изменение письма от 21 сентября 1994 г. N 9-3-1/130,
поручает ГП ЦПП вернуть КФХ "Крестьянская усадьба" кальки чертежей шифр 1010-2с.94,
выпуск 0-2.

Главпроект обращает внимание' руководства КФХ "Крестьянская усадьба" и разработчиков


документации на ответственность за результаты применения в практике проектирования и
строительства сейсмоизолирующего скользящего пояса по чертежам ( шифр 1010-2с.94,
выпуски 0-1 и 0-2.)

16
Приобрести Специальные технические условия на особое воздействие (СТУ ) для обеспечения
устойчивости существующих сооружений , зданий , трубопроводов , от ударной волны, за счет
использования сдвиговых упругопластических трубчатых, квадратных, кольцевых шарниров и
балочных энергопоглотителей на протяжных фрикционно-подвижных соединениях проф дтн ПГУП
А.М.здина , в том числе проведение лабораторных испытаний нелинейным методом расчета в ПК
SCAD,

их устойчивости существующих старых зданий, сооружений, мостов, гостиниц, отелей,


магистральных трубопроводов, на особые воздействия с использованием энергопоглотителей и
пластических шарниров и легко сбрасываемыхконструкций, за счет рассеивания энергии ШИФР
1.010.1-2с.94, выпуск 0-1, утвержден Главпроектом Мистрой России, письмо от 21.09.94 ; 9-3-1/130
за подписью Д.А.Сергеева, исп. Барсуков 930-54-87 согласно письма Минстроя № 9-3-1/199 от
26.12.94 и письма № 9-2-1/130 от 21.09.94) на взрывное воздействие ( 600 кг ) не приводящие
последствиям лавинообразному разрушению всех конструкций с, помощью компьютерного
моделирования в ПК SCAD , ANSYS, LS-DYNA ,

для существующих построенных старых зданий с использованием , упругопластических


балочных, струнных, трубчатых, квадратных упругопластичных шарниров и легко сбрасываемых
конструкций ( патент на полезную модель № 154506 «Панель противовзрывная»), за счет
использования упругопластичных энергопоглотителей в виде «гармошка» и прорезей в шахматном
порядке согласно изобретения полезная модель № 165076 «Опора сейсмостойкая» с
использованием фракционности, демпфирования для поглощение взрывной энергии по аналогу
Pall Dynanics cvjnhb yfexye. Ge,kbrfwb. Yf fyukbqcrjvz zpsrt
https://link.springer.com/content/pdf/10.1007%2F978-3-642-36197-5_312-1.pdf

согласно изобретения № 2010136746 «Способ защиты зданий и сооружение при взрыве с


использованием сдвигоустойчивых и легко сбрасываемых соединений , использующие систему
демпфирования и сейсмоизоляцию для поглощения взрывной и сейсмической энергии» на основе
изобретений проф. дтн ПГУПС Уздина А М №№ 1174616, 1143895, 1168755 , согласно расчетам
проф МГСУ О.В Мкртычева «Проблемы расчета зданий на особые воздействия» локальные
разрушения при взрыве заряда массой 600 кг при использовании эрегопоглотителей с
пластическим шарниром, закрепленных колоны с ригелем на фрикци –болтах с пропиленным
стальной шпильке пазе, куда забивается медный обожженный упругопластичный клин , или на
протяжных фрикционно –подвижных соединениях, не приводит к посредствующему
лавинообразному обрушении зданий и всей конструкции за счет поглощения пиковых ускорений и
поглощение взрывной энергии , за счет легко сбрасываемости наружных панелей и
упругоплатических узлов крепления колонны с ригелем в связи с податливостью и подвижности
фрикционно- подвижных соединениях.
http://www.palldynamics.com/pdf/Paper009511.pdf

Стоимость альбома (проекта ) со специальных технических решений, с использованием упругих


энергопоглотителей , пластических шарниров и легко сбрасываемости конструкций панелей зданий
, можно обратится к Мажиеву Хасан Нажоевичу по тел (999) 535-47-29 или по электронной почте
c9995354729@yandex.ru t3487810@interzet.ru

Стоимость альбома специальных технических условий (СТУ) на особые воздействия для


обеспечения устойчивости сооружений , от ударной волны, за счет использования сдвиговых
упругопластических шарниров и балочных энергопоглотителей, в том числе нелинейным
методом расчета в ПК SCAD с типовыми протяжными фрикционно –подвижными соединениями
(ФПС) и упругпастичными подвижными уздами креплениями лестничных маршей и легко
сбрасывемости.
http://www.palldynamics.com/pdf/Paper00901.pdf

17
Аванс 10 тр, после лабораторных испытаний методом численного (математического)
моделирования и испытания моделей и узлов крепления (расчета ) упругоплатических балочных,
квадратных, трубчатых, кольцевых, струнных (тросовых в оплетке) протяжных шарниров в ПК
SCAD, еще 10 тр за окончание лабораторных испытаний фрагментов и узлов крепления или
усиления существущих лестничных маршей

Карта Сбербанка 2202 2006 4085 5233 http://www.palldynamics.com/pdf/Paper001949.pdf

Электронный адрес О.В. Мкртычева МГСУ mkrtychev@yandex.ru t3487810@interzet.ru (999)


535-47-29, ( 993) 151-39-15, (996) 798-26-54 Мажиев Хасан Нажоевич Президент организации
«Сейсмофонд» ИНН 201400078, ОГРН 1022000000824
http://www.palldynamics.com/pdf/Paper001946.pdf
http://www.palldynamics.com/FrictionDampers.htm
https://dwg.ru/dnl/1380
https://dwg.ru/dnl/6479
http://www.palldynamics.com/pdf/Paper00901.pdf
В дополнение к ранее опубликованным

http://dwg.ru/dnl/5654
Серия 1.420.3-37.06 Каркасы стальные "УНИМАК-Р1"

Одноэтажных производственных зданий с применением одно- и многопролѐтных


рам переменного сечения пролѐтами 12, 15, 18, 24, 30 и 36 м для бескрановых
зданий, зданий с подвесными мостовыми кранами грузоподъѐмностью 1; 2; 3.2 и 5
т и зданий с опорными мостовыми кранами грузоподъѐмностью 5; 10; 16 и 20 т.

http://dwg.ru/dnl/5655
Серия 1.420.3-39.08 Каркасы стальные "УНИТЭКС-Р1"

Каркасы одноэтажных производственных зданий с применением однопролѐтных


рам из гнутых (в том числе оцинкованных) профилей пролѐтами 9, 12, 15, 18 и 21 м
для бескрановых зданий, зданий с подвесными мостовыми кранами
грузоподъѐмностью 1; 2 т и зданий с опорными мостовыми кранами
грузоподъѐмностью 5; 10; 16 т.
Серия 2.440-2Выпуск 1Шарнирные узлы балочных клеток и ...
https://dwg.ru/dnl/1752
27 ноя 2006 ... Выпуск 1 Узлы стальных конструкций производственных зданий промышленных
предприятий. Шарнирные узлы балочных клеток и ...
С пометкойDownload
Жесткие и шарнирные узлы.Запутался

18
https://forum.dwg.ru/showthread.php?t=29556
Конструкции зданий и сооружений Жесткие и шарнирные узлы. ... А в зоне сопряжения ригеля со
стойкой рамы зазор между ригелем и ...
С пометкойФорум
2.440-1 выпуск 1- Рамные и шарнирные узлы
https://dwg.ru/dnl/3487
15 фев 2008 ... Скачать 2.440-1 выпуск 1- Рамные и шарнирные узлы. Серии.
С пометкойDownload
Стык колонны с балкой покрытия - Страница 2
https://forum.dwg.ru/showthread.php?t=91850&page=2
Оба узла шарнирные. Или перерасчет рамы, или приведение узла в соответсвие с расчетной
схемой. Ответить. лариса К 15:04 30.11.
С пометкойФорум
Чем вызваны деформации связей покрытия (металл)?
https://forum.dwg.ru/showthread.php?t=7103
7 июн 2006 ... Крайние пролеты поперечных рам 6м, средний пролет 36м. ... Колонны рамы жестко
заделаны в фундамент и имеютшарнирные узлы ...
С пометкойФорум
Жесткий или шарнирный узел сопряжения пояса и стойки из ГСП.
https://forum.dwg.ru/showthread.php?t=150855&page=3
А если узлы рамы принять шарнирными, то система станет геометрически изменяемой.
Попробуйте посчитать ферму без шарниров и ...
С пометкойФорум
Расчетная длина колонн в плоскости поперечной рамы ...
https://forum.dwg.ru/showthread.php?t=96172&page=17
Рама считается свободной если нет связей препятствующих ... жесткие узлы фундамент-колонна
и шарнирные колонна - балка.
С пометкойФорум
Скачать - страница 293
https://dwg.ru/dnl/p293?ch=15&page=4&filtr=&sort=3
Ермолаев Н.Н., Каскевич. Компоновка поперечной рамы произв. здания. МУ. 1985 · 2.440-1 выпуск
1- Рамные и шарнирные узлы · Ремонт ведомого ...
С пометкойDownload
Помогите с разработкой механизма с гидравлическим приводом ...
https://forum.dwg.ru/showthread.php?t=136011&page=2
Верхняя часть рамы крепиться к нижней через шарнирные узлы. ... работу гидроцилиндров с
механизмами и шарнирными узлами рам.
С пометкойФорум
Жесткий узел сопряжения колонны с фундаментом в СКАДе.
https://forum.dwg.ru/showthread.php?t=62920
31 янв 2011 ... Жестким узел считается только в плоскости рамы. ... Зачастую колонны в
плоск рамы "заделаны", а из плоскости
4.402-9 в.5 Анкерные болты. Рабочие чертежи

19
https://dwg.ru/dnl/4221
24 июл 2008 ... Скачать 4.402-9 в.5 Анкерные болты. Рабочие чертежи. Серии.
С пометкойDownload
Ищу серию 4.402-9 выпуск 5 "Анкерные болты". Рабочие чертежи
https://forum.dwg.ru/showthread.php?t=23023
24 июл 2008 ... Поиск литературы, чертежей, моделей и прочих материалов Ищу серию 4.402-
9 выпуск 5 "Анкерные болты". Рабочие чертежи
https://www.quaketek.com/seismic-design/
https://www.iitk.ac.in/nicee/wcee/article/13_1955.pdf
Chertezh-gibkikh-podatlivikh-svyazey-frilktsi-unikon-ankernie-bolti-friktsionno-podvizhnie-
soedineniya-pall-friction-damper-spbgasu

20
21
22
23
13th World Conference on Earthquake Engineering
Vancouver, B.C., Canada August 1-6, 2004 Paper
No. 1955

PERFORMANCE-BASED DESIGN USING PALL FRICTION DAMPERS - AN


ECONOMICAL DESIGN SOLUTION.

Avtar PALLI and R. Tina PALL1

SUMMARY

Benefit-cost analysis approach suggests performance-based design for most modern buildings. The
conventional structural systems are highly unlikely to provide adequate performance in the event of a
major earthquake. With the emergence of Pall Friction Dampers, it has now become economically
feasible to design high performance structures. Their low cost and maintenance free characteristics
suggest wide application for new construction as well as for retrofit of existing buildings. Public sectors,
private sectors, developers and developing countries are all benefiting from this technology. They have
been used for the seismic protection of more than 80 major building projects, including the Boeing
Commercial Airplane Factory at Everett, WA - the world's largest building in volume.

INTRODUCTION

During a major earthquake, a large amount of energy is pumped into the building. The manner in which
this energy is consumed in a structure determines the level of damage. The building codes recognize that

I
Pall Dynamics Limited, Montreal, Canada. www.palldynamics.com, E-mail: info@palldynamics.com

24
it is economically not feasible to reconcile this energy within the elastic capacity of materials. The
criterion stipulated in conventional building codes is to design structures to resist moderate earthquakes
without significant damage and avoid collapse during major earthquakes. The primary emphasis is on life
safety. The reliance for survival is placed on ductility to dissipate energy during inelastic deformations
causing bending, twisting and cracking. This assumes permanent damage, repair costs of which could be
significant. Recent earthquakes have clearly demonstrated that conventional construction, even in
technologically advanced countries, is not immune to destruction.

In modern buildings, avoidance of structural collapse alone is not enough. The cost of non-structural
components and contents is much higher than the cost of the structure itself and must be protected. The
buildings of post disaster importance such as hospitals, telecommunications, police stations etc. must
remain operational. Benefit-cost analysis suggests performance-based design approach. Emphasis for
performance based seismic design is being given in recent standards, e.g., Vision 2000, ATC-57 and
FEMA-356/2000. While global deflections, story drifts, force and demand-capacity ratios are important to
a structural engineer; these have no meaning to a developer. What are important to an owner is how much

25
it will cost to build and how much it will cost to repair the damage, lost rental revenues from a future
earthquake. Early on, the structural engineers should resolve these issues with the owner and develop
performance-based design criteria consistent with the performance goals. In most cases, providing good
seismic performance yields good return on the investment.

If the amount of energy getting into the structure can be controlled and a major portion of the energy can
be dissipated mechanically independent of primary structure, the seismic response of the structure and
damage control potential can be considerably improved. These objectives can be delivered by adopting
new techniques of base isolation and energy dissipation devices. With the introduction of energy
dissipation devices, supplemental damping of 20-30% of critical can be easily achieved (inherent
damping is merely 1-5%). Thus forces exerted on the structure and the amplitudes of vibrations are
considerably reduced. The drifts of the structure can be reduced by a factor of about two to three and by
larger factors if the devices also add stiffness to the structure, refer Article C9.1 of FEMA-356 [27]. The
energy dissipation systems should be considered in somewhat broader context than isolation systems
(which are not feasible in taller buildings) as a design strategy when performance goals include damage
control. Also, the construction cost is less with energy dissipation systems than base isolation systems.

Seismic isolation and energy dissipation systems are relatively new and sophisticated concepts that
require more extensive design and detailed nonlinear time-history dynamic analysis than most
conventional schemes. With the availability of several commercial programs and powerful desktop
computers, the sophisticated analysis is no longer a daunting task and can be easily and quickly done in a
small design office environment. However, the benefits accrued outweigh the extra design expense and
resulting savings more than compensate the cost of dampers.

With the emergence of Pall Friction Dampers[1-3,6], the performance-based design of buildings is now
economically feasible. The first building with seismic dampers in North America was built with Pall
Friction Dampers. Their low cost and maintenance free characteristics suggest wide application for new
construction as well as for retrofit of existing buildings. Public sectors, private sectors and developers are
using and benefiting this technology. They have been used for the seismic protection of more than 80
major building projects, including the Boeing Commercial Airplane Factory at Everett, WA - the world's
largest building in volume. This technology is credited with for saving Boeing more than US$ 30 million.
The City and County of San Francisco chose Pall Friction Dampers for the seismic control of Moscone
West Convention Center and saved US$2.25 million, compared to alternate viscous dampers.

This paper describes the innovative structural system and several applications in the performance-based
design of major projects including cost analysis.

PALL FRICTION DAMPERS


State-of-the-Art
Of all the methods so far available to extract kinetic energy from a moving body, the most widely adopted
is undoubtedly the friction brake. Mechanical engineers have successfully used this concept for centuries
to stop the motion of equipment, automobiles, railway trains, airplanes etc. No other mean has been able
to replace the friction brake. Reason! It is the most effective, reliable and economical mean to dissipate
kinetic energy. Similar to automobiles, the motion of vibrating building can be slowed down by
dissipating seismic energy in friction.

Inspired by the principle of friction brake in mid 1970's, Pall Friction Dampers were pioneered for the
seismic control of buildings. Pall Friction Dampers significantly reduce the initial cost of construction
while dramatically increasing the earthquake resistance against damage.

26
Friction dampers for a building must possess a reliable and stable performance over the life of building.
Developing a reliable friction is very difficult and tricky. Over a period of more than a decade of research
and development, the common problems in friction were successfully overcome by using specially treated
surfaces and a unique manufacturing process. Over the years, Pall Dynamics has earned an international
reputation for excellence and is a world leader in friction dampers for seismic control of buildings.

Pall Friction Dampers have successfully undergone rigorous proof testing in the U.S and Canada. In 1985,
the National Research Council of Canada tested 3-story frame structures on a shaking table at the
University of British Columbia, Vancouver [4]. In 1986-1987, the U.S. National Science Foundation
tested a 9-story frame structure on a shaking table at the University of California at Berkeley [5]. The
structures were subjected to more than 20 different major earthquake records. Even for an earthquake 5
times stronger than the 1985-Mexico earthquake, the frames equipped with friction dampers remained
damage free. Pall Friction Dampers are well recognized and accepted by the building codes in Canada,
the U.S and many other countries.

Salient Features
Pall Friction Dampers are foolproof in construction. Basically, these consist of series of steel plates,
which are specially treated to develop very reliable friction. These plates are clamped together and
allowed to slip at a predetermined load. Decades of research and testing have led to perfecting the art of
friction. Their performance is reliable, repeatable and they possess large rectangular hysteresis loops with
negligible fade. Their performance is independent of velocity and hence exerts constant force for all
future earthquakes, design-based earthquake (DBE) or maximum credible earthquake (MCE). A much
greater quantity of energy can be dissipated in friction than any other method involving the yielding of
steel plates, viscous or viscoelastic dampers. Therefore, fewer Pall Friction Dampers are required to
provide the required amount of energy dissipation. Pall Friction Dampers are passive energy dissipation
devices and, therefore, need no energy source other than earthquake to operate it. They do not require any
repair or replacement after the earthquake and are always ready to do their job.

Friction Damper for Hystersis Loop Figure 1. Friction Damper in


Tension-Compression Tension-only Cross
Brace Pall Friction Dampers Brace

In a typical undamped structure, the inherent damping is merely 1-5% of critical. With the introduction of
Pall Friction Dampers, structural damping of 20-50% of critical can be easily achieved. As the dampers
dissipate a major portion of the seismic energy, forces and deformations on the structure are significantly
reduced. Pall Friction Dampers significantly reduce the initial cost of construction while dramatically
increasing the earthquake resistance against damage.

27
Pall Friction Dampers are customized to suit site conditions and allow greater adaptability than is possible
with other systems. These dampers can be bolted or welded into place.

Pall Friction Dampers are available for long slender tension-only cross bracing, single diagonal tension-
compression bracing and chevron bracing (Figure 1). The damper for cross bracing is a unique
mechanism. When one of the brace in tension forces the damper to slip, the damper mechanism forces the
other brace to shorten and thus avoid buckling. In this manner, the other brace is immediately ready to
slip the damper on reversal of cycle. These dampers have been used in 65 feet (22 m) long slender
bracing.

To avoid pounding at the expansion joints, Pall Friction Connectors are custom made to accommodate bi-
directional movements.

Slip Load of Friction Damper


The friction dampers are designed not to slip during wind. During a major earthquake, they slip prior to
yielding of structural members. In general, the lower bound is about 130% of wind shear and the upper
bound is 75% of the shear at which the members will yield. As seen in Figure 2, if the slip load is very
low or very high, the response is very high. Several parametric studies have shown that the slip load of
the friction damper is the principal variable with the appropriate selection of which it is possible to tune
the response of structure to an optimum value. Optimum slip load gives minimum response. Selection of
slip load should also ensure that after an earthquake, the building returns to its near original alignment
under the spring action of an elastic structure. Studies have also shown that variations up to ±20% of the
optimum slip load do not affect the response significantly. Therefore, small variations in slip load (810%)
over life of the building do not warrant any adjustments or replacement of friction damper.

C
D

t
n
с
о
Q.
№ Optimum
< Figure 2. Response versus Slip Load
D
{
Г

Sli
p
Lo
a
d

WHY PALL FRICTION DAMPERS ARE FAVORED?

• They are simple and foolproof in construction.


• Offer reliable and repeatable performance at low cost.

28
• Possess large rectangular hysteresis loops. Greater energy dissipation for a given force. Hence, fewer
Pall Friction Dampers are needed. Conversely, exert lesser force for a given damping.
• Provide supplemental damping and stiffness for added stability.
• Performance is independent of velocity and temperature.
• Constant force for all future earthquakes (DBE/MCE). Therefore, design of connections and
members is economical.

29
• They are not active during service loads and wind. Hence, no possibility of failure due to fatigue
before an earthquake.
• Need no repair or replacement before and after earthquake. There is nothing to damage or leak.
• Energy dissipation is through friction and not through the damaging process of yielding.
• After an earthquake, the building returns to its near original alignment due to spring action of an
elastic structure.
• Compact and narrow enough to be hidden in partitions.
• They can accommodate foundation settlements.
• Available for all types of bracing, including tension cross bracing, and expansion joints.
• Custom made. Easily adaptable to any site condition. Can be welded or bolted

COMPARISON OF HYSTERESIS LOOPS OF DIFFERENT DAMPERS

+p

Yielding Plate
Damper
Self-centering
Friction Viscoelastic
Damper Damper
Figure 3. Comparison of Hysteresis Loops of Different Dampers

COMPARISON WITH OTHER DEVICES

Viscous Damper
Viscous dampers are velocity dependent. The forces exerted by the damper are, therefore, different for
different earthquake records. Friction dampers are independent of velocity, therefore, exert constant force
for all future earthquakes (DBE / MCE). A friction-damped structure is an engineered structure in which
forces exerted are predetermined. This is a great technical and economic advantage over viscous dampers
in which forces are much higher at MCE level.

The hysteresis loop of viscous damper is elliptical compared to rectangular for friction damper. For a given
maximum force, the area of hysteresis loop (energy dissipation or damping) of viscous damper is about
70% of that for friction damper i.e. 70 friction dampers will achieve the same damping as 100 viscous

30
dampers of a given force. Conversely, for a given number and damping value, the forces exerted

31
by friction dampers are only 70% of those for viscous dampers. This leads to significant savings in cost of
dampers, bracing, connections, columns and foundations.

While supplemental damping is beneficial in reducing the earthquake forces and amplitudes of vibration,
added stiffness is beneficial for stability. A deflected building is similar to a stooped person. Similar to a
cane support for stooped person, additional stiffness helps the deflected building against overturning. Pall
Friction Dampers provide both added damping and added stiffness for stability - a complete structural
solution. Viscous dampers provide only damping and no stiffness - the structure is on its own to struggle
for stability.

Unbonded Brace
Unbonded brace is often called by other names like 'Yielding brace' or 'Buckling-Restrained brace'. It
consists of a slender steel brace (core steel), surrounded by a steel tube or pipe, which is then filled with
concrete or grout. The core steel is wrapped with plastic like material to separate it from the concrete
around. The concrete in steel tube prevents buckling of the steel brace.

Unbonded brace dissipates energy through the process of yielding. Yielding involves damage. No damage,
no energy dissipation or damping. After an earthquake, the brace may be damaged and needs to be
replaced. Replacement of brace after an earthquake is expensive and time consuming. It is too difficult to
replace before imminent aftershocks. Another problem is that the brace is enclosed in concrete and is not
visible for inspection to verify if it is broken or otherwise.

The cost of supply of core steel and wrapping material may appear to be small, but the overall cost of an
installed unbonded brace, including filling of outer tube with concrete and connections, is higher than the
installed friction damper. As friction dampers dissipate a large amount of energy mechanically, the forces
exerted are far less than those exerted by the unbonded braces. Besides, the structure with friction damper
is economical to design and always ready to resist earthquakes one after another without replacement.

DESIGN CRITERIA

The quasi-static design procedure given in most building codes are ductility based and do not explicitly
apply to buildings with supplemental damping. In the past few years, several guidelines on the analysis
and design procedure of passive energy dissipation devices have been developed in the U.S. The latest and
most comprehensive document is the "NEHRP Guidelines for the Seismic Rehabilitation of Buildings,
FEMA 356 / 357, issued in 2000" [27]. This equally applies to new buildings.

The Guidelines require that the structure be evaluated for response to two levels of ground shaking, a
design basis earthquake (DBE) and a maximum considered earthquake (MCE). The DBE is an event with
10% probability of exceedance in 50 years, while the MCE represents a severe ground motion of
probability of 2% in 50 years. Under the DBE, the structure is evaluated to ensure that the strength
demands on structural elements do not exceed their capacities and that the drift in the structure is within
the acceptable limits. For the MCE, the structure is evaluated to determine the maximum displacement and
overstress. It is presumed that if proper ductile detailing has been followed, the structure will have
sufficient reserve to avoid collapse during MCE.

Since different earthquake records, even of the same intensity, give widely varying structural responses,
results obtained using a single record may not be conclusive. Therefore, three time-history records,
suitable for the region should be used; one of which should be preferably site specific. The maximum
response is used for the design.

32
NEHRP guidelines require that friction dampers are designed for 130% MCE displacements and all
bracing and connections are designed for 130% of damper slip load. Variation in slip load from design
value should not be more that ± 15%.

NONLINEAR TIME-HISTORY DYNAMIC ANALYSIS


The slippage of friction damper in an elastic brace constitutes nonlinearity. Also, the amount of energy
dissipation or equivalent structural damping is proportional to the displacement. Therefore, the design of
friction-damped buildings requires the use of nonlinear time-history dynamic analysis. With these
analyses, the time-history response of the structure during and after an earthquake can be accurately
understood. Several nonlinear computer programs are now capable of modeling friction dampers. Some
of these programs are ETABS, SAP2000, DRAIN-TABS, DRAIN-2DX, DRAIN-3DX, ANSYS etc. With
the availability of powerful personal computers, the sophisticated nonlinear time history analysis can be
easily and quickly done in a small design office environment.

The modeling of friction dampers is very simple. Since the hysteresis loop of the damper is similar to the
rectangular loop of an ideal elasto-plastic material, the slip load of the friction damper can be considered
as a fictitious yield force.

PRACTICAL APPLICATIONS
The first building built with seismic dampers in North America was with Pall Friction Dampers (1987).
Pall Friction Dampers are finding increasing application worldwide for new construction as well as for
retrofit of existing buildings, including overhead water tanks. They have been used for the seismic
protection of more than 80 major buildings in Canada, the U.S., China and India. These are hospitals,
telecommunication buildings, educational institutions, police headquarters, defense installations,
convention centers, courthouses, office and residential buildings. Some projects are discussed below:
Seismic Upgrade of Boeing Commercial Airplane Factory, Everrett, WA, USA.

Figure 4. (a). Inside View of Boeing Factory. (b). Pall Friction Dampers

The mammoth Boeing plant, which could contain Disneyland under one roof, is the world's largest
building in volume [25]. It was built in phases from 1968-1991, for the assembly of wide-bodied 747
jetliners - world's largest commercial airplane. The steel frame building is 120 feet (37 m) high with clear
spans of 350 feet (107 m) and covers more than 98 acres.

33
In 1996, the Boeing engineers considered several seismic upgrade schemes for this structure. They chose
Pall Friction Dampers as they are foolproof in construction and offer reliable maintenance free
performance at low cost. Also, they possess stiffness for added stability. The performance of friction
dampers is independent of velocity, therefore, the forces on the connections remain constant for any future
earthquake. This is an engineered solution. Economy in the design of connections and easy installation of
dampers provided significant savings in construction cost and time.

Several types of Pall Friction Dampers, suitably modified to adapt to the site conditions, were incorporated
in different types of existing bracing. Friction dampers of capacity up to 200,000 lb. (900 KN) and stroke
up to 15 inch (380 mm) have been used. Pall Friction Dampers met or exceeded Boeing's stringent
specifications. Work on seismic upgrade was undertaken in 1998 and completed in 2002.

Cost of Retrofit: US$65 million. Savings: US$30 million compared to conventional construction.
Moscone West Convention Center, San Francisco, USA

Figure 5. (a). View of Moscone West (b). Pall Friction Damper

The Moscone West Convention Center is located in downtown San Francisco, between San Andrea's and
Hayward faults, which are about 19 kM in each direction. The four-story steel frame building is 112 feet
(34 m) high with clear spans of 45 to 90 feet (14-28 m) [26]. The US$186 million expansion of the 1.2
million sq. feet (111,700 m ) convention center was California's largest project of 2001.

Bids were invited on two alternate designs, one with viscous dampers and another with Pall Friction
Dampers. The scheme with friction dampers offered a saving of US$2.25 million compared to viscous
dampers. On the recommendations of peer reviewers and two specialist consultants from the University of
California at Berkeley, the City and County of San Francisco chose Pall Friction Dampers for seismic
control of this prestigious building. The construction of the convention center was completed in 2003. Pall
Friction Dampers up to 500,000 lb. (2250 kN) capacity and 9 inch (230 mm) stroke have been used.

The use of friction dampers reduced the story drifts from 2% without dampers to 1% with dampers. About
80% of the earthquake energy has been dissipated by the friction dampers leaving the building with little
damage. According to the project architects and engineers, the use of technology saved a couple of million
dollars in the initial construction costs and will save tens of millions of dollars in lowered repair costs in
the event of an earthquake.

34
Seismic Retrofit of Boeing Development Center, Cafeteria and Auditorium Buildings, Boeing Field,
Seattle, WA, USA

Figure 6. (a). View of Cafeteria, Auditorium and Fitness Building (b). Pall Friction Damper

There are three 2-story Boeing Development buildings and a 4-story Boeing Cafeteria, Auditorium and
Fitness Center building. These are steel frame buildings, built in 1980's. The foundations are on 70-80 feet
deep friction piles. Some of these buildings were damaged during February 28, 2001 Nisqually Earthquake
of magnitude 6.8. The epicenter was about 20 miles from the building site. Due to liquefaction of soil, the
differential settlements in pile foundations were 4-5 inches. This resulted in substantial non-structural and
structural damage to the Cafeteria and Auditorium building. The structural damage was primarily to rigid
steel bracing. Several bracing in the Cafeteria and Auditorium building buckled or broke.

Several structural schemes were considered for the seismic retrofit of these buildings. Friction dampers
were considered to be an ideal solution as they provide both damping and stiffness. Besides, they can be
easily modified to suit site conditions and designed to accommodate any future foundation settlements.

A total of 350 Pall friction Damper of 100-500 kip slip load and up to 10 inches stroke, were used in the
existing steel bracing. The seismic retrofit was completed in 2002. Savings are estimated to be more than
60% compared to conventional retrofit.
Ambulatory Care Center, Sharp Memorial Hospital, San Diego, California, USA

Figure 7. View of New Ambulatory Care Center Pall Friction Dampers in Chevron Brace

35
The Ambulatory Care Center complex consists of two buildings of four and five floors, connected in two
places by a pedestrian bridge and an elevator lobby. This facility will provide extensive medical services
to the community including surgery.

The project engineers selected steel-frame design for the cost efficiency, design flexibility and speedy
construction. Moment-resisting frames, in combination with Pall Friction Dampers in steel bracing, were
selected to resist lateral seismic forces. The structure was designed to meet 1997UBC and 1998 California
Code requirements for Seismic Zone 4. In this performance-based design, the Pall Friction Dampers
reduced building drift from 2% to 1%. The savings in construction cost due to reduced forces, more than
offset the cost of dampers. In case of a major seismic event, the dampers reduce building content damage
and increase safety of occupants. This resulted in long-term savings to the client through lower seismic
insurance premiums.

Seismic Retrofit of 3-Million Gallon Reservoir,


Sacramento, California, USA

The Freeport water tower, a distinctive landmark visible


from Interstate-5, was built in 1956 [22]. The steel
reservoir stands about 120 feet high. The supporting
structure consists of 27 steel columns with two levels of
60 feet long tension cross bracing. Of the several
seismic retrofit options, the scheme with Pall Friction
Dampers in tension cross bracing was chosen. When
tension in one of the brace forces the damper to slip, the
damper's mechanism shortens the other brace, thus
preventing buckling.

Due to high damping provided by the Pall Friction


Dampers, the strengthening of columns and foundations
was not necessary. Friction dampers up to 150 kip slip
capacity were used. Seismic retrofit was completed in
1999. Figure 8. 3-Million Gallon Water Tank.
Savings: More than 60% compared to conventional.
Concordia University Library Building, Montreal, Canada

Figure 9. (a). View of Concordia Library Building (b). Pall Friction Damper in Cross Bracing

36
The ten-story McConnell Library Building is a masterpiece in innovative structural design [7]. The
building was designed in 1987 and construction was completed in 1991. Pall Friction Dampers are
installed at the junction of steel cross bracing in rigid concrete frames. The use of steel bracing eliminated
the need of expensive concrete shearwalls and the introduction of supplemental damping provided by
friction dampers eliminated the need of dependence on ductility of structural members.

Unlike concrete shearwalls, the bracing were generally not continuous one over the other and thus
provided greater flexibility in space planning. Since the bracing do not carry any gravity load, they do not
go through the basement to the foundations. Therefore, they allow more space for parking. The architects
have boldly exposed several bracing with dampers as these add to the aesthetic appearance.

The innovative structural system provided an economical design solution to safeguard the building and its
valuable contents from earthquake damage.

Covered Area: 52,000 m2. Cost: $65 million. Savings: 6.5% of structural or 1.5% of building cost.

Seismic Rehabilitation of Justice Headquarters, Ottawa, Canada.

Figure 10. (a). View of Justice Headquarters (b). Pall Friction Damper

This eight-story concrete frame building is located in the nation's parliamentary district [21]. It was built in
1955 as a memorial to the Canadians killed during the Second World War. The stone clad building, with
sloping copper roof, has been designated a heritage structure. The existing structure was not capable of
resisting seismic forces specified in the Building Code. In 1997, seismic rehabilitation was undertaken
along with major renovations to protect the original and new investment.

Pall Friction Dampers offered the best solution for the seismic upgrade. Since the dampers dissipate a
major portion of the seismic energy, forces acting on the structure are considerably reduced. By staggering
the bracing at different story levels, overloading of columns and foundations was avoided. Hence,
expensive and time-consuming work of strengthening existing members and foundations was not required.

Unlike shearwalls, the friction-damped bracing need not be vertically continuous. This aspect was
particularly appealing to the architects as it offered flexibility in space planning. This structural solution
also facilitated construction scheduling as work could start at any floor level, depending on vacancy or
availability.
Covered Area: 50,565 m2. Savings: More than 40% in construction cost and time.

37
Canadian Space Agency Headquarters, St-Hubert, Canada

Figure 11. View of Canadian Space Agency Headquarters

The headquarters of the Canadian Space Agency is a building of national importance [11]. It contains
extremely sensitive and costly equipment / instrumentation. Therefore, it is of vital importance to protect
its valuable contents and electronically stored data in the event of a major earthquake.

The use of Pall Friction Dampers significantly increased the damage control potential of the building while
offering savings in the initial cost of construction. Construction was completed in 1992.

Covered Area: 50,000 m2. Cost $60,000,000. Savings: 1.25% of total cost.

Seismic Rehabilitation of Casino de Montreal, Canada

Figure 12. Casino de Montreal

38
In 1993, Casino de Montreal was housed in the existing French Pavilion built for EXPO'1967 [12]. The
lateral earthquake resistance of the existing eight-story braced steel structure was not adequate to meet the
requirements of the National Building Code of Canada.

Introduction of supplemental damping provided by Pall Friction Dampers was the most effective,
economical and hi-tech solution for the seismic rehabilitation of this building. The use of Pall Friction
Dampers in the existing steel bracing considerably reduced the forces on the structure. Hence, the
provision of additional bracing, strengthening of existing members and pile foundation was not required.

Covered Area: 38,000 m2. Savings: 50% in construction cost & time.

Quebec Provincial Police Headquarters, Montreal, Canada

Figure 13. (a). View of Police Headquarters (b). Pall Friction Dampers in Bracing

The existing sixteen-story office building, with two levels of basement, was built in 1964 [23]. Steel
moment frames and some braced bays provided lateral resistance to the existing structure. A change of
occupancy was planned in 1997 to house the provincial police (Surete du Quebec) headquarters. The
project structural engineers evaluated that the existing structure was not capable of resisting seismic forces
and the story drifts were excessively high, especially at the lover level due to soft story effect.

In 1999, the work on seismic rehabilitation was undertaken along with major renovations to protect the
original and new investment. Pall Friction Dampers offered the best solution for the seismic upgrade.
Since the dampers dissipate a major portion of the seismic energy, the story drifts and forces acting on the
structure are considerably reduced. Hence, expensive and time-consuming work of strengthening existing
member and pile foundations was not required. Pall Friction Dampers were incorporated in existing and
new bracing. The innovative structural scheme offered savings of more than 50% over conventional
retrofit scheme.

CONCLUSION

The use of Pall Friction Dampers has shown to provide a practical, economical and effective approach for
the performance-based design of new and retrofit of existing structures to resist major earthquakes. The
low cost and maintenance free characteristics of Pall Friction Dampers suggest wide application. Public
sector, private sector and developers, including developing countries, are using and benefiting from Pall
Friction Damper technology.

39
REFERENCES

1. Pall, A.S., Marsh, C. "Energy Dissipation in Large Panel Structures Using Limited Slip Bolted
Joints", AICAP/CEB Seismic Conference, Rome, Italy, 1979, 3:27-34.
2. Pall, A.S., Marsh, C. "Friction-Devices to Control Seismic Response", ASCE/EMD Speciality
Conference on Dynamic Response of Structures, Atlanta, USA, 1981, 809-818.
3. Pall, A.S., Marsh, C. "Seismic Response of Friction Damped Braced Frames", Journal of
Structural Division, ASCE, 1982, St. 9, 108:1313-1323. (ASCE "Raymond Reese Research Prize
1983").
4. Filiatrault, A., Cherry, S. "Seismic Tests of Friction-Damped Steel Frames", Third Conference on
Dynamic Response of Structures, ASCE, Los Angeles, USA, 1986.
5. Kelly, J.M., Aiken, I.D., Pall, A.S., "Seismic Response of a Nine-Story Steel Frame with Friction-
Damped Cross-Bracing", Report No. UCB / EERC-88/17. Earthquake Engineering Research
Center, University of California at Berkeley, 1988, 1-7.
6. Pall, A.S., Marsh, C., Fazio, P., "Friction Joints for Seismic Control of Large Panel Structures",
Journal of Prestressed Concrete Institute, 1980, No. 6, 25:38-61.
7. Pall, A.S., Verganelakis, V., Marsh, C., "Friction-Dampers for Seismic Control of Concordia
University Library Building", Fifth Canadian Conference on Earthquake Engineering, Ottawa,
1987,191-200.
8. Pall, A.S., Ghorayeb, F., Pall, R., "Friction-Dampers for Rehabilitation of Ecole Polyvalente at
Sorel, Quebec", Sixth Canadian Conference on Earthquake Engineering, Toronto, 1991, 389-396.
9. Vezina, S., Proulx, P., Pall, R., Pall, A., 'Friction-Dampers for Aseismic Design of Canadian
Space Agency", Tenth World Conference on Earthquake Engg., Madrid, Spain, 1992, 4123-4128.
10. Pall, A.S., Pall, R., "Friction-Dampers Used for Seismic Control of New and Existing Buildings in
Canada", ATC 17-1 Seminar on Base Isolation, Passive Energy Dissipation and Active Control,
San Francisco, USA, 1993a, 2:675-686.
11. Pall, A., Vezina, S., Proulx, Pall, R., "Friction-Dampers for Seismic Control of Canadian Space
Agency Headquarters", Journal Earthquake Spectra, 1993b, Number 3, 9:547-557.
12. Pasquin, C., Pall, A.S., Pall, R., "Hi-Tech Seismic Rehabilitation of Casino de Montreal", ASCE
Structures Congress, Atlanta, USA, 1994, 1292-1297.
13. Godin, D., Poirer, R., Pall, R., Pall, A., "Reinforcement Sismique du Nouveau Campus de l'Ecole
de Technologie Superieure de Montreal", Seventh Canadian Conference on Earthquake
Engineering, Montreal, 1995, 967-974.
14. Hale, T., Tokas, C., Pall, A., "Seismic Retrofit of Elevated Water Tanks at the University of
California at Davis", 7th. Canadian Conference on Earthquake Engg, Montreal, 1995, 959-966.
15. Savard, G., Lalancette, J.R., Pall, R., Pall, A., "High Tech Seismic Design of Maison 1 McGill",
Montreal, Seventh Canadian Conference on Earthquake Engineering, Montreal, 1995, 935-942.
16. Wagner, P., Vavak, L., Pall, R., Pall, A., "Seismic Rehabilitation of the New Hamilton Court
House", Seventh Canadian Conference on Earthquake Engineering, Montreal, 1995, 951-958.
17. Pall, A., Pall, R., "Friction-Dampers for Seismic Control of Buildings - A Canadian Experience",
Eleventh World Conference on Earthquake Engineering, Acapulco, Mexico, 1996, Paper No. 497.
18. Deslaurier, F., Pall, A., Pall, R., "Seismic Rehabilitation of Federal Building, Sherbrooke",
Canadian Society of Civil Engineers Annual Conference, Sherbrooke, 1997, 4:339-348.
19. Pasquin, C., Charania, H., Steele, R., Pall, R., Pall, A.S., "Friction-dampers for Seismic Control of
Selkirk Waterfront Offices, Victoria", Sixth U.S. National Conference on Earthquake Engineering,
Seattle, USA, 1998.
20. Pasquin, C., Leboeauf, N., Pall, R., Pall, A., "Seismic Rehabilitation of Hotel Dieu Hospital,
Sainte Hyacinthe, Quebec", Eighth Canadian Conference on Earthquake Engineering, Vancouver,
1999, 573-578.

40
21. Balazic, J., Guruswamy, G., Elliot, J., Pall, R., Pall, A., "Seismic Rehabilitation of Justice
Headquarters Building, Ottawa, Canada", Twelfth World Conference on Earthquake Engineering,
Auckland, New Zealand, 2000, Paper No. 2011.
22. Hale, T., Pall, R., "Seismic Upgrade of the Freeport Water Reservoir, Sacramento, California",
Twelfth World Conference on Earthquake Engg., Auckland, New Zealand, 2000, Paper No. 269.
23. Pall, R., Gauthier, G., Delisle, S., Pall, A., "Friction-Dampers for Seismic Upgrade of Quebec
Police Headquarters, Montreal", Twelfth World Conference on Earthquake Engineering,
Auckland, New Zealand, 2000, Paper No. 2014.
24. Chandra, R., Masand, M., Nandi, S., Tripathi, C., Pall, R., Pall, "Friction Dampers for Seismic
Control of La Gardenia Towers South City, Gurgaon, India", Twelfth World Conference on
Earthquake Engineering, Auckland, New Zealand, 2000, Paper No. 2008.
25. Vail, C., Hubbell, J., "Structural Upgrade of the Boeing Commercial Airplane Factory at Everett,
Washington", Proc. Structures Congress & Exposition, May 2003, Seattle, WA, Paper # 000529.
26. Sahai, R., Laws, J., Chen, D., Kong, F., Castillo, F., "Performance Based Design of 4-Story
Moscone Convention Center Expansion Using Steel Couple Girder Moment Resisting Frame and
Friction Dampers", Proceedings, SEAOC Annual Convention, Santa Barbara, California, 2000.
27. Prestandard and Commentary for the Seismic Rehabilitation of Buildings, FEMA 356/ November

41
13th World Conference on Earthquake Engineering
о w с^
Vancouver, B.C., Canada
August 1-6, 2004
Paper No. 1946

SEISMIC RETROFIT OF MUCTC BUILDING USING FRICTION


DAMPERS, PALAIS DES CONGRES, MONTREAL
2

Serge VEZINA2, R. Tina PALL

SUMMARY

The ten-storey MUCTC Building, built in 1928, is designated as a structure of heritage importance. In
2000, it was decided to expand the adjoining convention centre 'Palais des Congres'. The new expansion
is built around and integrated with the MUCTC Building. The expansion prompted the seismic upgrade
of the MUCTC Building. Conventional methods of seismic rehabilitation, with concrete shearwalls or
rigid steel bracing, were not suitable for the MUCTC Building. Supplemental damping in conjunction
with appropriate stiffness offered an innovative and attractive solution for the seismic rehabilitation of
this prestigious building. This was achieved by introducing Pall friction dampers in steel bracing.

INTRODUCTION

The MUCTC Building (Figure 1) is located in the heart of Montreal - close to historic Old Montreal and
in the area slated to be the most prestigious in town, the International District. This ten-storey concrete
frame office building was built in 1928. The floor slabs are one-way joist system and the foundations are
spread footings. As with the majority of buildings of this age, the earthquake resistance of the existing
structure was significantly less than that of current building code requirements. In 2000, it was decided to
expand the nearby Palais des Congres (Convention Centre). The new extension is built around and
integrated with the MUCTC Building, which had to be preserved as it is of historic significance (Figure
2). The expansion prompted the seismic upgrade of the MUCTC Building.

Conventional methods of seismic rehabilitation, with concrete shearwalls or rigid steel bracing, were not
considered suitable for the MUCTC Building as upgrade with these methods would have required
expensive and time consuming foundation work. Also, the shearwalls would have interfered with the
heritage character of the structure. The tight budget and schedule made these conventional options
unfeasible. Supplemental damping in conjunction with appropriate stiffness offered an innovative and
attractive solution for the seismic rehabilitation of this prestigious building. This was achieved by
introducing friction dampers in steel bracing. In contrast to shearwalls, friction-damped bracing need not
be vertically continuous. This aspect was particularly appealing to the architectural designers as it offered
flexibility in space planning. Since friction-damped bracing do not carry any gravity load, these do not

2
Dessau Soprin, Laval, Quebec, Canada. Email: serge.vezina@dessausoprin.com

42
Figure 1. MUCTC Building
Figure 2. Palais des Congres extension integrating
MUCTC Building

42m

Figure 4. Friction damper at bottom of


Figure 3. Typical single diagonal brace
floor

43
Figure 5. Typical detail of connections

need to go down through the basement to the foundation. At the ground floor level, the lateral shear from
the bracing is transferred through the rigid floor diaphragm to the perimeter retaining walls of the
basement. A typical floor plan is shown in Figure 3.

A total of 88 friction dampers were required to extract sufficient energy to safeguard the structure and its
contents from damage. A typical friction damper in steel bracing and connection details is shown in
Figures 4 and 5, respectively.

This paper describes the state-of-the-art, analysis, design and construction details of the seismic upgrade.
A brief review on Pall friction dampers has also been included so that the use of the novel solution can be
better appreciated.

CONVENTIONAL CONSTRUCTION

The design criteria stipulated in all building codes, including the National Building Code of Canada
(NBCC 1995), is based on the philosophy of designing structures to resist moderate earthquakes without
significant damage and to avoid structural collapse during a major earthquake. In general, reliance for
survival is placed on the ductility of the structure to dissipate energy while undergoing large inelastic
deformations causing bending, twisting and cracking. This results in permanent damage. Repair costs can
often be as significant as the costs of collapse of the structure. Recent earthquakes have clearly shown
that conventional construction even in technologically advanced and industrialized countries is not
immune to destruction.

While the minimum design provisions of the building codes were adequate in the past, in modern
buildings, avoidance of structural collapse alone is not enough. The cost of finishes, contents, sensitive
instrumentation and electronically stored records can be much higher than the cost of the structure itself
and these must be protected. In view of the massive financial losses and social sufferings, highlighted by
recent earthquakes, building officials, structural engineers, developers, owners, bankers and insurers have
started giving due consideration to performance based design rather than life safety alone.

Braced steel frames are known to be economical and effective in controlling lateral deflections due to
wind and moderate earthquakes. During a major earthquake, these structures do not perform that well. A
brace in tension stretches during severe shock and buckles in compression during reversal of load. On the
next application of load in the same direction, this elongated brace is not effective even in tension until it
is taut again and is stretched even further. As a result, the energy dissipation degrades very quickly and
the structure may collapse. The 1995 Kobe earthquake demonstrated several failures of braced buildings.

44
Concrete shearwalls or steel bracing is often used to add rigidity to moment-resisting frames. Generally,
stiffer structures attract higher ground accelerations thus exert higher forces on supporting members and
foundations. Therefore, any advantage gained by added stiffness is negated by increased amount of
energy input. Ductility in a reinforced concrete wall is extremely sensitive to detailing and quality control
and is often viewed with suspicion. Besides the high cost of construction, the use of shearwalls severely
restricts the flexibility of space planning. Once located, they have to continue from top to foundation.

The problems created by the dependence on ductility of the structure can be reduced if a major portion of
the seismic energy is dissipated mechanically, independent of the primary structure. With the emergence
of friction dampers, it has become economically feasible to significantly increase the earthquake
resistance and damage control potential of buildings.

PALL FRICTION DAMPERS

Of all the methods available to extract kinetic energy from a moving body, the most widely adopted is
undoubtedly the friction brake. It is the most effective, reliable and economical mean to dissipate energy.
For centuries, mechanical engineers have successfully used this concept to control the motion of
machinery and automobiles. In the late 1970's, the principle of friction brake inspired the development of
friction dampers, Pall [1,2]. Similar to automobiles, the motion of a vibrating building can be controlled.

Friction dampers suitable for use in different types of bracing have been developed, Pall [3]. They are
available for tension cross bracing, single diagonal bracing, and chevron bracing. Pall friction dampers
are simple and foolproof in construction and inexpensive in cost. Basically, these consist of series of steel
plates specially treated to develop most reliable friction. The plates are clamped together with high
strength steel bolts. Slippage is without any stick-slip phenomenon. Friction dampers are designed not to
slip during service load and windstorms. During a major earthquake, they slip at a predetermined
optimum load before yielding occurs in other structural members and dissipate a major portion of the
seismic energy. By properly selecting the slip load, it is possible to 'tune' the response of the structure to
an optimum value. This allows the building to remain elastic or at least yielding is delayed to be available
during maximum credible earthquakes. Parametric studies have shown that the optimum slip load is
independent of earthquake record and is rather a structural property. Also, within a variation of ± 20% of
slip load, the seismic response is not significantly affected. After the earthquake, building returns to its
near original alignment under the spring action of an elastic structure.

These particular friction dampers have successfully gone through rigorous proof testing on shake tables in
Canada and the United States. In 1985, a three-storey frame equipped with friction dampers was tested on
a shake table at the University of British Columbia, Vancouver, Cherry [4]. Even an earthquake record
with a peak acceleration of 0.9g did not cause any damage to the friction-damped braced frame, while the
conventional frames were severely damaged at much lower seismic levels. In 1987, a nine-storey three-
bay frame, equipped with friction dampers, was tested on a shake table at Earthquake Engineering
Research Centre of the University of California at Berkeley, Kelly [5]. All members of the friction damped
frame remained elastic for 0.84g acceleration, while the moment-resisting frame would have yielded.

These friction dampers possess large rectangular hysteresis loops, similar to an ideal elasto-plastic
behaviour, with negligible fade over several cycles of reversals, Pall [6], Filiatrault [4]. Unlike viscous or
visco-elastic devices, the performance of friction dampers is independent of temperature and velocity. For
a given force and displacement in a damper, the energy dissipation of a friction damper is the largest
compared to other damping devices (Figure 6). Therefore, fewer friction dampers are required to provide
a given amount of supplemental damping.

45
Unlike systems that dissipate energy through the process of yielding - causing permanent damage,
friction dampers dissipate seismic energy in friction. The maximum force in a friction damper is well
defined and remains constant for any future ground motion. Hence, the design of bracing and connections
is straightforward and economical. Since they are not active during wind or service load conditions, there
is no danger of failure due to fatigue. There is nothing to leak or damage. Therefore, they do not need
regular inspection, maintenance, repair or replacement before and after the earthquake. Friction dampers
are also very compact in design and can be easily hidden within drywall partitions. These friction
dampers meet a high standard of quality control. Every damper is load tested to ensure proper slip load
before it is shipped.

These friction dampers have found many applications. They have been used in both new construction and
seismic retrofit of existing buildings, Pall [7-11,17,23], Pasquin [12,19,20], Vezina [9,11], Godin [13],
Hale [14,22], Savard [15], Wagner [16], Deslaurier [18], Balazic [21], Chandra [24]. Boeing Commercial
Airplane Factory at Everett - the world's largest building in volume and Boeing Development Center
Buildings at Seattle have been retrofitted with these friction dampers, Vail [25]. Compared to
conventional retrofit, Boeing saved more than US$30 million. The City and County of San Francisco
chose Pall friction dampers for seismic control of Moscone Convention Center as it saved US$2.25
million compared to alternate viscous dampers, Sahai [26]. To date, more than eighty buildings have
already been built and several are under design or construction. For more details refer:
www.palldynamics.com.

Friction damper Viscous damper Viscoelastic damper Self-centering


Friction damper
Figure 6. Comparison of hysteresis loops of different dampers

DESIGN CRITERIA

The quasi-static design procedure given in the NBCC is ductility based and does not explicitly apply to
friction-damped buildings. However, structural commentary - J of the NBCC, allows the use of friction
dampers for seismic control of buildings. It requires that nonlinear analysis must demonstrate that a
building so equipped will perform equally well in seismic events. In the past few years, several guidelines
on the analysis and design procedure of passive energy dissipation devices have been developed in the
U.S. The latest and most comprehensive document is the "NEHRP Guidelines for the Seismic
Rehabilitation of Buildings", FEMA 356 / 357, issued in 2000. These guidelines and provisions of
NBCC, served as basis for the analysis and design of the MUCTC Building.

The Guidelines require that the structure with energy dissipating devices be evaluated for response to two
levels of ground shaking - a design basis earthquake (DBE) and a maximum considered earthquake
(MCE). The DBE is an event with 10% probability of exceedance in 50 years, while the MCE represents
a severe ground motion of probability of 2% in 50 years. Under the DBE, the structure is evaluated to
ensure that the strength demands on structural elements do not exceed their capacities and that the drift is
within the tolerable limits. For the MCE, the structure is evaluated to determine the maximum
displacement requirement of the damping device. It is presumed that if proper ductile detailing has been
followed, the structure will have sufficient reserve to resist any overstress conditions that occur during
MCE.

46
NEHRP guidelines require that friction dampers are designed for 130% MCE displacements and all
bracing and connections are designed for 130% of damper slip load. Variation in slip load from design
value should not be more that ±15%.

NONLINEAR TIME-HISTORY DYNAMIC ANALYSIS

The slippage of friction damper in an elastic brace constitutes nonlinearity. Also, the amount of energy
dissipation or equivalent structural damping is proportional to the displacement. Hence, the design of
friction-damped buildings requires the use of nonlinear time-history dynamic analysis. With these
analyses, the time-history response of the structure during and after an earthquake can be accurately
understood.

Three-dimensional nonlinear time-history dynamic analyses were carried out using the computer program
ETABS (Nonlinear version), developed by Computers and Structures Inc. Analytical computer model is
shown in Figure 7. Several other programs are also available on which friction dampers can be easily
modeled. The modeling of friction dampers is very simple. Since the hysteretic loop of the damper is
similar to the rectangular loop of an ideal elasto-plastic material, the slip load of the friction-damper can
be considered as a fictitious yield force.

Since different earthquake records, even of the same intensity, give widely varying structural responses,
results obtained using a single record may not be conclusive. Therefore, three time-history records,
suitable for the region, were used to ensure that possible coincidence of ground motions and building
frequencies was not missed. The earthquake record based on the Whittier earthquake of 1987 provided
maximum response and was used for the design. Analyses were carried out for ground motions
simultaneously 100% along x-direction and 30% along y-direction, and also for ground motions 100%
along y-direction and 30% along x-direction. Viscous damping of 2% of critical was assumed in the
initial elastic stage to account for the presence of non-structural elements. Several iterations were made to
determine the optimum slip load to achieve minimum response. A total of 88 friction dampers of 500-600
kN slip load were used in diagonal and chevron bracing (Figure 8).

47
Figure 7. Analytical computer model Figure 8. Friction dampers in diagonal
and chevron bracing
0.12

0.08

0.04

-0.04 -

-0.08

-0.12
0 5 30 35 40
10 15 20 25

Time, seconds

Figure 9. Time histories of displacements at roof

8lH

------------------------------SGB-l-
DfcptaCamefils (mm)

Figure 10. Hysteretic loop of a typical 600kN friction damper

48
In order to compare the effectiveness of friction-damped frames (FDF), analyses were also conducted on
the building using rigid bracing in frames (RBF). The rigid bracing with twice the area of friction damped
bracing gave the best response.
DISCUSSION OF RESULTS

1. Time-histories of deflections at the top of building are shown in Figure 9. Maximum defection at roof
is 100mm for FDF and 105 for RBF. The permanent offset of the FDF building at the end of ground
motion is negligible, about 2mm.
2. Maximum storey drift in FDF is less than 0.7%. At this low level of deformations, no damage is
expected during a major earthquake. In conventional construction, the building codes allow up to 2%.
3. Hysteretic loop of a typical damper in bracing is shown in Figure 10. The slippage in the damper is
about 8 mm. The slope in the hysteretic loop is due to the elastic shortening of brace. Unlike rigid
bracing, the maximum force developed in the friction-damped bracing and connections are constant
for all earthquake records. This results in an engineered solution i.e. the forces are predetermined by
the engineer and not by the earthquake.
4. Time-histories of deformations in friction-damped bracing is shown in Figure 11. The permanent
offset in the damper after the earthquake is less than 0.5mm.

5. Maximum envelopes for axial load in a column of a FDF and RBF are shown in Figure 12. The axial
loads in FDF are about 60% of those for the BMF. The use of rigid bracing would have resulted in
significant strengthening of columns and foundations.
CONCLUSION

The use of friction dampers has shown to provide a practical and economical solution for the seismic
upgrade of the MUCTC building. As the seismic forces exerted on the structure are significantly reduced,
the system offered savings in upgrade costs. The analytical studies have shown that the friction- damped
structure should perform well in the event of a major earthquake.

REFERENCES

1. Pall, A.S., Marsh, C. "Energy Dissipation in Large Panel Structures Using Limited Slip
Bolted Joints", AICAP/CEB Seismic Conference, Rome, Italy, 1979, 3:27-34.
2. Pall, A.S., Marsh, C. "Friction-Devices to Control Seismic Response", ASCE/EMD Speciality
Conference on Dynamic Response of Structures, Atlanta, USA, 1981, 809-818.
3. Pall, A.S., Marsh, C. "Seismic Response of Friction Damped Braced Frames", Journal of
Structural Division, ASCE, 1982, St. 9, 108:1313-1323. (ASCE "Raymond Reese Research Prize
1983").
4. Filiatrault, A., Cherry, S. "Seismic Tests of Friction-Damped Steel Frames", Third Conference on
Dynamic Response of Structures, ASCE, Los Angeles, USA, 1986.
5. Kelly, J.M., Aiken, I.D., Pall, A.S., "Seismic Response of a Nine-Story Steel Frame with
Friction- Damped Cross-Bracing", Report No. UCB / EERC-88/17. Earthquake Engineering
Research Center, University of California at Berkeley, 1988, 1-7.

49
>
ф
C
O

Figure 11. Time history of slippage of a typical 600kN friction damper


Time, seconds

□ FDF 1BMF

5 2000
4 Column Axial Force, kN
3
Figure 12. Envelope of column axial force
2

1
0 1000 3000 4000

50
6. Pall, A.S., Marsh, C., Fazio, P., "Friction Joints for Seismic Control of Large Panel Structures",
Journal of Prestressed Concrete Institute, 1980, No. 6, 25:38-61.
7. Pall, A.S., Verganelakis, V., Marsh, C., "Friction-Dampers for Seismic Control of Concordia
University Library Building", Fifth Canadian Conference on Earthquake Engineering, Ottawa,
1987, 191-200.
8. Pall, A.S., Ghorayeb, F., Pall, R., "Friction-Dampers for Rehabilitation of Ecole Polyvalente at
Sorel, Quebec", Sixth Canadian Conference on Earthquake Engineering, Toronto, 1991, 389-396.
9. Vezina, S., Proulx, P., Pall, R., Pall, A., 'Friction-Dampers for Aseismic Design of Canadian
Space Agency", Tenth World Conference on Earthquake Engineering, Madrid, Spain, 1992,
4123-4128.
10. Pall, A.S., Pall, R., "Friction-Dampers Used for Seismic Control of New and Existing Buildings
in Canada", ATC 17-1 Seminar on Base Isolation, Passive Energy Dissipation and Active
Control, San Francisco, USA, 1993a, 2:675-686.
11. Pall, A., Vezina, S., Proulx, Pall, R., "Friction-Dampers for Seismic Control of Canadian Space
Agency Headquarters", Journal Earthquake Spectra, 1993b, Number 3, 9:547-557.
12. Pasquin, C., Pall, A.S., Pall, R., "Hi-Tech Seismic Rehabilitation of Casino de Montreal", ASCE
Structures Congress, Atlanta, USA, 1994, 1292-1297.
13. Godin, D., Poirer, R., Pall, R., Pall, A., "Reinforcement Sismique du Nouveau Campus de l'Ecole
de Technologie Superieure de Montreal", Seventh Canadian Conference on Earthquake
Engineering, Montreal, 1995, 967-974.
14. Hale, T., Tokas, C., Pall, A., "Seismic Retrofit of Elevated Water Tanks at the University of
California at Davis", Seventh Canadian Conference on Earthquake Engineering, Montreal, 1995,
959-966.
15. Savard, G., Lalancette, J.R., Pall, R., Pall, A., "High Tech Seismic Design of Maison 1 McGill",
Montreal, Seventh Canadian Conference on Earthquake Engineering, Montreal, 1995, 935-942.
16. Wagner, P., Vavak, L., Pall, R., Pall, A., "Seismic Rehabilitation of the New Hamilton Court
House", Seventh Canadian Conference on Earthquake Engineering, Montreal, 1995, 951-958.
17. Pall, A., Pall, R., "Friction-Dampers for Seismic Control of Buildings - A Canadian Experience",
Eleventh World Conference on Earthquake Engineering, Acapulco, Mexico, 1996, Paper No.
497.
18. Deslaurier, F., Pall, A., Pall, R., "Seismic Rehabilitation of Federal Building, Sherbrooke",
Canadian Society of Civil Engineers Annual Conference, Sherbrooke, 1997, 4:339-348.
19. Pasquin, C., Charania, H., Steele, R., Pall, R., Pall, A.S., "Friction-dampers for Seismic Control
of Selkirk Waterfront Offices, Victoria", Sixth U.S. National Conference on Earthquake
Engineering, Seattle, USA, 1998.
20. Pasquin, C., Leboeauf, N., Pall, R., Pall, A., "Seismic Rehabilitation of Hotel Dieu Hospital,
Sainte Hyacinthe, Quebec", Eighth Canadian Conference on Earthquake Engineering, Vancouver,
1999, 573-578.
21. Balazic, J., Guruswamy, G., Elliot, J., Pall, R., Pall, A., "Seismic Rehabilitation of Justice
Headquarters Building, Ottawa, Canada", Twelfth World Conference on Earthquake Engineering,
Auckland, New Zealand, 2000, Paper No. 2011.
22. Hale, T., Pall, R., "Seismic Upgrade of the Freeport Water Reservoir, Sacramento, California",
Twelfth World Conference on Earthquake Engineering, Auckland, New Zealand, 2000, Paper
No. 269.
23. Pall, R., Gauthier, G., Delisle, S., Pall, A., "Friction-Dampers for Seismic Upgrade of Quebec
Police Headquarters, Montreal", Twelfth World Conference on Earthquake Engineering,
Auckland, New Zealand, 2000, Paper No. 2014.

51
24. Chandra, R., Masand, M., Nandi, S., Tripathi, C., Pall, R., Pall, "Friction Dampers for Seismic
Control of La Gardenia Towers South City, Gurgaon, India", Twelfth World Conference on
Earthquake Engineering, Auckland, New Zealand, 2000, Paper No. 2008.
25. Vail, C., Hubbell, J., "Structural Upgrade of the Boeing Commercial Airplane Factory at Everett,
Washington", Proceedings, Structures Congress & Exposition, May 2003, Seattle, WA, Paper #
000529.
26. Sahai, R., Laws, J., Chen, D., Kong, F., Castillo, F., "Performance Based Design of 4- Story
Moscone Convention Center Expansion Using Steel Couple Girder Moment Resisting Frame and
Friction Dampers", Proceedings, SEAOC Annual Convention, Santa Barbara, California, 2000.

52
13th World Conference on Earthquake Engineering
Vancouver, B.C., Canada
August 1-6, 2004
Paper No. 1949

FRICTION DAMPERS FOR SEISMIC REHABILITATION OF EATON'S


BUILDING, MONTREAL

Claude PASQUIN3, Normand LEBOEUF 1, R. Tina PALL4, Avtar PALL

SUMMARY

The ten-storey Eaton's building was originally built in 1925 and was expanded in several stages. The
structure derived its lateral stability from partial frame action and infill walls of unreinforced masonry.
The existing structure was not adequate to resist the lateral seismic forces specified in the current building
code. In 2000, the new ownership planned major redevelopment of this building. Therefore, it was decided
that seismic rehabilitation work be undertaken along with major renovations to protect the existing and
new investment.

Of the several schemes, the introduction of supplemental damping using friction dampers in steel bracing
was considered to be the most effective and economical solution for the seismic upgrade. This novel
approach significantly reduced the drifts and base shear while greatly minimizing the strengthening of the
existing members. This paper discusses the design criteria, seismic analysis and its results.

INTRODUCTION

The existing Eaton's building is located in the heart of the shopping district on St. Catherine Street in
Montreal (Figures 1, 2). It is a vast structure of over one million square feet distributed over ten levels and
two basements. It was built in several stages from 1925 to 1959. The different construction phases
introduced different structural systems. It has both concrete and steel frames with concrete slabs. The
Eaton's building has a beautiful fagade. As the building has been used to house a department store, there
are large open spaces at each level. The structure derived its lateral stability mainly from the perimeter
masonry walls, a few interior walls and the concrete frame built in the last construction phase of 1959. A
typical floor plan is shown in Figure 3. The preliminary analysis indicated that the existing structure was
not adequate to resist the lateral seismic forces specified in the National Building Code of Canada 1995.

3
. Pasquin St-Jean & Associates Inc., Montreal, Canada. Email: psa@psa.ca
4
. Pall Dynamics Limited, Montreal, Canada. Email: info@palldynamics.com

53
In 2000, the new building owner decided to proceed with important renovations to welcome a major tenant
"Les Ailes de la Mode". The planned renovations lead to significant changes in the interior and further
diminished its seismic resistance capacity. Among these, a new large egg-shaped atrium with a glass roof
and punching of openings in the lower floors for new elevators and escalators made this project

54
a daunting challenge for the structural engineers. It was necessary that seismic rehabilitation work be
undertaken along with the architectural renovations to protect the existing and new investment.

SEISMIC UPGRADE

There were several alternative methods open for the seismic upgrade of Eaton's building. The conventional
methods of stiffening consist of adding concrete shearwalls or rigid steel bracing. During a major
earthquake, these structures tend to attract higher ground accelerations causing higher inertial forces on the
supporting structure. Therefore, any advantage gained with the added stiffness may be negated by the
increased amount of seismic energy input. Addition of new shearwalls would have interfered with the
open character of the interior plan. In a conventional braced frame, the energy dissipation capacity of a
brace is very limited. Several rigid braced buildings have failed in Kobe earthquake. Both conventional
methods of upgrade require expensive and time-consuming work of strengthening the existing columns
and foundations. The tight budget and "Fast Track" schedule made these conventional options unfeasible.

Supplemental damping in conjunction with appropriate stiffness offered an innovative solution for the
seismic rehabilitation of the Eaton's building. This was achieved by incorporating friction dampers in new
steel bracing. As soon as the structure undergoes small deformations, the friction dampers are activated
and start dissipating energy. However, repairable cracks in the masonry may have to be accepted. Since
the dampers dissipate a major portion of the seismic energy, the forces acting on the structure are
considerably reduced. In contrast to shearwalls, the friction-damped bracing need not be vertically
continuous. Since the damped bracing do not carry gravity load, they do not need to go down through the
basement to the foundation. At the ground floor level, the lateral shear from the bracing is transferred
through the rigid floor diaphragm to the perimeter retaining walls of the basement. These aspects were
particularly appealing to the project architects as they offered great flexibility in space planning. By
staggering the bracing at different story levels, the overloading on some columns and foundations was
reduced. Hence, expensive and time-consuming work on strengthening of foundations was not required.
Higher energy dissipation capacity of friction dampers compensates the lack of ductility and mitigates
damage to other nonstructural components.

A total of 161 friction dampers were installed in the Eaton's building. Typical friction dampers in single
diagonal and chevron bracing are shown in Figures 4 and 5, respectively.

This paper describes the design criteria, seismic analysis and its results. A brief review on Pall friction
dampers has also been included so that its practical application can be better appreciated.

FRICTION DAMPERS

Of all the methods available to extract kinetic energy from a moving body, the most widely adopted is
undoubtedly the friction brake. It is the most effective, reliable and economical mean to dissipate energy.
In late seventies, the principle of friction brake inspired the development of friction dampers, Pall et al. [8-
11]. Similar to automobiles, the motion of vibrating building can be slowed down by dissipating energy in
friction. Several types of friction dampers have been developed, Pall et al. [11]. For frame buildings, these
are available for tension cross bracing, single diagonal bracing, chevron bracing, and friction connectors at
expansion joints to avoid pounding.

Pall friction dampers are simple and foolproof in construction and inexpensive in cost. They consist of
series of steel plates specially treated to develop most reliable friction. The plates are clamped together
with high strength steel bolts. Friction dampers are designed not to slip during wind. During severe
seismic excitations, friction dampers slip at a predetermined optimum load before yielding occurs in other
structural members and dissipate a major portion of the seismic energy. Another feature of friction damped

55
buildings is that their natural period varies with the amplitude of vibration. Hence the phenomenon of
resonance is avoided. After the earthquake, the building returns to its near original alignment under the
spring action of an elastic structure.

These particular friction dampers have successfully gone through rigorous proof testing on shake tables in
Canada and the United States. In 1985, a three-storey frame equipped with friction dampers was tested on
a shake table at the University of British Columbia, Vancouver, Cherry et al. [3]. Even an earthquake
record with a peak acceleration of 0.9g did not cause any damage to friction damped braced frame, while
the conventional frames were severely damaged at lower seismic levels. In 1987, a nine storey three bay
frame, equipped with friction dampers, was tested on a shake table at the Earthquake Engineering
Research Center of the University of California at Berkeley, Kelly et al. [7]. All members of the friction
damped frame remained elastic for 0.84g acceleration, while the moment-resisting frame would have
yielded at about 0.3g acceleration.

Friction dampers possess large rectangular hysteresis loops, similar to an ideal elasto-plastic behavior,
with negligible fade over several cycles of reversals Pall et al. [9], Filiatrault et al. [3]. Unlike viscous or
visco-elastic devices, the performance of friction dampers is independent of temperature and velocity. For
a given force and displacement in a damper, the energy dissipation of friction damper is the largest
compared to other damping devices (Figure 6). Therefore, fewer friction dampers are required to provide a
given amount of supplemental damping. Unlike other devices, the maximum force in a friction damper is
pre-defined and remains the same for any future ground motion. Therefore, the design of bracing and
connections is simple and economical. There is nothing to yield and damage, or leak. Thus, they do not
need regular inspection, maintenance, repair or replacement before and after the earthquake. These friction
dampers are also compact in design and can be easily hidden within drywall partitions.

Friction dampers manufactured by Pall Dynamics Limited have found many applications in new
construction and seismic retrofit of existing buildings, Pall et al. [12-17], Vezina et al. [23], Pasquin et al.
[18-20], Godin et al. [4], Savard et al. [21], Wagner et al. [25], Deslaurier et al. [2], Balazic et al. [2000],
Hale et al. [5,6]. Boeing's Commercial Airplane Factory in Everett WA - world's largest building in
volume has been retrofitted with these friction dampers, Vail et al. [24]. Boeing saved more than US$30
million by using this technology. The City and County of San Francisco chose Pall friction dampers for
the seismic control of Moscone Convention Center as it saved them US$2.25 million compared to
alternate viscous dampers, Sahai et al. [22]. To date, more than eighty buildings have already been built
and several are under design or construction. For more details refer www.palldynamics.com.

DESIGN CRITERIA

The quasi-static design procedure given in the NBCC is ductility based and does not explicitly apply to
friction-damped buildings. However, structural commentary - J of the NBCC, allows the use of friction
dampers for seismic control of buildings. It requires that nonlinear analysis must show that a building so
equipped will perform equally well in seismic events as the same building designed following the NBCC
seismic requirements. In the past few years, various guidelines on the analysis and design procedure of
passive energy dissipation devices have been developed in the U.S. The latest and most comprehensive
document is the "NEHRP Guidelines for the Seismic Rehabilitation of Buildings", FEMA 356 / 357,
issued in 2000. These guidelines and provisions of NBCC, served as basis for the analysis and design of
the Eaton's building.
The guidelines require that the structure with energy dissipating devices be evaluated for response to two
levels of ground shaking - a design basis earthquake (DBE) and a maximum considered earthquake
(MCE). The DBE is an event with 10% probability of exceedance in 50 years, while the MCE represents a
severe ground motion of probability of 2% in 50 years. Under the DBE, the structure is evaluated to ensure

56
that the strength requirements on structural elements do not surpass their capacities and that the drift in the
structure is within the permissible limits. For the MCE, the structure is assessed to ascertain the maximum
displacement requirement of the damping device. It is presumed that if proper ductile detailing has been
followed, the structure will have sufficient reserve to resist any overstress conditions that occur during
MCE.

NEHRP guidelines require that friction dampers are designed for 130% MCE displacements and all
bracing and connections are designed for 130% of damper slip load. Variation in slip load from design
value should not be more that ±15%. The friction dampers used in this project meet high standard of
quality control. Before delivery to site, each damper is load tested to ensure proper slip load.

NONLINEAR TIME-HISTORY DYNAMIC ANALYSIS

The slippage of friction damper in an elastic brace constitutes nonlinearity. Also, the amount of energy
dissipation or equivalent structural damping is proportional to the displacement. Therefore, the design of
friction-damped buildings requires the use of nonlinear time-history dynamic analysis. With these
analyses, the time-history response of the structure during and after an earthquake can be accurately
understood.

Three-dimensional nonlinear time-history dynamic analyses were carried out using the computer program
ETABS (Nonlinear version), developed by Computers and Structures Inc. The analysis model is shown in
Figure 7. The modeling of friction dampers is very simple. Since the hysteretic loop of the damper is
similar to the rectangular loop of an ideal elasto-plastic material, the slip load of the friction damper can be
considered as a fictitious yield force. In the analyses, friction dampers in single diagonal brace are
modeled as damped braces having member stiffness equal to brace stiffness and nonlinear axial yielding
equal to the slip load. Friction damped chevron braces are modeled as braces plus dampers. These dampers
have nonlinear yield force in shear equal to the slip load.

A series of analyses were made to determine the optimum slip load of friction dampers to achieve
minimum response. The damper slip loads are 700kN at ground storey, 600kN for the next five storeys,
and 300kN at upper storeys.

Since different earthquake records, even of the equal intensity, give widely varying structural responses,
results attained using a single record may not be conclusive. Therefore, different time-history records,
suitable for the Montreal region, were used to ensure that possible coincidence of ground motions and
building frequencies was not missed. The earthquake record based on the Whittier earthquake of 1987
gave maximum response and was used for the design. Analyses were carried out for ground motions
applied 100% along x and y directions. Viscous damping of 4% of critical was assumed in the initial
elastic stage to account for the presence of non-structural elements.

Analyses were also conducted on frames with concentric rigid bracing in moment frames. The
effectiveness of friction dampers in improving the seismic response is seen in comparison of the results of
two types of frames. The friction damped frames (FDF) and the concentrically braced moment frames
(BMF) have the same member properties, except that the BMF has twice the area of brace than that in the
FDF. For smaller or larger brace areas tried, the response of the BMF was higher. The results compared
are for the maximum response of the DBE record.
DISCUSSION OF RESULTS
1. The total energy input in the structure for BMF and FDF and the energy dissipated by friction dampers
is shown in Figure 8. The FDF has a dual advantage over the BMF. Firstly, the FDF attracts only 50%
of the seismic energy input of the stiffer BMF. Secondly, the friction dampers in the FDF dissipate

57
about 50% of the input energy. So the remaining energy left in the FDF is approximately 25% of the
energy in the BMF.

2. Time-histories of deflections at the top of building are shown in Figure 9. The peak amplitude in x
direction is 156mm, about 54% of BMF. Maximum storey drifts in FDF are less than 1%. At this low
level of drift, no damage is expected during a major earthquake.

3. Hysteretic loop of a 700kN friction-damped single diagonal brace is shown in Figure 10. The
maximum amplitude of slippage is -25mm. Time history of slippage in this damped brace is shown in
Figure 11. The permanent offset in the damper after the earthquake was less than 2mm.

4. Maximum envelopes for axial loads in a column of a braced bay are shown in Figure 12. The column
axial forces in FDF are about 50% of that for the BMF. While the forces in brace s for FDF remained
constant for all earthquake records, the forces in BMF were different for different earthquakes. In fact
use of friction dampers provides an engineered solution i.e. the engineer controls the maximum forces
rather than the earthquake.

5. The base shears in FDF are 16000kN and 13000kN in x and y direction, respectively. For BMF, the
base shears are 47000kN and 27000kN in x and y direction, respectively. In general, the use of friction
dampers resulted in an overall improvement in seismic response.

CONCLUSIONS
The analytical studies have shown that the friction-damped structure should perform well in the event of a
major earthquake. The seismic performance of the structure is far superior to the requirements of the
building code. As the seismic forces exerted on the structure are significantly reduced, the system offered
savings in upgrade costs. Besides savings in the upgrade cost, the saving in life cycle cost could be
significant as damage to the building and its content is minimised. The use of friction dampers has shown
to provide a practical and economical solution for the seismic upgrade of the Eaton's building.

REFERENCES
1. Balazic, J., Guruswamy, G., Elliot, J., Pall, R., Pall, A. "Seismic Rehabilitation of JusticeHeadquarters
Building, Ottawa, Canada", Twelfth World Conference on Earthquake Engineering, Auckland, New
Zealand, 2000 , Paper No. 2011.
2. Deslaurier, F., Pall, A., Pall, R. "Seismic Rehabilitation of Federal Building, Sherbrooke", Canadian
Society of Civil Engineers Annual Conference, Sherbrooke, 1997, pp.339-348.
3. Filiatrault, A., Cherry, S. (1986) Seismic Tests of Friction-Damped Steel Frames, Third Conference on
Dynamic Response of Structures, ASCE, Los Angeles, USA.
4. Godin, D., Poirer, R., Pall, R., Pall, A. (1995) Reinforcement Sismique du Nouveau Campus de
l'Ecole de Technologie Superieure de Montreal, Seventh Canadian Conference on Earthquake
Engineering, Montreal, 967-974.
5. Hale, T., Tokas, C., Pall, A. (1995) Seismic Retrofit of Elevated Water Tanks at the University of
California at Davis, Seventh Canadian Conference on Earthquake Engineering, Montreal, 959-966.
6. Hale, T., Pall, R. (2000) Seismic Upgrade of the Freeport Water Reservoir, Sacramento, California,
Twelfth World Conference on Earthquake Engineering, Auckland,New Zealand, Paper No. 269.
7. Kelly, J.M., Aiken, I.D., Pall, A.S. (1988) Seismic Response of a Nine-Story Steel Frame with
Friction-Damped Cross-Bracing, Report No. UCB / EERC-88/17. Earthquake Engineering Research
Center, University of California at Berkeley, 1-7.
8. Pall, A.S., Marsh, C. (1979) Energy Dissipation in Large Panel Structures Using Limited Slip Bolted
Joints, AICAP/CEB Seismic Conference, Rome, Italy, 3: 27-34.

58
9. Pall, A.S., Marsh, C., Fazio, P. (1980) Friction Joints for Seismic Control of Large Panel Structures,
Journal of Prestressed Concrete Institute, No. 6, 25:38-61.
10. Pall, A.S., Marsh, C. (1981) Friction-Devices to Control Seismic Response, ASCE/EMD Speciality
Conference on Dynamic Response of Structures, Atlanta, USA, 809-818.
11. Pall, A.S., Marsh, C. (1982) Seismic Response of Friction Damped Braced Frames, Journal of
Structural Division, ASCE, St. 9, 108:1313-1323. (ASCE "Raymond Reese Research Prize1983").
12. Pall, A.S., Verganelakis, V., Marsh, C. (1987) Friction-Dampers for Seismic Control of Concordia
University Library Building, Fifth Canadian Conference on Earthquake Engineering, Ottawa, 191200.
13. Pall, A.S., Ghorayeb, F., Pall, R. (1991) Friction-Dampers for Rehabilitation of Ecole Polyvalente at
Sorel, Quebec, Sixth Canadian Conference on Earthquake Engineering, Toronto, 389-396.
14. Pall, A.S., Pall, R. (1993a), Friction-Dampers Used for Seismic Control of New and Existing
Buildings in Canada, ATC 17-1 Seminar on Base Isolation, Passive Energy Dissipation and Active
Control, San Francisco, USA, 2:675-686.
15. Pall, A., Vezina, S., Proulx, Pall, R. (1993b), Friction-Dampers for Seismic Control of Canadian Space
Agency Headquarters, Journal Earthquake Spectra, Number 3, 9:547-557.
16. Pall, A., Pall, R. (1996) Friction-Dampers for Seismic Control of Buildings - A Canadian Experience,
Eleventh World Conference on Earthquake Engineering, Acapulco, Mexico, Paper No. 497.
17. Pall, R., Gauthier, G., Delisle, S., Pall, A. (2000) Friction-Dampers for Seismic Upgrade of Quebec
Police Headquarters, Montreal, Twelfth World Conference on Earthquake Engineering, Auckland,
New Zealand, Paper No. 2014.
18. Pasquin, C., Pall, A.S., Pall, R. (1994), Hi-Tech Seismic Rehabilitation of Casino de Montreal, ASCE
Structures Congress, Atlanta, USA, 1292-1297.
19. Pasquin, C., Charania, H., Steele, R., Pall, R., Pall, A.S. (1998) Friction-dampers for Seismic Control
of Selkirk Waterfront Offices, Victoria, Sixth U.S. National Conference on Earthquake Engineering,
Seattle,USA.
20. Pasquin, C., Leboeuf, N., Pall, R., Pall, A. (1999) Seismic Rehabilitation of Hotel-Dieu Hospital,
Sainte Hyacinthe, Quebec, Eighth Canadian Conference on Earthquake Engineering, Vancouver, 573-
578.
21. Savard, G., Lalancette, J.R., Pall, R., Pall, A. (1995) High Tech Seismic Design of Maison 1 McGill,
Montreal, Seventh Canadian Conference on Earthquake Engineering, Montreal, 935-942.
22. Sahai, R., Laws, J., Chen, D., Kong, F., castillo, F., "Performance Based Design of 4-Story Moscone
Convention Center Expansion Using Steel Couple Girder Moment Resisting Frame and Friction
Dampers", Proceedings, SEAOC Annual Convention, Santa Barbara, California, 2000.
23. Vezina, S., Proulx, P., Pall, R., Pall, A. (1992) Friction-Dampers for Aseismic Design of Canadian
Space Agency, Tenth World Conference on Earthquake Engineering, Madrid, Spain, 4123-4128.
24. Vail, C., Hubbell, J., "Structural Upgrade of the Boeing Commercial Airplane Factory at Everett,
Washington", Proceedings, Structures Congress & Exposition, May 2003, Seattle, WA, Paper #
000529.
25. Wagner, P., Vavak, L., Pall, R., Pall, A., "Seismic Rehabilitation of the New Hamilton Court House",
Seventh Canadian Conference on Earthquake Engineering, Montreal, 1995, pp. 951-958

59
Figure 1. Eaton Building. ._. „ ._ . ... .... ... ,
Figure 2. Eaton Building - View from
atrium during construction.

11
i /


<a

63m
a V
Яe
i •F

Y
«е- 1 L5 -> X
m
Figure 3. Typical floor plan

60
Friction damper Viscous damper Viscoelastic damper Self-centering friction damper

Figure 6. Comparison of hysteresis loops of different dampers

Figure 7. 3 dimensional analytical model of Eaton building.

61
15000

/ INPUT EQ ENERGY,
RBF

у
10000 - INPUT EQ ENERGY,

то ф
ш 5000 -
FDF
— ENERGY DISSIPATED
BY DAMPERS
/ •>------------------------------------------
----------------- — --------------------------------------------

0-
0 10 20 30 40
Time, seconds
Figure 8. Time histories of energy input and energy dissipated

0.2

с 0.1 -
ф
я0-
E
gф -0.1
ЕС
о -0.2 -

я -0.3
а
20
. 0.3
Q Time, seconds

Figure 9. Time histories of displacements at roof

62
800 -
600 -
400 -
200

e, 0
о
о -200
LL
-400
-600
-800 0.015
-0.025 -0.02 -0.015 Figure 10. Hysteretic-0.01
loop-0.005 0
of a 700kN friction damper in a0.005 0.01 brace
diagonal 0.02
Displacement, m

63
0.02

0.01

I0

— Elastic elongation of brace


-0.02

-0.03
0 10 20 30 40
0.03 Time, seconds

Figure 11. Time history of deformation of damped bracing

64
9 □ FDF bBMF

6
25
о 5
CO
4

1000 2000 3000 4000 5000 6000 7000


Column Axial Force, kN 8000

0
Figure 12. Envelope of column axial force.

65
13th World Conference on Earthquake Engineering
Vancouver, B.C., Canada August 1-6, 2004 Paper No. 1955

PERFORMANCE-BASED DESIGN USING PALL FRICTION


DAMPERS - AN ECONOMICAL DESIGN SOLUTION.

Avtar PALL5 and R. Tina PALL1

SUMMARY

Benefit-cost analysis approach suggests performance-based design for most modern buildings. The
conventional structural systems are highly unlikely to provide adequate performance in the event of a
major earthquake. With the emergence of Pall Friction Dampers, it has now become economically feasible
to design high performance structures. Their low cost and maintenance free characteristics suggest wide
application for new construction as well as for retrofit of existing buildings. Public sectors, private sectors,
developers and developing countries are all benefiting from this technology. They have been used for the
seismic protection of more than 80 major building projects, including the Boeing Commercial Airplane
Factory at Everett, WA - the world's largest building in volume.

INTRODUCTION

During a major earthquake, a large amount of energy is pumped into the building. The manner in which
this energy is consumed in a structure determines the level of damage. The building codes recognize that it
is economically not feasible to reconcile this energy within the elastic capacity of materials. The criterion
stipulated in conventional building codes is to design structures to resist moderate earthquakes without
significant damage and avoid collapse during major earthquakes. The primary emphasis is on life safety.
The reliance for survival is placed on ductility to dissipate energy during inelastic deformations causing
bending, twisting and cracking. This assumes permanent damage, repair costs of which could be
significant. Recent earthquakes have clearly demonstrated that conventional construction, even in
technologically advanced countries, is not immune to destruction.

In modern buildings, avoidance of structural collapse alone is not enough. The cost of non-structural
components and contents is much higher than the cost of the structure itself and must be protected. The
buildings of post disaster importance such as hospitals, telecommunications, police stations etc. must
remain operational. Benefit-cost analysis suggests performance-based design approach. Emphasis for
performance based seismic design is being given in recent standards, e.g., Vision 2000, ATC-57 and
FEMA-356/2000. While global deflections, story drifts, force and demand-capacity ratios are important to
a structural engineer; these have no meaning to a developer. What are important to an owner is how much

5
Pall Dynamics Limited, Montreal, Canada. www.palldynamics.com, E-mail: info@palldynamics.com

66
it will cost to build and how much it will cost to repair the damage, lost rental revenues from a future
earthquake. Early on, the structural engineers should resolve these issues with the owner and develop
performance-based design criteria consistent with the performance goals. In most cases, providing good
seismic performance yields good return on the investment.

If the amount of energy getting into the structure can be controlled and a major portion of the energy can
be dissipated mechanically independent of primary structure, the seismic response of the structure and
damage control potential can be considerably improved. These objectives can be delivered by adopting
new techniques of base isolation and energy dissipation devices. With the introduction of energy
dissipation devices, supplemental damping of 20-30% of critical can be easily achieved (inherent damping
is merely 1-5%). Thus forces exerted on the structure and the amplitudes of vibrations are considerably
reduced. The drifts of the structure can be reduced by a factor of about two to three and by larger factors if
the devices also add stiffness to the structure, refer Article C9.1 of FEMA-356 [27]. The energy dissipation
systems should be considered in somewhat broader context than isolation systems (which are not feasible
in taller buildings) as a design strategy when performance goals include damage control. Also, the
construction cost is less with energy dissipation systems than base isolation systems.

Seismic isolation and energy dissipation systems are relatively new and sophisticated concepts that require
more extensive design and detailed nonlinear time-history dynamic analysis than most conventional
schemes. With the availability of several commercial programs and powerful desktop computers, the
sophisticated analysis is no longer a daunting task and can be easily and quickly done in a small design
office environment. However, the benefits accrued outweigh the extra design expense and resulting
savings more than compensate the cost of dampers.

With the emergence of Pall Friction Dampers[1-3,6], the performance-based design of buildings is now
economically feasible. The first building with seismic dampers in North America was built with Pall
Friction Dampers. Their low cost and maintenance free characteristics suggest wide application for new
construction as well as for retrofit of existing buildings. Public sectors, private sectors and developers are
using and benefiting this technology. They have been used for the seismic protection of more than 80
major building projects, including the Boeing Commercial Airplane Factory at Everett, WA - the world's
largest building in volume. This technology is credited with for saving Boeing more than US$ 30 million.
The City and County of San Francisco chose Pall Friction Dampers for the seismic control of Moscone
West Convention Center and saved US$2.25 million, compared to alternate viscous dampers.

This paper describes the innovative structural system and several applications in the performance-based
design of major projects including cost analysis.

PALL FRICTION DAMPERS


State-of-the-Art
Of all the methods so far available to extract kinetic energy from a moving body, the most widely adopted
is undoubtedly the friction brake. Mechanical engineers have successfully used this concept for centuries
to stop the motion of equipment, automobiles, railway trains, airplanes etc. No other mean has been able to
replace the friction brake. Reason! It is the most effective, reliable and economical mean to dissipate
kinetic energy. Similar to automobiles, the motion of vibrating building can be slowed down by dissipating
seismic energy in friction.

Inspired by the principle of friction brake in mid 1970's, Pall Friction Dampers were pioneered for the
seismic control of buildings. Pall Friction Dampers significantly reduce the initial cost of construction
while dramatically increasing the earthquake resistance against damage.

67
Friction dampers for a building must possess a reliable and stable performance over the life of building.
Developing a reliable friction is very difficult and tricky. Over a period of more than a decade of research
and development, the common problems in friction were successfully overcome by using specially treated
surfaces and a unique manufacturing process. Over the years, Pall Dynamics has earned an international
reputation for excellence and is a world leader in friction dampers for seismic control of buildings.
Pall Friction Dampers have successfully undergone rigorous proof testing in the U.S and Canada. In 1985,
the National Research Council of Canada tested 3-story frame structures on a shaking table at the
University of British Columbia, Vancouver [4]. In 1986-1987, the U.S. National Science Foundation tested
a 9-story frame structure on a shaking table at the University of California at Berkeley [5]. The structures
were subjected to more than 20 different major earthquake records. Even for an earthquake 5 times
stronger than the 1985-Mexico earthquake, the frames equipped with friction dampers remained damage
free. Pall Friction Dampers are well recognized and accepted by the building codes in Canada, the U.S and
many other countries.
Salient Features
Pall Friction Dampers are foolproof in construction. Basically, these consist of series of steel plates, which
are specially treated to develop very reliable friction. These plates are clamped together and allowed to slip
at a predetermined load. Decades of research and testing have led to perfecting the art of friction. Their
performance is reliable, repeatable and they possess large rectangular hysteresis loops with negligible fade.
Their performance is independent of velocity and hence exerts constant force for all future earthquakes,
design-based earthquake (DBE) or maximum credible earthquake (MCE). A much greater quantity of
energy can be dissipated in friction than any other method involving the yielding of steel plates, viscous or
viscoelastic dampers. Therefore, fewer Pall Friction Dampers are required to provide the required amount
of energy dissipation. Pall Friction Dampers are passive energy dissipation devices and, therefore, need no
energy source other than earthquake to operate it. They do not require any repair or replacement after the
earthquake and are always ready to do their job.

Friction Damper for Hystersis Loop Friction Damper in


Tension-Compression Brace Tension-only Cross Brace
Figure 1. Pall Friction Dampers
In a typical undamped structure, the inherent damping is merely 1-5% of critical. With the introduction of
Pall Friction Dampers, structural damping of 20-50% of critical can be easily achieved. As the dampers
dissipate a major portion of the seismic energy, forces and deformations on the structure are significantly
reduced. Pall Friction Dampers significantly reduce the initial cost of construction while dramatically
increasing the earthquake resistance against damage.
Pall Friction Dampers are customized to suit site conditions and allow greater adaptability than is possible
with other systems. These dampers can be bolted or welded into place.

68
Pall Friction Dampers are available for long slender tension-only cross bracing, single diagonal tension-
compression bracing and chevron bracing (Figure 1). The damper for cross bracing is a unique
mechanism. When one of the brace in tension forces the damper to slip, the damper mechanism forces the
other brace to shorten and thus avoid buckling. In this manner, the other brace is immediately ready to slip
the damper on reversal of cycle. These dampers have been used in 65 feet (22 m) long slender bracing.

To avoid pounding at the expansion joints, Pall Friction Connectors are custom made to accommodate bi-
directional movements.

Slip Load of Friction Damper


The friction dampers are designed not to slip during wind. During a major earthquake, they slip prior to
yielding of structural members. In general, the lower bound is about 130% of wind shear and the upper
bound is 75% of the shear at which the members will yield. As seen in Figure 2, if the slip load is very low
or very high, the response is very high. Several parametric studies have shown that the slip load of the
friction damper is the principal variable with the appropriate selection of which it is possible to tune the
response of structure to an optimum value. Optimum slip load gives minimum response. Selection of slip
load should also ensure that after an earthquake, the building returns to its near original alignment under
the spring action of an elastic structure. Studies have also shown that variations up to ±20% of the
optimum slip load do not affect the response significantly. Therefore, small variations in slip load (810%)

(
U
с
l/J

о
Li-

over life of the building do not warrant any adjustments or replacement of friction damper.

WHY PALL FRICTION DAMPERS ARE FAVORED?

• They are simple and foolproof in construction.


• Offer reliable and repeatable performance at low cost.
• Possess large rectangular hysteresis loops. Greater energy dissipation for a given force. Hence, fewer
Pall Friction Dampers are needed. Conversely, exert lesser force for a given damping.
• Provide supplemental damping and stiffness for added stability.
Figure 2. Response versus Slip Load
• Performance is independent of velocity and temperature.
• Constant force for all future earthquakes (DBE/MCE). Therefore, design of connections and members
is economical.
• They are not active during service loads and wind. Hence, no possibility of failure due to fatigue
before an earthquake.
• Need no repair or replacement before and after earthquake. There is nothing to damage or leak.
• Energy dissipation is through friction and not through the damaging process of yielding.

69
• After an earthquake, the building returns to its near original alignment due to spring action of an elastic
structure.
• Compact and narrow enough to be hidden in partitions.
• They can accommodate foundation settlements.
• Available for all types of bracing, including tension cross bracing, and expansion joints.
• Custom made. Easily adaptable to any site condition. Can be welded or bolted

COMPARISON OF HYSTERESIS LOOPS OF DIFFERENT DAMPERS

Yielding Plate Self-centering Viscoelastic


Damper Friction Damper Damper
Figure 3. Comparison of Hysteresis Loops of Different Dampers

COMPARISON WITH OTHER DEVICES

Viscous Damper
Viscous dampers are velocity dependent. The forces exerted by the damper are, therefore, different for
different earthquake records. Friction dampers are independent of velocity, therefore, exert constant force
for all future earthquakes (DBE / MCE). A friction-damped structure is an engineered structure in which
forces exerted are predetermined. This is a great technical and economic advantage over viscous dampers
in which forces are much higher at MCE level.

The hysteresis loop of viscous damper is elliptical compared to rectangular for friction damper. For a
given maximum force, the area of hysteresis loop (energy dissipation or damping) of viscous damper is
about 70% of that for friction damper i.e. 70 friction dampers will achieve the same damping as 100
viscous dampers of a given force. Conversely, for a given number and damping value, the forces exerted
by friction dampers are only 70% of those for viscous dampers. This leads to significant savings in cost of
dampers, bracing, connections, columns and foundations.

While supplemental damping is beneficial in reducing the earthquake forces and amplitudes of vibration,
added stiffness is beneficial for stability. A deflected building is similar to a stooped person. Similar to a
cane support for stooped person, additional stiffness helps the deflected building against overturning. Pall

70
Friction Dampers provide both added damping and added stiffness for stability - a complete structural
solution. Viscous dampers provide only damping and no stiffness - the structure is on its own to struggle
for stability.

Unbonded Brace
Unbonded brace is often called by other names like 'Yielding brace' or 'Buckling-Restrained brace'. It
consists of a slender steel brace (core steel), surrounded by a steel tube or pipe, which is then filled with
concrete or grout. The core steel is wrapped with plastic like material to separate it from the concrete
around. The concrete in steel tube prevents buckling of the steel brace.

Unbonded brace dissipates energy through the process of yielding. Yielding involves damage. No damage,
no energy dissipation or damping. After an earthquake, the brace may be damaged and needs to be
replaced. Replacement of brace after an earthquake is expensive and time consuming. It is too difficult to
replace before imminent aftershocks. Another problem is that the brace is enclosed in concrete and is not
visible for inspection to verify if it is broken or otherwise.

The cost of supply of core steel and wrapping material may appear to be small, but the overall cost of an
installed unbonded brace, including filling of outer tube with concrete and connections, is higher than the
installed friction damper. As friction dampers dissipate a large amount of energy mechanically, the forces
exerted are far less than those exerted by the unbonded braces. Besides, the structure with friction damper
is economical to design and always ready to resist earthquakes one after another without replacement.

DESIGN CRITERIA

The quasi-static design procedure given in most building codes are ductility based and do not explicitly
apply to buildings with supplemental damping. In the past few years, several guidelines on the analysis
and design procedure of passive energy dissipation devices have been developed in the U.S. The latest and
most comprehensive document is the "NEHRP Guidelines for the Seismic Rehabilitation of Buildings,
FEMA 356 / 357, issued in 2000" [27]. This equally applies to new buildings.

The Guidelines require that the structure be evaluated for response to two levels of ground shaking, a
design basis earthquake (DBE) and a maximum considered earthquake (MCE). The DBE is an event with
10% probability of exceedance in 50 years, while the MCE represents a severe ground motion of
probability of 2% in 50 years. Under the DBE, the structure is evaluated to ensure that the strength
demands on structural elements do not exceed their capacities and that the drift in the structure is within
the acceptable limits. For the MCE, the structure is evaluated to determine the maximum displacement and
overstress. It is presumed that if proper ductile detailing has been followed, the structure will have
sufficient reserve to avoid collapse during MCE.

Since different earthquake records, even of the same intensity, give widely varying structural responses,
results obtained using a single record may not be conclusive. Therefore, three time-history records,
suitable for the region should be used; one of which should be preferably site specific. The maximum
response is used for the design.
NEHRP guidelines require that friction dampers are designed for 130% MCE displacements and all
bracing and connections are designed for 130% of damper slip load. Variation in slip load from design
value should not be more that ± 15%.

NONLINEAR TIME-HISTORY DYNAMIC ANALYSIS

The slippage of friction damper in an elastic brace constitutes nonlinearity. Also, the amount of energy
dissipation or equivalent structural damping is proportional to the displacement. Therefore, the design of

71
friction-damped buildings requires the use of nonlinear time-history dynamic analysis. With these
analyses, the time-history response of the structure during and after an earthquake can be accurately
understood. Several nonlinear computer programs are now capable of modeling friction dampers. Some
of these programs are ETABS, SAP2000, DRAIN-TABS, DRAIN-2DX, DRAIN-3DX, ANSYS etc. With
the availability of powerful personal computers, the sophisticated nonlinear time history analysis can be
easily and quickly done in a small design office environment.

The modeling of friction dampers is very simple. Since the hysteresis loop of the damper is similar to the
rectangular loop of an ideal elasto-plastic material, the slip load of the friction damper can be considered
as a fictitious yield force.

PRACTICAL APPLICATIONS

The first building built with seismic dampers in North America was with Pall Friction Dampers (1987).
Pall Friction Dampers are finding increasing application worldwide for new construction as well as for
retrofit of existing buildings, including overhead water tanks. They have been used for the seismic
protection of more than 80 major buildings in Canada, the U.S., China and India. These are hospitals,
telecommunication buildings, educational institutions, police headquarters, defense installations,
convention centers, courthouses, office and residential buildings. Some projects are discussed below:

Seismic Upgrade of Boeing Commercial Airplane Factory, Everrett, WA, USA.

Figure 4. (a). Inside View of Boeing Factory. (b). Pall Friction Dampers

The mammoth Boeing plant, which could contain Disneyland under one roof, is the world's largest
building in volume [25]. It was built in phases from 1968-1991, for the assembly of wide-bodied 747
jetliners - world's largest commercial airplane. The steel frame building is 120 feet (37 m) high with clear
spans of 350 feet (107 m) and covers more than 98 acres.
In 1996, the Boeing engineers considered several seismic upgrade schemes for this structure. They chose
Pall Friction Dampers as they are foolproof in construction and offer reliable maintenance free
performance at low cost. Also, they possess stiffness for added stability. The performance of friction
dampers is independent of velocity, therefore, the forces on the connections remain constant for any future
earthquake. This is an engineered solution. Economy in the design of connections and easy installation of
dampers provided significant savings in construction cost and time.

Several types of Pall Friction Dampers, suitably modified to adapt to the site conditions, were incorporated
in different types of existing bracing. Friction dampers of capacity up to 200,000 lb. (900 KN) and stroke

72
up to 15 inch (380 mm) have been used. Pall Friction Dampers met or exceeded Boeing's stringent
specifications. Work on seismic upgrade was undertaken in 1998 and completed in 2002.

Cost of Retrofit: US$65 million. Savings: US$30 million compared to conventional construction.
Moscone West Convention Center, San Francisco, USA

Figure 5. (a). View of Moscone West (b). Pall Friction Damper

The Moscone West Convention Center is located in downtown San Francisco, between San Andrea's and
Hayward faults, which are about 19 kM in each direction. The four-story steel frame building is 112 feet
(34 m) high with clear spans of 45 to 90 feet (14-28 m) [26]. The US$186 million expansion of the 1.2
million sq. feet (111,700 m2) convention center was California's largest project of 2001.

Bids were invited on two alternate designs, one with viscous dampers and another with Pall Friction
Dampers. The scheme with friction dampers offered a saving of US$2.25 million compared to viscous
dampers. On the recommendations of peer reviewers and two specialist consultants from the University of
California at Berkeley, the City and County of San Francisco chose Pall Friction Dampers for seismic
control of this prestigious building. The construction of the convention center was completed in 2003. Pall
Friction Dampers up to 500,000 lb. (2250 kN) capacity and 9 inch (230 mm) stroke have been used.

The use of friction dampers reduced the story drifts from 2% without dampers to 1% with dampers. About
80% of the earthquake energy has been dissipated by the friction dampers leaving the building with little
damage. According to the project architects and engineers, the use of technology saved a couple of million
dollars in the initial construction costs and will save tens of millions of dollars in lowered repair costs in
the event of an earthquake.

73
Seismic Retrofit of Boeing Development Center, Cafeteria and Auditorium Buildings, Boeing Field,
Seattle, WA, USA

Figure 6. (a). View of Cafeteria, Auditorium and Fitness Building (b). Pall Friction
Damper

There are three 2-story Boeing Development buildings and a 4-story Boeing Cafeteria, Auditorium and
Fitness Center building. These are steel frame buildings, built in 1980's. The foundations are on 70-80 feet
deep friction piles. Some of these buildings were damaged during February 28, 2001 Nisqually Earthquake
of magnitude 6.8. The epicenter was about 20 miles from the building site. Due to liquefaction of soil, the
differential settlements in pile foundations were 4-5 inches. This resulted in substantial non-structural and
structural damage to the Cafeteria and Auditorium building. The structural damage was primarily to rigid
steel bracing. Several bracing in the Cafeteria and Auditorium building buckled or broke.
Several structural schemes were considered for the seismic retrofit of these buildings. Friction dampers
were considered to be an ideal solution as they provide both damping and stiffness. Besides, they can be
easily modified to suit site conditions and designed to accommodate any future foundation settlements.
A total of 350 Pall friction Damper of 100-500 kip slip load and up to 10 inches stroke, were used in the
existing steel bracing. The seismic retrofit was completed in 2002. Savings are estimated to be more than
60% compared to conventional retrofit.

Ambulatory Care Center, Sharp Memorial Hospital, San Diego, California, USA

Figure 7. View of New Ambulatory Care Center


Pall Friction Dampers in Chevron Brace

74
The Ambulatory Care Center complex consists of two buildings of four and five floors, connected in two
places by a pedestrian bridge and an elevator lobby. This facility will provide extensive medical services
to the community including surgery.

The project engineers selected steel-frame design for the cost efficiency, design flexibility and speedy
construction. Moment-resisting frames, in combination with Pall Friction Dampers in steel bracing, were
selected to resist lateral seismic forces. The structure was designed to meet 1997UBC and 1998 California
Code requirements for Seismic Zone 4. In this performance-based design, the Pall Friction Dampers
reduced building drift from 2% to 1%. The savings in construction cost due to reduced forces, more than
offset the cost of dampers. In case of a major seismic event, the dampers reduce building content damage
and increase safety of occupants. This resulted in long-term savings to the client through lower seismic
insurance premiums.

Seismic Retrofit of 3-Million Gallon Reservoir,


Sacramento, California, USA

The Freeport water tower, a distinctive landmark visible


from Interstate-5, was built in 1956 [22]. The steel
reservoir stands about 120 feet high. The supporting
structure consists of 27 steel columns with two levels of
60 feet long tension cross bracing. Of the several
seismic retrofit options, the scheme with Pall Friction
Dampers in tension cross bracing was chosen. When
tension in one of the brace forces the damper to slip, the
damper's mechanism shortens the other brace, thus
preventing buckling.

Due to high damping provided by the Pall Friction


Dampers, the strengthening of columns and foundations
was not necessary. Friction dampers up to 150 kip slip
capacity were used. Seismic retrofit was completed in
1999. Figure 8. 3-Million Gallon Water Tank.
Savings: More than 60% compared to conventional.

Concordia University Library Building, Montreal, Canada

Figure 9. (a). View of Concordia Library Building (b). Pall Friction Damper in Cross Bracing

75
The ten-story McConnell Library Building is a masterpiece in innovative structural design [7]. The
building was designed in 1987 and construction was completed in 1991. Pall Friction Dampers are
installed at the junction of steel cross bracing in rigid concrete frames. The use of steel bracing eliminated
the need of expensive concrete shearwalls and the introduction of supplemental damping provided by
friction dampers eliminated the need of dependence on ductility of structural members.
Unlike concrete shearwalls, the bracing were generally not continuous one over the other and thus
provided greater flexibility in space planning. Since the bracing do not carry any gravity load, they do not
go through the basement to the foundations. Therefore, they allow more space for parking. The architects
have boldly exposed several bracing with dampers as these add to the aesthetic appearance.
The innovative structural system provided an economical design solution to safeguard the building and its
valuable contents from earthquake damage.

Covered Area: 52,000 m2. Cost: $65 million. Savings: 6.5% of structural or 1.5% of building cost.

Seismic Rehabilitation of Justice Headquarters, Ottawa, Canada.

111

fi jj
ill

m
vjaHl
(b). Pall Friction Damper
foWiifi.

Figure 10. (a). View of Justice Headquarters

This eight-story concrete frame building is located in the nation's parliamentary district [21]. It was built
in 1955 as a memorial to the Canadians killed during the Second World War. The stone clad building,
with sloping copper roof, has been designated a heritage structure. The existing structure was not capable
of resisting seismic forces specified in the Building Code. In 1997, seismic rehabilitation was undertaken
along with major renovations to protect the original and new investment.
Pall Friction Dampers offered the best solution for the seismic upgrade. Since the dampers dissipate a
major portion of the seismic energy, forces acting on the structure are considerably reduced. By
staggering the bracing at different story levels, overloading of columns and foundations was avoided.
Hence, expensive and time-consuming work of strengthening existing members and foundations was not
required.
Unlike shearwalls, the friction-damped bracing need not be vertically continuous. This aspect was
particularly appealing to the architects as it offered flexibility in space planning. This structural solution

76
also facilitated construction scheduling as work could start at any floor level, depending on vacancy or
availability.
Covered Area: 50,565 m2. Savings: More than 40% in construction cost and time.

77
Canadian Space Agency Headquarters, St-Hubert, Canada

Figure 11. View of Canadian Space Agency Headquarters

The headquarters of the Canadian Space Agency is a building of national importance [11]. It contains
extremely sensitive and costly equipment / instrumentation. Therefore, it is of vital importance to protect
its valuable contents and electronically stored data in the event of a major earthquake.

The use of Pall Friction Dampers significantly increased the damage control potential of the building
while offering savings in the initial cost of construction. Construction was completed in 1992.

Covered Area: 50,000 m2. Cost $60,000,000. Savings: 1.25% of total cost.

Seismic Rehabilitation of Casino de Montreal, Canada

Figure 12. Casino de Montreal

78
In 1993, Casino de Montreal was housed in the existing French Pavilion built for EXPO'1967 [12]. The
lateral earthquake resistance of the existing eight-story braced steel structure was not adequate to meet the
requirements of the National Building Code of Canada.

Introduction of supplemental damping provided by Pall Friction Dampers was the most effective,
economical and hi-tech solution for the seismic rehabilitation of this building. The use of Pall Friction
Dampers in the existing steel bracing considerably reduced the forces on the structure. Hence, the
provision of additional bracing, strengthening of existing members and pile foundation was not required.

Covered Area: 38,000 m2. Savings: 50% in construction cost & time.

Quebec Provincial Police Headquarters, Montreal, Canada

Figure 13. (a). View of Police Headquarters (b). Pall Friction Dampers in Bracing

The existing sixteen-story office building, with two levels of basement, was built in 1964 [23]. Steel
moment frames and some braced bays provided lateral resistance to the existing structure. A change of
occupancy was planned in 1997 to house the provincial police (Surete du Quebec) headquarters. The
project structural engineers evaluated that the existing structure was not capable of resisting seismic
forces and the story drifts were excessively high, especially at the lover level due to soft story effect.

In 1999, the work on seismic rehabilitation was undertaken along with major renovations to protect the
original and new investment. Pall Friction Dampers offered the best solution for the seismic upgrade.
Since the dampers dissipate a major portion of the seismic energy, the story drifts and forces acting on the
structure are considerably reduced. Hence, expensive and time-consuming work of strengthening existing
member and pile foundations was not required. Pall Friction Dampers were incorporated in existing and
new bracing. The innovative structural scheme offered savings of more than 50% over conventional
retrofit scheme.

CONCLUSION

The use of Pall Friction Dampers has shown to provide a practical, economical and effective approach for
the performance-based design of new and retrofit of existing structures to resist major earthquakes. The
low cost and maintenance free characteristics of Pall Friction Dampers suggest wide application. Public
sector, private sector and developers, including developing countries, are using and benefiting from Pall
Friction Damper technology.

79
REFERENCES

1. Pall, A.S., Marsh, C. "Energy Dissipation in Large Panel Structures Using Limited Slip
Bolted Joints", AICAP/CEB Seismic Conference, Rome, Italy, 1979, 3:27-34.
2. Pall, A.S., Marsh, C. "Friction-Devices to Control Seismic Response", ASCE/EMD Speciality
Conference on Dynamic Response of Structures, Atlanta, USA, 1981, 809-818.
3. Pall, A.S., Marsh, C. "Seismic Response of Friction Damped Braced Frames", Journal of
Structural Division, ASCE, 1982, St. 9, 108:1313-1323. (ASCE "Raymond Reese Research Prize
1983").
4. Filiatrault, A., Cherry, S. "Seismic Tests of Friction-Damped Steel Frames", Third Conference on
Dynamic Response of Structures, ASCE, Los Angeles, USA, 1986.
5. Kelly, J.M., Aiken, I.D., Pall, A.S., "Seismic Response of a Nine-Story Steel Frame with
Friction- Damped Cross-Bracing", Report No. UCB / EERC-88/17. Earthquake Engineering
Research Center, University of California at Berkeley, 1988, 1-7.
6. Pall, A.S., Marsh, C., Fazio, P., "Friction Joints for Seismic Control of Large Panel Structures",
Journal of Prestressed Concrete Institute, 1980, No. 6, 25:38-61.
7. Pall, A.S., Verganelakis, V., Marsh, C., "Friction-Dampers for Seismic Control of Concordia
University Library Building", Fifth Canadian Conference on Earthquake Engineering, Ottawa,
1987,191-200.
8. Pall, A.S., Ghorayeb, F., Pall, R., "Friction-Dampers for Rehabilitation of Ecole Polyvalente at
Sorel, Quebec", Sixth Canadian Conference on Earthquake Engineering, Toronto, 1991, 389-396.
9. Vezina, S., Proulx, P., Pall, R., Pall, A., 'Friction-Dampers for Aseismic Design of Canadian
Space Agency", Tenth World Conference on Earthquake Engg., Madrid, Spain, 1992, 4123-4128.
10. Pall, A.S., Pall, R., "Friction-Dampers Used for Seismic Control of New and Existing Buildings
in Canada", ATC 17-1 Seminar on Base Isolation, Passive Energy Dissipation and Active
Control, San Francisco, USA, 1993a, 2:675-686.
11. Pall, A., Vezina, S., Proulx, Pall, R., "Friction-Dampers for Seismic Control of Canadian Space
Agency Headquarters", Journal Earthquake Spectra, 1993b, Number 3, 9:547-557.
12. Pasquin, C., Pall, A.S., Pall, R., "Hi-Tech Seismic Rehabilitation of Casino de Montreal", ASCE
Structures Congress, Atlanta, USA, 1994, 1292-1297.
13. Godin, D., Poirer, R., Pall, R., Pall, A., "Reinforcement Sismique du Nouveau Campus de l'Ecole
de Technologie Superieure de Montreal", Seventh Canadian Conference on Earthquake
Engineering, Montreal, 1995, 967-974.
14. Hale, T., Tokas, C., Pall, A., "Seismic Retrofit of Elevated Water Tanks at the University of
California at Davis", 7th. Canadian Conference on Earthquake Engg, Montreal, 1995, 959-966.
15. Savard, G., Lalancette, J.R., Pall, R., Pall, A., "High Tech Seismic Design of Maison 1 McGill",
Montreal, Seventh Canadian Conference on Earthquake Engineering, Montreal, 1995, 935-942.
16. Wagner, P., Vavak, L., Pall, R., Pall, A., "Seismic Rehabilitation of the New Hamilton Court
House", Seventh Canadian Conference on Earthquake Engineering, Montreal, 1995, 951-958.
17. Pall, A., Pall, R., "Friction-Dampers for Seismic Control of Buildings - A Canadian Experience",
Eleventh World Conference on Earthquake Engineering, Acapulco, Mexico, 1996, Paper No.
497.
18. Deslaurier, F., Pall, A., Pall, R., "Seismic Rehabilitation of Federal Building, Sherbrooke",
Canadian Society of Civil Engineers Annual Conference, Sherbrooke, 1997, 4:339-348.
19. Pasquin, C., Charania, H., Steele, R., Pall, R., Pall, A.S., "Friction-dampers for Seismic Control
of Selkirk Waterfront Offices, Victoria", Sixth U.S. National Conference on Earthquake
Engineering, Seattle, USA, 1998.

80
20. Pasquin, C., Leboeauf, N., Pall, R., Pall, A., "Seismic Rehabilitation of Hotel Dieu Hospital,
Sainte Hyacinthe, Quebec", Eighth Canadian Conference on Earthquake Engineering,
Vancouver,
1999, 573-578.
21. Balazic, J., Guruswamy, G., Elliot, J., Pall, R., Pall, A., "Seismic Rehabilitation of Justice
Headquarters Building, Ottawa, Canada", Twelfth World Conference on Earthquake Engineering,
Auckland, New Zealand, 2000, Paper No. 2011.
22. Hale, T., Pall, R., "Seismic Upgrade of the Freeport Water Reservoir, Sacramento, California",
Twelfth World Conference on Earthquake Engg., Auckland, New Zealand, 2000, Paper No. 269.
23. Pall, R., Gauthier, G., Delisle, S., Pall, A., "Friction-Dampers for Seismic Upgrade of Quebec
Police Headquarters, Montreal", Twelfth World Conference on Earthquake Engineering,
Auckland, New Zealand, 2000, Paper No. 2014.
24. Chandra, R., Masand, M., Nandi, S., Tripathi, C., Pall, R., Pall, "Friction Dampers for Seismic
Control of La Gardenia Towers South City, Gurgaon, India", Twelfth World Conference on
Earthquake Engineering, Auckland, New Zealand, 2000, Paper No. 2008.
25. Vail, C., Hubbell, J., "Structural Upgrade of the Boeing Commercial Airplane Factory at Everett,
Washington", Proc. Structures Congress & Exposition, May 2003, Seattle, WA, Paper # 000529.
26. Sahai, R., Laws, J., Chen, D., Kong, F., Castillo, F., "Performance Based Design of 4-Story
Moscone Convention Center Expansion Using Steel Couple Girder Moment Resisting Frame and
Friction Dampers", Proceedings, SEAOC Annual Convention, Santa Barbara, California, 2000.
27. Prestandard and Commentary for the Seismic Rehabilitation of Buildings, FEMA 356/ November
FRICTION DAMPERS FOR SEISMIC UPGRADE OF A 14-STORY PATIENT TOWER
WITH A 36-FOOT TALL SOFT-STORY

Dihong Shao, P.E., S.E.


6
Avtar Pall, PH.D. P.E.
7
Bharat Soli, P.E., S.E. 8

ABSTRACT

The 14-story tower is a 1970's concrete shear wall building located in the greater Seattle
area. Between the top of its podium (level-3) and level-5, all interior shear wall cores are
supported by rectangular concrete columns located at corners of these cores; all exterior
concrete shear walls stop at level-5 and are supported by concrete columns. This creates a
2-story soft-story. The lateral resistant system of the soft-story is composed of multi-
strand post-tensioned "X" braces between the adjacent columns that support the interior
shear wall cores.

Studies indicated that such a soft-story would result in service disruptions after a 10%-50
year seismic event. Various seismic upgrade schemes were evaluated. The scheme with
supplemental damping offered substantial cost savings on the foundation work. Long
bracing prompted the use of tension-only cross-braces with friction dampers at the brace

6
Principal, Andersen Bjornstad Kane Jacobs Inc., Seattle, Washington
7
Principal, Pall Dynamics Limited, Montreal, Canada
8
Senior Principal, Andersen Bjornstad Kane Jacobs, Inc., Seattle, Washington

81
intersection. With the implementation of this upgrade, FEMA356 structural immediate-
occupancy performance level is achieved and soft-story story drift is reduced in half.

This paper describes the design implementation and construction of this seismic upgrade.

Introduction

The patient tower, designed and constructed in the 1970s, is a 14-story concrete building with a
rigid 2-story concrete podium. The lateral system of the tower is composed of interior concrete shear wall
cores and curved exterior concrete shear walls. Between the top of the podium (level-3) and level-5, all
interior shear wall cores are transferred to rectangular concrete columns located at corners of these cores;
the curved exterior concrete shear walls stop at level-5 and are supported by sixteen (16) 36" diameter
concrete columns. This creates a 2-story, 36-foot tall soft-story. The lateral resistant system of the soft-
story is composed of multi-strand post- tensioned "X" tension braces between the adjacent columns that
support the interior shear wall cores.

Studies indicated that such a soft-story would result in structural damages and service disruptions
after a 10%-50 year seismic event. Various seismic upgrade schemes were evaluated. The scheme with
supplemental damping offered substantial cost savings on the foundation work. Long bracing prompted
the use of tension-only cross-braces with friction dampers at the brace intersection. Twelve (12) cross-
brace friction damper bays are created in the soft-story by coupling adjacent 36" diameter concrete
columns. FEMA356 structural immediate-occupancy performance level is achieved by providing two (2)
200-kip cross-brace friction dampers at each of the twelve (12) bays. Prototype dampers were designed,
built, and tested to ensure their performance. Damper fabrication was completed and damper installation
is being completed as this paper is being written. This paper describes the use of cross-brace friction
dampers for the seismic upgrade in the 36-foot tall soft-story of this patient tower.

Building Description

The patient tower was designed and constructed in the 1970s. It has been a well-received and
celebrated concrete building in the concrete industry ever since it was built. In fact, it was featured in the
flyer for the ACI Centennial Celebration in 2004 that was widely published in numerous architectural and
engineering magazines (Figure-1).

Figure-1. A Well-Celebrated Concrete Building

82
The tower has a total of fourteen (14) stories, including a 2-story podium. The tower and podium
have an out-to-out dimension of 136'-0" and 238'-0", respectively, in both directions. The typical floor-
to-floor height is 11 '-0" with the exception of floor-to-floor height between Level-3 and Level-5, which
is 18'-0". See Figure-1 and Figure-2.

83
The tower floors are constructed with 15" thick flat concrete slabs without column capitals. The

нЧ

Figure-2. Existing Building

flat slab at each floor is supported by the 10" thick perimeter curved concrete walls and the interior
concrete shear wall cores. The podium floors are constructed with waffle slabs using 4 У" slabs over 14"
deep 4'-0" square pans. The waffle slabs are supported by 24" square concrete columns at 34'-0" on
center each way outside of the tower footprint, sixteen (16) 36" diameter concrete columns that support
the tower perimeter curved concrete walls, and podium interior and perimeter 12" thick concrete shear
walls. Combined with the tower interior shear wall cores this creates a very stiff concrete box podium
between Level-1 and Level-3. All walls and columns are supported by strip or pad concrete footings with
the allowable soil bearing capacity of 12 ksf.

The lateral system of the tower is composed of 10" thick interior concrete shear wall cores and
10" thick curved exterior concrete walls. Between the top of the podium (level-3) and level-5, all interior
shear wall cores are transferred to rectangular concrete columns located at the corners of these cores; the
curved exterior concrete shear walls stop at level-5 and are supported by sixteen (16) 36" diameter
concrete columns. This creates a 2-story 36-foot tall soft-story (Figure-3). The lateral resistant system of
the soft-story is composed of eight (8) bays of multi- strand post-tensioned "X" tension braces between
the adjacent columns that support the interior shear wall cores with four (4) bays in each direction. Each
brace is composed of thirty-six (36) 0.6" diameter post-tensioned strands in three (3) tendons that were
tensioned at 50-kips per tendon with a total post-tension force of 150-kips in each brace. The author
believes that such a unique lateral system at the soft-story was invented with its intent to prolong the
building periods and reduce the seismic forces imposed onto this very heavy concrete tower. This
concept is now widely used with the inventions of various base-isolation systems.

84
Figure-3. Existing Building Soft-Story

Figure-4 (top) shows the entire building lateral system; Figure-4 (middle) shows the full extent of
tower concrete cores from Level-1 to roof along with post-tensioned "X" tension braces at the soft-story
(Level-3 to Level-5); Figure-4 (bottom) shows the entire building lateral system below Level-5.

Seismic Concerns and Upgrade Challenges

Detailed studies indicated that such a soft-story would result in structural damage and service
disruptions after a code-defined 10%-50 year seismic event. The primary concerns are as follows:
Figure-4. Lateral System
In order to minimize post-earthquake damages to the building structure and reduce building service
disruptions after a seismic event, various seismic upgrade schemes to reduce or eliminate the soft-story effects
were studied. They are as follows:

1. Construct shear walls at locations where existing post-tensioned "X" tension braces are located. This
will effectively reduce or eliminate the soft-story effects. However, the building foundation will have
to be strengthened and enlarged. Such foundation work in an occupied building can be very costly and
affect the building's current normal function during construction. Consequently, this scheme was
eliminated.

85
2. Construct steel braces at locations where existing post-tensioned "X" tension braces are located. This
will also effectively reduce the soft-story effects. However, connections between the steel braces and
existing concrete columns will be very difficult to construct due to existing post-tensioned "X" tension
braces. Again, costly foundation work is also unavoidable.
3. Coupling adjacent 36" diameter columns with steel "X" braces. This will result in large size steel
braces due to longer bracing length, which will affect the building appearance. Tension-only braces
were considered and they are not as effective as tension- compression braces. In addition, costly
foundation work is also unavoidable due to increased seismic forces caused by these braces that
shorten the building periods.

Seismic Upgrade With Cross-Brace Friction Dampers

The above-mentioned third scheme led to the consideration of using cross-brace friction dampers
(Figure-5), notice pencil on the damper for scale) for the building seismic upgrade. The main advantages of
using this type of damper are as follows:

Figure-5. Friction Damper In the Lab Friction Dampers Installed

Based on the analyses, a total of 12 cross-brace friction damper bays will be required in the soft-story.
This is achieved by coupling adjacent 36" diameter concrete columns with HSS10x6 tube "X" braces. Two (2)
200-kip cross-brace friction dampers will be required at each of the 12 bays. Dampers will be located at the
intersection of these "X" braces with one on each side of the braces. Dampers were sized to avoid column
footing enlargement and footing uplift under the code defined 10%-50 year seismic event. Dampers were also
sized to reduce the soft- story drift in half.

Two (2) prototype dampers were designed, built, and tested by Pall Dynamics Inc. to ensure their
performance (Figure-5 and Figure-6). Damper fabrication was completed and dampers were delivered to the
site. Damper installation was completed in October of 2005.

Figure -6. Friction Damper Under Testing

86
In order to implement the damper installation and increase the structural integrity, the following
structural components were also added for various structural reasons. They are as follows:
2. All sixteen (16) 36" diameter concrete columns below the soft-story are enlarged to 52" diameter from
the foundation up with 8" thick concrete jackets. This is to avoid the need to enlarge the existing
column footings and to increase column axial load carrying capacity for the increased downward
seismic forces imposed to columns due to added seismic dampers at the soft-story.
3. Drag struts at the top of the building podium (Level-3, which is the bottom of damper braces) are
added to allow positive load transfer between damper braces and rigid-box building podium. See
Figure-7 with comparison to Figure-4 (bottom).
4. 3-D steel trusses are added between the bottom of Level-5 floor slab and the top of damper braces to
allow load transfer between Level-5 floor slab and the damper braces. This is also to avoid requiring
the enlarged 36" diameter columns to transfer all the damper forces with potential shear failures. See
Figure-7 and Figure-8.
5. Concrete shear walls and shear wall footings are added along the building podium exterior to complete
the podium "box". See Figure-7 with comparison to Figure-4 (bottom).

Figure-7. Drag Struts, 3-D Trusses and Damper Braces

Construction of the seismic upgrade is nearly complete. Figure 9 shows the before and after of the
column enlargement for columns below the soft story. Figure 10 shows various stages of the damper
installation including concrete column steel jacket installation and Figure 11 shows dampers and steel braces
installed. The construction of concrete drag struts are well underway as this paper is being written.

87
Figure-8. 3-D Trusses and Damper Braces (One
Quadrant)

Figure-9. Concrete Column Enlargement Before and After

Figure-10. Steel Jacket and Damper Installation

88
Figure-11. Completed Damper Installation Conclusions
This paper presents a unique solution for the seismic upgrade of a 14-story patient tower with a 36-
foot tall soft-story. Cross-brace seismic dampers are utilized at the soft-story and they consequently reduce
the story drift in half. The upgraded the building achieves the FEMA356 structural immediate-occupancy
performance level under code defined 10%-50 year seismic event.

This unique solution has the following advantages:

1. It is very cost effective and resulted in approximately $1 million savings on the foundation work
compared to the conventional concrete shear wall seismic upgrade scheme.
2. It simplifies the construction process with minimum disturbances to an occupied hospital facility
during construction.
3. It is aesthetically acceptable.
4. It adds structural integrity and improves building life safety.
5. It helps minimize post-earthquake structural and non-structural damage and reduce potential down
time and repair costs after a seismic event.

Acknowledgements

The authors would like to acknowledge the teamwork provided by Callison Architecture and Sellen
Construction. The editorial comments provided by Mr. Dennis Baerwald of ABKJ Consulting Engineers are
also greatly appreciated.

89
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_312-1
© Springer-Verlag Berlin Heidelberg 2021

Friction Dampers for Seismic Protections of Steel Buildings Subjected to


Earthquakes: Emphasis on Structural Design
Lucia Tirca9
Department of Building, Civil and Environmental Engineering, Concordia University, Montreal, QC, Canada

Synonyms

Damping; Earthquake; Energy dissipation; Friction dampers; Seismic design; Steel buildings

Introduction
To reduce the seismic demand, researchers have proposed to incorporate supplemental energy dissipation
devices into the structural system of buildings. According to the primary dissipation mechanism, supplemental
energy dissipation devices are grouped into two categories: hysteretic and viscoelastic. Hysteretic devices rely
on the relative displacements of components within the device and are typically based on either metallic
yielding or frictional sliding, while viscoelastic devices are velocity dependent. More specifically, friction
devices dissipate energy through the relative sliding developed between two solid interfaces. Depending on the
type of friction devices, they could be installed in line with single-diagonal or chevron steel braces, at the
intersection of X-bracing system, and in parallel with the beam located at the top of chevron bracing system.
The activation of slip forces that characterize the designed friction dampers occur simultaneously with the
maximum internal forces allowable to develop in the system during the ground motion excitation. The building
reaches the peak interstorey drift when the available slip distance provided by friction damper devices was
consumed. "The forces generated by these devices installed in the structural members are usually in phase with
the internal forces resulting from ground motion shaking" (Christopoullos and Filiatrault 2006).
The total input energy, Ej, induced by a seismic event into a structural system can be expressed as a
summation of kinetic energy, Ek, cumulative strain energy, ES, inherent damping, ED, and the hysteretic
damping Eh of the seismic force resisting system (SFRS). In this study, Eh is the damping induced by friction
devices. The energy balance equation is:

Ej = Ek + Es + ED + Eh (1)

The kinetic and cumulative strain energy are accumulated into the primary structural system and rely on
structural damage (Akiyama 2000; Tirca 2009), while the system is damped by both ED and Eh, which are
amplitude-dependent. In general, the contribution of ED and Eh is related to the amount of post-yielding
response and Eq. 1 can be rearranged as follows:

E k + E s = E j - ( E D + Eh )
(2
)

In Eq. 1, the term (Ek + ES) expresses the vibrational energy, Ev or more specifically the potential damage
energy, while (ED + Eh) is the energy dissipated by viscous damping in the structural members and
supplemental devices. Thus, by adding damping into a structural system the elastic vibration energy

9Email: Lucia.Tirca@concordia.ca

Page 1 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_312-1 ©
Springer-Verlag Berlin Heidelberg 2021

80%
70%

60%

50%
LLI
s- 40%
liu 30%

20%

10%

0%
0% 20% 40% 60% 80% Damping

Fig. 1 Vibrational energy versus supplemental damping

diminishes, while structural members are protected from damage associated with permanent deformations. In
this light, the main design objective is to minimize the difference between the seismic input energy and that
dissipated by the dampers (Christopoullos and Filiatrault 2006). The variation of Ev as a function of
supplemental damping is depicted in Fig. 1.
Pioneering work on friction devices was conducted by Pall (1979) and Pall and Marsh (1981). Since then,
several types of friction dampers have been developed and studied in the literature. These devices differ in their
mechanical shape and materials used for the sliding surfaces.
Pall friction dampers dissipate energy through friction developed by the relative sliding within two surfaces
in contact which are clamped by posttensioned bolts. In order to obtain stable rectangular hysteresis loops
similar to that of Coulomb friction, different types of surface treatment and lining material were studied
experimentally. After investigating the response of slip bolted joints under monotonic loading, Pall reported
that the most stable behavior was obtained when brake lining pads in contact with mill scale surface on plate
was chosen (Fig. 2a). Nevertheless, minor differences between the static and dynamic friction coefficient were
observed. However, under large seismic excitations, the posttensioned bolts of friction dampers may impact
into the end of slotted hole and undergo bearing or even bolt shear failure. The hysteretic behavior of friction
damper shown in Fig. 2b is similar to that of an elasto-perfectly plastic system. Herein, the backbone curve is
composed of four segments: elastic, slipping, bearing, and bolt shear failure. When the demand is higher than
the available slip length which is equal to the length of slotted hole, a sudden increment in storey shear forces
accompanied by decreasing of Coulomb damping is encountered. The hysteresis behavior depicted in Fig. 2c in
terms of slip load versus the slip length, A, shows rectangular symmetrical loops which are largely influenced
by the fluctuation of friction coefficient during the slipping stage. The elastic stiffness, ko, shown in Fig. 2c, is
the stiffness of the attached brace member.
Due to their efficiency (Pall and Pall 2004), Pall friction dampers were employed in more than 40 new and
retrofit buildings in Canada, United States, and India (Christopoullos and Filiatrault 2006). Examples of Pall
friction damper installed in an X-bracing system and single-diagonal brace are showed in Fig. 3.
In the last two decades, the following types of friction devices have been developed: slotted bolted
connections (Fitzgerald et al. 1989; Grigorian et al. 1993; Tremblay 1993), Sumitomo devices developed by
Sumitomo Metal Industries Ltd. of Japan and reported by Aiken and Kelly (1990), energy dissipating restraint
damper (EDR) developed by Flour Daniel Inc. (Nims et al. 1993), friction variable damper developed based on
the EDR damper (Zhou and Peng 2009), friction damper developed by Mualla (2000), etc.

Page 2 of 14
Encyclopedia of Earthquake Engineering

Page 3 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_312-1
© Springer-Verlag Berlin Heidelberg 2021

Fig. 2 Response of slip-bolted joint: (a) monotonic test, (b) back-bone curve, and (c) hysteretic behavior
(After Pall 1979)

Fig. 3 Pall friction dampers manufactured by Pall Dynamics Lmt. Montreal: (a) installed in X-bracing at
Concordia Library, Montreal, and (b) installed in single-diagonal brace at Boeing Commercial Airplane
Factory Everrett, USA (Courtesy Dr. Pall)

The energy dissipated by the Sumitomo friction damper is due to friction generated when friction pads, made
of cooper alloy with graphite plug inserts, slide directly on the inner surface of the outer cylinder. The
precompressed internal spring, incorporated into the device, applies a force that is converted through the action
of inner and outer wedges into a normal force on the friction pads. Sumitomo device can be installed in parallel
to the beam located on top of chevron bracing system or in-line with steel braces. As reported by Aiken and
Kelly (1990), Sumitomo dampers were used in two high-rise buildings in Japan in order to resist small intensity
earthquakes and ground floor vibrations.
The EDR damper is similar to Sumitomo device. It includes the following components: an internal spring,
steel compression wedges, bronze friction wedges, stops at both spring's ends, and the outer steel cylinder. Its
functioning comprising the generating slip force depends on the length of the internal spring.
In addition, the normal force acting on the cylinder wall and the friction force developed between the bronze
friction wedges and the inner surface of the outer cylinder determine the slip force in the device. As noted by

Page 4 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_312-1
© Springer-Verlag Berlin Heidelberg 2021

Nims et al. (1993), the EDR device can produce a wide range of hysteretic behavior such as double-flag shape
and triangular lobed shape. Both types of hysteresis loops are self-centering. By using the same principles,
Zhou and Peng (2009) proposed a new friction variable damper where a sliding shaft and a friction ring were
used to replace the spring and wedges. In addition, two zones with high and low friction coefficient have been
defined in the internal walls of the outer cylinder.
The friction damper proposed by Mualla (2000) was added to connect the top part of chevron braces to the
mid-span of a moment resisting frame beam. This friction damper, patented by DAMPTECH Denmark,
consists of three steel plates that are able to rotate against each other around a pre-stressed bolt which passes
through. Two circular friction pad discs inserted between the aforementioned steel plates provide dry friction
lubrication and ensure stable friction forces. As reported by Liao et al. (2004), a full scale test conducted on a
three-storey moment resisting frame with added chevron bracing system and Mualla's dampers was
investigated. The total weight of the frame was about 34 t. The bracing system consisted of bar members that
were pre-tensioned in order to avoid buckling. Due to the application of lower scaling simulated ground
motions, no damage of structural members was reported during the conducted experimental tests.

Design Provisions
Since 1980, seismic response control techniques have been used as complementary solutions to the existing
SFRSs. Despite of their wide-spread applications, design guidelines for structures equipped with supplemental
energy dissipation devices (hysteretic and viscoelastic) are still in evolving phases. This study refers to the
design philosophy of structures with incorporated friction damper devices (FDD). As mentioned above, FDDs
are hysteretic devices. For example, buckling-restrained braces are also known as hysteretic damper devices
and are designed to dissipate part of vibration energy through yielding of the core plate.
In general, hysteretic devices are installed in braces and are able to undergo large inelastic response, while
they dissipate most of the hysteretic energy. The purpose of installing these devices into the structural system is
to maintain the main structure either elastic or within low inelastic deformations. Both, the main frame and the
supplemental energy dissipation system, share the same deformation, which in turn is that of the entire system.
It is important to assure stable response of these devices under dynamic loading. Thus, the added dampers
installed in new or retrofit buildings should yield or slip before the shear resistance of the main structure is
reached.
In North America, the first design guidelines addressing provisions for passive energy dissipated devices
were introduced in FEMA 356 (2000) and FEMA 450 (2003) that refer to the seismic rehabilitation of
buildings and seismic design for new buildings, respectively. Later on, FEMA 450 was replaced by FEMA-P
750 (2009). The aforementioned design guidelines were incorporated in Chapter 14 of ASCE/SEI 41-13
(Seismic evaluation and retrofit of existing buildings), as well as in Chapter 18 of ASCE/SEI 7-10 standard. It
is noted that Commentary J of NBCC 2010 (National Building Code of Canada) includes two clauses in regard
to seismic design with supplemental energy dissipated devices. However, both CSA/S16-2009 and ANSI/AISC
360-2010 standard provide design regulations for the "ductile buckling-restrained braced frames" system.
In Europe, the Eurocode 8 part 1 (CEN 2004) contains Chapter 10 entitled Base isolation, but this chapter
does not cover regulations for passive energy dissipation devices installed into several storeys of building's
structure.
The aforementioned guidelines specify general requirements, analysis procedures, required testing
program, etc. A brief review of these guidelines with highlights on friction damper devices design is
presented below.

FEMA P-750 and ASCE/SEI 7-10 Provisions

Page 5 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_312-1
© Springer-Verlag Berlin Heidelberg 2021

Design Philosophy
General requirements:

- Structures with a damping system have a seismic force resisting system that provides a complete load path.
The base shear used to design the employed SFRS shall not be less than V min, where Vmin is the grater of:
Vmin = V/Bv+1 and Vmin = 0.75 V.

Herein, Vis the seismic base shear in the direction of calculation determined in accordance with the code
procedure corresponding to the selected SFRS and Bv+1 is the numerical coefficient set for effective damping
reduction factor that is equal to the damping provided by supplemental dampers in addition to the inherent
damping (2 % for steel structures). For irregular building structures and for damping system that has less than
four damper devices per floor disposed in the direction of loading, the minimum base shear will not be less
than 1.0 V.
- Damping system may be used in addition to the SFRS in order to meet interstorey drift limits and it may be
located external or internal to the structure, while it shares or not the members of SFRS. In Fig. 4 is illustrated
a SFRS consisting of one bay moment resisting frame (MRF) and a damping system composed of diagonal
braces with installed FDs. The FD system is located independently to the MRF system.
- Energy dissipating devices must be tested before installation for displacement and slip force corresponding
to seismic demand that is required by code.
- Linear static and response spectrum analysis methods for design are accepted for regular structures that
have at least four damper devices per each floor located in one principal direction of the building. Like MRF
or braced frame structures, damped structures yield during the design ground motions, while they need to
behave elastically under the wind loads.
- For detailed design and seismic response investigations, nonlinear dynamic time-history analyses are
recommended.

Fig. 4 Configuration of SFRS and damping system (no shared


elements)
- Dampers are modeled to perform in the plastic range. The computer model should be calibrated based on
experimental test results. For designs in which the SFRS is expected to yield, the postyield behavior of the
structural elements must be modeled explicitly.

In the ASCE/SEI 7-10 provisions, damping system is defined as being "the collection of structural elements
that includes all the individual damping devices, all structural elements or bracing required to transfer forces
from damping devices to the base of the structure, and the structural elements required to transfer forces from
damping devices to the seismic force-resisting system." For details of design procedure, the reader is referred to
the ASCE/SEI 7-10 document. It is noted that other design methodologies can be applied to design structures
equipped with friction dampers.

Design of Structures Equipped With Pall Friction Dampers

Page 6 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_312-1
© Springer-Verlag Berlin Heidelberg 2021

As noted above, metallic dampers and friction dampers are hysteretic dampers. The design of structures
equipped with metallic dampers or friction dampers is similar. The difference consists of changing the yield
load corresponding to metallic dampers with the slip load of friction dampers. As mentioned above, friction
dampers are installed in braces and are designed not to slip under the wind forces. On the other hand, in order
to increase the building performance under earthquake loads, systems composed of friction- damped braces are
added to the MRF structures, while the MRF system should posses enough strength, stiffness, and ductility.
Furthermore, all members of the SFRS and those attached to FDs should be designed to carry the internal
forces developed when slip forces are activated. Thus, the application of capacity design checks to structural
members is required. According to Christopoullos and Filiatrault (2006) the design of structures equipped with
friction dampers can be divided in four steps:

• Calculate the demanded slip force and size the in-line brace by employing simplified methods (e.g., the
equivalent static force procedure). Then, the demanded peak interstorey drift (e.g. linear dynamic analysis
methods) should be evaluated.
• Optimize the design of friction dampers and adjacent members.
• Apply the capacity design and size all structural members to carry the designed slip forces.
• Use the nonlinear time-history dynamic analysis to check the design performance of building structure under
an ensemble of selected ground motions scaled to match the design response spectrum corresponding to 2 %
probability of exceedance in 50 years (NBCC 2010).

A review of different design procedures from the literature is discussed below. However, in all design
procedures, the building structure is composed of a bare frame system and a supplemental damping system,
whilst its response is analyzed before and after the friction dampers are activated.

Design Procedures from the Literature


The first design procedure for MRF system equipped with friction damped braces was proposed by Baktash
and Marsh (1986). They considered a single storey MRF structure with friction damped braced bays. Based on
this procedure, the slip force is activated when yielding in the moment resisting frame brace is reached. For a
single bay MRF frame that was designed based on the principle "strong column weak beam", the shear force
resisted by the frame action, Vs is related to the plastic moment capacity of the MRF beam, Mp. Herein, Mp =
Vsh/2, where h is the storey height. Thus, the proposed design method is based on the assumption that the
lateral shear force leading the damper in-line with brace to slip must be equal to the shear force causing the
MRF to yield and is expressed by the following equation:

Vs = 2Mp =h (3)
The corresponding shear deflection of the storey, As is:
As = (Vt - V br ) / k u (4)
where Vt is the total shear force exerted by the frame and braces, Vbr is the total shear force exerted by braces
alone, and ku is the lateral stiffness of the bare MRF.
The hysteretic energy dissipated by friction devices Ef is:

E f = V br As = V br ( V t - V br ) / k u (5)

By differentiating Eq. 5 with respect to Vbr and setting it equal to zero, the maximum energy dissipated by
friction devices is obtained:

d E f / d V br = V t / k u - 2 V br / k u = 0 (6)

From Eq. 6, the shear force exerted by braces is:

Page 7 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_312-1
© Springer-Verlag Berlin Heidelberg 2021

Vbr = 0.5 Vt (7)


Thus, to maximize the energy dissipated by friction devices, the total shear force must be equally shared
between friction devices and the bare frame (MRF). In addition, it was reported that the optimum slip force of
friction devices installed in structure depends on the structural characteristics only and not on the ground
motion intensity (Baktash and Marsh 1987).
Filiatrault and Cherry (1988) proposed to determine the optimum activation of friction devices based on
minimizing the Relative Performance Index, RPI, devised from the application of the energy concept.

RPI = V2[SEA/SEAo + Umax / Umax0 ]


(8
)
where SEA is the strain energy area of all structural members of a friction damped system, SEA0 is the strain
energy area corresponding to a zero activation force, Umax is the maximum strain energy stored in all structural
members of a friction damped system, and Umax0 is the maximum strain energy for a zero slip force. The values
resulted for the RPI yield to three cases: (i) RPI = 1 corresponds to the response of a bare frame structure; (ii)
RPI < 1 corresponds to the response of a damped structure which is smaller than the response of the bare frame
structure; (iii) RPI > 1 corresponds to the response of a damped structure which is larger than the response of
the bare frame structure. Authors recommended the selection of diagonal braces to comply to Tb/Tu < 0.4,
where Tb and Tu are the fundamental period of the braced frame structure and bare frame structure,
respectively. Herein, the fundamental periods Tb and Tu can be expressed as Tb = 2n/ob and Tu = 2p/ou, where ob
is the natural circular frequency of the fully braced frame before dampers are activated, ob = (kb/m)05 and ou is
the natural circular frequency of bare frame, ou = (ku/m)05. In addition, kb and ku are the lateral stiffness of the
braced frame structure and bare frame structure, respectively. In these expressions, m is the total seismic mass
of the system. Later on, Filiatrault and Cherry (1990) proposed the following equation to calculate the total
shear force, Vo that is required to activate all friction damper devices in a structure:

V o/ W = ( a g /g) x Q { T b / T g , T b / T u, N f) (9)

where Nf is the number of floors, ag is the design peak ground acceleration, g is the gravity acceleration, Tg is

the predominant period of the selected design ground motion, and Q is an unknown single valued function that

is given in Eq. 10a for 0 < Tg Tu < 1 and Eq. 10b for Tg Tu > 1.

Q = ( T g / T u ) x [(-1.24Nf - 0.31)Tb /Tu + 1.04Nf + 0.43] (10a)

Q = (Tb /Tu) x [(0.01Nf + 0.02)Tg /Tu - 1.25Nf - 0.32] + (Tg /Tu) x (0.002 - 0.002Nf )
+ 1.04Nf + 0.42 (10b)

Equations 9 and 10a or 10b can be used directly to calculate the total base shear force V0. It is noted that for a
single diagonal brace located at floor i, the slip force is Pi and the activation shear at that floor is Vi = Pi cosdi,
where di is the angle of brace inclination reported to a horizontal line.
In the meantime, studies on hysteretic bracing systems were carried out by Ciampi et al. (1990, 1992, 1995,
etc.) who considered that both MRF and bracing system are arranged in parallel. They concluded that the
activation of friction damper devices should occur before yielding of moment resisting frame. In addition,
Ciampi et al. (1995) referred to four key parameters such as:

Page 8 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_312-1
© Springer-Verlag Berlin Heidelberg 2021

• Tf, which is the fundamental period of the bare moment resisting frame. A deductable parameter is a = Tb /Tf
where Tb is the fundamental period of the bracing system.
• l = kb/kf which is the ratio between the stiffness of bracing system and the stiffness of the bare moment
resisting frame.
• fi = dby/df y which is the ratio between the displacement that cause yielding of bracing system and that that
causes yielding of bare MRF system.
• Vf = Ffy /mag which is the yield strength of the frame normalized to the mass of the structure multiplied by
the design peak ground acceleration. Similarly, it can be defined vb = Fby/mag, where Fby is the yield force of
bracing system. The total normalized force is Vn = Vf + Vb.

The selection of the above parameters suggests the variation effect of fi and l on the building response. When
fi = 1, bracing system and MRF system yield for the same lateral displacement. When fi = 0, the structural
system corresponds to the MRF system only. They recommended an optimal value offi around 0.5.
The application of this design methodology to MDOF systems requires uniform distribution of stiffness and
slip forces along the building height, aiming at uniform activation of friction devices in order to avoid damage
concentration within a specific floor. In addition, they proposed to maintain at each storey the stiffness of
braces proportional to the bare frame stiffness. Also, they recommended keeping proportionality between the
horizontal projection of slip forces per floor and the corresponding static or dynamic distributed storey shear
over the building height. In addition, two design assumptions were made: (i) the MRF system is designed to
respond elastically while the friction damper devices behave in the non-linear range and (ii) the MRF system
may yield when dy was reached, therefore after friction damper devices were activated. The tri-linear force-
displacement constitutive low is depicted in Fig. 5.
Fu and Cherry (1999, 2000) proposed a simplified design method that leads to a code compatible procedure
for a friction-damped steel bracedframe system. First, they developed the method for a SDOF system based on
establishing an equivalent ductility-related force modification factor R, which reflects the capacity of the
system to dissipate energy. They have developed a trilinear model similar with that illustrated in Fig. 5, where
V parameter was replaced by the restoring force of a SDOF system, f ( t ) . In addition, Vi and Vu parameters
were replaced by f and f, where f is the slip restoring force (after dampers were activated) and f y is the yielding
restoring force developed when f ( t ) exceeds f y and the frame members yield. Thus, to design the friction-
damped steel braced frame system, the design base shear is

Fig. 5 Trilinear force-displacement constitutive


low (After Ciampi et al. 1995)

first calculated based on the code procedure (e.g. equivalent static force procedure). In this case, the value
assigned to the R factor (Rd in the current NBCC code) is evaluated based on the selected system parameters.
Then, all members are designed based on forces resulted from the distribution of design base shear over the
building height, and the brace stiffness and slip forces are evaluated at each floor. In this study, friction-damped
braces were installed in a ductile steel MRF system. The response of building designed based on the employed

Page 9 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_312-1
© Springer-Verlag Berlin Heidelberg 2021

R factor should be verified such that the demanded lateral frame deflection under the wind and earthquake load
to be maintained below the code limits. For example, according to the NBCC 2010 requirements, the system
must respond elastically under the wind load, while the interstorey drift should not exceed hs /500. Under
earthquake loads, the maximum interstorey drift should be less than 2.5%hs for a building classed under the
normal category. Herein, hs is the storey's height.
Tirca (2009) proposed a design methodology to upgrade the seismic resistance of an existing steel moment
frame building designed before 1970. The method complies with FEMA 356 (2000) provisions. First, the
available elastic base shear, Vf, provided by the existing steel moment frame system was evaluated by using
modal response spectrum method and a three-dimensional building model. Then, the required base shear
carried by friction-damped braces Vbr was computed as being the difference between the code demand, V, and
Vf. Herein, V is the minimum lateral earthquake force computed based on NBCC provisions by setting Rd = 1.0.
The effect of torsion was included. By considering the nonlinear behavior of friction dampers, the resulted base
shear assigned to friction-damped braces, Vbr was divided by Bv+1 as per FEMA 356 and then was distributed
over the structure height. According to FEMA 356 recommendation, minimum four friction-damped braces
displaced in the direction of loading were installed in each floor. The slip force of each device located at storey
i was computed by divided the corresponding storey shear force resulted from a dynamic analysis to the
number of friction-damped braces and to the cosdh where в{ was defined above. All steel braces were designed
to behave elastically while sustaining the nonlinear response of attached friction damper. Thus, braces were
sized such that 130%Cr to be larger than the slip force, whereas Cr is the brace compressive resistance. To
preserve the existing columns of MRF system in elastic range, it was proposed to stagger the installation of
friction- damped braces such that to minimize torsion. To prevent the concentration of damage at specific
storeys, it was proposed to maintain a constant ratio between the horizontal projection of slip forces and
dynamic distributed storey shear over the building height. However, the available slip length should be
recommended in the design phase in order to avoid sudden failure. The displacement at yield, Dy, depends on
the stiffness of friction-damped braces, kbd, as shown in Fig. 2c. As noted above, the stiffness of the

Page 10 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_312-1 ©
Springer-Verlag Berlin Heidelberg 2021

Nonlinear и Differential Formulation of the Hysteresis Variable z =


restoring force it[A0 — \z\ i/j(u,z)]
n

\f s & z ) FCt) I
Mapping Function: f s ( u , z ) = ak0u + (1 —
a)k0z

Fig. 6 Mechanical model of a SDOF system

frame and the stiffness of friction-damped braces are in parallel (kf + kbd), whereas Dy = Fs/(kf + kbd) and Fs is
the slip restoring force. In this study, the seismic response of retrofitted building equipped with friction-
damped braces was verified to remain within the interstorey drift code limits and forces developed in structural
members to be below the members' resistance. It was concluded that friction-damped braces installed in
moment frame buildings are able to control the interstorey drift and floor acceleration.

Numerical Modeling of Friction-Damped Braces


In general, researchers select the bilinear force-deformation model for simulating the behavior of friction
damper devices. When performing nonlinear time-history analyses, convergence problems may arise due to the
sharp transitions from the elastic to inelastic stages during the loading, unloading, and reloading cycles. Due to
the large number of friction devices installed in a building structure, the bilinear model "can become
computationally inefficient" when these devices are in different phases of stiffness transition (Moreschi and
Singh 2003). To overpass this modeling drawback, the Bouc-Wen model is recommended. In addition, in the
case of Bouc-Wen model, it is the same differential equation that governs its response (Eq. 11) and more
specifically its behavior during different transition stages. Therefore, the Bouc-Wen model is able to simulate
the highly nonlinear Coulomb friction and has the ability to represent different hysteresis shapes according to
the values of the parameters involved (Morales Ramirez 2011). Since the desired shape of the Coulomb dry
friction law is symmetric and strength and stiffness degradation is neglected, the Bouc-Wen model is reduced
to a nonlinear restoring force (Eq. 11) of a SDOF system shown in Fig. 6. The evolutionary variable z, given in
Eq. 11, is defined in Eq. 12.

f s ( u , z ) = akou + (1 - a ) k o z (11)

(12)

In the above equations, a is the participation ratio of the initial


stiffness in the nonlinear response, ko is the initial stiffness of the system, u is the displacement of the SDOF
system, and z is the hysteresis variable. In Eq. 12, g and b are parameters controlling the shape of the hysteresis
cycle and the exponent n influences the sharpness of the model in the transition zones. The remaining
parameters A, v, and V control the degradation process in stiffness and strength. When the degradation process
is neglected, the aforementioned parameters are A = Ao, v = 1, and V = 1. The considered SDOF system is
characterized by the restoring force fs(du/dt, z) that has a linear and a nonlinear component, as defined in Eq.
11. Using n = 1 might yield to a flexible behavior, while increasing the period of vibration and reducing the
inertial forces. The smooth transition toward a rectangular hysteresis shape is obtained for a large value of
variable n which might become computationally expensive, because the transient analysis requires the
calculation

Page 11 of14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_312-1
© Springer-Verlag Berlin Heidelberg 2021

a of the evolutionary
b Ps + fd
s

kbr

Bearing No
Slipping Ps slipping

No uy u
u
a + uy
slipping
Ps Slipping variable z at each
step of time for a
Bearing
-(Ps + fd) single element. By
Fig. 7 Friction-damped brace: (a) Schematic model and (b) Hysteresis model plus the bearing state
using n = 10, it gives
an acceptable level of
prediction because the difference is quickly reduced throughout the evolution of the slipping stage.
To simulate the slip and slip-lock phases of friction-damped brace device in OpenSees (McKenna and
Fenves 2004), Morales Ramirez (2011) proposed a novel equivalent uniaxial material composed of BoucWen
material (BW) in parallel with gap springs made of Elastic-Perfectly Plastic Gap material (EPPGap), which are
already available in the OpenSees library (Mazzoni et al. 2006). The friction damper was arranged in series
with the brace member made of steel material. The brace was modeled with a truss element. The schematic
representation of friction-damped brace device is shown in Fig. 7a. In the illustrated model, BoucWen material
was selected to replicate the smooth hysteresis behavior of friction damper (stick-slip and slip phase), which
was activated when the axial stress ss generated by the slip force Ps was reached (ss = Ps/Abr and Abr is the gross
area of the attached brace). Each friction damper is characterized by its slip force and slip length that are
resulted from calculation. However, under strong ground motions, the provided slip length, ua, may be limited
and the pretensioned edge bolt may drive the friction damper into the slip-lock phase as depicted in Fig. 7b
(Morales Ramirez and Tirca 2012). Herein, uy is the elastic axial deformation of the brace. To simulate the slip-
lock phase, one equivalent EPPGap spring made of three bilinear gap springs arrange in parallel each other are
defined to act in tension and another set in compression. Each uniaxial ElasticPPGap material (OpenSees
notation) has a defined stress-strain or force-deformation relationship either in tension or in compression. Thus,
one equivalent EPPGap spring is activated when the displacement demand exceeds the available slip length,
+ua, and the other is activated when the -ua slip length is reached. Once activated, this component is able to
limit the displacement and to increase the force experienced by the friction damper. The threshold force of
these gap elements is related to the maximum force that the device is able to withstand (Ps + fd) without
reaching failure (Fig. 7b). To control the failure phase, the MinMax material was set to decouple the friction
device when the strain of the ElasticPPGap material exceeds the predefined bounds either in tension or
compression. In addition, when the MinMax material is activated (e.g., t = t) the device is decoupled from t = ti
until the end of analysis.
Detailed design examples of 4-, 8-, and 12-storey MRF building equipped with Pall friction dampers
installed in-line with braces are given in Morales Ramirez (2011) reference. The building is located in
Montreal, Canada, and was subjected to ten simulated and historical ground motions scaled to match the design
response spectrum over the period of interest 0.2 T1-1.5 T1, where T1 is the fundamental period of the building.
All analyses were performed using OpenSees.
To evaluate the seismic response of structures equipped with friction-damped braces, the following
parameters should be investigated: interstorey drift, lateral deflection, and residual interstorey drift.

Page 12 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_312-1
© Springer-Verlag Berlin Heidelberg 2021

For an accurate OpenSees model of friction-damped brace, it is required to calibrate the model against
experimental test results. Presently, experimental tests conducted on Pall friction dampers installed in-line with
braces are carried out and the obtained results will be disseminated. Furthermore, the development of a code-
based design method for structures equipped with friction-damped braces is required.

Summary
Since 1980, seismic response control techniques have been used as complementary solutions to the existing
seismic force resisting systems and several types of friction damper devices were developed. Among them, Pall
friction damper devices are the most popular. However, despite of their wide-spread applications, design
guidelines for structures equipped with supplemental energy dissipation devices are still in an evolving phase.
This study refers to the design philosophy of structures with incorporated friction dampers that in most cases
are installed in-line with steel braces. A brief review of design guidelines and other seismic design provisions
that emphasize on the design of friction-damped braces is presented. In addition, a brief discussion on the
numerical modeling of friction-damped braces was conducted.

Cross-References
► Nonlinear Dynamic Seismic Analysis

References
Aiken ID, Kelly JM (1990) Earthquake simulator testing and analytical studies of two energy-absorving
systems for multistory structures. Report No UBC/EERC-90/03, Earthquake Engineering Rresearch Center,
University of California, Berkeley Akiyama H (2000) Evaluation of fractural mode of failure in steel structures
following Kobe lessons.
J Construct Steel Res 55:211-227 American Society of Civil Engineers (2010) Minimum design loads for
buildings and other structures ASCE Reston. SEI 7-10
American Society of Civil Engineers (2013) Seismic evaluation and retrofit of existing buildings ASCE Reston.
SEI 41-13
Baktash P, Marsh C (1986) Seismic behavior of friction damped braced frames. In: Proceedings of the 3rd
U.S. National Conference on Earthquake Engineering, Charleston, pp 1099-1105 Baktash P, Marsh C (1987)
Damped moment-resistant braced frames: A comparative study. Can J Civil Eng 14:342-346
CEN (2004) Eurocode 8: Design provisions for earthquake resistance structures. ENV 1998-2. Brussels,
Begium
Christopoullos C, Filiatrault A (2006) Passive supplemental damping and seismic isolation. Instituto
Universitario di Studi Superiori di Pavia, IUSS Press, Pavia Ciampi V, Ferretti A (1990) Energy dissipation
in buildings using special bracing systems. In: Proceedings of the 9th European conference on earthquake
engineering, vol 3. Moscow, pp 9-18 Ciampi V, Paolone A, De Angelis M (1992) On the seismic design of
dissipative bracing. In: 10th world conference on earthquake engineering, vol 7. Madrid, pp 4133-4138
Ciampi V, De Angelis M, Paolone F (1995) Design of yielding or friction-based dissipative bracing for
seismic protection of buildings. Eng Struct 17:381-391 Federal Emergency Management Agency (FEMA)
(2000) Prestandard and commentary for the seismic

Page 13 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_312-1
© Springer-Verlag Berlin Heidelberg 2021

rehabilitation of buildings. FEMA-356, Washington, D.C. Federal Emergency Management Agency (FEMA)
(2003) NEHRP Recommended seismic provisions for
new buildings and other structures. FEMA-450, Washington, DC Federal Emergency Management Agency
(FEMA) (2009) NEHRP recommended seismic provisions for
new buildings and other structures. FEMA-P-750, Washington, DC Filiatrault A, Cherry S (1988) Seismic
design of friction damped braced steel plane frames by energy methods. Earthquake Engineering Research
Laboratory report UBC-EERL-88-01. Department of Civil Engineering, University of British Columbia,
Vancouver Filiatrault A, Cherry S (1990) Seismic design spectra for friction damped structures. ASCE J
Struct Eng 116(5):1334-1355
Fitzgerald TF, Anagnos T, Goodson M, Zsutty T (1989) Slotted bolted connections in a seismic design of
concentrically braced connections. Earthquake Spectra, EERI 5(2):383-91 Fu Y, Cherry S (1999) Simplified
seismic code design procedure for friction damped steel frames.
Canadian J Civil Eng 26:55-71 Fu Y, Cherry S (2000) Design of friction damped structures using lateral force
procedure. Earthq Eng Struct Dyn 29:989-1010
Grigorian CE, Yang TS, Popov EP (1993) Slotted bolted connection energy dissipators. Earthq Spectra
9(3):491-504
Liao WI, Mualla I, Loh CH (2004) Shaking table test of a friction-damped frame structure in. Struct Des Tall
Spec 13:45-54
Mazzoni S, McKenna F, Scott M, Fenves G (2006) Opensees command language manual. PEER,
University of California, Berkeley McKenna F, Fenves GL (2004) Open system for earthquake engineering
simulation (OpenSees), PEER,
University of California, Berkeley Morales Ramirez JD (2011) Numerical simulations of steel frames
equipped with friction-damped diagonal-bracing devices. MASc thesis, Building, Civil and Environmental
Engineering, Concordia University, Montreal
Morales Ramirez JD, Tirca L (2012) Numerical simulation and design of friction-damped steel frame
structures. In: The 15th world conference in earthquake engineering, Lisbon, paper #2538 Moreschi LM,
Singh MP (2003) Design of yielding metallic and friction dampers for optimal seismic
performance. Earthq Eng Struct Dyn 32:1291-1311 Mualla IH (2000) Experimental and computational
evaluation of a novel friction damper device. PhD
thesis, Department of Structural Engineering and Materials, Technical University of Denmark National
Research Council of Canada (2010) National building code of Canada. National Research
Council of Canada, Ottawa Nims D, Ritcher R, Bachman R (1993) The use of the energy dissipating restraint
for seismic hazard
mitigation. Earthq Spectra 9(3):467-489 Pall A (1979) Limited slip bolted joints, a device to control the
seismic response of large panel structures.
PhD thesis, Building, Civil and Environmental Engineering, Concordia University, Montreal Pall A, Marsh C
(1981) Response of friction damped braced frames. J Struct Div ASCE 108(ST6):1313-1323
Pall A, Pall R (2004) Performance based design using Pall friction dampers - An economical design
solution. In: 13th world conference on earthquake engineering, Vancouver, paper #1955 Tirca L (2009) A
simple approach for the seismic retrofit of buildings with staggered friction dampers. In: Mazzolani FM (ed)
Proceedings of the first international conference on: protection of historical buildings. CRC Press, Italy, pp 761-
768 Tremblay R (1993) Seismic behavior and design of friction concentrically braced frames for steel
buildings. PhD thesis, University of British Columbia, Vancouver Zhou X, Peng L (2009) A new type
of damper with friction. Earthq Eng Eng Vibrat 8(4):507-520

Page 14 of 14
THE USE OF FRICTION DAMPERS FOR SESIMIC RETROFIT OF THE MONTEREY COUNTY
GOVERNMENT CENTER

Carlos Y.L. Chang, S.E.10, Avtar Pall, Ph.D., P.Eng.11, and Jason J.C. Louie, S.E.1

ABSTRACT
Built in 1966, the existing North Wing of the Monterey County Government Center is a 3-story
steel structure with one basement level. It is a free-standing structure with a gross area of 99,000 sf.
The North Wing houses eleven courtrooms, as well as offices and support facilities.

The building's steel-framed system and precast concrete panels did not satisfy current building code
seismic requirements. The framing, the panels and their connections would most likely suffer
severe damage in a major seismic event.

The building is being retrofitted to alleviate structural deficiencies and meet life safety performance
levels required by FEMA 356 guidelines. The retrofit is taking place while the courtrooms at the
third level remain operational.

This paper describes the application of friction dampers and their impact on the seismic
performance of the retrofitted building.

Introduction

This project started as a simple tenant improvement project without any seismic retrofit. In 2003, the
California Court Building Seismic Assessment Program of Trial Court Facilities Act of 2002 (SB 1732)
mandated a seismic evaluation of the building using the ASCE 31-02 guidelines. Based on the evaluation
results, the building was assigned a seismic risk level of 5. The evaluation summary is presented in Figure 1
and the seismic risk level definitions are presented in Figure 2.

Under the Trial Court Facilities Act, Monterey County was required to seismically retrofit the building to
a seismic risk level of 4 before transferring ownership and management responsibility to the State of California.
As a result, the project scope expanded to include seismic strengthening of the building.

This building is the first court building to move forward with a seismic retrofit design and construction
under the state's Seismic Assessment Program. The County's project manager is Nova Partners. The architect is
HOK San Francisco and the general contractor is Skanska USA.

10
Associate Principal, Middlebrook + Louie, San Francisco, CA. www.mplusl.com
11
Principal, Pall Dynamics Limited, Montreal, Canada. www.palldynamics.com

1
RISKLEVEL ASPECT ANTICIPATED RESULTS
, California Court Building Seismic Assessment Program I Tier 1 Evaluation Summary Sheet Explanation of Tier 2 Objective:
Building ID: 27-A1 By/Firm: Kenneth T. Tarn/ Simpson Gumpertz & Date: 3 September
None recommended
__________ Heger Inc. _______________________ _2003 _______________
Building: Potentially no structural damage: repairable, if any,
Bldg. Name: Salinas Courthouse- North Wing Page: 4 of 19
Negligible non-structural damage: repairable.
Risk to Life: Negligible.
a result of frame drift. Some cracking of panels at connections is evident, currently.
Systems: All systems will probably remain operational. Occupancy:
Immediate, with only negligible disruption during clean-up.
Damage to main building could include extensive cracking of some exterior panels, failure of some panels, with
hazards created for pedestrians and severe damage to the frame. (V) The series of 2-story bridges which span to 8uilding: Negligible structural damage: repairable.
adjacent structures are of nonductile concrete frame construction and are judged to be potentially hazardous. Minor non-structural damage: repairable, Risk to Life:
Retrofit, replacement, and or demolition of these bridge structures should be considered. Retrofit could include Negligible. Systems: Minor disruptions for hours to days. Occupancy:
provisions of bracing, wall elements, or, potentially, confinement. Minor disruptions, return within hours.

Building: Minor structural damage: repairable.


DSA Seismic Risk Level (Tier 1): []l П11 El111 П^ DVI D^I
Moderate non-structural damage: extensive repair.
RECOMMENDATION ^ No Further Study, Assign Risk Level From Tier 1 Risk to Life: Minor
Due to overwhelming number of deficiencies related to moment frames, we do not anticipate Tier-2 Systems: Disruption of systems for days to months. Occupancy: Return within
evaluation to improve the rating of this building. □ Perform Tier 2 Evaluation (Check applicable box weeks, with minor disruptions.
below)
□ Risk Level Can Be Refined Building: Moderate structural damage: substantial repair.
□ Retrofit Concept Can be Refined Substantial non-structural damage: extensive repair.
□ Field Exploration Required Risk to Life: Moderate
□ Other (Explain) ________________________________________________________________________________________ Systems: Disruption of systems for months to years. Occupancy: Partially to
totally vacated during repairs,

Building: Substantial structural damage: partial collapse likely: repair may not

be cost effective Extensive non-structural damage: repair mav not be cost


Figure 1. Seismic Evaluation Summary effective. Risk to Life: Substantial.
Systems: Total disruption of systems: repair may not be cost effective.
Occupancy; ___ Tolalfy vacated during repairs ______________________________
Figure 2. Seismic Risk Level Definition

Retrofit Program Description

As-Built Condition

The 4 story court house (1 story basement plus 3 stories above ground) was constructed in 1966.
The building is a steel frame structure clad with precast concrete panels. The typical floor dimensions are
218 ft x 108 ft with an approximate floor area of 23,500 ft2. Typical story height is +/- 14'-6".

The typical floor framing consists of metal deck and concrete fill with supporting steel beams and
columns. The building is found on a 3 foot thick concrete mat. A typical floor framing plan is shown in
Figure 3.

The building skin consists of precast concrete panels which are two story high (from ground to the
3rd floor) and one story high (from the 3rd floor to the roof) as shown in Figures 4 and 5. These panels are
interlocked with solid push-pull connections that do not allow for inter- story drift.

2
Figure 3. 3rd Floor Framing Plan

Figureelevation
Figure 4. Typical precast panel 5. Typical Precast Panel Section

3
The existing lateral load resisting system is steel moment frames at the building perimeter. The moment frame
beams have top and bottom flange cover plates as shown in Figure 6. As concluded in the seismic evaluation
shown in Figure 7, the moment frames do not conform to the current building code. The effectiveness of the
moment frames is further hindered by the inability of the precast panels to accommodate inter-story drift. The
stiffness of these precast panels is much higher than that of the steel moment frames. In a major seismic event,
the panels and their connections to the structure will most likely attract a large amount of seismic force and
suffer severe damage.

3.7.3S SUPPLEMENTAL STRUCTURAL CHECKLIST FOR BUILDING TYPE SI: STEEL MOMENT FRAMES WITH RIGID OR STIFF
DIAPHRAGMS

LATERAL FORCE RESISTING SYSTEM


MOMENT RESISTING CONNECTIONS: All Connections utilize flange plates and double angles at
ir panel zones, strength of the adjacent beams. (Tier 2: Sec. 4.4.1.3.3)
PANEL ZONES: All panel zones shall have the shear Beam (18WF45): shear demand from 0.8E Mb = 386 capacity to resist the shear demand required to k, (Mb=RyFyZ,
Ry = 1.5). Column (14WF95): panel develop 0.8 times the sum of the flexural strengths of zone shear capacity per AISC seismic provision, Rv = the girders framing in at
the face of the column. (Tier 0.6Fvdot [1 +3bcf tcr Vd„d0t ] = 171 k. (N.C.) 2: Sec. 4.4.1.3.4)
COLUMN SPLICES: All column splice details Only flanges are connected at splice
located in moment resisting frames shall include
connection of both flanges and the web for Life Safety
and the splice shall develop the strength of the column
for Immediate Occupancy. (Tier 2: Sec. 4.4.1.3.5)
1
STRONG COLUMN/WEAK.3 BEAM: The percent of Beam (16WF36): Z = 64 in , M„ = LlRyF^Z = 3802k- strong column/weak beam joints in each story of each in.
Column (12WF65): Z = 97 in . Mp = F,Z = 3492 line of moment-resisting frames shall be greater than k-in. 50% for Life Safety and Immediate Occupancy. (Tier 2: Sec.
4.4.1.3.6)
COMPACT MEMBERS: All frame elements shall W14x30 is non-compact as a beam, meet section requirements set forth by Table 1-9-1 of Seismic Provisions for
Structural Steel Buildings (AISC, 1997). (Tier 2: Sec. 4.4.1.3.7)

Figure 7. Moment Frame Deficiency Summary

Retrofit Schemes
Figure 6. Typical As-built Moment The seismic retrofit scheme originally proposed was to
Connection place steel braced frames from the 1st floor to the roof at selected
locations. However, since incorporating any new braces
rd
on the 3 floor would interfere with the on-going court
proceedings, the scheme was rejected. A second scheme, in which the braces on the 3rd floor were eliminated,
was analyzed. The results indicated the added stiffness from the remaining braces would attract a large seismic
force and cause the structure above the 3rd floor to fail.
rd
To satisfy the seismic retrofit requirement without interrupting the 3 floor court
rd
proceeding, a third scheme to install seismic dampers below the 3 floor was proposed and accepted by the
County. Seismic dampers provide the benefits of reducing seismic forces and movement in the structure by
absorbing part of the seismic forces generated in the ground. In this scheme, twenty four (24) 250kip friction
dampers are placed at the ground level and twenty four (24) 200kip friction dampers are placed at the 2nd floor.
As a result, it
rd
was no longer necessary to seismically reinforce the structural framing at the 3 level.

There are four major groups of seismic dampers currently used in the United States. They are fluid
viscous dampers (FVDs), viscoelastic dampers (VEDs), friction dampers and

4
metallic yielding dampers (MYDs). For this project, friction dampers as shown in Figures 8 and 9 were
chosen because of the following features:

1. Friction dampers provide both damping and stiffness, and therefore, the least story drift.
2. The energy dissipation of friction dampers is the greatest, for a given force level.
3. Friction dampers are less expensive and do not require maintenance and post installation inspection.

Figure 8. 250 kip friction damper in crate

Figure 9. Installed 250 kip friction damper


To accommodate the inter-story drift and allow the damper scheme to function properly, the
interlocking precast panels were separated. The existing roof connection (connection C1 in Figure 5) was
modified to allow for in-plane movement as shown in Figure 10. The existing 3rd floor connection

5
(connection C2 in Figure 5) locking the upper and lower panels was separated to allow for independent
in-plane movement as shown in Figure 11. An example of the connection C2 before and after separation
condition is shown in Figure 12.

Figure 10. C1 retrofit scheme


Figure 11. C2 retrofit scheme

As built condition prior to separation


Condition after separation

Figure 12. Precast connection C2 retrofit scheme


Analysis Procedure
The analysis follows the guidelines of FEMA 351 - Steel Moment Frame Buildings: Evaluation and
Upgrade Criteria for Existing Buildings and FEMA 356 - Prestandard and Commentary for The Seismic
Rehabilitation of Buildings. The target building performance level is life safety (3-C) under basic safety
objective (section 1.4.1) per FEMA 356. Existing structural components are analyzed per chapters 4, 5 and 6 of
FEMA 356 and the existing precast panels and their connections to the building were analyzed per chapter 11 of
FEMA 356. The friction dampers and their connections to the building were designed per chapter 9 of FEMA
356.

3-D models were created using ETABS to analyze the building's as-built and retrofitted
rd

6
conditions. All structural elements from 3 level down, except the friction dampers, were
rd
assumed to behave linearly and were analyzed accordingly. All structural elements above 3 level were assumed
to behave nonlinearly and were analyzed accordingly. Members below the 3rd floor were evaluated against
Table 5-5 and members above the 3rd floor were evaluated against Table 5-6 of FEMA 356.

The models were analyzed with both BSE-1 and BSE-2 time histories as required in FEMA 356. The
results from the BSE-1 analysis were checked against the Life Safety (LS) Performance Level as required in
FEMA 356.

Performance Assessment
The analysis results indicate that the retrofitted structural framing is able to dissipate the seismic energy
in a controllable manner due to the friction dampers' capability to absorb a constant force with varying story
displacement. Graphic output of a 250 kip friction damper's force and displacement is presented in Figures 13
and 14 respectively.

Figure 13. 250 kip damper force

7
D$Hripe* Displacement Legend
VerHcai Axis Link

Ш
L41 g[ Fnd't

Щ
240.
/ ?

$ м
2ND

~ LEV
z EL Awial

//:
/
1 во ^ Force Min is -

7 iV
2.50ue+02

/
// f
120 *
/
ф
/ о
at 4 330e+00 Мак is
2.499e+C2

feci, 1
,
//
с

/
at 4 730e+0G

7 Ш
LL

/
CI; "j i-ionzontal Axis
//
о.

O/ LL.inkL41-1

f
/ / 7I // 8 п
9
With the addition of the friction dampers, the story shear and the story displacement are reduced on all
floors compared to the as-built condition. A comparison of the story displacement and story shear of the three
analysis schemes - braced frames on all levels (scheme A), braced frames with friction dampers (scheme C),
and existing moment frames on all levels (scheme D) are shown in Figures 15 and 16 respectively.

Story Displacement Comparison - E/W Direction

-Scheme A - Braces on all


levels

)( Scheme C - Braces w ith


dampers below level 3

-Scheme D - Existing
moment frames

Level

Figure 15. Story Displacement Comparison

Story Shear Comparison - E/W Direction


7000

6000
-Scheme A - Braces on all
levels

—H—Scheme C - Braces w ith


dampers below level 3

-Scheme D - Existing
moment frames

Level

Figure 16. Story Shear Comparison

10
Conclusions

Based on the comparison of the existing steel moment frame and the friction damper
performance, the friction damper retrofit scheme will significantly improve the structural
performance of the building, both in terms of story displacement and story shear. The story
displacement is reduced by 50% and the story shear is reduced by 20% to 30%.

With the story displacement and shear reduction, the seismic force being transferred to
rd
the existing framing above the 3 floor is greatly reduced. As a result, the existing framing above the
3rd floor has sufficient capacity to resist the reduced seismic force and can be preserved without any
interruption to the on-going court operations as required by the County.

References

American Society of Civil Engineers, 2002. "ASCE 31-02 Seismic Evaluation of Existing Buildings ",
Reston, Virginia

Federal Emergency Management Agency, 2000. "FEMA 351 Recommended Seismic Evaluation and
Upgrade Criteria for Existing Welded Steel Moment-Frame Buildings", Washington, D.C.

Federal Emergency Management Agency, 2000. "FEMA 356 Prestandard and Commentary for The
Seismic Rehabilitation of Buildings", Washington, D.C.

11
2
Pall Dynamics Limited, Montreal, QC, Canada. E-mail: info@palldynamics.com
1. The non-ductile large diameter existing concrete columns at the soft-story will suffer severe
damage after repeated large lateral movements during the code-defined 10%-50 year seismic
event.
2. Damage to the non-structural building systems and components due to excessive story drift at the
soft-story during the seismic event. This issue is very well addressed by the construction industry
nowadays for base-isolated buildings. However, it becomes a real concern for this building.
1. Dampers can be strategically sized and placed to avoid foundation work.
2. Such dampers are driven by tension-only braces. Consequently, longer while smaller size braces
can be used to reduce impacts to the building appearance.
3. Such dampers are cost effective compared to other damping devices, which do not have the
features mentioned in item 2 above.
4. Such dampers have been widely accepted and utilized in the engineering and construction
industries, especially in building seismic upgrade and retrofit projects.
5. The existing post-tensioned "X" braces provide a building re-center mechanism after a seismic
event.
1. All sixteen (16) 36" diameter concrete columns at the soft-story are enlarged to 48" diameter along
their entire length with 3/8" thick steel plate jackets and infill grout. This is to increase column
ductility and to ease design difficulty for the connections between concrete columns and damper
braces.
3
Principal, Middlebrook + Louie, San Francisco, CA. www.mplusl.com

13th World Conference on Earthquake Engineering


Vancouver, B.C., Canada August 1-6, 2004 Paper No. 1955
PERFORMANCE-BASED DESIGN USING PALL FRICTION DAMPERS - AN
ECONOMICAL DESIGN SOLUTION.
Avtar PALL1 and R. Tina PALL1
SUMMARY
Benefit-cost analysis approach suggests performance-based design for most modern
buildings. The conventional structural systems are highly unlikely to provide adequate
performance in the event of a major earthquake. With the emergence of Pall Friction
Dampers, it has now become economically feasible to design high performance
structures. Their low cost and maintenance free characteristics suggest wide application

1 Pall Dynamics Limited, Montreal, Canada. www.palldynamics.com, E-mail:


info@palldynamics.com

12
for new construction as well as for retrofit of existing buildings. Public sectors, private
sectors, developers and developing countries are all benefiting from this technology.
They have been used for the seismic protection of more than 80 major building projects,
including the Boeing Commercial Airplane Factory at Everett, WA - the world's largest
building in volume.
INTRODUCTION

During a major earthquake, a large amount of energy is pumped into the building. The
manner in which this energy is consumed in a structure determines the level of damage.
The building codes recognize that it is economically not feasible to reconcile this energy
within the elastic capacity of materials. The criterion stipulated in conventional building
codes is to design structures to resist moderate earthquakes without significant damage
and avoid collapse during major earthquakes. The primary emphasis is on life safety.
The reliance for survival is placed on ductility to dissipate energy during inelastic
deformations causing bending, twisting and cracking. This assumes permanent
damage, repair costs of which could be significant. Recent earthquakes have clearly
demonstrated that conventional construction, even in technologically advanced
countries, is not immune to destruction.
In modern buildings, avoidance of structural collapse alone is not enough. The cost of
non-structural components and contents is much higher than the cost of the structure
itself and must be protected. The buildings of post disaster importance such as
hospitals, telecommunications, police stations etc. must remain operational. Benefit-cost
analysis suggests performance-based design approach. Emphasis for performance
based seismic design is being given in recent standards, e.g., Vision 2000, ATC-57 and
FEMA-356/2000. While global deflections, story drifts, force and demand-capacity ratios
are important to a structural engineer; these have no meaning to a developer. What are
important to an owner is how much

13
it will cost to build and how much it will cost to repair the damage, lost rental revenues
from a future earthquake. Early on, the structural engineers should resolve these issues
with the owner and develop performance-based design criteria consistent with the
performance goals. In most cases, providing good seismic performance yields good
return on the investment.
If the amount of energy getting into the structure can be controlled and a major portion
of the energy can be dissipated mechanically independent of primary structure, the
seismic response of the structure and damage control potential can be considerably
improved. These objectives can be delivered by adopting new techniques of base
isolation and energy dissipation devices. With the introduction of energy dissipation
devices, supplemental damping of 20-30% of critical can be easily achieved (inherent
damping is merely 1-5%). Thus forces exerted on the structure and the amplitudes of
vibrations are considerably reduced. The drifts of the structure can be reduced by a
factor of about two to three and by larger factors if the devices also add stiffness to the
structure, refer Article C9.1 of FEMA-356 [27]. The energy dissipation systems should
be considered in somewhat broader context than isolation systems (which are not
feasible in taller buildings) as a design strategy when performance goals include
damage control. Also, the construction cost is less with energy dissipation systems than
base isolation systems.
Seismic isolation and energy dissipation systems are relatively new and sophisticated
concepts that require more extensive design and detailed nonlinear time-history
dynamic analysis than most conventional schemes. With the availability of several
commercial programs and powerful desktop computers, the sophisticated analysis is no
longer a daunting task and can be easily and quickly done in a small design office
environment. However, the benefits accrued outweigh the extra design expense and
resulting savings more than compensate the cost of dampers.
With the emergence of Pall Friction Dampers[1-3,6], the performance-based design of
buildings is now economically feasible. The first building with seismic dampers in North
America was built with Pall Friction Dampers. Their low cost and maintenance free
characteristics suggest wide application for new construction as well as for retrofit of

14
existing buildings. Public sectors, private sectors and developers are using and
benefiting this technology. They have been used for the seismic protection of more than
80 major building projects, including the Boeing Commercial Airplane Factory at Everett,
WA - the world's largest building in volume. This technology is credited with for saving
Boeing more than US$ 30 million. The City and County of San Francisco chose Pall
Friction Dampers for the seismic control of Moscone West Convention Center and
saved US$2.25 million, compared to alternate viscous dampers.
This paper describes the innovative structural system and several applications in the
performance-based design of major projects including cost analysis.
PALL FRICTION DAMPERS
State-of-the-Art
Of all the methods so far available to extract kinetic energy from a moving body, the
most widely adopted is undoubtedly the friction brake. Mechanical engineers have
successfully used this concept for centuries to stop the motion of equipment,
automobiles, railway trains, airplanes etc. No other mean has been able to replace the
friction brake. Reason! It is the most effective, reliable and economical mean to
dissipate kinetic energy. Similar to automobiles, the motion of vibrating building can be
slowed down by dissipating seismic energy in friction.
Inspired by the principle of friction brake in mid 1970's, Pall Friction Dampers were
pioneered for the seismic control of buildings. Pall Friction Dampers significantly reduce
the initial cost of construction while dramatically increasing the earthquake resistance
against damage.

15
Friction dampers for a building must possess a reliable and stable performance over the
life of building. Developing a reliable friction is very difficult and tricky. Over a period of
more than a decade of research and development, the common problems in friction
were successfully overcome by using specially treated surfaces and a unique
manufacturing process. Over the years, Pall Dynamics has earned an international
reputation for excellence and is a world leader in friction dampers for seismic control of
buildings.
Pall Friction Dampers have successfully undergone rigorous proof testing in the U.S and
Canada. In 1985, the National Research Council of Canada tested 3-story frame
structures on a shaking table at the University of British Columbia, Vancouver [4]. In
1986-1987, the U.S. National Science Foundation tested a 9-story frame structure on a
shaking table at the University of California at Berkeley [5]. The structures were
subjected to more than 20 different major earthquake records. Even for an earthquake 5
times stronger than the 1985-Mexico earthquake, the frames equipped with friction
dampers remained damage free. Pall Friction Dampers are well recognized and
accepted by the building codes in Canada, the U.S and many other countries.
Salient Features
Friction Damper in Tension-only Cross Brace
Hystersis Loop Figure 1. Pall Friction Dampers
Friction Damper for Tension-Compression Brace
In a typical undamped structure, the inherent damping is merely 1-5% of critical. With
the introduction of Pall Friction Dampers, structural damping of 20-50% of critical can be
easily achieved. As the dampers dissipate a major portion of the seismic energy, forces
and deformations on the structure are significantly reduced. Pall Friction Dampers
significantly reduce the initial cost of construction while dramatically increasing the
earthquake resistance against damage.
Pall Friction Dampers are foolproof in construction. Basically, these consist of series of
steel plates, which are specially treated to develop very reliable friction. These plates
are clamped together and allowed to slip at a predetermined load. Decades of research

16
and testing have led to perfecting the art of friction. Their performance is reliable,
repeatable and they possess large rectangular hysteresis loops with negligible fade.
Their performance is independent of velocity and hence exerts constant force for all
future earthquakes, design-based earthquake (DBE) or maximum credible earthquake
(MCE). A much greater quantity of energy can be dissipated in friction than any other
method involving the yielding of steel plates, viscous or viscoelastic dampers. Therefore,
fewer Pall Friction Dampers are required to provide the required amount of energy
dissipation. Pall Friction Dampers are passive energy dissipation devices and, therefore,
need no energy source other than earthquake to operate it. They do not require any
repair or replacement after the earthquake and are always ready to do their job.

17
Pall Friction Dampers are customized to suit site conditions and allow greater
adaptability than is possible with other systems. These dampers can be bolted or
welded into place.
Pall Friction Dampers are available for long slender tension-only cross bracing, single
diagonal tension- compression bracing and chevron bracing (Figure 1). The damper for
cross bracing is a unique mechanism. When one of the brace in tension forces the
damper to slip, the damper mechanism forces the other brace to shorten and thus avoid
buckling. In this manner, the other brace is immediately ready to slip the damper on
reversal of cycle. These dampers have been used in 65 feet (22 m) long slender
bracing.
To avoid pounding at the expansion joints, Pall Friction Connectors are custom made to
accommodate bidirectional movements.
Slip Load of Friction Damper
The friction dampers are designed not to slip during wind. During a major earthquake,
they slip prior to yielding of structural members. In general, the lower bound is about
130% of wind shear and the upper bound is 75% of the shear at which the members will
yield. As seen in Figure 2, if the slip load is very low or very high, the response is very
high. Several parametric studies have shown that the slip load of the friction damper is
the principal variable with the appropriate selection of which it is possible to tune the
response of structure to an optimum value. Optimum slip load gives minimum response.
Selection of slip load should also ensure that after an earthquake, the building returns to
its near original alignment under the spring action of an elastic structure. Studies have
also shown that variations up to ±20% of the optimum slip load do not affect the
response significantly. Therefore, small variations in slip load (810%) over life of the
building do not warrant any adjustments or replacement of friction damper.

C
D

18
tn
с
о
Q.

Q

O
pti
m
u
m
Fi
gu
re
2.
R
es
po
ns
e
ve
rs
us
Sli
p

19
Lo
ad
Sli
p
Lo
ad
tr

WHY PALL FRICTION DAMPERS ARE FAVORED?


They are simple and foolproof in construction.
Offer reliable and repeatable performance at low cost.
Possess large rectangular hysteresis loops. Greater energy dissipation for a given force.
Hence, fewer Pall Friction Dampers are needed. Conversely, exert lesser force for a given
damping.
Provide supplemental damping and stiffness for added stability.
Performance is independent of velocity and temperature.
Constant force for all future earthquakes (DBE/MCE). Therefore, design of connections
and members is economical.

20
They are not active during service loads and wind. Hence, no possibility of failure due to
fatigue before an earthquake.
Need no repair or replacement before and after earthquake. There is nothing to damage
or leak.
Energy dissipation is through friction and not through the damaging process of yielding.
After an earthquake, the building returns to its near original alignment due to spring action
of an elastic structure.
Compact and narrow enough to be hidden in partitions.
They can accommodate foundation settlements.
Available for all types of bracing, including tension cross bracing, and expansion joints.
Custom made. Easily adaptable to any site condition. Can be welded or bolted

COMPARISON OF HYSTERESIS LOOPS OF DIFFERENT DAMPERS

+p

21
Yielding Plate Self-centering
Damper Friction Damper
Viscoelastic
Damper

Figure 3. Comparison of Hysteresis Loops of Different Dampers


COMPARISON WITH OTHER DEVICES
Viscous Damper
Viscous dampers are velocity dependent. The forces exerted by the damper are,
therefore, different for different earthquake records. Friction dampers are independent of
velocity, therefore, exert constant force for all future earthquakes (DBE / MCE). A friction-
damped structure is an engineered structure in which forces exerted are predetermined.
This is a great technical and economic advantage over viscous dampers in which forces
are much higher at MCE level.
The hysteresis loop of viscous damper is elliptical compared to rectangular for friction
damper. For a given maximum force, the area of hysteresis loop (energy dissipation or
damping) of viscous damper is about 70% of that for friction damper i.e. 70 friction
dampers will achieve the same damping as 100 viscous dampers of a given force.
Conversely, for a given number and damping value, the forces exerted

22
by friction dampers are only 70% of those for viscous dampers. This leads to significant
savings in cost of dampers, bracing, connections, columns and foundations.
While supplemental damping is beneficial in reducing the earthquake forces and
amplitudes of vibration, added stiffness is beneficial for stability. A deflected building is
similar to a stooped person. Similar to a cane support for stooped person, additional
stiffness helps the deflected building against overturning. Pall Friction Dampers provide
both added damping and added stiffness for stability - a complete structural solution.
Viscous dampers provide only damping and no stiffness - the structure is on its own to
struggle for stability.
Unbonded Brace
Unbonded brace is often called by other names like 'Yielding brace' or 'Buckling-
Restrained brace'. It consists of a slender steel brace (core steel), surrounded by a steel
tube or pipe, which is then filled with concrete or grout. The core steel is wrapped with
plastic like material to separate it from the concrete around. The concrete in steel tube
prevents buckling of the steel brace.
Unbonded brace dissipates energy through the process of yielding. Yielding involves
damage. No damage, no energy dissipation or damping. After an earthquake, the brace
may be damaged and needs to be replaced. Replacement of brace after an earthquake
is expensive and time consuming. It is too difficult to replace before imminent
aftershocks. Another problem is that the brace is enclosed in concrete and is not visible
for inspection to verify if it is broken or otherwise.
The cost of supply of core steel and wrapping material may appear to be small, but the
overall cost of an installed unbonded brace, including filling of outer tube with concrete
and connections, is higher than the installed friction damper. As friction dampers
dissipate a large amount of energy mechanically, the forces exerted are far less than
those exerted by the unbonded braces. Besides, the structure with friction damper is
economical to design and always ready to resist earthquakes one after another without
replacement.
DESIGN CRITERIA
The quasi-static design procedure given in most building codes are ductility based and
do not explicitly apply to buildings with supplemental damping. In the past few years,

23
several guidelines on the analysis and design procedure of passive energy dissipation
devices have been developed in the U.S. The latest and most comprehensive document
is the "NEHRP Guidelines for the Seismic Rehabilitation of Buildings, FEMA 356 / 357,
issued in 2000" [27]. This equally applies to new buildings.
The Guidelines require that the structure be evaluated for response to two levels of
ground shaking, a design basis earthquake (DBE) and a maximum considered
earthquake (MCE). The DBE is an event with 10% probability of exceedance in 50 years,
while the MCE represents a severe ground motion of probability of 2% in 50 years.
Under the DBE, the structure is evaluated to ensure that the strength demands on
structural elements do not exceed their capacities and that the drift in the structure is
within the acceptable limits. For the MCE, the structure is evaluated to determine the
maximum displacement and overstress. It is presumed that if proper ductile detailing has
been followed, the structure will have sufficient reserve to avoid collapse during MCE.
Since different earthquake records, even of the same intensity, give widely varying
structural responses, results obtained using a single record may not be conclusive.
Therefore, three time-history records, suitable for the region should be used; one of
which should be preferably site specific. The maximum response is used for the design.
NEHRP guidelines require that friction dampers are designed for 130% MCE
displacements and all bracing and connections are designed for 130% of damper slip
load. Variation in slip load from design value should not be more that ± 15%.
NONLINEAR TIME-HISTORY DYNAMIC ANALYSIS
The slippage of friction damper in an elastic brace constitutes nonlinearity. Also, the
amount of energy dissipation or equivalent structural damping is proportional to the
displacement. Therefore, the design of friction-damped buildings requires the use of
nonlinear time-history dynamic analysis. With these analyses, the time-history response
of the structure during and after an earthquake can be accurately understood. Several
nonlinear computer programs are now capable of modeling friction dampers. Some of
these programs are ETABS, SAP2000, DRAIN-TABS, DRAIN-2DX, DRAIN-3DX,
ANSYS etc. With the availability of powerful personal computers, the sophisticated
nonlinear time history analysis can be easily and quickly done in a small design office
environment.

24
The modeling of friction dampers is very simple. Since the hysteresis loop of the damper
is similar to the rectangular loop of an ideal elasto-plastic material, the slip load of the
friction damper can be considered as a fictitious yield force.
PRACTICAL APPLICATIONS
The first building built with seismic dampers in North America was with Pall Friction
Dampers (1987). Pall Friction Dampers are finding increasing application worldwide for
new construction as well as for retrofit of existing buildings, including overhead water
tanks. They have been used for the seismic protection of more than 80 major buildings in
Canada, the U.S., China and India. These are hospitals, telecommunication buildings,
educational institutions, police headquarters, defense installations, convention centers,
courthouses, office and residential buildings. Some projects are discussed below:
Seismic Upgrade of Boeing Commercial Airplane Factory, Everrett, WA, USA.

Figure 4. (a). Inside View of Boeing Factory. (b). Pall Friction Dampers

The mammoth Boeing plant, which could contain Disneyland under one roof, is the
world's largest building in volume [25]. It was built in phases from 1968-1991, for the
assembly of wide-bodied 747 jetliners - world's largest commercial airplane. The steel
frame building is 120 feet (37 m) high with clear spans of 350 feet (107 m) and covers
more than 98 acres.
In 1996, the Boeing engineers considered several seismic upgrade schemes for this
structure. They chose Pall Friction Dampers as they are foolproof in construction and

25
offer reliable maintenance free performance at low cost. Also, they possess stiffness for
added stability. The performance of friction dampers is independent of velocity, therefore,
the forces on the connections remain constant for any future earthquake. This is an
engineered solution. Economy in the design of connections and easy installation of
dampers provided significant savings in construction cost and time.
Several types of Pall Friction Dampers, suitably modified to adapt to the site conditions,
were incorporated in different types of existing bracing. Friction dampers of capacity up
to 200,000 lb. (900 KN) and stroke up to 15 inch (380 mm) have been used. Pall Friction
Dampers met or exceeded Boeing's stringent specifications. Work on seismic upgrade
was undertaken in 1998 and completed in 2002.
Cost of Retrofit: US$65 million. Savings: US$30 million compared to conventional
construction.
Moscone West Convention Center, San Francisco, USA

Figure 5. (a). View of Moscone West (b). Pall Friction Damper

The Moscone West Convention Center is located in downtown San Francisco, between
San Andrea's and Hayward faults, which are about 19 kM in each direction. The four-
story steel frame building is 112 feet
(34 m) high with clear spans of 45 to 90 feet (14-28 m) [26]. The US$186 million
expansion of the 1.2
2

26
million sq. feet (111,700 m ) convention center was California's largest project of 2001.
Bids were invited on two alternate designs, one with viscous dampers and another with
Pall Friction Dampers. The scheme with friction dampers offered a saving of US$2.25
million compared to viscous dampers. On the recommendations of peer reviewers and
two specialist consultants from the University of California at Berkeley, the City and
County of San Francisco chose Pall Friction Dampers for seismic control of this
prestigious building. The construction of the convention center was completed in 2003.
Pall Friction Dampers up to 500,000 lb. (2250 kN) capacity and 9 inch (230 mm) stroke
have been used.
The use of friction dampers reduced the story drifts from 2% without dampers to 1% with
dampers. About 80% of the earthquake energy has been dissipated by the friction
dampers leaving the building with little damage. According to the project architects and
engineers, the use of technology saved a couple of million dollars in the initial
construction costs and will save tens of millions of dollars in lowered repair costs in the
event of an earthquake.
Seismic Retrofit of Boeing Development Center, Cafeteria and Auditorium Buildings,
Boeing Field, Seattle, WA, USA

Figure 6. (a). View of Cafeteria, Auditorium and Fitness Building (b). Pall Friction Damper

There are three 2-story Boeing Development buildings and a 4-story Boeing Cafeteria,
Auditorium and Fitness Center building. These are steel frame buildings, built in 1980's.
The foundations are on 70-80 feet deep friction piles. Some of these buildings were
damaged during February 28, 2001 Nisqually Earthquake of magnitude 6.8. The
epicenter was about 20 miles from the building site. Due to liquefaction of soil, the

27
differential settlements in pile foundations were 4-5 inches. This resulted in substantial
non-structural and structural damage to the Cafeteria and Auditorium building. The
structural damage was primarily to rigid steel bracing. Several bracing in the Cafeteria
and Auditorium building buckled or broke.
Several structural schemes were considered for the seismic retrofit of these buildings.
Friction dampers were considered to be an ideal solution as they provide both damping
and stiffness. Besides, they can be easily modified to suit site conditions and designed to
accommodate any future foundation settlements.
A total of 350 Pall friction Damper of 100-500 kip slip load and up to 10 inches stroke,
were used in the existing steel bracing. The seismic retrofit was completed in 2002.
Savings are estimated to be more than 60% compared to conventional retrofit.
Ambulatory Care Center, Sharp Memorial Hospital, San Diego, California, USA

Figure 7. View of New Ambulatory Care Center Pall Friction Dampers in Chevron
Brace

28
The Ambulatory Care Center complex consists of two buildings of four and five floors,
connected in two places by a pedestrian bridge and an elevator lobby. This facility will
provide extensive medical services to the community including surgery.
The project engineers selected steel-frame design for the cost efficiency, design flexibility
and speedy construction. Moment-resisting frames, in combination with Pall Friction
Dampers in steel bracing, were selected to resist lateral seismic forces. The structure
was designed to meet 1997UBC and 1998 California Code requirements for Seismic
Zone 4. In this performance-based design, the Pall Friction Dampers reduced building
drift from 2% to 1%. The savings in construction cost due to reduced forces, more than
offset the cost of dampers. In case of a major seismic event, the dampers reduce
building content damage and increase safety of occupants. This resulted in long-term
savings to the client through lower seismic insurance premiums.

Figure 8. 3-Million Gallon Water Tank.


Seismic Retrofit of 3-Million Gallon Reservoir, Sacramento, California, USA
The Freeport water tower, a distinctive landmark visible from Interstate-5, was built in
1956 [22]. The steel reservoir stands about 120 feet high. The supporting structure
consists of 27 steel columns with two levels of 60 feet long tension cross bracing. Of the
several seismic retrofit options, the scheme with Pall Friction Dampers in tension cross

29
bracing was chosen. When tension in one of the brace forces the damper to slip, the
damper's mechanism shortens the other brace, thus preventing buckling.
Due to high damping provided by the Pall Friction Dampers, the strengthening of
columns and foundations was not necessary. Friction dampers up to 150 kip slip capacity
were used. Seismic retrofit was completed in 1999.
Savings: More than 60% compared to conventional.
Concordia University Library Building, Montreal, Canada

Figure 9. (a). View of Concordia Library Building (b). Pall Friction Damper in Cross
Bracing

30
The ten-story McConnell Library Building is a masterpiece in innovative structural design
[7]. The building was designed in 1987 and construction was completed in 1991. Pall
Friction Dampers are installed at the junction of steel cross bracing in rigid concrete
frames. The use of steel bracing eliminated the need of expensive concrete shearwalls
and the introduction of supplemental damping provided by friction dampers eliminated
the need of dependence on ductility of structural members.
Unlike concrete shearwalls, the bracing were generally not continuous one over the other
and thus provided greater flexibility in space planning. Since the bracing do not carry any
gravity load, they do not go through the basement to the foundations. Therefore, they
allow more space for parking. The architects have boldly exposed several bracing with
dampers as these add to the aesthetic appearance.
The innovative structural system provided an economical design solution to safeguard
the building and its valuable contents from earthquake damage.
Covered Area: 52,000 m2. Cost: $65 million. Savings: 6.5% of structural or 1.5% of
building cost. Seismic Rehabilitation of Justice Headquarters, Ottawa, Canada.

Figure 10. (a). View of Justice Headquarters (b). Pall Friction Damper

This eight-story concrete frame building is located in the nation's parliamentary district
[21]. It was built in 1955 as a memorial to the Canadians killed during the Second World
War. The stone clad building, with sloping copper roof, has been designated a heritage
structure. The existing structure was not capable of resisting seismic forces specified in

31
the Building Code. In 1997, seismic rehabilitation was undertaken along with major
renovations to protect the original and new investment.
Pall Friction Dampers offered the best solution for the seismic upgrade. Since the
dampers dissipate a major portion of the seismic energy, forces acting on the structure
are considerably reduced. By staggering the bracing at different story levels, overloading
of columns and foundations was avoided. Hence, expensive and time-consuming work of
strengthening existing members and foundations was not required.
Unlike shearwalls, the friction-damped bracing need not be vertically continuous. This
aspect was particularly appealing to the architects as it offered flexibility in space
planning. This structural solution also facilitated construction scheduling as work could
start at any floor level, depending on vacancy or availability.
Covered Area: 50,565 m2. Savings: More than 40% in construction cost and time.

32
Figure 11. View of Canadian Space Agency Headquarters

The headquarters of the Canadian Space Agency is a building of national importance


[11]. It contains extremely sensitive and costly equipment / instrumentation. Therefore, it
is of vital importance to protect its valuable contents and electronically stored data in the
event of a major earthquake.
The use of Pall Friction Dampers significantly increased the damage control potential of
the building while offering savings in the initial cost of construction. Construction was
completed in 1992.
Covered Area: 50,000 m2. Cost $60,000,000. Savings: 1.25% of total cost.
Canadian Space Agency Headquarters, St-Hubert, Canada
Seismic Rehabilitation of Casino de Montreal, Canada

33
Figure 12. Casino de Montreal

34
In 1993, Casino de Montreal was housed in the existing French Pavilion built for
EXPO'1967 [12]. The lateral earthquake resistance of the existing eight-story braced
steel structure was not adequate to meet the requirements of the National Building Code
of Canada.
Introduction of supplemental damping provided by Pall Friction Dampers was the most
effective, economical and hi-tech solution for the seismic rehabilitation of this building.
The use of Pall Friction Dampers in the existing steel bracing considerably reduced the
forces on the structure. Hence, the provision of additional bracing, strengthening of
existing members and pile foundation was not required.
Covered Area: 38,000 m2. Savings: 50% in construction cost & time.
Quebec Provincial Police Headquarters, Montreal, Canada

Figure 13. (a). View of Police Headquarters (b). Pall Friction Dampers in Bracing

The existing sixteen-story office building, with two levels of basement, was built in 1964
[23]. Steel moment frames and some braced bays provided lateral resistance to the
existing structure. A change of occupancy was planned in 1997 to house the provincial
police (Surete du Quebec) headquarters. The project structural engineers evaluated that
the existing structure was not capable of resisting seismic forces and the story drifts were
excessively high, especially at the lover level due to soft story effect.
In 1999, the work on seismic rehabilitation was undertaken along with major renovations
to protect the original and new investment. Pall Friction Dampers offered the best

35
solution for the seismic upgrade. Since the dampers dissipate a major portion of the
seismic energy, the story drifts and forces acting on the structure are considerably
reduced. Hence, expensive and time-consuming work of strengthening existing member
and pile foundations was not required. Pall Friction Dampers were incorporated in
existing and new bracing. The innovative structural scheme offered savings of more than
50% over conventional retrofit scheme.
CONCLUSION
The use of Pall Friction Dampers has shown to provide a practical, economical and
effective approach for the performance-based design of new and retrofit of existing
structures to resist major earthquakes. The low cost and maintenance free
characteristics of Pall Friction Dampers suggest wide application. Public sector, private
sector and developers, including developing countries, are using and benefiting from Pall
Friction Damper technology.
REFERENCES
Pall, A.S., Marsh, C. "Energy Dissipation in Large Panel Structures Using Limited Slip
Bolted Joints", AICAP/CEB Seismic Conference, Rome, Italy, 1979, 3:27-34.
Pall, A.S., Marsh, C. "Friction-Devices to Control Seismic Response", ASCE/EMD
Speciality Conference on Dynamic Response of Structures, Atlanta, USA, 1981, 809-
818.
Pall, A.S., Marsh, C. "Seismic Response of Friction Damped Braced Frames", Journal of
Structural Division, ASCE, 1982, St. 9, 108:1313-1323. (ASCE "Raymond Reese
Research Prize 1983").
Filiatrault, A., Cherry, S. "Seismic Tests of Friction-Damped Steel Frames", Third
Conference on Dynamic Response of Structures, ASCE, Los Angeles, USA, 1986.
Kelly, J.M., Aiken, I.D., Pall, A.S., "Seismic Response of a Nine-Story Steel Frame with
Friction- Damped Cross-Bracing", Report No. UCB / EERC-88/17. Earthquake
Engineering Research Center, University of California at Berkeley, 1988, 1-7.
Pall, A.S., Marsh, C., Fazio, P., "Friction Joints for Seismic Control of Large Panel
Structures", Journal of Prestressed Concrete Institute, 1980, No. 6, 25:38-61.

36
Pall, A.S., Verganelakis, V., Marsh, C., "Friction-Dampers for Seismic Control of
Concordia University Library Building", Fifth Canadian Conference on Earthquake
Engineering, Ottawa, 1987,191-200.
Pall, A.S., Ghorayeb, F., Pall, R., "Friction-Dampers for Rehabilitation of Ecole
Polyvalente at Sorel, Quebec", Sixth Canadian Conference on Earthquake Engineering,
Toronto, 1991, 389-396.
Vezina, S., Proulx, P., Pall, R., Pall, A., 'Friction-Dampers for Aseismic Design of
Canadian Space Agency", Tenth World Conference on Earthquake Engg., Madrid,
Spain, 1992, 4123-4128.
Pall, A.S., Pall, R., "Friction-Dampers Used for Seismic Control of New and Existing
Buildings in Canada", ATC 17-1 Seminar on Base Isolation, Passive Energy Dissipation
and Active Control, San Francisco, USA, 1993a, 2:675-686.
Pall, A., Vezina, S., Proulx, Pall, R., "Friction-Dampers for Seismic Control of Canadian
Space Agency Headquarters", Journal Earthquake Spectra, 1993b, Number 3, 9:547-
557.
Pasquin, C., Pall, A.S., Pall, R., "Hi-Tech Seismic Rehabilitation of Casino de Montreal",
ASCE Structures Congress, Atlanta, USA, 1994, 1292-1297.
Godin, D., Poirer, R., Pall, R., Pall, A., "Reinforcement Sismique du Nouveau Campus de
l'Ecole de Technologie Superieure de Montreal", Seventh Canadian Conference on
Earthquake Engineering, Montreal, 1995, 967-974.
Hale, T., Tokas, C., Pall, A., "Seismic Retrofit of Elevated Water Tanks at the University
of California at Davis", 7th. Canadian Conference on Earthquake Engg, Montreal, 1995,
959-966.
Savard, G., Lalancette, J.R., Pall, R., Pall, A., "High Tech Seismic Design of Maison 1
McGill", Montreal, Seventh Canadian Conference on Earthquake Engineering, Montreal,
1995, 935-942.
Wagner, P., Vavak, L., Pall, R., Pall, A., "Seismic Rehabilitation of the New Hamilton
Court House", Seventh Canadian Conference on Earthquake Engineering, Montreal,
1995, 951-958.

37
Pall, A., Pall, R., "Friction-Dampers for Seismic Control of Buildings - A Canadian
Experience", Eleventh World Conference on Earthquake Engineering, Acapulco, Mexico,
1996, Paper No. 497.
Deslaurier, F., Pall, A., Pall, R., "Seismic Rehabilitation of Federal Building, Sherbrooke",
Canadian Society of Civil Engineers Annual Conference, Sherbrooke, 1997, 4:339-348.
Pasquin, C., Charania, H., Steele, R., Pall, R., Pall, A.S., "Friction-dampers for Seismic
Control of Selkirk Waterfront Offices, Victoria", Sixth U.S. National Conference on
Earthquake Engineering, Seattle, USA, 1998.
Pasquin, C., Leboeauf, N., Pall, R., Pall, A., "Seismic Rehabilitation of Hotel Dieu
Hospital, Sainte Hyacinthe, Quebec", Eighth Canadian Conference on Earthquake
Engineering, Vancouver,
573-578.
Balazic, J., Guruswamy, G., Elliot, J., Pall, R., Pall, A., "Seismic Rehabilitation of Justice
Headquarters Building, Ottawa, Canada", Twelfth World Conference on Earthquake
Engineering, Auckland, New Zealand, 2000, Paper No. 2011.
Hale, T., Pall, R., "Seismic Upgrade of the Freeport Water Reservoir, Sacramento,
California", Twelfth World Conference on Earthquake Engg., Auckland, New Zealand,
2000, Paper No. 269.
Pall, R., Gauthier, G., Delisle, S., Pall, A., "Friction-Dampers for Seismic Upgrade of
Quebec Police Headquarters, Montreal", Twelfth World Conference on Earthquake
Engineering, Auckland, New Zealand, 2000, Paper No. 2014.
Chandra, R., Masand, M., Nandi, S., Tripathi, C., Pall, R., Pall, "Friction Dampers for
Seismic Control of La Gardenia Towers South City, Gurgaon, India", Twelfth World
Conference on Earthquake Engineering, Auckland, New Zealand, 2000, Paper No. 2008.
Vail, C., Hubbell, J., "Structural Upgrade of the Boeing Commercial Airplane Factory at
Everett, Washington", Proc. Structures Congress & Exposition, May 2003, Seattle, WA,
Paper # 000529.
Sahai, R., Laws, J., Chen, D., Kong, F., Castillo, F., "Performance Based Design of 4-
Story Moscone Convention Center Expansion Using Steel Couple Girder Moment
Resisting Frame and Friction Dampers", Proceedings, SEAOC Annual Convention,
Santa Barbara, California, 2000.

38
Prestandard and Commentary for the Seismic Rehabilitation of Buildings, FEMA 356/
November

39
13th World Conference on Earthquake Engineering
Vancouver, B.C., Canada August 1-6, 2004 Paper No. 3446
SEISMIC DESIGN OF STRUCTURES USING FRICTION DAMPER
BRACINGS
Babak ESMAILZADEH HAKIMI 13 , Alireza RAHNAVARD2, Teymour HONARBAKHSH14
SUMMARY
A Pall based friction damper located at intersection point of X or chevron bracing has been
designed concerning Iranian workmanship. Tests have been performed to investigate the
energy dissipation capacity of dampers with different types of slipping surface. Efficiency of
the system has been investigated through time history nonlinear analysis performed on 2
and 3 story steel structure with a logical range of period under sets of 9 accelerograms
matching with S2 and S3 types of soil.
Due to high performance of the system, damage index of the whole structure has been
found to be governed rather by non-structural damage potential or casual p-delta effect
leading to potential instability. The aspect ratio of the friction devices has been judged to
be very effective in energy dissipation, so that with aspect ratios near to main frame one,
the energy dissipation reached its maximum level. For aspect ratios far from the structure
one, rigid body motion of device resulted in reduction of energy dissipation potential.
Keywords: Friction Damper, Energy Dissipation, Damage Index, Hystersis
INTRODUCTION
Energy dissipation capacity and capability of structures subjected to severe earthquakes
plays a vital role to prevent them from catastrophic instability and failure. Inelastic behavior
of structural material is the main source of energy dissipation in conventional structures.
Potentially high cost of structural retrofitting, aftermath of a severe earthquake along with

13 Nirou Research Institute, Tehran, Iran , (besmailzadeh @ nri.ac.ir)


14 Sarzamin Consulting Engineers, Tehran, Iran , (sarzamin @ neda.net)

40
the need to predict the actual seismic behavior of structures, are the main causes to
localize the energy dissipation capability by devices installed in structures. Viscous
dampers (Constantinou [1]), ADAS damper (Kelly [2]), Friction dampers (Filiatrault [3,4])
and Viscoelastic Elastomeric dampers (Mahmoodi [5]) are within this category of structural
systems. It seems that the statistical and probabilistic perspective of upcoming
earthquakes which may just be judged only doubtfully due to their long return period, can
be considered as a source of encouraging implementation of methods with higher
efficiency and unlimited capacity of energy absorption.

41
Pall dampers
It has been a common practice among mechanical engineers to apply friction-based
brakes to absorb kinetic energy in machines and devices. This has encouraged the
development of Pall friction based damper (Chandra [6]). Pall friction dampers have
successfully gone through sophisticated experimental studies on shake tables in Canada
(Filiatrault [3]) and the United States (Aiken [7]). Pall friction dampers have been very
attractive due to their simplicity and low cost of construction. High seismic performance of
Pall's damper has been a great motivation to extend its application out of its origin to other
countries, specially to United states.
General structure of Pall type dampers is made up of some steel plates layed on each
other with high strength bolts pressing them together, generating friction between them.
Contrary to Viscoelastic systems, Pall system is not sensitive to environmental
temperature and state of loading. Pall system's hysteretic behaviors are almost rectangular
and completely similar to ideal elastoplastic behavior. Due to high dissipation energy
capacity and stability of hyteresis loops, Pall system seems to show higher seismic
performance, than other damping systems.
Friction dampers mechanism
As mentioned before, a Pall damper is very popular due to its low cost of construction and
its simplicity. Furthermore, considering architectural restrictions, such elements are very
easy to be hidden in internal partitions.
Friction based dampers are formed of steel plates tightened together by means of high
strength bolts with either axial or rotational deformation mechanism leading to
transformation of kinetic energy to thermal one.
Pall friction based bracing has utilized the lever mechanism wisely, to transform relatively
low global displacements of braced frames to local high deformations of dampers (figure 1-
a). A transformation ratio in order of 10 may be easily reached, leading to high efficiency of
the system.

42
Potential of energy dissipation is affected by damper aspect ratio and its relation with
braced frame aspect ratio, so that a maximum potential of energy dissipation may be
reached in the case of aspect ratio of damper be equal to the frame one. Formation of 2
instantaneous centers of rotation in the case of different aspect ratios is the source of
temporary instability of the system (figure 1-b) and reduction of energy dissipation
potential.

(a) Behavior of damper within structure (b) Formation of instantaneous centers of rotation
Figure (1): Performance mechanism of Pall Friction Damper

43
Tests15 have been performed on 28 samples under 50 loading cycles. 3 types of contact
surface, as smooth surface, rough surface and pad sandwiched system have been
assessed through displacement controlled cyclic tests with amplitudes of 10 and 20 mm.

Displacement (mm)|
All of 3 types of contact surface have shown complete elasto - plastic hysteretic behavior,
but it is obvious that the added pad has significant influence on stability of hysteretic loops
(figure 2-c). The pad sandwiched system has been chosen to be studied further in this
program.
Hysteresis Behavior of Friction Damper
Test No. a
50

g»-

30 -2 ^^^^^^^^fcgg^g^g^ 20 3
-30
Displacement (mm)|

(a) Smooth surface (b) Rough surface


Hysteresis Behavior of

15 In structural laboratory, civil faculty, Sharif University of Technology, Tehran, Iran.

44
Friction Damper
Test No. 18

z
О
u
О
" 0
O -10 3
Q 0 -2 10 J 0 30
.
a
<

Displacement (mm)|
(c) Pad sandwiched system Figure (2): Hysteretic curves drawn out of tests

Case studies
The program consists of structural rehabilitation of 2 schools with 2 and 3 stories, against
seismic effects. Due to symmetry of structural plans, only 2 dimensional models of
structural systems have been investigated under unidirectional accelerograms with
assumption of rigidity of floors. Elevation of structures have been shown in figures (3).
Structural configuration
Structure (A) consists of a 2 story steel frame with 5 m of span in each direction. Story
heights are 4 and 3m and floor masses are 360 and 340 tons for first and second floors
respectively. Structure B is a 3 story steel structure as structure A in which, story heights
are 4,3,3 m and floor masses are 360, 340, 340 tons for first, seconds and third floors
respectively. The structure have been assumed with no lateral strength by itself, in order to
make sense in investigation of the pure behavior of friction bracings.

45
(a): Structure (A)

(b): Structure (B) Figure (3): Structural elevations

Generalities
A nonlinear time-history analysis have been performed using Sap 2000 software. The 0-
Wilson nonlinear time history analysis method has been used and the nonlinear behavior
of links which model the friction behavior of interconnected plates, has been assumed to
be complete elasto - plastic with 1% of hard straining according to results derived from
tests. A damping of 5% of critical damping has been assumed and the following strategy
has been chosen as the main discipline to select accelerograms.
Selection of accelerograms
Selection of accelerograms has been subject of discussions in the past (Naeim [8]). No
general agreement has been reached about how to select accelerograms to use in time
history analysis of structures. This is due to the fact that no absolute logical link may be
found between average properties of few selected accelerograms and a specific,

46
developed, modified and smoothed design spectrum. However a general rule of selection
may be considered as:
To make sure the average response spectra of selected accelerograms matching with
design spectrum in the period range under consideration.
Selecting of accelerograms with highest energy content in the period range under
consideration with due attention to effective period of nonlinear structure. This may be
provided by using power spectra of considered accelerograms.

47
3. Assessment of sensitivity of nonlinear response of structures through either frequency
scalling of selected accelerograms or by using artificial accelerograms, with due
consideration of the fact that the later may lead to nonlogical high responses of nonlinear
structures.
In these studies, the recommendations of UBC 97 have been used in addition to same
extra considerations to sweep the whole range of period under consideration. 9
accelerograms (acc1 to acc9) have been selected (table 1) with following procedure to
draw out design values from (figure 5):

average( accl: acc7)


acc8
Design quantity=
max
(1)
acc9

2 levels of performance objective have been adopted under 2 levels of seismic hazard;
Life safety performance objective under 5% probability of exceedence in 100 years
earthquake.

48
Serviceability performance objective under 50% probability of exceedence in 50 years
earthquake.
Table (1): Accelerograms used in time history analysis
No. ■ ■ Record Location Date IStatbn DEG hp LP
Name ■ Name :
1 holuster city hollister 4/9/1 usgs1028 181 0.25 11
hall 931
2 holuster city hollister 4/9/1 usgs1028 271 0.11 11
hall 961
3 gilroy array #3 ilomaprieta 10/18 cdmg 47381 90 0.1 W
/1989
4 gilroy array #4 lomaprieta 10/18 cdmg 57382 90 0.2 30
/1989
5 holuster city Lomaprieta 10/18 usgs1028 180 0.1 30
hall /1989
6 salinas john & lomaprieta 10/18 cdmg 47179 160 01 30

49
work' /1989
7 sunnyvalcolton lomaprieta 10/18 usgs1695 270 0.1 40
ave /1989
■8 ' parkheld- ■■ - Ш® ..36 o-2- Щ
cholmew:' ? .coalINGA.: 0:'
-'.
9 HOLLETER «Ш ■ljsgs 1028' 1 ' ■ V"
ClT f HA.L о.г::' i:9:.
S

1.5 2 2.5 3 3.5


PERIOD (sec)
Figure (4): Selected accelerograms acceleration spectra

50
1.5 2 2.5

PERIOD (sec)
Figure (5): A case of design and resulted spectrum from formula (1)
Preliminary design
(2)
Preliminary design of friction dampers has been performed by equivalent linear static method
using reduction factor R=12. The value of reduction factor has been chosen according to high
performance of friction damper compared to highly ductile structural systems recognized by
seismic codes. The following formulation has been developed to estimate the plastic resisting
moment of each of friction connections, Pu being resulted from equivalent static method.
Pub2l
MP =
4 (b2 +12 )
Parameters in formula (2) are shown in figure (6).
Pu-

51
1

Figure (6): Configuration of friction braced frames and related parameters

52
Nonlinear time - history analysis results
Analysis of the 2 story structure has been performed considering 2 cases of with / without P-
Л (large deformation) effect. According to these studies 26 and 32% of increase in floor
displacements have been resulted from P- Л in 1st and 2nd levels respectively. Furthermore a
decrease of about 5% in energy dissipation of dampers has been resulted from P- Л effects.
The essential parameter to control the performance of structures under severe earthquake
effect scenarios, is the allowable or maximum damage index, modeling the maximum
robustness of structures in their post elastic status. Regarding the fact that an unlimited
energy absorption capacity may be considered actually for Pall dampers, according to
observation of tests, the limitations suggested by Iranian code to control non - structural
damages have been accepted as allowable damage index. 3% and 0.5% relative
drift I = Interstory Drift have been considered as allowable damage indices for 5% and 50% of
probability
^ Story Height J

of exceedence hazard levels respectively. In order to provide the structure with an almost
uniform relative drift demand along its' height, a balancing of stories strength has been
performed. The stories drift demand, before and after balancing of strengths has been shown
in figure (7).

N \
4 \ \

53
RELATIVE FLOOR DRIFT
§5
оt*

(a): 2 story structure (b): 3 story structure


Figure (7): Distribution of relative story drift (indices beside MP are respectively from left to
right, indicating the percent of initial strength of 1st and 2nd stories used in balancing process)
As it is shown in figure (8) the balancing in story strength leads to a balancing in story energy
dissipation demand, while the total energy dissipation demand remains almost the same in all
strength arrangements. It is notable that all results have been derived using nonlinear
analysis under 9 accelerograms processed using procedure of formula (1).

(a): 2 story structure (b): 3 story structure


Figure (8): Distribution of energy dissipation demand of stories

54
In order to model the variation of energy imparting of accelerograms, due to variation in
effective structural period resulted from nonlinear behavior of structures or due to errors of
period calculation, ± 20% changes in structural period have been performed and analyses
have been repeated for each case. Analyses showed variations less than 0.5% and 5% in
relative story drifts and dissipation energy demands respectively.

(b): Cyclic test of damper (c): A residential building structure


equipped with Friction Damper [9] Figure (9): Friction damper device application (Iran)

55
CONCLUSIONS
It is obvious that increases in energy dissipation of structural systems is an essential strategy
to withstand the effect of probable earthquakes. Due to limited energy dissipation capacity of
actual steel material, friction connections with unlimited energy dissipation capacity observed
in cyclic tests, maybe considered as a favorite alternative for conventional structural systems.
An increase of about 30% in relative drifts has been resulted from large deformation (P- Л)
effects.
Decrease in higher story strengths provides them with higher potential of energy dissipation.
An optimum arrangement of story strengths provides the structure with an almost uniform
relative drift and energy dissipation demands in stories of the structure.
The equivalent static method along with a modified distribution of seismic loading resulted
from balancing of strengths (in order to reach to a uniform relative drift demand) has been
evaluated very effective and accurate, regarding results drawn out of nonlinear time history
analyses.
A dimensional ratio (aspect ratio) of damper frame (figure 6) proportional to the braced frame
one, will lead to highest level of energy dissipation potential. An unproportional aspect ratio
will lead to an instantaneous instability and lack of energy dissipation potential.
Neither a single nonlinear dynamic method of analysis nor unique parameters of a nonlinear
method (like в in Wilson - в method) shall be used to analyse the nonlinear behaviors of
structures. A try and error strategy shall be used to inquire the closest responses to exact
quantities. A balance of costs (i.e. time consumation) and benefits (most accurate responses)
shall be provided regarding requirements of projects. Convergence objectives are not
sufficient to satisfy the efficiency and accuracy requirements of analysis. In addition, the
difficulties and uncertainties of accelerogram selection through different strategies, turn the
nonlinear dynamic method to a rather doubtful way of structural analysis than an exact
approach.
AKNOWLEDGEMENT
This Research has been conducted at Niroo Research Institute and has been supported by
Sarzamin Consulting Engineers.

56
REFERENCES
Constantinou, M.C.; Symons, M.P.; Tsopelas, P.; Taylor, D.P.; "Fluid Viscous Dampers in
Applications of Seismic Energy Dissipation and Seismic Isolation", ATC - 17 - 1, ATC, 1993
Kelly, J.M.; Skinner, R.I.; Heine, A.J.; "Mechanisms of Energy Absorption in Special Devices
for Use in Earthquake - Resistant Structures", Bull. N.Z. Nat. Soc. for Earthquake Eng., 5(3),
1972
Filiatrault, A.; Cherry, S.; "Performance Evaluation of Friction Damped Braced Steel Frames
under Simulated Earthquake Loads", Earthquake Spectra, 3 (1), EERI, 1987
Filiatrault, A.; Cherry, S.; "A Simplified Design Procedure for Friction Damped Structures",
Proc., 4USNCEE, Palm Springs, Calif., 1990
Mahmoodi, P.; Keel, G.J.; "Analysis and Design of Multi Layer Viscoelastic Dampers for Tall
Structures", Proc., 7th ASCE str. Cong., San Francisco, Calif., 1989
Chandra, R.; Masand, M.; Naudi, S.K.; Tripathi, C.P.; Pall, R.; Pall, A.; "Friction - Damper for
Seismic Control of La Gardenia Towers South City, Gurgaon, India", 12 WCEE, 2000.
Aiken, I.D.; Kelly, J.M.; Pall, A.S.; "Seismic Response of an Nine-Story Steel Frame with
Friction Damped Cross-Bracing", Report No. UCB/EERC - 88/17, EERC, Univ. of California
Berkeley, 1988
Farzad Naeim.; "Design Ground Motion Criteria for Seismic-Isolated California Hospitals", A
Report to the County of LosAngeles, 1993
Esmailzadeh Hakimi, B.; Tehran's Municipality Project No. 5632, 2000

57
See discussions, stats, and author profiles for this
publication at:
https://www.researchgate.net/publication/266499580
Guidelines for Structural Control
Article ■ January 2006
CITATIONS READS
18 2,056
3 authors, including:
ф MaurerSwitzerland GmbH
75 PUBLICATIONS 1,088 CITATIONS
SEE PROFILE
All content following this page was uploaded by Felix
Weber on 24 November 2014.
ResearchGate
The user has requested enhancement of the
downloaded file.

58
OAMCO
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Guidelines for Structural Control


by
Felix Weber, Glauco Feltrin, and Olaf Huth
Structural Engineering Research Laboratory, Swiss Federal Laboratories for
Materials Testing and Research Dubendorf, Switzerland
Dubendorf, March 2006

О AMCO
CONTENTS
LIST OF FIGURES 4
LIST OF TABLES 9
INTRODUCTION 10
Goal 10
Definition of the term "Structural Control" 10
Structure of guidelines 10
INCREASING DAMPING 11
Damping devices 11
Passive dampers 11
Linear viscous damper 11
Nonlinear viscous damper 17
Viscoelastic damper 20
Coulomb friction damper 26
Structural friction damper 31
Hysteretic damper 33
Shape memory alloy damper 39
Passive tuned mass damper 47
Tuned liquid damper 67
Semi-active devices 71
Defintion of term "semi-active" 71
Magnetorheological/electrorheological fluid damper 73
Controlled shape memory alloy 97
Actuators 100
Actuators for increase of damping 100
Hydraulic aggregate 102
Piezo actuator 105
CONTROL ALGORITHMS113
Control strategies 113
Feedback control 113
Feed forward control 114
Passive control 114

www.samco.org Page 1 of 155


OAMCO
SAMCO Final Report 2006
F05 Guidelines for Structural Control

References 114
Notations 115
Active damping with collocated pairs 117
Linear compensators 117
Active damping with collocated actuator sensor pairs 120
References 126
Notations 126
Optimal control 128
State space representation 128
Linear quadratic regulator 129
Linear Quadratic Gaussian noise control 131
References 136
Notations 137
Other control algorithms 140
Fuzzy control 140
Neural network control 142
References 144
Notations 144
3 VIBRATION ISOLATION 146
Passive vibration isolation 146
Theoretical background 146
Design issues 147
Low tuning 147
High tuning 148
Base isolation 148
Testing and validation 148
Implementation 149
References 149
Notations 150
ACTIVE VIBRATION ISOLATION 151
Scope 151
Passive isolator 151
Active isolator 153
Velocity feedback 153
Force feedback 154
Flexible clean body 154
References 155
Notations 155

www.samco.org Page 2 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

LIST OF FIGURES
Figure 1: Linear viscous damping: a) force displacement and b) force velocity
trajectories for
periodic excitation 12
Figure 2: Nonlinear viscous damping at small velocities due to seal friction
12
Figure 3: Viscous damping device (Connor (2001)) 14
Figure 4: Viscous damper, 450 kN (http://www.taylordevices.com/3seismic.htm)
14
Figure 5: Schematic test set-up 15
Figure 6: Measurement of damper behaviour at Empa 15
Figure 7: Mean power equivalent nonlinear viscous damper 18
Figure 8: Behaviour of: a) pure elastic materials, b) and c) of viscoelastic
materials 20
Figure 9: Example of a viscoelastic damping device according to Connor (2001)
22
Figure 10: Single bearing test machine (Aiken et al. (1989)) 23
Figure 11: Cyclic hysteresis loop of a bolted rubber bearing (Aiken et al. (1989))
23
Figure 12: Bearing system design proposed by Aiken et al. (1989) 24
Figure 13: Coulomb friction: a) force displacement trajectory, b) force velocity
trajectory; c) force velocity trajectory of real Coulomb friction dampers (e.g. MR
dampers at constant
current) 26
Figure 14: Limited Slip Bolted Joints (Pall et al. (1980)) 28
Figure 15: X-braced Friction Damper (Pall and Marsh (1982)) 28
Figure 16: Sumitomo Friction Damper (Aiken and Kelly (1990)) 29
Figure 17: Energy Dissipating Restraint (EDR) (Nims et al. (1993)) 29
Figure 18: Structural damping: a) force displacement trajectory and b) force
velocity
trajectory 31
Figure 19: Hysteretic damping: a) stress strain relation, b) real hysteretic damper
behaviour,
c) idealized hysteretic damper behaviour 33
Figure 20: Ideal elastic-plastic damping device according to Jones (2001) 34
Figure 21: X-shaped Plate Damper 34
Figure 22: Stress strain curves and transformation temperature profiles for
"actuator like"
change of strain and stiffness at constant stress (Janke et al. (2005)) 39

www.samco.org Page 3 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Figure 23: Stress strain curves and transformation temperature profiles for shape
memory effect in the case of free strain recovery and constraint strain recovery
(Janke et al.
(2005)) 40
Figure 24: Stress strain curves and transformation temperature profiles for the
so-called
superelastic behaviour (Janke et al. (2005)) 40
Figure 25: Stress strain curves and transformation temperature profiles for
martensitic
hysteretic damping (Janke et al. (2005)) 41
Figure 26: Hysteresis loops for: a) superelastic SMA behaviour and b) martensitic
hysteretic
damping (Janke et al. (2005)) 41
Figure 27: Torsion bar design (Witting and Cozzarelli (1992)) 41
Figure 28: Stress strain diagram used for bilinear damper model (Witting and
Cozzarelli
(1992)) 44
Figure 29: Force displacement loops (Witting and Cozzarelli (1992)) 44
Figure 30: Typical implementation of a TMD for vertical vibration mitigation of
bridges 48
Figure 31: Two degree of freedom model of a TMD attached to a primary
structure. Left: Excitation force f (t) acting on the primary mass. Right: Excitation
through base
acceleration Hg (t) 49
Figure 32: Methods for H„ norm optimization of a TMD applying the fixed points
method 51
Figure 33: Methods for H„ norm optimization of a TMD applying the real roots
method 51
Figure 34: Effects of poor frequency tuning of a TMD on the dynamic
magnification factor of the displacement of the primary system 52

Figure 35: Effects of poor damping tuning of a TMD on the dynamic magnification
factor of the
displacement of the primary system 52
Figure 36: Optimal TMD frequency for minimizing the displacement of the primary
system with
respect to the H. norm 53
Figure 37: Optimal TMD damping ratios for minimizing the displacement of the
primary

www.samco.org Page 4 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

system with respect to the H. norm 53


Figure 38: Maximum of the non-dimensional transfer function of the displacement
of the primary system for parameters minimizing the displacement of the primary
system with
respect to the H. norm 53
Figure 39: Maximum of the non-dimensional transfer function of the relative
displacement for parameters minimizing the displacement of the primary system
with respect to the H.
norm 53
Figure 40: Mitigation factor of a TMD optimized for minimizing the displacement
of the primary
system with respect to the H. norm 53
Figure 41: Amplification due to a perturbation of frequency tuning of a TMD
optimized for minimizing the displacement of the primary system with respect to
the H. norm
(S = Vt 1, H = 0.02) 53
Figure 42: Normalized variance of the displacement of the primary system
55
Figure 43: Optimal TMD frequency for minimizing the variance of the
displacement of the
primary system with respect to the H2 norm 56
Figure 44: Optimal TMD damping ratios for minimizing the variance of the
displacement of the
primary system with respect to the H2 norm 56
Figure 45: Minimum of the non-dimensional variance of the displacement of the
primary
system for parameters minimizing the variance of the displacement of the primary
system
with respect to the H 2 norm 56
Figure 46: Non-dimensional variance of the relative displacement of the mass of
the TMD for parameters minimizing the variance of the displacement of the
primary system with
respect to the H 2 norm 56
Figure 47: Mitigation factor of a TMD optimized for minimizing the displacement
of the primary
system with respect to the H2 norm 56
Figure 48: Amplification due to a perturbation of frequency tuning of a TMD
optimized for minimizing the displacement of the primary system with respect to
the H2 norm
(S = Vt / П.-1, M = 0-02) 56

www.samco.org Page 5 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Figure 49: Schematic illustration of a sloshing (a) and a column tuned liquid
damper (b)
coupled to a primary structure 67
Figure 50: Principal parameters of sloshing and column TLD 67
Figure 51: Definitions and characteristics of damping devices often used in
structural control.71
Figure 52: Rotational MR damper, force range approximately [10, 320] N 73
Figure 53: Cylindrical MR damper with gas pocket (accumulator) for
compensation of piston
volume; force range approximately [70, 1800] N 73
Figure 54: Schematic steady-state damper behaviour of MR dampers 74
Figure 55: Measured MR damper behaviour in pre- and postyield regions (MR
damper of
Maurer Sohne) 74
Figure 56: Force displacement trajectories including transient measurement data
from
preyield to postyield regions (MR damper of Maurer Sohne) 74
Figure 57: Simple Bouc-Wen modelling approach for MR-dampers 76
Figure 58: Fitted inverse MR damper function 76
Figure 59: Testing damping devices 77
Figure 60: Transient measurement data due to MR damper behaviour in preyield
and
postyield regions with force overshoot, fairly small MR fluid viscosity 77
Figure 61: Transient measurement data due to MR damper behaviour in preyield
and
postyield regions, no force overshoot, fairly large MR fluid viscosity 77
Figure 62: MR damper force response on approximate velocity step at 0 A
78
Figure 63: MR damper force response on approximate velocity step at 4 A
78
Figure 64: Cable-stayed "Dongting Lake Bridge" in China, equipped with MR
dampers of
LORD Corporation 79
Figure 65: Cable-stayed "Eiland Bridge" nearby Kampen, The Netherlands,
equipped with one
MR damper of Maurer Sohne for long-term field tests and damping
measurements 79
Figure 66: Cable-stayed „Dubrovnik Bridge" nearby Dubrovnik, Croatia 80
Figure 67: Flow chart of damper design procedure 82
Figure 68: Cable/damper system with distributed disturbance force according to
the shape of

www.samco.org Page 6 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

the excited mode (here shown for the first vibration mode) 82
Figure 69: Characteristics of linear viscous and friction dampers 83
Figure 70: Force displacement trajectories of MR damper of the "Eiland Bridge"
84
Figure 71: Force current relation of MR damper of the "Eiland Bridge" 84
Figure 72: Instrumentation of MR damper on the "Eiland Bridge" nearby Kampen,
The
Netherlands 85
Figure 73: Taking decay measurements on the "Eiland Bridge" nearby Kampen,
The
Netherlands 85
Figure 74: Sketch of decay measurements on the "Eiland Bridge" nearby
Kampen, The
Netherlands 86
Figure 75: Clamping of cables 86
Figure 76: Measured vibration decay at 12% cable length and damper position
with MR
damper at 0 A 86
Figure 77: Measured vibration decay at 12% cable length and damper position
with MR
damper at 0.4 A 86
Figure 78: Measured vibration decay at 12% cable length and damper position
with MR
damper at 2.0 A 87
Figure 79: Measured vibration decay at 12% cable length and damper position of
free cable.87 Figure 80: Exponential approximation of linear decay within the
amplitude range from 95% to
50% 88
Figure 81: Evaluated mean logarithmic decrement 89
Figure 82: Schematic controller structure 89
Figure 83: Displacement sensor and MR damper on the "Eiland Bridge" nearby
Kampen, The
Netherlands 90
Figure 84: Solar panel and controller box mounted on the lightning pylon on the
"Eiland
Bridge" nearby Kampen, The Netherlands 90
Figure 85: Emulation of desired damping characteristics using feedback
controlled actuators.92
Figure 86: Characteristics of real "linear viscous dampers" 92
Figure 87: Stress-strain curves for the superelastic behaviour of SMAs at different
temperatures98

www.samco.org Page 7 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Figure 88: Different moduli of elasticity for martensite and austenite 98


Figure 89: Hydraulic aggregate, mounted on a steel frame 102
Figure 90: Schematic of a piezo-electric single layer element
(http://www.piezomechanik.com/)106
Figure 91: Schematic of an axially acting multilayer piezo stack
(http://www.piezomechanik.com/)106
Figure 92: Schematic of a stacked ring actuator (http://www.piezomechanik.com/)
107
Figure 93: Schematic of an actuator with prestress
(http://www.piezomechanik.com/)107
Figure 94: Schematic of a piezo actuated system
(http://www.piezomechanik.com/)108
Figure 95: Schematic voltage/stroke diagram of a stack actuator
(http://www.piezomechanik.com/)108
Figure 96: Schematic voltage force relation of a piezo stack
(http://www.piezomechanik.com/)109

Figure 97: Maximum piezo actuator stroke versus maximum (blocking) piezo
actuator force
(http://www.piezomechanik.com/)109
Figure 98: Approximately linear relation between stroke and charge content of
piezo actuators
(http://www.piezomechanik.com/)110
Figure 99: Creep of piezo actuators (http://www.piezomechanik.com/) 110
Figure 100: Closed-loop structure of feedback control 113
Figure 101: Closed-loop structure of feedback control including actuator and
sensor
dynamics 113
Figure 102: Structure of feed forward control 114
Figure 103: Open-loop structure of passive control 114
Figure 104: Structure of parallel PID controller with inverse plant model NL-1
117
Figure 105: Bode diagram of P, I, D elements, and of low pass filter of 2nd order
117
Figure 106: Bode diagram of a parallel PID controller with low pass filter of 2nd
order for a
good "roll-off" 118
Figure 107: Characteristics of Lead and Lag elements shown in the Bode
Diagram 118

www.samco.org Page 8 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Figure 108: Plant of 3rd order with proportional gain only 120
Figure 109: Root Locus of 3rd order plant with feedback proportional gain KP
only 120
Figure 110: Closed-loop structure of direct velocity feedback for collocated
actuator sensor
pairs 121
Figure 111: Four mass system 121
Figure 112: Transfer function of four mass system with force input and velocity
output at mass
no.1 122
Figure 113: Placement of closed-loop pole for maximum additional damping of
mode 1 (direct
velocity feedback) 122
Figure 114: Comparison between very low damped plant and plant with optimally
damped
mode 1 123
Figure 115: Location of closed-loop poles of all four modes for maximum
damping of
mode 1 123
Figure 116: State space representation of the plant 129
Figure 117: Plant in SSR, all state variables measured, full state feedback with
stochastic LQ
Regulator 129
Figure 118: Real plant with full state observer and regulator 135
Figure 119: Observer and regulator design 135
Figure 120: Schematic operating range of magnetorheological fluid dampers
136
Figure 121: Inverted pendulum on a cart 141
Figure 122: Structure of a fuzzy controller 141
Figure 123: The steps of fuzzification, inference, and defuzzification 142
Figure 124: Black box identification by training of neural networks 143
Figure 125: Neural network consisting of neurons (weights and transfer function)
and
layers 144
Figure 126: Schematic representation of vibration isolation systems. Left:
Isolation reducing the effects of the excitation force acting on the structure. Right:
Isolation reducing the
effects of a forced excitation of the support 146
Figure 127: Force transmissibility for an isolator consisting of a linear spring and
dashpot
acting in parallel for different damping ratios 147

www.samco.org Page 9 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Figure 128: Rubber bearing applied for base isolation149


Figure 129: Shear displacement test of a rubber bearing for base isolation
149
Figure 130: i) Operating system (1) generating disturbance forces. ii) Sensitive
device (1) excited by vibrating support structure (2). iii) Active vibration isolation
with velocity
feedback and force actuator ("skyhook damper") 151
Figure 131: Magnitude of transmissibility for different damping ratios 153

www.samco.org Page 10 of 155


OAMCO
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Figure 132: i) Velocity feedback for flexible clean body. ii) Force feedback for
flexible clean body 153
LIST OF TABLES
Table 1: Non-dimensional transfer functions for various excitations and response
parameters 62
Table 2: Approximate solutions of the H. optimization problem for primary
systems with
vanishing structural damping (Z p = 0) 63
Table 3: Closed form expressions of the non-dimensional H2 norm for various
excitations and
response parameters 64
Table 4: Exact solutions of the H2 optimization problem for primary systems with
vanishing
structural damping (Zp = 0) 65
Table 5: Approximate solutions of the H 2 optimization problem for primary
systems with
structural damping (Zp > 0) 66
Table 6: Properties of longest stay cable of "Eiland Bridge" 84
Table 7: Possible fuzzy rules for inverted pendulum 141
Table 8: Fuzzy logic operations 142
Table 9: Properties of critically damped active isolator 154

6 AMCO

1 INTRODUCTION
Goal
The goal of this guideline is to give an overview of damping devices and control
algorithms that are often used in the field of structural control of civil structures.
From this point of view, the guideline may represent a helpful document and a
source of ideas and references for engineers and scientists working in this
research area.
Definition of the term "Structural Control"
This guideline focuses on "Structural Control" of civil and large scale mechanical
structures but not on macro or micro sized structures. Prestressing of structures
using strain actuators in order to produce a desired shape or prestress of the
structure will not be treated within this document although shape control belongs
to the wide field of structural control. The presented guideline describes the
common damping devices and control methods applied for suppression of

www.samco.org Page 1 of 155


OAMCO
SAMCO Final Report 2006
F05 Guidelines for Structural Control

undesired structural vibrations that occur after the erection of the structure.
Hence, this guideline describes the tools for vibration mitigation of civil structures.
Structure of guidelines
In the second chapter, the guideline describes control devices and control
methods that are implemented in order to increase the damping of structures.
This is the case when additional dampers or actuators, respectively, are attached
to vibrating structures and are controlled by a control law. The control algorithms
may be part of an open loop system or may work in a closed loop, which allows
for controlling the device force according to the actual vibration state.
In contrast to the second chapter, the third chapter gives an overview of vibration
isolation systems. Here, the target is not to enhance the structural damping using
external damping devices but to isolate sensitive structures from disturbances.
This chapter is divided into two parts. First, passive isolators are described, then,
active vibration isolation is presented.

www.samco.org Page 2 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

2 INCREASING DAMPING
Within the first subchapter, an overview of existing damping devices is given.
These damping devices include passive dampers, controllable dampers, and
actuators. Passive dampers are characterized by constant damping properties
given by their materials and construction. Controllable dampers are often called
semi-active devices since their damping force may be adjusted or controlled,
respectively. Actuators comprise such devices that are able not only to dissipate
but also to generate energy. Thus, actuators are able to put energy into the
vibrating structure.
The second subchapter describes the control algorithms that are usually applied
when controllable dampers and actuators are used for vibration control. First, the
main control strategies such as feedback, feed forward and passive control are
shortly introduced. Then, active damping with collocated actuator sensor pairs is
described, followed by the optimal control approach within the third subchapter.
Finally, the most common, other linear control approaches used for structural
control are presented.
2.1 Damping devices
In the following, the physics of the most common damping devices are described.
The damping devices are split up in the sections passive damping devices, semi-
active damping devices, and actuators.
2.1.1 Passive dampers 2.1.1.1 Linear viscous damper Theoretical background
The equation of motion of a single degree of freedom system (SDOFS) with
external damper force fd and external disturbance force fw is
m ■ x + c ■ x + к ■ x = fd + fw (1)
where m denotes the SDOFS mass, c the SDOFS eigendamping and к the
SDOFS stiffness. The viscous damper is a hypothetical device that creates a
damping force which is directly proportional to the SDOFS velocity but with
opposite sign (Fig. 1)
fd =~cd ■ X (2)
The proportionality factor cd is the damping coefficient of the external viscous
damper. The reason why pure linear viscous dampers may be a preferable
damping device demonstrates the following equation, which results from the
substitution of Eq. (2) in Eq. (1)
m ■ X + (c-bcj-X + к ■ x = fw (3)
Ctot
This equation is again linear and therefore analytically solvable. Moreover, the
main benefit of external viscous dampers is that they seem to enhance the
material damping of the structure, in this case of the SDOFS, which is in good

www.samco.org Page 3 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

approximation pure viscous damping. The external viscous damper cannot


compensate for the disturbance force fw since the disturbance force fw
representing excitation forces caused by wind, earthquakes and other impacts is
unknown.
Viscous damping devices are based on material damping which may be assumed
to behave linear viscous. However, it must be mentioned that most dampers
classified as "viscous dampers" do not behave completely linear over the entire
velocity range due to sealing friction and nonlinear material behaviour which ends
up in a nonlinear viscous behaviour at small velocities (Fig. 2).
The work W dissipated within the time interval [t1,12] can be formulated as
h
Wi_2 = J fd ■ x ■ dt (4)
ti
For periodic excitation, the energy dissipated per cycle becomes (Bachmann et
al. (1995), Weber (2002))
W = П ■ f max ■ Xmax = П ' Cd ' Xmax (5)
linear viscous damper'

Fig. 1: Linear viscous damping: a) force displacement and b) Fig. 2: Nonlinear


viscous damping at force velocity trajectories for periodic excitation. small
velocities due to seal friction.

Design issues
In practice, in order to produce damping devices with almost linear viscous
characteristics, dampers on the base of fluids are produced. Basic principles of
balance laws and explanations about inviscid fluids, Newtonian fluids, and
temperature dependent Newtonian fluids are given in the books of Soong and
Dargush (1997) and Constan- tinou et al. (1998).
A possible design for a viscous damping device is shown in Fig. 3 (Connor
(2001)). The gap between plunger and external plates is filled with a linear
viscous fluid which can be characterized by
t = -Gv 'Y (6)

www.samco.org Page 4 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

The variables т and у are the shear stress and strain rate, respectively, and Gv is
the viscosity coefficient. In case that slip does not exist between fluid and
plunger, the shear strain is related to the plunger motion by
Y =x (7)
t,
d
where td represents the thickness of the viscous layer. With L as the wetted
length and w as the width of the plunger, the damping force may be written as
fd = 2wLt (8)
Substituting т by Eq. (6) and у by Eq. (7) in Eq. (8) yields

{ 2wLGv Л
V td J
fd =-
x (9)

According to Eq. (2), the viscous damping coefficient of the damping device
becomes
2wLG
cd = ^ (10)
td
Hence, the design parameters of such linear viscous dampers based on fluids
are:
thickness td of the viscous layer,
the wetted length L,
the width w of the plunger, and
the fluid viscosity coefficient Gv.
The design of viscous dampers (Fig. 3, Fig. 4) in order to enhance the overall
viscous damping of, e.g., civil structures, comprises the following steps and may
need iterations to meet the requirements:
Determination of structural properties of the civil structure and analysis of the
structure.
Determination of the desired damping ratio.
Spatial placement of the damping devices.
Design of the viscous fluid damper.

www.samco.org Page 5 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Estimation of structural damping ratio.


Analysis of the overall structural behaviour with the enhanced damping ratio.

www.samco.org Page 6 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

www.samco.org Page 7 of 155


OAMCO
SAMCO Final Report 2006
F05 Guidelines for Structural Control

If the dampers are well distributed over the structure, the design process can be
reduced primarily to the selection of the desired damping ratio. The optimal
locations are found if the results of structural analysis show that the targeted
mode/modes are reduced maximally in amplitude. However, to find the optimal
damper locations is a challenging and nontrivial issue. Since the implementation
of dampers requires bracings, the structural stiffness and therefore its resonance
frequencies are also affected. The design methodology has to be modified in
presence of nonlinearities.
The physical properties of thick viscoelastic fluids are often quite sensitive to tem-
perature changes. Therefore, temperature compensation is often included in the
form of a bi-metallic orifice working similarly to a thermostat (Fig. 4). The thermal
stability is combined with the steel construction that has internally threaded joints
and no welded or bonded parts.
A critical issue in the design of orifice dampers concerns the durability of the high
strength seals. Additionally, these of the high strength seals may be responsible
for nonlinear viscous damping at small piston velocities since the sealing friction
behaves like the force offset of friction dampers (Fig. 2).
Usually, typical buildings have internal critical structural damping of 1% to 3%.
The critical damping of buildings should be increased by external viscous
dampers to approximately 20% to 25% in order to guarantee that the building is
not susceptible to vibrations of large amplitudes anymore. As a comparison, most
conventional passenger cars use dampers with 20% to 30% percent of critical
damping.

www.samco.org Page 8 of 155


OAMCO
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Testing and validation


The damper force characteristics are obtained with a simple testing procedure.
The damping device is clamped between an actuator and a fixed panel, with a
force transducer in series connected (Fig. 5, Fig. 6). Due to the sinusoidal
displacement control of the actuator, the damping device is tested from zero
velocity to maximum velocity. The first derivation of the measured displacement
delivers the velocities. In conjunction with the measured force, the damping
characteristics of the tested device are known. In case that the force transducer is
moving together with the damping device, the measured force value has to be
compensated for the inertia term due to the sensor's mass.

Fig. 5: Schematic test set-up. Fig. 6: Measurement of damper behaviour at


Empa.

There exist dampers like, e.g., the orifice fluid damper (Constantinou and Symans
(1993)), which behave as linear viscous dampers if operating at low frequencies.
At high frequencies, many dampers based on fluids show also elastic force values
caused by the fluid stiffness. These dampers are then called viscoelastic dampers
(Jones (2001), see chapter 2.1.1.3).
Implementation

www.samco.org Page 9 of 155


OAMCO
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Viscous fluid dampers have been successfully used for structural control in the
fields of:
Automotive industries.
Bridge design, foundation of superstructures.
High buildings for the alleviation of wind and seismic vibrations (Martinez-Romero
and Romero (2003)).
References
Bachmann, H., et al. (1995), Vibration Problems in Structures: Practical
Guidelines, Birk- hauser Verlag Basel, ISBN 3-7643-5148-9.
Connor, J. J. (2001), Introduction to Structural Motion Control,
http://www.mit.edu/afs/athena/course/171.561/Book/WebBook-July-2001.
Constantinou, M. C., and Symans, M. D. (1993), "Experimental Study of Seismic
Response of Buildings with Supplemental Fluid Dampers", Structural design Tall
Buildings, 2.
Constantinou, M. C., Soong, T. T., and Dargush, G. F. (1998), "Passive Energy
Dissipation Systems for Structural Design and Retrofit", Multidisciplinary Center
for Earthquake Engineering Research.
Jones, D. I. G. (2001), Handbook of Viscoelastic Vibration Damping, John Wiley &
Sons LTD.

www.samco.org Page 10 of 155


OAMCO
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Martinez-Romero, M., and Romero, M. L. (2003), "An optimum retrofit strategy for
moment resisting frames with nonlinear viscous dampers for seismic
applications", Engineering Structures, 25, 913-925.
Soong, T. T., and Dargush, G. F. (1997), Passive Energy Dissipation Systems in
Structural Engineering, John Wiley & Sons, New York.
Weber, B. (2002), "Damping of vibrating footbridges", Proceedings of the
International Conference on Footbridge, Paris, France, 20-22 November 2002, on
CD, AFGC - OTUA (eds.).
http://www.taylordevices.com/3seismic.htm
Notations
Symbol Description Unit
Gv viscosity coefficient Ns/m2
L length m
W work, energy J
c viscous damping coefficient kg/s
f force N
к stiffness kg/s2
m mass kg
t thickness m
w width m
x displacementm
Y shear strain -
т shear stress N/m2
Subscripts
d
max
tot
w
damper
maximum
total

www.samco.org Page 11 of 155


OAMCO
SAMCO Final Report 2006
F05 Guidelines for Structural Control

disturbance

www.samco.org Page 12 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

2.1.1.2 Nonlinear viscous damper Theoretical background


The force fd of a damper with nonlinear viscous damping behaviour may be
expressed as follows
I IП
fd _—Cd—nonlin ■ N • Sgn(X) (1)
where x is the relative velocity between the two damper ends, cd—nonlin is the
viscosity of the
nonlinear damper and n is a frictional power law exponent (Fig. 7, Tsai and
Popov (1998),
Weber (2002)). The case of n _ 1 describes the linear viscous damper behaviour.
Velocity
exponents smaller than one are often chosen because of the fast increase of the
damping force at low velocities (Fig. 2).
The main drawback of this kind of damper is that it may clamp the structure at
small vibration velocities due to the relatively large damper force in the case of n
< 1. If the structure is clamped at damper position, damper relative displacement
and therefore energy dissipation become zero. Then, the structure is undamped
and the structure at damper position becomes a nodal point. Consequently, the
entire excitation energy excites the rest of the structure.
Design issues
In the case of linear viscous dampers, the optimum viscosity may be determined
using the Matlab tool "root locus" (Preumont (2002)), which determines the
optimum viscosity depending on:
the structural properties,
damper position, and
the target mode that shall be damped maximally.
However, this tool fails for nonlinear viscous dampers. In this case, two possible
design strategies are presented.
Martinez-Rodrigo and Romero (2003) proposed a trial and error method, where
the maximum damper force of the nonlinear viscous damper is chosen to be

www.samco.org Page 13 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

smaller than its counterpart of the linear viscous damper. The relation is
expressed by the so-called force index FI.

fd—nonlin— max fd—lin—


max
1—
FI
:100 [%] (2)

In a first step, the maximum required damping force is estimated assuming


linear viscous damping. Then, the maximum force of the nonlinear viscous
damper is determined using Eq. (2). As a next step, the structure is damped with
the nonlinear viscous damper with a chosen value of the nonlinear exponent n.
Both damper parameters, FI and n (note: the parameter
clin is already given by the optimum linear viscous damper, Krenk (2000)), are
varied within
an optimization procedure until the optimization criterion is fulfilled. Possible
optimization criteria the may be:
maximum damping of one target mode,
maximum damping of several target modes, or
minimum tip displacement of a high rise building for a given excitation time
history (earthquake).

www.samco.org Page 14 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 7: Mean power equivalent nonlinear viscous damper.

Another design procedure is proposed by Pekcan et al. (1999). The idea is to


determine a mean power equivalent nonlinear viscous damper based on the
optimum linear viscous damper. The damper energy per cycle is
W = J fd ■ x ■ dt
0
The mean power over one cycle is the first time derivative, hence
(3)
P = ^Tt = Jt \jfd x ■ dt\ = Jfd ' dx =4 ■ dJfd ' dx =4 ■ (grey area ^

(4)

The mean power over one cycle of the optimal linear viscous
damper is

Plin = 4 ■
(5)

www.samco.org Page 15 of 155


SAMCO Final
Report 2006
F05
Guidelines for
Structural
Control

Copt ■ x16
d-lin lin-max

16

www.samco.
org Page
16 of 155
SAMCO Final Report 2006
F05 Guidelines for Structural Control

The mean power over one cycle of the nonlinear viscous damper is
- C ■ xv+1
(6)
л d-nonlin nonlin-max
2
(7)
2
П +1
nonlin — 4 ■
П +1
Equating both mean power values yields the tuning of mean power equivalent
nonlinear viscous dampers
■ n+1

www.samco.org Page 17 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

The formula shown in Eq. (7) indicates that the mean power equivalent nonlinear
viscous damper still needs the choice of one of its design parameters and the as-
sumption of equal maximum damper velocities.
References
Krenk, S. (2000), "Vibrations of a Taut Cable With an External Damper", Journal
of Applied Mechanics, 67, 772-776.
Martinez-Romero, M., and Romero, M. L. (2003), "An optimum retrofit strategy
for moment resisting frames with nonlinear viscous dampers for seismic
applications", Engineering Structures, 25, 913-925.
Pekcan, G., Mander, J. B., and Chen, S. S. (1999), "Fundamental considerations
for the design of nonlinear viscous dampers", Earthquake Engineering and
Structural Dynamics, 28(11), 1405-1425.
Preumont, A. (2002), Vibration Control of Active Structures, Kluwer Academic
Publishers, Dordrecht.
Tsai, K. C., and Popov, E. P. (1998), "Steel beam-column joints in seismic mo-
ment resisting frames", Berkeley, CA, Earthquake Engineering Research Center,
University of California, 1998, report No. UCB/EERC-88/19.
Weber, B. (2002), "Damping of vibrating footbridges", Proceedings of the
International Conference on Footbridge, Paris, France, 20-22 November 2002,
on CD, AFGC - OTUA (eds.).
Notations
Symbol Description Unit
FI force index %
P power W
W work, energy J
c viscous damping coefficient kg/s
f force N
X displacement m
п frictional power law exponent -

Superscripts

www.samco.org Page 18 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

opt optimum
a mean value of a

Subscripts

d damper
lin linear
nonlin nonlinear
max maximum

www.samco.org Page 19 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

2.1.1.3 Viscoelastic damper Theoretical background


The total force of a viscoelastic damper is the some of its viscous and stiffness
parts (Jones (2001), Weber (2002))
fd = kd ■x - cd ■x (1)
The stiffness part causes that the ellipse curve in the force displacement map
has a mean slope of kd (Fig. 8). The force velocity trajectory is also an ellipse
curve. Its
mean slope still indicates the damper viscosity cd . The dissipated energy over
one cycle is

Td
(2)
E
= J fd ■x ■dt
0

In contrast to pure linear viscous dampers, the force velocity map does not only
show dissipative but also active force values due to the damper's stiffness (within
quadrants I and III in Fig. 8). Viscoelastic behaviour is typical for material
dampers, not fluid dampers, since materials can withstand constant deformations.

www.samco.org Page 20 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 8: Behaviour of: a) pure elastic materials, b) and c) of viscoelastic materials.

The stress strain relationship of a linear elastic material is


(3)
т = Ge ■y

www.samco.org Page 21 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

where т and у are the shear stress and strain, respectively, and Ge is the elastic
shear modulus. The stress strain curve is depicted in Fig. 8. For the elastic
behaviour, time lag between stress and strain and force and displacement,
respectively, does not exist. The proportionality factor of course is Ge. In contrast,
stress is п/2 radians out of phase with strain for linear viscous damping only. The
linear combination of both produces the behaviour shown by the viscoelastic
trajectories in Fig. 8 with a phase between stress and strain between 0 and п/2
according to the linear combination.
The hysteresis loops deliver an initial base for analytical modelling the
viscoelastic material behaviour as a function of strain, frequency and
temperature. For the periodic excitation, the basic strain stress relations may be
written as follows
Y=Ymax • sin(ffl-1) (4)
T = Ymax • [Gs • sin(ffl-1) + Gl • cos(ffl-1)] (5)
where Gs is the storage modulus and Gl is the loss modulus. The ratio of loss
modulus and storage modulus is called loss factor
G
n = -f = tan(5) (6)
G
The angle 5 signs the phase shift between stress and strain and ranges from 0
for elastic behaviour to п/ 2 for pure viscous behaviour. Equation Eq. (5) written in
an alternative form yields
T = Ymax • Gmax • Sm(^ t + 5) (7)
Gmax (8)
Chang et al. (1993) performed several tests of three types of viscoelastic
dampers. They concluded that the storage modulus and the loss modulus are
functions of ambient temperature, material temperature, excitation frequency and
shear strain. During damper operation, the energy is dissipated in form of heat
producing temperature rise in the viscoelastic material. Therefore, the damper
properties depend - up to a certain degree - on the number of loading cycles and
the range of deformation, especially under large strain due to temperature
increase of the damper material.

www.samco.org Page 22 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Bagley and Torvik (1983) characterized the frequency and temperature


dependent shear storage modulus and shear loss modulus for most viscoelastic
materials. Ferry (1980) proposed the method of reduced variables to determine
the dependence of Gl and Gs on ambient temperature. This method is based on
a simplification in
separating the two variables, frequency and temperature, on which the
viscoelastic properties depend. These properties are expressed in terms of a
single function of each.

www.samco.org Page 23 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

www.samco.org Page 24 of 155


OAMCO
SAMCO Final Report 2006
F05 Guidelines for Structural Control

side view top view

Fig. 9: Example of a viscoelastic damping device according to Connor (2001).

A serious problem with viscoelastic devices is an increase in force at low


temperatures coupled with an accompanying overloading of the bonding agent
used to glue the viscoelastic material to its steel attachments.
Design issues
There exists a large variety of different damper design procedures, which are col-
lected in Soong and Dargush (1997).
The damper design procedure for application in structures follows approximately
the steps described for pure viscous dampers. Aiken et al. (1989) reported that
the general design methodology for elastomeric bearings for seismic isolation is
substantial different from the currently used for bridge bearings, in particular in
aspects of slen- derness, rated loads and lateral displacements.
(9)
A possible viscoelastic damper design with bonding steel sheets of a viscoelastic
material to steel plates is illustrated in Fig. 9 (Connor (2001)). The equation of
shear strain rewritten becomes
X
Y=
t
d
For a given shear strain у, one determines т with the stress strain relation. Then,
the damping force may be estimated as
fd = 2 ■ w ■ L т (10)
For periodic excitation and substituting т from Eq. (4), one obtains
(11)
(12)
fd = 2 ■W ■L ■ Gs ■ Xmax - [sin(ffl-1)+n- cos(ffl-1)]
t

www.samco.org Page 25 of 155


Ьамсо
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Then, the dissipated energy per cycle becomes


TI_ 2 ■ w ■ L 2
W =— ПП^ Gs ■ xma
t

www.samco.org Page 26 of 155


Ьамсо
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Testing and validation


The testing procedure of a viscoelastic damper needs one step more compared
to the testing of pure viscous dampers because the elasticity has to be identified
too. The damping device is clamped between actuator and fixed panel with a
force transducer connected in series (Fig. 10). The spring constant kd can be
easily identified
by applying a constant load and measuring the resulting displacement. With the
known spring constant, the testing procedure follows the one of viscous dampers
(Fig. 11).
If the actuator is displacement controlled, then the resulting force represents the
measured system output. The first derivation of the displacement delivers the
velocity. Together with the measured force, the damping coefficient cd may be
identified
as the mean slope of the force velocity trajectory (Fig. 8). Please note that the
measured total damper force has to be reduced by the inertia force of the sensor
if the force transducer is moving together with the damping device.
Implementation
Basically, most material dampers behave viscoelastic. Even some fluid dampers
show viscoelastic behaviour due to the fluid stiffness although the force due to
their stiffness is small compared to the force resulting from their viscosity.
Therefore, numbers of viscoelastic dampers have been implemented on real civil
structures for structural control. They have been used in the fields of:
vibration control and vibration mitigation of sensitive machines (e.g. nuclear
industry and aerospace technology),
base isolation systems, also foundation of superstructures (Huffmann (1985)),
damping of vibrations of high rise buildings, and
damping of bridge deck and cable vibrations (elastomeric bearings).
Elastomeric bearings
W

www.samco.org Page 27 of 155


Ьамсо
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 10: Single bearing test machine (Aiken et al. (1989)).


Fig. 11: Cyclic hysteresis loop of a bolted rubber bearing (Aiken et al. (1989)).
The bearing systems which have been implemented in buildings are either
systems which include additional devices to enhance the overall damping of the
isolation system or do not have extrinsic devices which increase the system
damping. The latter type can be of the lead-rubber type, in which elastomeric
bearings contain a lead-

www.samco.org Page 28 of 155


Ьамсо
SAMCO Final Report 2006
F05 Guidelines for Structural Control

www.samco.org Page 29 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

plug insert to supplement damping. Another realization is the filled type, in which
a filler material is added to the rubber to enhance the damping and stiffness
properties of the compound. A bearing design, proposed by Aiken et al. (1989), is
shown in Fig. 12. The bearing consists of three rubber layers separated by two
steel shims and two end plates. An external rubber cover layer is used for
protection. The material used for the bearings is a filled, natural rubber with high
damping.
Kelly and Edgardo (1992) reported that if an isolation system uses high strength
elastomers, such as polychloroprene rubber, connected with high quality bonding
techniques, the ultimate capacity can be accurately predicted and very
substantial safety margins can be established.

Fig. 12: Bearing system design proposed by Aiken et al. (1989).

References
Aiken, I. D., Kelly, J. M., and Tajirian, F. F. (1989), "Mechanics of Low Shape
Factor Elastomeric Seismic Isolation Bearings," Report No. UCB/EERC-89/13,
Earthquake Engineering Research Center, College of Engineering, University of
California at Berkeley.

www.samco.org Page 30 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Bagley, R. L., and Torvik, P. J. (1983), "Fractional Calculus - A Different


Approach of the Analysis of Viscoelastic Damped Structures", AIAA Journal,
1983.
Connor, J. J. (2001), Introduction to Structural Motion Control,
http://www.mit.edu/afs/athena/course/1/1.561/Book/WebBook-July-2001.
Chang, K. C., Lai, M. L., Soong, T. T., Hao, D. S., and Yeh, Y. C. (1993),
"Seismic Behaviour and Design Guidelines for Steel Frame Structures with
Added Viscoelastic Dampers", NCEER 93-0009, National Center for Earthquake
Engineering Research, Buffalo.
Ferry, J. D. (1980), Viscoelastic Properties of Polymers, John Wiley, New York.

6 AMCO

Huffmann, G. K. (1985), "Full Base Isolation for Earthquake Protection by Helical


Springs and Viscodampers", Nuclear Engineering Design, 84(2).
Jones, D. I. G. (2001), Handbook of Viscoelastic Vibration Damping, John Wiley
& Sons LTD.
Kelly, J. M., and Edgardo, Q. (1992), "Mechanical Characteristics of Neoprene
Isolation Bearings", Report No. UCB/EERC-92/11, Earthquake Engineering Re-
search Center, College of Engineering, University of California at Berkeley.
Soong, T. T., and Dargush, G. F. (1997), Passive Energy Dissipation Systems in
Structural Engineering, John Wiley & Sons, New York.
Weber, B. (2002), "Damping of vibrating footbridges", Proceedings of the Interna-
tional Conference on Footbridge, Paris, France, 20-22 November 2002, on CD,
AFGC - OTUA (eds.).
Notations
Symbol Description Unit
elastic shear modulus N/m2
Gs storage modulus N/m2

Gt loss modulus N/m2


L length m
W work, energy J
c viscous damping coefficient kg/s
f force N
k stiffness kg/s2

www.samco.org Page 31 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

t thickness m
w width m
x displacement m
8 phase angle rad
Y shear strain -
n loss factor -
T shear stress N/m2
а radial frequency rad/s

Subscripts
d damper
max maximum

6 AMCO

2.1.1.4 Coulomb friction damper Theoretical background


In friction dampers, irrecoverable work is done by tangential force required to
slide one surface of solid body across to another. Contacting surfaces are
intended to remain dry during operation. No additional hydrodynamic lubricating
layer is needed. More detailed information in conjunction with numerous
references can be found in Larssen-Basse (1992) and Taylor (1981).
Basic theory of solid friction is inferred from physical experiments involving planar
sliding of rectangular blocks:
The total frictional force is independent of the apparent surface area of contact
The total frictional force is proportional to the total normal force acting across the
interface.
In case of sliding with low relative velocities, the total frictional force is in-
dependent of velocity.
Hence, the behaviour of a Coulomb friction damper may be described as
fd =ju- N •(- sign(X)) (1)
where fd represents the damper force, N the normal force and u the friction coef-
ficient for the two layers. In some cases, the friction coefficient is somewhat
higher when slippage is imminent than during sliding. This is indicated by the two
separate friction coefficients of static uta and kinetic Ukm friction coefficients.
The Coulomb friction force acts always against the direction of movement and
therefore behaves dissipative like any pure damping force (Fig. 13). The Coulomb

www.samco.org Page 32 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

friction force is zero in the case of zero velocity since relative displacement
between the two surfaces vanishes (Fig. 13). Therefore, the force of an ideal
Coulomb friction damper "jumps" from its positive value to zero and then to its
negative value when the velocity changes its sign. In contrast, real dampers that
are seen as Coulomb friction dampers (e.g. MR dampers at constant current)
show a force velocity trajectory as depicted in Fig. 13 c). Assuming periodic
excitation, the work per one full cycle is represented by the rectangular area
depicted in Fig. 13 (Weber (2002))
W = 4 • fd-max • Xmax (2)

Fig. 13: Coulomb friction: a) force displacement trajectory, b) force velocity


trajectory; c) force velocity trajectory of real Coulomb friction dampers (e.g. MR
dampers at constant current).

6 AMCO

In order to generalize and extend the theory, involving non-uniform distributions


or non-planar surfaces, the basic assumptions mentioned above are often
abstracted to the infinitesimal limit. The generalization of Eq. (1) becomes
T =^-тп •(-sign(X)) (3)
where Tt represents the tangential traction and тп the normal traction. This
equation can be also used for determining nominal contact stresses.
Design issues
In practice, the force trajectory of Coulomb friction dampers will follow
qualitatively the trajectory depicted Fig. 13 c) due to sealing friction and viscous
material behaviour. However, the main and only design variable of Coulomb
friction dampers is their friction force level fd. This force level has to be chosen
the way that the damper

www.samco.org Page 33 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

does not clamp the main structure due to the constant damper force. However,
the friction force level of the originally designed and manufactured Coulomb
friction damper may change during lifetime for several reasons:
The damper force level may change due to varying conditions of the sliding
interfaces. Bimetal interfaces are susceptible to this behaviour because ongoing
physical and chemical processes change the friction coefficient considerably. At
microscopic level one finds that natural and engineered surfaces are not smooth.
These irregularities are often categorized as waviness and roughness and they
are typically present over a wide range of scales. It should be mentioned that true
contact does occur directly between metals, adhesive bonds form across the
interface often producing a friction coefficient larger than one (ц> 1).
Aging and corrosion of the surfaces may also affect the originally produced
friction damper. The investigation of friction processes under such circumstances
is a challenging issue since the mechanical characteristics of oxide films are not
well understood.
Finally, local deformational processes which occur in the vicinity of surfaces may
change the friction force behaviour (Soong and Dargush (1997)). During the
slippage, the dissipated energy will cause local heating along the interface of the
constituent materials. The thermal effects can result in an aging process of
surfaces caused by material softening or promoting oxidation. However, it may be
assumed that the system response is not sensitive to relatively small variations of
ambient temperature. The attention has to be more directed on physiochemical
processes, often triggered by atmospheric moisture or containments. These
processes may change the physical and chemical character of the surfaces,
caused by formation of oxide layers. In more aggressive environments the
interfaces tend to corrode. The formation of the geometry of the component is of
prime importance to attenuate corrosion. For instance, exposed surfaces become
a surplus of oxygen and other inaccessible regions have much less dissolved
oxygen. Therefore, a mass transfer does occur between anode and cathode
(Soong and Dargush (1997)).
Implementation
In the following, several types of friction damping devices are presented.

www.samco.org Page 34 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

a) Limited Slip Bolted Joints


Pall et al. (1980) started with the realization of friction dampers as Limited Slip
Bolted Joints varying simple sliding elements which have different surface
treatments. Several surfaces like mill scale, sand blasted, inorganic metallic zinc-
rich paint, brake lining pads and polyethylene coating were investigated by static
and dynamic tests evaluating load displacement response under constant cyclic
loading (Fig. 14). The maximum static slip is obtained for metallic surfaces. The
cyclic response is quite erratic, with considerable stick-slip associated with the
transition from static to kinetic frictional response.
Anagnostides and Hargreaves (1990) reported experimental results from a
number of other frictional materials. Brake lining materials would perform well in
dynamic tests. The characterization of their simple brake lining frictional system in
terms of an elastic perfectly plastic model is quite appropriate. A mathematical
model is given in Soong and Dargush (1997).

Fig. 14: Limited Slip Bolted Joints (Pall et al. (1980)).


Fig. 15: X-braced Friction Damper (Pall and Marsh (1982)).

b) X-braced Friction Damper


A much more effective operation can be reached special damper mechanism,
proposed by Pall and Marsh (1982) (Fig. 15). These devices utilize the brake
lining pads. One principle of a typical X-braced system is based on braces that
are designed to buckle at relatively low compressive loads, so the braces
contribute only when subjected to tension. Installing uni-axial friction elements
within each brace, slippage would only occur in the tensile direction and little

www.samco.org Page 35 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

energy dissipation would result during cyclic loading. The mechanism tends to
straighten buckled braces and also enforces slippage in both tensile and
compressive directions.
Pall and Marsh (1982) used a simple elastoplastic model to describe the
behaviour of this X-braced friction damper. Filiatrault and Cherry (1988)
proposed a more refined macroscopic model for this device, since the Pall-Marsh
model overestimates the energy dissipation and the simple elastoplastic model is
only valid if the device slips every cycle and if that the slippage is always
sufficient to straighten completely the buckled braces. Wu et al. (2005) published
an improved variant of the Pall-typed fric- tional dampers.

www.samco.org Page 36 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

c) Sumitomo Friction Damper


The uni-axial friction damper made by Sumitomo Metal Industries Ltd. utilizes a
more sophisticated design (Fig. 16). A precompressed internal spring exerts a
force that is converted trough the action of inner and outer wedges into a normal
force on the friction pads. The use of graphite plug inserts ensures dry lubrication
in order to maintain a consistent friction coefficient between the pads and the
inner surface of the steel casing. Aiken and Kelly (1990) reported that the
response of this damper is extremely regular and repeatable with rectangular
hysteresis loops. Effects of loading frequency, number of cycles, and ambient
temperature are indicated as not sensitive.

Fig. 16: Sumitomo Friction Damper (Aiken and Fig. 17: Energy Dissipating
Restraint (EDR) (Nims Kelly (1990)). et al. (1993)).

d) Energy Dissipating Restraint


The Energy Dissipating Restraint (EDR), designed by Fluor Daniel, Inc., is similar
to the Sumitomo concept, since this device also includes an internal spring and
wedges encased in a steel cylinder (Fig. 17). A detailed description of the
damper design is provided in Nims et al. (1993). The EDR uses steel
compression wedges and bronze friction wedges in order to transform the axial
spring force into normal pressure acting outward on the cylinder wall. The
frictional surface is formed by the interface between the bronze edges and the
steel cylinder. Internal stops are ensured within the cylinder in order to create the
tension and compression gaps. It should be mentioned that the length of the
internal spring can be altered during operation, providing a variable frictional slip

www.samco.org Page 37 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

force. Other realized friction dampers can be found in the book of Soong and
Dargush (1997).
References
Aiken, I. D., and Kelly, J. M. (1990), "Earthquake Simulator Testing and
Analytical Studies of Two Energy-Absorbing Systems for Multistory Structures",
Report No. UCB/EERC-90/03, University of California, Berkeley, CA.
Anagnostides, G., and Hargreaves, A. C. (1990), "Shake Table Testing On an
Energy Absorption Device for Steel Braced Frames", Soil Dynamics Earthquake
Engineering, 9(3).
Filiatrault, A., and Cherry, S. (1988), "Comparative Performance and Friction
Damped Systems and Base Isolation Systems for Earthquake Retrofit and
Aseismic Design", Earthquake Engineering Structural Dynamics, 16.

6 AMCO

Larsen-Basse, J. (1992), Basic Theory of Solid Friction, in Friction, Lubrication


and Wear Technology, ASM International Handbook Committee, American
Society for Metals, Materials Park, Ohio.
Pall, A. S., Marsh, C., and Fazio, P. (1980), "Friction Joints for Seismic Control of
Large Panel Structures", Journal Prestressed Concrete Inst., 26(6).
Pall, A. S., and Marsh, C. (1982), "Response of Friction Damped Braced
Frames", Journal Structural Division, ASCE, 108(ST6).
Nims, D. K., Richter, P. J., and Bachmann, R. E. (1993), "The Use of the Energy
Dissipating Restraint for Seismic Hazard Mitigation", Earthquake Spectra, 9(3).
Soong, T. T., and Dargush, G. F. (1997), Passive Energy Dissipation Systems in
Structural Engineering, John Wiley & Sons, New York.
Taylor, D. (1981), "Friction - The Present State of Our Understanding", Journal
Lubr. Technics, ASME, 103.
Weber, B. (2002), "Damping of vibrating footbridges", Proceedings of the Interna-
tional Conference on Footbridge, Paris, France, 20-22 November 2002, on CD,
AFGC - OTUA (eds.).
Wu, B, Zhang, J., Williams, M. S., and Ou, J. (2005), "Hysteretic behaviour of im-
proved Pall-typed Frictional dampers", Engineering Structures, 27(3), 1258-1267.
Notations
Symbol Description Unit

www.samco.org Page 38 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

W work, energy J
N normal force N
f force N
x displacementm
И friction coefficient -
т traction N/m2
Subscripts
d damper
kin kinetic
max maximum
n normal
sta static
t tangent

www.samco.org Page 39 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

2.1.1.5 Structural friction damper Theoretical background


(1)
Structural damping is the use of internal friction in a material to transform
structural vibration energy into heat produced within the damper material. This
reduces vibration amplitudes of the main structure. The definition equation of
friction dampers is (Weber (2002))
fd = kd • W •(- sgn(X))
where ks is called pseudo-stiffness factor. For periodic excitation, Fig. 18 shows
the
force displacement and force velocity trajectories of structural friction dampers.
Please notice that the damper force "jumps" just before reaching its maximum
value to zero (see the path direction in Fig. 18 indicated by arrows) because of
the damper velocity that is zero at t = П (2 • a) and at t = 3 • П (2 •с). The
dissipated energy per full cycle is equal to

W=4•
2>
max
: 2 • kd • Xmax
Fig. 18: Structural damping: a) force displacement trajectory and b) force
velocity trajectory.
2

www.samco.org Page 40 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

f
k•x
(2)

Testing and validation


The goal of testing structural friction dampers is to identify the single damper pa-
rameter kd. Therefore, the testing procedure of structural friction dampers
follows the
one of linear viscous dampers except that not the damper's viscosity but the
damper's stiffness is identified. The identification can be realized by simply
measuring pairs of force deformation values. Several pairs have to be measured
in order to be able to identify nonlinear stiffness behaviour.

www.samco.org Page 41 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

www.samco.org Page 42 of 155


OAMCO
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Implementation
Structural friction dampers are often implemented as vibration mitigation devices
directly within pylons and struts in order to dissipate vibration energy and
therefore to mitigate the structure. Especially, they are used in:
base isolation systems,
damping of belfries,
gear boxes as damping materials,
hollow parts or profiles as damping material, and
automotive applications.
References
Weber, B. (2002), "Damping of vibrating footbridges", Proceedings of the
International Conference on Footbridge, Paris, France, 20-22 November 2002,
on CD, AFGC - OTUA (eds.).
Notations
Symbol
W
Description work, energy force
Unit
JN
kg/s2
f
k
stiffness

m
x
displacement
Subscripts

www.samco.org Page 43 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

d
damper maximum
max

www.samco.org Page 44 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

2.1.1.6 Hysteretic damper Theoretical background


The inelastic deformation capability of metallic substances represents an
effective energy dissipation mechanism for damping of engineering structures.
The stress strain curve depicted in Fig. 19 a) is typical for most metals. For
strain values less or equal than eY , the stress strain relation is linear and the
initial state O is fully recoverable if the applied load is removed. For strain
values larger than eY, the metal yields and deforms irreversible. The plastic
deformation occurs directly with energy dissipation. If the load in point B is
removed, static deformation remains due to the inelastic deformation from Y to
B, therefore denoted as eine. Only the elastic part Јela is recovered. At point M,
the maximum load is reached. The ultimate failure occurs at point Q (Soong and
Dargush (1997)).

Fig. 19: Hysteretic damping: a) stress strain relation, b) real hysteretic damper
behaviour, c) idealized hysteretic damper behaviour.

The shape of the force displacement trajectory depends on the stress strain
relationship of the material and the device design (Fig. 19 b)). For the case of an
ideal elastic-plastic material behaviour, the relation between damping force and
displacement is as shown in Fig. 19 c). The dissipated energy per cycle of an
ideal elastic-plastic material becomes
W = 4 • fy -(Jmax - Jy )= 4 \ky • Xy Mjmax " Xy )= 4 • ky • X2y • (f - l)
(1)

www.samco.org Page 45 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

where fy is the yield force, xy the displacement at which the material starts to
yield, xmax the maximum displacement, ky the elastic damper stiffness and the
ductility ratio f denotes the ratio between maximum displacement and yield
displacement (Jones (2001))
f = Xma^ (2)
Jy

www.samco.org Page 46 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Design issues An example


(3)
(4)
Jones (2001) presents an example how to design a hysteretic damping device.
The damping device consists of a cylindrical rod of length L and cross sectional
area A (Fig. 20). Provided that the damper material behaves ideal elastic-
plastic, the yield force may be expressed as follows
fy = A = ky • Xy
Considering the relationship between strain and deformation
(5)
x„ = L •£
the linear stiffness describing the elastic behaviour becomes
ky = =A•£
yL£L

Equation Eq. (5) describes the known relation between stiffness and elasticity
modulus E. Hence, by appropriate choice of material (E) and geometry (A, L),
the desired damper with maximum expected deformation and maximum
tolerable damper force may be designed.

Fig. 20: Ideal elastic-plastic


damping device according to Jones
(2001). Fig. 21: X- shaped Plate
Damper.

У /

www.samco.org Page 47 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

"V

о
The idea of utilizing separate metallic hysteretic dampers in order to improve
the resistance of structures in the case of earthquakes was formulated firstly
by the conceptual work of Kelly et al. (1972) and Skinner et al. (1975). In
recent years, many new designs of metallic dampers have been produced.
Two examples are shown in Fig. 21. An X-shaped plate damper or ADAS
(Added Damping And Stiffness) device has been studied via experiments by
Bergman and Goel (1987), Whittaker et al. (1989), Whittaker et al. (1993), and
subsequently employed in the seismic retrofit projects discussed by Martinez-
Romero (1993) and Perry et al. (1993). The hourglass shape produces nearly
uniform curvature throughout the plate during infinitesimal deformation.
Similar reasoning has also led to the development of triangular plate systems
by Tsai et al. (1993).
Modelling approaches

In order to characterize the metallic damper response, a couple of different


suitable force displacement models can be taken. The first approach involves
the direct

www.samco.org Page 48 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

use of experimental data from damper tests. The whole damper behaviour is
described by the static damper map, which consists of the fitted force
displacement trajectories.
Another way of modelling is that the damper model is constructed from an
appropriate constitutive relationship for metals by applying the principles of
mechanics. From a constitutive model of a metal, the force displacement
relationship can be developed introducing a geometric description of the
device and employing principles of mechanics. The geometric description may
require a finite element discretization or can be adequately modelled, for
instances by simple strength of materials representation. For the example of a
triangular plate damper, Tsai et al. (1993) introduced a set of equations to
describe the following force displacement model

fd =
(6)
x
( N • E-w0 •h3 ^ 6-
L3

where fd represents the damper force, N the number of identical triangular struc-
tural steel plates, which are positioned in parallel, x the damper displacement, h
the thickness of the cantilevered plate, L the length and w0 the base width. This
model
agrees reasonably well with performed experiments. However, the model is valid
only for the elastic response of the damper. Thus, the dissipated energy of the
device cannot be estimated using this model.
Soong and Dargush (1997) recommended a finite element analysis methodology.
For the example of triangular plate damper model, development and subsequent

www.samco.org Page 49 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

calculations were described. Generally, in the same way, a wide range of metallic
dampers, including the X-shaped plate dampers, can be developed. A more
detailed description of a mechanic based modelling approach for triangular plate
dampers is provided in Dargush and Soong (1990) and in Tsai and Tsai (1995).
In the analysis of single-degree-of-freedom systems (SDOFS), the addition of a
pure viscous device increases the system damping independently of the
excitation forces. Although the insertion of metallic dampers into SDOFS
generally reduces the response by the portion of the dissipated energy, the
SDOFS response may increase for some specific seismic inputs due to the
added stiffness of the hysteretic damper. Soong and Dargush (1997) have
illustrated a particular case in which an elastoplastic damper is either ineffective
or even detrimental under special ground motion. This experience suggests that
a detailed analysis of the nonlinear transient dynamics is required to evaluate the
effectiveness of metallic dampers for seismic protection of civil structures.
On the basis of the finite element method, the Newton-Raphson time domain ap-
proach is directly applicable for structures that include metallic dampers.
Applications of nonlinear structural analysis to building frames using X-shaped
dampers are provided by Xia et al. (1990) and for the case of using triangular
plate dampers by Tsai et al. (1993). Another approach is provided by Soong
(1990), which allows for rewriting the governing equations in state space
representation and solving in an efficient and accurate way by application of first
order differential equation solvers.
Effect of viscoplasticity
For steel at approximately room temperature and in the case of moderate strain
rates, it can be assumed that the plastic flow occurs instantaneously compared to
the time variation of the applied load. In cases of lead dampers or steel at high
tempera

www.samco.org Page 50 of 155


OAMCO
SAMCO Final Report 2006
F05 Guidelines for Structural Control

ture or under very high strain rates, the creep and relaxation phenomenon must
be considered. Creep signifies increasing strain with time under constant stress.
Relaxation is defined as continual reduction of stress with time for a material
under constant strain. In order to model the motion of dislocations, it is necessary
to incorporate the physics of time independent plastic strains and time dependent
creep strains to a unified theory. Ozdemir (1976) proposed the following uni-axial
model

where ё is the strain decrement, т and (b represent the relaxation time and back
stress, respectively. The drag stress (d is a material constant. The quantity (-(b
is denoted as overstress. This allows the modelling of the kinematical hardening
effect. A detailed discussion can be found in the book of Soong and Dargush
(1997).
Effect of temperature
The mechanical properties of steel in structures at approximately room
temperature are both consistent and stable. Therefore, steel is often used as a
building material. In case of impacts by major earthquakes, the structural steel
within metallic dampers will cycle into the inelastic range. A significant portion of
dissipated energy will be converted into heat. The surrounding metal will be
heated. The quantity of temperature increase can be estimated by considering
the energy balance. The dissipated energy represents a heat source, while
conduction and convection processes lead to a redistribution of energy. For
reasonable large steel damper design, it may be assumed that temperature
increase does not significantly alter the mechanical properties of the device.
The behaviour of lead dampers is much more sensitive to moderate increases
beyond room temperatures. Thus, in this case, thermal effects must be
considered.
Consideration of failure
The theories of plasticity and viscoplasticity describe the behaviour of metals in
the inelastic range during cycling loading but not the fail due to fatigue. The
phenomenon of the low-cycle fatigue is of main interest. Low cycle fatigue results
from a limited number of excursions into the inelastic range. Growth and
interconnection of micro sized cracks lead eventually to failure at macroscopic
level. In practice, a more phe- nomenological approach is used based upon the

www.samco.org Page 51 of 155


OAMCO
SAMCO Final Report 2006
F05 Guidelines for Structural Control

concept of material damage. It should be recognized that these plate dampers


represent critical elements in the overall seismic protective system and therefore
must be engineered to a high level of reliability. Dargush and Soong (1995) used
a simple modelling approach for the mechanics of materials in order to study the
behaviour of triangular metallic plate dampers under cyclic load.
Testing and validation
Soong and Dargush (1995) reported that, from a review of literature, much effort
has been expended on experimental testing of metallic dampers and test
structures. For determination of the mechanical characteristics of the damping
devices, an in-plane testing frame is employed and the horizontal displacement
and the vertical force as well are controlled. The investigations should include:
the force response at several displacement levels,
fatigue history, and
temperature rise of the specimen during cycling loading.
Implementation
Hysteretic dampers have been successfully applied for structural control in the
fields of:
design of seismic passive energy dissipation systems (Constantinou et al. (1998),
Paulay and Priesley (1992)), and
base isolation systems and foundation of superstructures (Bhatti et al. (1978)).
References
Bergman, D. M., and Goel, S. C. (1987), "Evaluation of Cyclic Testing of Steel
Plate Devices for Added Damping and Stiffness", Report No. UMCE 87-10,
University of Michigan, Ann Arbor, MI.
Bhatti, M. A., Pister, K. S., and Polek, E. (1978), "Optimal design of an
Earthquake Isolation System", Report No. UCB/EERC-78/22, University of
California.
Constantinou, M. C., Soong, T. T., and Dargush, G. F. (1998), "Passive Energy
Dissipation Systems for Structural Design and Retrofit", MCEER Monograph
Series 1, Multidisciplinary Center for Earthquake Engineering Research, Buffalo,
NY.
Dargush, G. F., and Soong, T. T. (1995), "Behaviour of Metallic Plate Dampers in
Seismic Passive Energy Dissipation Systems", Earthquake Spectra, 9(3).
Jones, D. I. G. (2001), Handbook of Viscoelastic Vibration Damping, John Wiley
& Sons LTD.

www.samco.org Page 52 of 155


OAMCO
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Kelly, J. M., Skinner, R. I., and Heine, A. J. (1972), "Mechanisms of Energy Ab-
sorption in Special devices for Use in Earthquake Resistant Structures", Bull.
N.Z. Society Earthquake Engineering, 5(3), 63-88.
Martinez-Romero, E. (1993), "Experiences on the Use of Supplemental Energy
Dissipators on Building Structures", Earthquake Spectra, 9(3).
Ozdemir, H. (1976), Nonlinear Transient Dynamic Analysis of Yielding Structures,
Ph.D. Dissertation, University of California, Berkeley, CA.
Paulay, T., and Priesley, M. J. N. (1992), Seismic Design of Reinforced Concrete
and Masonry Buildings, John Wiley & Sons, New York.
Perry, C. L., Fierro, E. A., Sedarat, H., and Scholl, R. E. (1993), "Seismic
Upgrade in San Francisco Using Energy Dissipation Devices", Earthquake
Spectra, 9(3).
Skinner, R. I., Kelly, J. M., and Heine, A. J. (1975), "Hysteresis Dampers for
Earthquake-Resistant Structures", Earthquake Engineering and Structural
Dynamics, 3, 287-296.
Soong, T. T. (1990), Active Structural Control: Theory and Practise, Wiley, New
York.

www.samco.org Page 53 of 155


OAMCO
SAMCO Final Report 2006
F05 Guidelines for Structural Control

www.samco.org Page 54 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

Soong, T. T., and Dargush, G. F. (1997), Passive Energy Dissipation Systems in


Structural Engineering, John Wiley & Sons, New York.
Tsai, K. C., Chen, H. W., Hong, C. P., and Su, Y. F. (1993), "Design of Steel Tri-
angular Plate Energy Absorbers for Seismic-Resistant Construction", Earthquake
Spectra, 9(3).
Tsai, C. S., and Tsai, K. C. (1995), "TPEA Device as Seismic Damper for High-
Rise Buildings", Journal Engineering Mechanics, ASCE, 121(10).
Whittaker, A., Bertero, V., Alonso, J., and Thompson, C. (1989), "Earthquake
Simulator Testing of Steel Plate Added Damping and Stiffness Elements", Report
No. UCB/EERC-89/02, College of Engineering, University of California at
Berkeley.
Whittaker, A., Aiken, I., Bergman, D., Clark, P., Cohen, J., Kelly, J., and Scholl,
R. (1993), "Code Requirements for the Design and Implementation of Passive
Energy Dissipation Systems", Proc. ATC 17-1 on Seismic Isolation, Energy
Dissipation, and Active Control, 2, 497-508.
Xia, C., Hanson, R. D., and Wight, J. K. (1990), "A Study of ADAS Element Pa-
rameters and Their Influence on Earthquake Response of Building Structures",
Report No. UMCE 87-10, The University of Michigan, Ann Arbor, MI.
Notations
Symbol Description Unit
A cross sectional area m2
E elasticity modulus N/m
L length m
W work, energy J
f force N
к stiffness kg/s
x displacement m
£ strain m
и ductility ratio -
а stress N/m
т relaxation time s
2
2

www.samco.org Page 55 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

2
Subscripts
d
max
у
damper
maximum
yield

www.samco.org Page 56 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

2.1.1.7 Shape memory alloy damper Theoretical background


Shape memory alloys (SMAs) are metallic materials that mainly exist in two
different crystal structure types, namely martensite and austenite or a
combination of both. Martensite is characterized by a body centred tetragonal
crystal structure and aus- tenite by a body centred cubic crystal structure.
Since the properties of SMAs change according to the two input variables
temperature and strain, SMAs are often called "smart materials". The phase
transformation of SMAs is reversible. Profound information about material
behaviour of SMAs and their applications may be found in Duerig (1990),
Funakubo (1987), Humbeeck (2001), Janke et al. (2005), Otsuka and Way-
man (1999), and in Otsuka and Kakeshita (2002).
Due to the fairly high damping characteristics of SMAs, these materials are
also used as dampers. If the material properties are changed according to the
actual material temperature and material strain, respectively, SMAs represent
controllable dampers or semi-active damping devices, respectively. Within this
chapter, SMAs used as passive dampers are introduced, whereas the use of
controllable SMAs is described in chapter 2.1.2.2.
According to Janke et al. (2005), primarily, one may distinguish between four
different behaviours of SMAs:
SMAs may behave "actuator like" (Fig. 22),
SMAs show the so-called "shape memory effect" (Fig. 23),
SMAs behave super elastically (Fig. 24), and
SMAs may act as hysteretic material dampers in pure martensitic state (Fig.
25).

g-8 curve of the G Л—


austenite z

www.samco.org Page 57 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

A
-20 W heating cooling toAf
100% below Mf
ambient temperatures
0% for outdoor civil
60 applications
T[°C]

Fig. 22: Stress strain curves and transformation temperature profiles for
"actuator like" change of strain and stiffness at constant stress (Janke et al.
(2005)).
If SMAs behave "actuator like", ambient temperature is below the temperature
"martensite finish" (Mf). After pure martensite has been deformed, the crystal
structure of the SMA may be changed from martensite to austenite and back
by heating and cooling (horizontal dashed line in Fig. 22). Since austenite has
a larger elasticity modulus than martensite, a device with two different
elasticity moduli may be produced. If a spring is produced with such an SMA
and connected with another steel

www.samco.org Page 58 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

spring, the connecting point will move when heating and cooling. This is the
reason why this SMA behaviour is called "actuator like".
If ambient temperature lies between "martensite finish" and "austenite start"
(As), the so-called "shape memory effect" of SMA occurs. First, pure
martensite is deformed plastically. Then, the load is removed and plastic
deformation results (Fig. 23). From this point,
two

scenarios are feasible. If the SMA is heated


without geometrical constraints, martensite will change to austenite which
ends up in a full shape recovery of the initial shape in point A. If the SMA is
heated at constant strain, stress increase results due to the phase change to
austenite, which has a larger elasticity modulus.

0%--
-20
T[°C]
100%
A
60
heating toAf s 1 s =
free

www.samco.org Page 59 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 23: Stress strain curves and transformation temperature profiles for shape
memory effect in the case of free strain recovery and constraint strain
recovery (Janke et al. (2005)).
If ambient temperature lies above "austenite finish" (Af), the SMA behaves
"super elastically" (Fig. 24). Pure austenite is deformed until it reaches plastic
state. Then, austenite will transform to martensite with increasing strain (see
plateau in Fig. 24). When the state of pure martensite is reached, the plateau
ends and the stress strain curve goes up with smaller slope due to smaller
elasticity modulus of martensite relatively to austenite. During unloading,
martensite will change to austenite on a lower stress plateau. Finally, the SMA
consists again of pure austenite. The area enclosed by the stress strain
trajectory is equivalent to the dissipated energy per cycle. Therefore, SMA
working in superelastic mode may be used as dampers.

100%--
0%--
Mfv As
-20
unloading
austenite
60
T[°C]
A
stress induced
martensite

Fig. 24: Stress strain curves and transformation temperature profiles for the
so-called superelastic behaviour (Janke et al. (2005)).

www.samco.org Page 60 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

If ambient temperature lies between "martensite finish" and "austenite start",


SMAs may work as pure martensitic, hysteretic dampers if temperature is
constant (Fig. 25). Basically, also pure austenitic SMAs may work as hysteretic
dampers. However, the main advantages of martensitic hysteretic dampers
are:
The elasticity modulus of martensite is smaller than of austenite. Therefore,
the stress strain trajectory encloses a larger area for the same stress which
results in larger damping for the target structure.
Due to the reorientation of martensite variants,
martensite can withstand many more deformation
cycles than austenite without failure for the same
dissipated energy.
z

A
MfA As
УЛУ/

Af
100% —
-20

0%-
60
A
T[°C]

www.samco.org Page 61 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 25: Stress strain curves and transformation temperature profiles for
martensitic hysteretic damping (Janke et al. (2005)).
Due to the larger hysteresis loop of SMAs working as pure martensitic hysteretic
dampers compared to the hysteresis area of SMAs operating in superelastic
mode (Fig. 26), SMA passive dampers are usually based on the former effect.
Fig. 26: Hysteresis loops for: a) superelastic SMA Fig. 27: Torsion bar design
(Witting and Cozzarelli behaviour and b) martensitic hysteretic damping (1992)).
(Janke et al. (2005)).

www.samco.org Page 62 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Design issues
Witting and Cozzarelli (1992) investigated four principal design mechanisms for
producing an SMA based damping device. They used a Cu-Zn-Al alloy. It was a
bar in torsion and beam in bending, which was axially loaded, and the clamped
plate was loaded in the centre (Fig. 27). They found that the annular clamped
plate was much too stiff and resulted in very small strains. The axially loaded
beam was also found unsuitable because the constraints of stiffness and strain
would cause the beam to buckle. A direct comparison between torsion and
bending beam showed that more energy was put into higher strain regions in the
torsion bar design than in the bending beam design. The larger strain resulted in a
bigger amount of energy absorbed. Therefore, the torsion bar design apparently
resulted in a more effective damper.
For the first assessment of the damper capabilities, a linear constitutive mode was
used. The basic dimensions of the damper are torsion arm length D, radius of tor-
sion bar R, and length of torsion bar L (Fig. 27). The angle of twist © of a solid
round bar of radius R, length L, with torque T applied at the end of the bar and
shear modulus G is
2TL
© = 2-TJl (1) GR4
The angle of twist © due to the displacement x is expected to be small, which al-
lows the following simplification
— _ sin(©) ~ © (2)
D
The shear strain exy may be expressed in terms of L, D, x and the radial distance
r from the centre of the bar as follows
_ ©• r _ x-r
_ 2-L " 2-LD ( )
The damper force is given by
GR 4-x (4)
fd LD2W

www.samco.org Page 63 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Considering that the damper force is the product of damper stiffness and damper
displacement, the damper stiffness can be expressed as
к G R4 (5)
kd _ L D2 (5)
The maximum shear strain occurs at r _ R if the displacement x reaches its maxi-
mum xmax . This yields for the maximum strain
^max _ xmax R (6) max 2-LD
Solving Eqs. (5) and (6) for D yields

www.samco.org Page 64 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

2-п-G- R3 —„
D= ^ (7)
xmax - kd
which leads to the length L as follows
L= kd ' Xm2ax 2 (8)
4-n-G ■ R2-e^
Upon substitution of Eq. (7) into Eq. (3) leads to the shear strain as follows
e-r
exy =-mx— (9)
R
The equation of strain energy density is
Uo = 2 - G ---у (10)
This equation must be integrated over the high strain region. Since the strain is
independent of y, the integration limits for © become 0 and 2n. The maximum of r
is
R . The minimum value of r can be calculated by substituting -min for -xy in Eq. (9)
which leads to
R—
= шп. (и)
in
-
max
Using Eq. (7) in order to eliminate L, the total strain energy may be estimated by
2n a R rs 2
W = 2- JJJ - -2 -r3 -dr-dy-d© (12)
x 2 -k
x max k
4-n-G R2 —
00bR
with: a = - ™ 2d 2 (13)

www.samco.org Page 65 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

ax
R—
b= пш (14)
—max
For the final design, a more accurate nonlinear model is developed. The
constitutive law used in the analysis is bilinear
Ty = 2 - G - —xy +{2 - G - G2) - (— - —xy)}- U (—xy - —) (15)
The two shear moduli G1 and G2 represent the elastic and inelastic shear
modulus. The variable — is the strain value at which the stress strain trajectory
changes its slope. The term U(exy - et) denotes the step function as follows
, ч f0: x < 0 U (x ) = j (16)
11: x > 0

www.samco.org Page 66 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

www.samco.org Page 67 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

гртрг
Ьамсо
The stress strain curve described by Eq. (15) is depicted in Fig. 28. The torque
produced from the two torsion bars within the damper (Fig. 27) may be calculated
as
2nR
T _fd • D _ 2-\\Txy-r)-r-dr-d© (17)
00
Substituting Txy from the constitutive law in Eq. (17) and using Eq. (3) for the
shear strain, the damper force becomes

I
x
-W (г
fd _ ~ и
G - G2)
r 2 dr d©
+
r
xy
2-LD (18)
4 П \ G2 x2
D 00 I2-l D<

Assuming R > 2 - L D - Јt /x, integration of Eq. (18) simplifies


to
R 3 - 2-(LD-Јt)
n-G2 Rl LD2
8-n-Јt 3-D
-(G -G2)-
fd _
x+

www.samco.org Page 68 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

(19)
The two presented design approaches delivered two estimates of the damper
force. By choosing the geometrical design parameters L, D, and R and selecting
the appropriate material, which defines G1, G2, and Јt, the damper force may be
estimated using the simplified approach of Eq. (4) or the more accurate
approach given by Eq. (19).

Fig. 28: Stress strain diagram used for bilinear damper model (Witting and Cozzarelli
(1992)).
Fig. 29: Force displacement loops (Witting and Cozzarelli (1992)).

The application of the SMA dampers is limited due to dimensions of civil structures
which makes relatively high damping forces necessary. Therefore, in case of
damping civil structures using SMAs, large amount of material is needed
compared to other applications as in the case of, e.g., automotive engineering or
medical micro technologies. Probably one of the most important reasons why SMA
dampers are not so often used in civil engineering for vibration mitigation is that
the increase of damping is low compared to the financial investment. Another
reason might be that a value

www.samco.org Page 69 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

of 40% material damping is quite high but only a part of all mechanical energy can
be transmitted to the damping part in devices (Humbeeck and Kustov (2005)).
Testing and validation
SMA dampers are tested similarly to other damper types; see the test set-up
configuration depicted in Fig. 5 and Fig. 6. An example of measured force
displacement trajectories of SMA dampers is depicted in Fig. 29 (Witting and
Cozzarelli (1992)). The tested SMA is a CU-Zn-Al alloy.
Implementation
Due to the fairly large material costs of SMAs, such dampers are rarely
implemented in civil structures in order to enhance structural damping. If SMA
based passive dampers shall be used for vibration reduction, the SMA operating
as a martensitic hysteretic damper should be used because this operating mode
provides maximum additional damping to the structure.
Basically, the main motivation to use SMA as damping element is to have a semi-
active damping device available, where, e.g., the force may be adapted to the
structure by heating and cooling the SMA device (Li et al. (2004), Rustighi et al.
(2005), Williams et al. (2002)). Such kind of applications is shortly described in
chapter 2.1.2.3.
References
Duerig, T. W. (1990), Engineering Aspects of Shape Memory Alloys, Butterworth-
Heinemann, London.
Funakubo, H. (1987), "Precision Machinery and Robotics, Vol. 1 - Shape Memory
Alloys", Gordon and Breach.
Humbeeck, J. V. (2001), "Shape memory alloys: A material and a technology",
Adv. Eng. Mater, 3(11), 837-850.
Humbeeck, J. V., and Kustov, S. (2005), "Active and passive damping of noise and
vibrations through shape memory alloays: applications and mechanisms", Smart
materials Structures, 14, 171-815.
Janke, L., Czaderski, C., Motavalli, M., and Ruth, J. (2005), "Applications of shape
memory alloys in civil engineering structures - Overview, limits and new ideas",
Journal of Materials and Structures, 38(279), 578-592.
Li, H., Liu, M., and Oh, J. (2004), "Vibration mitigation of a stay cable with one
shape memory alloy damper", Journal of Structural Control and Health Monitoring,
11(1), 21-36.
Otsuka, K., and Wayman, C. M. (1999), Shape Memory Materials, Cambridge
University Press.

www.samco.org Page 70 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Otsuka, K., and Kakeshita, T. (2002), "Science and technology of shape-memory


alloys. New developments", Mrs Bulletin 27(2), 91-100.
Rustighi, E., Brennan, M. J., and Mace, B. R. (2005), "A shape memory alloy
adaptive tuned vibration absorber: design and implementation", Journal of Smart
Materials and Structures, 14, 19-28.

6 AMCO

Williams, K., Chiu, G., and Berhard, R. (2002), "Adaptive-passive absorbers using
shape-memory alloys", Journal of Sound and Vibration, 249(5), 835-848.
Witting, P.R., and Cozzarelli, F.A. (1992), Shape Memory Structural Dampers:
Material Properties, Design and Seismic Testing, Report No. NCEER-92-0013,
State University of New York at Buffalo.
Notations
Symbol Description Unit
D torsion arm length m
G shear modulus N/m2
L length m
R radius m
T temperature; torque K; Nm
W work, energy J
f force N
k stiffness kg/s2
r radial distance m
x displacement m
© angle of twist rad
— shear strain -
T shear stress N/m2

Subscripts
d damper
max maximum
2.1.1.8 Passive tuned mass damper Introduction
A passive tuned mass damper (TMD) or tuned vibration absorber is basically an
energy dissipation device that in its simplest form consists of a mass that is
attached to a structure (primary system) with spring and damper elements (Fig.

www.samco.org Page 71 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

30). Due to the damper, energy dissipation occurs whenever the mass of the TMD
oscillates with non-vanishing displacement or velocity relative to the primary
system. This is achieved by transferring as much energy as possible from the
primary system to the TMD by a careful tuning of the natural frequency and
damping ratio of the TMD. Since the mass of the TMD is significantly smaller that
that of the primary system, transferring energy from the primary system to the
TMD generates a great relative oscillation of the mass of the TMD.
Contrary to a standard damper which generally provides additional energy
dissipation in a wide frequency band, a TMD operates efficiently only in a narrow
frequency band. This behaviour is closely related to the mechanism of energy
transfer from the primary system to the TMD. High energy transfer arises
whenever the natural frequency of the TMD is tuned to the natural frequency of the
primary structure. Therefore, if attached to a continuous structure, the TMD
mitigates only one specific vibration mode.
The concept of a TMD without integrated damping device was invented by Frahm
in 1909 (Frahm (1909)) to reduce the rolling motion of ships. A first theory of the
TMD with integrated damping device was presented years later by Ormondroyd
and Den Hartog (Ormondroyd and Den Hartog (1928)). In Den Hartog's
monograph (Den Har- tog (1947)), a detailed analysis of optimal parameters of a
TMD is presented. There, the model of a primary structure with an attached TMD
is represented by the two degree of freedom system shown in Fig. 31. Den Hartog
used the fixed-points method for obtaining an accurate approximate solution of the
optimal parameters, natural frequency ft and damping ratio Z, of a TMD that
minimizes the displacement of the primary structure where the latter has vanishing
structural damping. A list of optimal parameters for different minimization
objectives obtained by the fixed-points method is given in Warburton (1982) and in
the textbook of Korenev and Reznikov (1993). Recently, a method for computing
exact closed form solutions of optimal parameters that minimizes the maximum of
transfer functions was presented by Nishihara and Asami (2002). For structures
with vanishing structural damping, exact solutions with respect to different
minimization objectives are given in Asami and Nishihara (2003).
If the primary system is subjected to random white noise excitation, that is, an
excitation that contains any frequency with exactly the same amplitude, the design
of a TMD is based on the minimization of the integral of the square of the absolute
value of the transfer function (Crandall and Mark (1963)). Exact closed form
solutions of the optimal parameters of a TMD for different minimization objectives

www.samco.org Page 72 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

and undamped primary system are given by Warburton (1982) and Korenev and
Reznikov (1993).
In practical applications, the primary system has non-vanishing structural damping
and therefore many attempts have been made to find exact, closed form solutions
of the optimal parameters. Unfortunately, so far, no exact algebraic solutions to
this problem are known. Empirical formulas for several minimization objectives,
that are based on numerical optimization results, were given by Ioi and Ikeda
(1978). Recently, analytical approximations obtained by perturbations techniques
were constructed by Asami et al. (2002).

www.samco.org Page 73 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

www.samco.org Page 74 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 30: Typical implementation of a TMD for vertical vibration mitigation of


bridges.

Basic equations of the passive TMD


The standard model of a linear elastic structure with a linear TMD is the two
degree of freedom model displayed in Fig. 31. The equations of motion of this
model are given by
mpup (t) + (cp + ct)up (t) + (kp + kt)up (t) - ctut(t) - ktut(t) = f (t) - fhpUg (t), mtut (t)
+ ctut (t) + ktut (t) - ctUp (t) - ktUp (t) = -mtug (t),
(1)
where up (t) is the displacement of the primary system ut (t) the displacement of
the
mass of the TMD (see Fig. 31). Both displacements are measured relative to the
base. mp, cp, kp are the modal mass, damping constant and stiffness of the
primary
system, and mt,ct, kt are the mass, damping constant and stiffness of the TMD. f(t)
is the force acting the primary system, ug (t) is the base acceleration and mp is the
participating mass of the primary system (see the end of this section for further de-
tails).
(2)
When formulated with respect to up(t) and ur(t) = ut(t) -up(t), the relative dis-
placement of the TMD mass with respect to the displacement of the primary
system, the equations of motion are

www.samco.org Page 75 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

mpUp (t) + cpUp (t) + kpUp (t) - ctur (t) - ktur (t) = f (t) - nipUg (t), mtii p (t) + mtiir (t)
+ ctur (t) + ktur (t) = -mtu g (t),
The standard model of a structure with a TMD is obtained by modelling the
structure as a one-degree-of-freedom system using modal displacements. This
dramatic simplification is justified by the particularity of properly tuned TMDs to
operate effectively only in a narrow frequency band. TMDs are therefore designed
to mitigate the oscillations of a specific vibration mode. A linear elastic structure
with TMD can be mod-

www.samco.org Page 76 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

elled by multi-degree-of-freedom system with N degree-of-freedoms where the


TMD is connected with the k'th degree-of-freedom of the structure. The N +1
equations of motion are given by

Fig. 31: Two degree of freedom model of a TMD attached to a primary structure.
Left: Excitation force f (t) acting on the primary mass. Right: Excitation through
base acceleration Ug (t).

MpUp (t) + CpUp (t) + KpUp (t) - (ctut(t) + ktut (t)) Ek = F(t) - (MpE)ug (t), mtup k (t)
+ mtiir (t) + ct& (t) + ktur (t) = -mtii g (t),

where Mp,Cp and Kp are the mass, damping and stiffness matrices of the multi-
degree-of-freedom model of the structure, E is the column vector with a one in
each row, Ek is the column vector with a one in the k'th row and a zero in all other
rows
and upk is the displacement of the k'th degree-of-freedom of the structure. We as-
sume that the TMD shall be designed to mitigate the vibrations associated to the
mode shape y/k, where y/k is a solution of the eigenvalue problem
MpYko>l - Kpyk = 0 and is normalized in such a way to display a one in its k'th
row. Using the mode shape y/k, the displacement of the primary structure
associated to this mode shape can be described by Upk = pkupk. The two-degree-
of-freedom
model is obtained by multiplying the first equation in Eq. (3) from the left with the
transpose of the mode shape y/k. By this operation, the relationship between the

www.samco.org Page 77 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

parameters describing the structure of the two-degree-of-freedom system and the


N- degree-of-freedom system are given by
mp = YkMpYk, cp = VkCpVk, kp = v[Kpyk, f (t) = y[F(t), rhp = y^MpE .

Observe that y[Ek = 1 because of the specific normalization of the mode shape
y/k. In general, the participating mass is not equal to the modal mass of the
primary system: mip ф mp. By introducing the participation factor

r _ VTkMpE _ 1 + ҐTkMp(E-Ґk) (5)


1 m _ T ,, _ 1 + T ,, (5)
ҐkMpҐk ҐkMpҐk
mp can be expressed as mp _rmmp. In general, because TMDs are connected to
the degree-of-freedom with maximum modal amplitude, the participation factor rm
is greater than unity. rm has an effect on optimal tuning parameters of the TMD in
the cases of base excited structures (e.g. earthquake loadings). Since the ratio of
the mass of the TMD and the modal mass mp is small, in general, this effect is
also
small.
H„ optimization
Until now, many optimization criteria for the TMD have been proposed. The most
common criteria are the H„ and H2 norm optimization criteria. The H„ norm of a
scalar transfer function G(z) is defined by
||G(z)|| _ max|G(z)|, (6)
II lira z1 1
that is, the H„ norm represents the maximum amplitude of the absolute value of
the transfer function. The absolute value of the complex transfer function |G( z)| is
also
known as the dynamic magnification factor. The H„ norm provides a bound of the
output gain, where the gain is defined in terms of the H2 norm:
11x4 <||G(z)|| J f (4. (7)
That is, given a square integrable excitation f (t), the H„ norm provides a bound of
the integral of the square response. Near equality is achieved for a harmonic
excitation f (t) _ Fsin(fit) provided that the integral is taken over a finite time interval

www.samco.org Page 78 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

0...tmax, since any harmonic function is not square integrable over an infinite time
interval. Therefore, Eq. (7) provides a sharp bound for harmonic excitations acting
over a long but finite period of time. It is important to note that Eq. (7) does not pro-
vide any information about the maximum amplitude of the output y(t) within a time
interval for an arbitrary excitation f (t). Dimensionless complex transfer functions
G( z) of various response parameters of the primary system up (t) and the relative
displacement ur (t) of the TMD mass with respect to the primary system are given
in Table 1.
The objective of the H„ norm optimization is to find the frequency ratio n, and
damping ratio Zt of the TMD that minimize ||G(z)||:
mini|G(z)|| _ minmax|G(z)|. (8)
n, Z, n, Z z 1 1
With respect to the graph of the absolute value of the transfer function G( z), the
minimum of ||G(z)||is achieved if and only if there exist two local maxima and both
have exactly the same amplitude (see Fig. 32 and Fig. 33).
Den Hartog's fixed-points method (Den Hartog (1947)) for computing
approximations of the optimal parameters n and is based on the observation that
for primary

www.samco.org Page 79 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

systems with vanishing damping (Zp = 0), there exist at least two frequencies z1
and z2 where |G(z)| is invariant with respect to a variation of Z, (v, is fixed), that is,
(9)
dZt
28
15
28

15
18
8.8
1.2
5
18
о
О
1.4
д |G(z,) d |G( Z2)
= 0.
dZ,
h> h
-h h (zfhJ (z2>hJ -
=
h< h

www.samco.org Page 80 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

0.8
1.2
5

1.4
Fig. 32: Methods for l-L norm optimization of a Fig. 33: Methods for l-L norm
optimization of a TMD TMD applying the fixed points method. applying the
real roots method.
Hence, all graphs of |G(z)| with fixed -q, and variable Z, cross the points (z1;|G(^)|)
and (z2,|G(z2)|). The approximations of the optimal parameters qt^ and are ob-
tained by maximizing |G(z1) and |G(z2)| using the condition |G(z1) = |G(z2)|. The
fixed-points method provides simple closed form approximations of the optimal pa-
rameters qt^ and Z,,^ that are accurate to a few percent within the range of mass
ratios that are typical for civil engineering applications of a TMD (0.01 <j< 0.1) (Fig.
32). Unfortunately, the fixed-points method is limited to primary systems with
vanishing damping since the presence of structural damping destroys the fixed
points.
A method that provides exact closed form solutions of the H„ norm optimization
problem was recently proposed by Nishihara and Asami (2002). Their method is
based on the observation that the function H(z) = h - |G(z)|, where h is an arbitrary
real constant, has exactly two positive real roots z1 and z2 of multiplicity two if and
only if h = h^ = min||G(z)||^. This concept is sketched in Fig. 33. For h = hx and qt^
and Z,,^
the graph of |G(z)| touches the horizontal line hx at the two points (z1;|G(z1)|) and
(z2,|G(z2)|). The algebraic expressions of the exact closed form solutions are more
complicated than those obtained by the fixed-points method.
Table 2 lists approximate solutions of the H„ norm optimization problem for a
primary structure with vanishing damping. The formulas are obtained by
simplifying the exact solutions by using a rational functions approximation
technique. The error of the formulas with respect to the exact solutions is smaller
than 1% for ju< 0.1. The formulas
regarding h = mini|G(z)|| show that the effectiveness of the TMD increases with
increasing mass ratio j . The effectiveness is very sensitive to an inaccurate fre-
quency tuning. A tuning error of ±5% with respect to the optimal frequency
increases significantly the peak value of the transfer function of the primary
structure (see Fig.

www.samco.org Page 81 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

34). Much less critical is the correct tuning of the damping ratio of the TMD. A
tuning error of ±5% has little influence on the peak value of the transfer function of
the primary structure (see Fig. 35).

О 0.8
О 1.2
0 0.6
0 0.6 1.4
50 no
no TMD
TMD — Zt = ^ = 0.02
nt = V | = 0.02 ■ --- Zt = i n, = v -
■ = 0-95 v Г: Z = Z - o.95 Z,m
: 't Ч» -- Zt = 1.05
._._ п, = -1- Zt,»
05 nti„
i ■ '. 1 ■■ ' ■ \

\/. , 11
'.,. -
0.8
1.2
iii'
40 1.4
30
20
10

Fig. 34: Effects of poor frequency magnification factor of the dis-


tuning of a TMD on the dynamic placement of the primary system.

www.samco.org Page 82 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 35: Effects of poor damping magnification factor of the dis-


tuning of a TMD on the dynamic placement of the primary system.

In most practical applications, the primary structure has non-vanishing damping


so that the equations given in Table 2 represent an approximation of the optimal
parameters. Unfortunately, no simple equations exist for non-vanishing damping.
However, if the damping ratio of the primary system Zp is less than 1%, the
optimal parameters given in Table 2 are still a reasonable good approximation for
engineering applications. With increasing damping of the primary system, the
equations of Table 2 become increasingly inaccurate. Fig. 36 and Fig. 37 display
the optimal parameters П and that achieves the minimum of the displacement of
the primary system
up(z) for several damping ratios Zp of the structure. The H„ norm of the non-
dimensional transfer function of the displacement of the primary system up (z)
and of the relative displacement of the TMD mass ur (z) is displayed in Fig. 38
and Fig. 39. The figures show that the maximum displacement of the primary
system as well as the maximum relative displacement decreases with increasing
mass ratio. It is important to note that the maximum amplitude of the relative
displacement ur (z) is large for small mass ratios ц.
The effectiveness of a TMD can be described by using the mitigation factor that
is defined as the ratio of the non-dimensional H„ norm of the primary system
without TMD ||Gp (z)|| , to the non-dimensional H„ norm of the primary system
with optimally
tuned TMD ||G(z)||„ :
G (z)|| 1
Я=,p ^—n- . (10)
" G(z) 2Z G(z)
II ||rc> * p N llrc>
Fig. 40 shows that that the mitigation factor of an optimally tuned TMD increases
with increasing mass ratio and decreases with increasing damping ratio of the
structure. The increase with increasing mass ratio is particularly strong for small
structural damping. In the range of practically realizable mass ratios, a TMD
provides good performance only for lightly damped structures (Zp ^0.01). The

www.samco.org Page 83 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

smaller the damping

www.samco.org Page 84 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

ratio the better is the performance of a TMD. In structures with a structural


damping ratio of 5%, the dynamic response of the primary system can only be
reduced by approximately a factor of two.

Fig. 37: Optimal TMD damping ratios for minimizing the displacement of the
primary system with respect to the H»
norm.

Fig. 36: Optimal TMD frequency for minimizing the displacement of the primary
system with respect to the H» norm.
IX
I

200
0.01
—Z=0p
--- Z = 0.01 p
.-.- Z = 0.02

www.samco.org Page 85 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Z = 0.05 _=p
0.02 0.03 I
0.04
0
0
Fig. 39: Maximum of the non-
dimensional transfer function of the
relative displacement for parameters
minimizing the displacement of the
primary system with respect to the H»
norm.
0
0.05
0
0.05

Fig. 41: Amplification due to a


perturbation of frequency tuning of a
TMD optimized for minimizing the
displacement of the primary system
with respect to the H» norm (8 = rjt
lqt<x- 1, | = 0.02).
25
1\1\ — zp = 0

\\ ---Zp =
o.o1
.-.- Z =
0.02 p

www.samco.org Page 86 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Z = 0.05

\\X
\4:4^

20
= 15 U 10
0.01
0.02 0.03 I
0.04
5

Fig. 38: Maximum of the non-


dimensional transfer function of the
displacement of the primary system for
parameters minimizing the displace-
ment of the primary system with
respect to the H» norm.

Fig. 40: Mitigation factor of a TMD


optimized for minimizing the
displacement of the primary system
with respect to the H» norm.

www.samco.org Page 87 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

The performance of a TMD with respect to H„ norm is very sensitive to incorrect


frequency tuning. The frequency detuning amplification with respect to an
optimally tuned damper can be defined as the ratio of the non-dimensional H„
norm of the response of the detuned primary system ||Ge(z)||, to the non-
dimensional H„ norm of
the primary system with optimally tuned TMD jGopt(z)|| :
. (11)
Gopt (z) »
Aexi increases with increasing frequency detuning and decreasing mass ratio,
and
decreases with increasing damping of the structure. Fig. 41 displays the
frequency detuning amplification due to a perturbation of frequency tuning of a
TMD optimized for minimizing the displacement of the primary system with
respect to the H„ norm. The amplification is considerable and must be accounted
for in the design since live loads or environmental effects induce changes of the
natural frequencies of a structure. Frequency detuning increases also the H„
norm of the relative displacement. However, the detuning amplification is much
smaller than that of the response of the primary system.
Hp optimization
The H2 norm of a transfer function G( z) is defined by

!т ГГ Iz7°,
! G(t)2dt =Vn f G(z)\dz (12)
llG( z)||2 =
Л
=—то ■ z=—1X1

and is essentially a measure of the area between the square of the absolute
value of

www.samco.org Page 88 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

I |2
the transfer function |G(z)| and the z -axis. If the system is subjected to a
random, white noise excitation, the H2 norm of the transfer function G( z) is
directly related to the variance of the output associated to the transfer function.
That is
C7y = E[(y(t) — E[y(t)])2] = I|G(z)|2 Sf , (13)
where y(t) is the response of the structure defined by y(t) = G(t) ® f(t), ® is sym-
bolizing the convolution of the transfer function G(t) with the excitation f (t), E[y(t)]
is the average of the output that is usually zero, and Sf is the spectral density of
the excitation f (t). For ergodic processes, the variance can be estimated via
temporal averages
1 t=T 2
ay = lim— f (y(t) — E[y(t)])2dt = ||G(z)||2 Sf . (14)
T 2T J—T
The objective of the H2 norm optimization is to find parameters nt2 and Zt2 that
minimize ||G( z)|2:

min| |G( z )|2 =


minj^
2ж\
n Z, 2 Vt Z, z
f |G(z)|2dz . (15)
z=

www.samco.org Page 89 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Fig. 42: Normalized variance of the displacement of the primary system.

||G(z)\\2 is a function of n,, Z, Zp and ^. Generally, closed form expressions of


||G(z)||2 can be obtained by evaluating the integral in Eq. (12) (Crandall and
Mark (1963)) or by computing the observability grammian of the linear system
described by
ii ii2
the transfer function G(z). Table 3 lists closed form expressions of ||G(z)||2 for
the most common minimization objectives. Fig. 42 displays the normalized
variance of the displacement of the primary system. Minimization of ||G( z)||2
is achieved if
d| |G( z)|L Э1 |G( z)|L
11 "2 = 0 and 11 "2 = 0 (16)
дП, К V;
hold. These conditions are usually sufficient and yields two nonlinear
equations defining the optimal parameters n,2 and Z,2.
Exact closed form solutions have been obtained for systems with vanishing
structural damping (Zp = 0). Table 4 lists exact solutions for various excitations
and response
parameters. For primary systems with structural damping, accurate closed
form expressions can be obtained in form of a power series with respect to the

www.samco.org Page 90 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

damping ratio Zp. A list of first and second order terms for various excitations
and response parameters can be found in Table 5. In general, the optimal
parameters with respect to the H2 norm are less sensitive to structural
damping Zp than the optimal parameters
with respect to the H„ norm (see Fig. 43 and Fig. 44). For small mass ratios,
the optimal damping ratio Z,2 which minimizes the variance of the
displacement of the primary system with respect to the H2 norm is virtually
independent of the structural damping Zp. Fig. 45 and Fig. 46 display the non-
dimensional variance the displacement of the primary system up (z) and of the
relative displacement of the TMD mass
ur (z) of a TMD optimally tuned with respect to minimization of displacement of
the primary system.

www.samco.org Page 91 of 155


SAMCO Final
Report 2006
F05 Guidelines
for Structural
Control

— Vе1
r = 0.01

С
—-;; = o
= 0.02
--- ;. =
ч!
o.oi
U '-'
.— L =
0.02 ^
=
D.9B
□.96
0.1
□.92
□.05
0.01
0.01
D.02
□.02
0.04
0.03
0.04
D.B
0ДЗ
0.05
0.05
□.15

Fig. 43: Optimal TMD frequency for


minimizing the variance of the
displacement of the primary system
with respect to the H2 norm.

www.samco.or
g Page 92
of 155
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 44: Optimal TMD damping ratios


for minimizing the variance of the
displacement of the primary system
with respect to the H2 norm.
11' 1 — с =o -
111 --, =0.01
1 1 ', ._._ i;
\ =0.02 p
■1V ц
=0.05
P
н1\ \
11
■■ ч\
'у\

3
r i v.
E 400

Fig. 45: Minimum of the non-


dimensional variance of the
displacement of the primary
system for parameters
minimizing the variance of the
displacement of the primary

www.samco.org Page 93 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural
Control

system with respect to the H2


norm.

0.05
Fig. 46: Non-dimensional
variance of the relative
displacement of the mass of the
TMD for parameters minimizing
the variance of the displacement
of the primary system with
respect to the H2 norm.
J*
15
^
= 0 DC 5
— ^ = 0-
01 -.- с =
0 oi
С
= D.DS
-"T ......

--■-

10
0.01
0.:: 0.03
0.04

www.samco.org Page 94 of
155
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 47: Mitigation factor of a TMD a TMD optimized for minimizing the
optimized for minimizing the displacement of the primary system
displacement of the primary system with respect to the H2 norm (S = tft
with respect to the H2 norm. ltft^ -1, Ц = 0.02).
Fig. 48: Amplification due to a
perturbation of frequency tuning of

Similar to the case of H„ norm optimization, the mitigation factor is defined as the
ratio of the non-dimensional H2 norm of the primary system without TMD \Gp (z)\
to

www.samco.org Page 95 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

the non-dimensional H2 norm of the primary system with optimally tuned TMD
G( z)
\\GP (Щ 1
X=,p2 -i . (17)
G(z)2 4ZP G(z)2
The effects of mass ratio and damping ratio of the structure on the mitigation
factor X2 are equal to those observed for X^ (see Fig. 47). For small structural
damping, the overall performance is approximately 20% smaller than that of a
TMD optimized with respect to H» norm.
Similar to the H» norm, the H2 norm is sensitive to changes of the frequency ratio
n, and much less sensitive to changes of the damping ratio Z, ■ This is evident by
observing the shape of the H2 norm shown in Fig. 42. Defining the frequency
detuning amplification factor with respect to an optimally tuned damper as
X, 2 , (18)
' \\GoP' (Z)ll2
where ||Ge(z)||2 is the non-dimensional H2 norm of the response of the detuned
primary system and jGopt (z)||2 is the non-dimensional H2 norm of the primary
system
with optimally tuned TMD. Xe2 increases with increasing frequency detuning and
decreasing mass ratio, and decreases with increasing damping of the structure.
Fig. 48 displays the frequency detuning amplification factor due to a perturbation
of frequency tuning of a TMD optimized for minimizing the displacement of the
primary system with respect to the H2 norm. Since the H2 norm is defined via an
integral, the amplification is significantly smaller than the amplification observed
for TMDs optimized with respect to the H» norm. Frequency detuning increases
also the H 2 norm of the relative displacement. The amplification factor is of
similar magnitude of the amplification of the response of the primary system.
Design consideration
Based on the theory developed in the last sections, a design procedure can be
developed. The very first step in designing a TMD is to take a decision according
to which norm the TMD should to be optimized. An optimization with respect to H»
norm is recommended for structures that are excited by loads exhibiting mainly

www.samco.org Page 96 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

periodical time components (e.g. loads generated by human activities like walking,
running, jumping, dancing etc. or machines). For loads having mainly a wide band
stochastic character (e.g. wind loads, earthquake loads), an optimization with
respect to H2 norm is more appropriate.
2
The second step is to identify the vibration mode of the structure that shall be
damped. Using the shape of this mode, an equivalent single-degree-of-freedom
model of the structure is generated by computing the associated modal mass and
stiffness. For obtaining a scaling of the modal mass that is compatible with Eq. (1)
or (2), the amplitude of the mode shape at the position where the TMD is attached
to the primary structure has to be chosen equal to unity in the direction of action of
the TMD. The modal stiffness is estimated using the modal mass and the natural
frequency of the vibration mode. The natural frequency should be determined with
field tests because the natural frequencies computed from analytical or numerical
models

www.samco.org Page 97 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

are too unreliable to be used for design purposes. As already pointed out, an
inaccurate frequency tuning reduces significantly the effectiveness of a TMD.
Although much less critical as the natural frequency, the structural modal damping
should also be determined by field tests or estimated conservatively based on
experience.
The third step regards the choice of the mass of the TMD. For a given damping
ratio Zp of the structure, the mass ratio i determines the optimal parameters nt,opt
and
Ztp as well as the response of the structure. The mass ratio is chosen to satisfy
the
maximum acceptable response level of the structure. In principle, the latter can al-
ways be achieved since the response of the primary structure decreases
monotoni- cally with increasing mass ratio. The maximum acceptable response
level may be given by codes, guidelines, provisions or generally accepted rules.
In the design of a TMD, a number of issues have to be considered:
A TMD is effective if a sufficiently large TMD mass (mass ratio i>0.02) can be
accommodated and the vibration mode exhibit small structural damping 0.01 <ZP.
Beyond a mass ratio of i = 0.05 a further increase of
TMD mass results only in a modest further reduction of the response of the
primary system.
A TMD is only effective when properly tuned to a specific natural frequency or,
equivalently, to a specific vibration mode. Structures with several closely spaced
natural frequencies which are all inside the frequency band of the excitation
require a TMD for each natural frequency for achieving an effective vibration
mitigation. Since each natural frequency is associated to a vibration mode and the
coupling between the vibration modes is usually negligible, the design procedure
for single modes can be applied.
The relative displacement, the displacement of the mass of the TMD with respect
to the structure, can be considerable and requires space that must be
accommodated within the structure. If the relative displacement of an optimally
tuned TMD exceeds existing space constraints, the relative displacement can be

www.samco.org Page 98 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

reduced by increasing the mass of the TMD or, if the latter is not feasible, by
increasing the damping ratio Z, of the TMD. For
structures with structural damping (Zp > 0) a limitation of relative displacement
can always be achieved. However, any restriction of the relative displacement that
can not be accommodated by an increase of the mass of the TMD reduces the
effectiveness of the TMD.
The mass of the TMD with respect to the modal mass of the structure determines
the effectiveness of the TMD. In existing structures, the mass of the TMD may be
restricted by the load bearing capacity of the structure and the installation of a
single TMD may require a strengthening of the structure. A design with multiple
TMDs having smaller masses may avoid a strengthening. However, the
effectiveness decreases if not all TMDs can be installed at the locations with
maximum amplitude of the mode shape. Furthermore, because of the smaller
mass, the TMDs will exhibit a larger relative displacement and hence require more
space than a single TMD and, finally, the costs of multiple TMDs is generally
larger than the costs of a single TMD.
Natural frequency and damping ratio of structures may change because of the
effect of live loads (e.g. pedestrians on footbridges) or environmental parameters
(e.g. temperature). Since the effectiveness and response of

www.samco.org Page 99 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

TMDs are very sensitive to an incorrect frequency tuning, changes of natural


frequencies have to be considered in the design. Bad frequency tuning increases
the response of the structure as well as the relative displacement of the TMD.
6. Since the relative displacement can be considerable, the fatigue behaviour of
the TMD's springs has to be considered.
Testing and validation
For determining reliably the relevant absorber parameters, the natural frequencies
and the damping ratios of the structure's relevant modes have to be established
experimentally. This can be accomplished by ambient or forced vibration tests. A
reliable estimation of the damping ratios can usually only be established by forced
vibration tests. For lightweight structures, a simple and effective method is an
impact test (e.g. by dropping a sandbag).
A TMD should be tuned in the laboratory before proceeding to it's installation on a
structure. Natural frequency and damping ratio of a TMD are established by free
vibration tests. Notice that the damping devices may contribute to the stiffness of
the TMD and therefore affect its natural frequency. This has to be considered
during the tuning of a TMD because of its high sensitivity with respect to
frequency tuning.
The final tuning of the TMD's natural frequency is performed after installation. The
easiest method for frequency tuning is to increase or reduce TMD's mass to fit the
correct natural frequency. For lightweight structures, the validation can be
performed by impact test (e.g. by dropping a sandbag). The impulsive force
covers a wide frequency band so that the response of the structure is similar to
the dynamic magnification curve shown in Fig. 34. Structures with an optimally
tuned TMD show an amplification curve with two humps of equal height. Another
useful test is to compare the response of the structure with locked and unlocked
TMD. The effectiveness of the TMD can be obtained by computing the ratio of the
maximum response with locked and unlocked TMD. Obviously, a reliable
estimation requires a test with controllable excitation forces.
Implementation examples
The most common implementation of a TMD consists of a mass either hanging or
being supported by several springs as shown in Fig. 30. This design is often used
for controlling bridge and floor motions. However, there exist a large variety of
different designs. Pendulum TMD are very often used for motion control of steel
smoke stacks (Petersen (2001)). Solutions for space-limited applications include
the tuned roller pendulum damper and the multi stage pendulum damper (Soong

www.samco.org Page 100 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

and Dargush (1997)). TMDs have been successfully installed for vibration
response control of:
Pedestrian bridges, stairs, spectator stands excited by walking or jumping people
(Bachmann and Weber (1995)).
Lightweight factory floors excited in one of their natural frequencies by machines.
Tall free-standing structures (bridges, pylons of bridges, smoke stacks, towers)
excited by wind-induced loads (Kwok (1984), Kwok and Samali (1995), Ohtake et
al. (1992), Petersen (2001), Petersen (1980), Ueda et al. (1992)).
References
Asami, T., et al. (2002), "Analytical solutions to H-infinity and H-2 optimization of
dynamic vibration absorbers attached to damped linear systems", Journal of
Vibration and Acoustics-Transactions of the Asme, 124(2), 284-295.
Asami, T., and Nishihara, O. (2003), "Closed-form exact solution to H-infinity opti-
mization of dynamic vibration absorbers (Application to different transfer functions
and damping systems)", Journal of Vibration and Acoustics-Transactions of the
Asme, 125(3), 398-405.
Bachmann, H., and Weber, B. (1995), "Tuned Vibration Absorbers for "Lively"
Structures", Structural Engineering International, 5(1), 31-36.
Crandall, S. H., and Mark, W. D. (1963), Random vibration in mechanical
systems, New York: Academic Press.
Den Hartog, J. P. (1947), Mechanical Vibrations, Third ed, New York and London:
McGraw-Hill Book Company.
Frahm, H. (1909), Device for Damped Vibrations of Bodies. U. S. Patent No.
989958.
Ioi, T., and Ikeda, K. (1978), "On the Dynamic Vibration Damped Absorber of the
Vibration System", Bulletin of the Japanese Society of Mechanical Engineering,
21, (151), 64-71.
Korenev, B., and Reznikov, L. M. (1993), Dynamic Vibration Absorbers: Theory
and Technical Applications, Chichester, Eng-land: John Wiley & Sons Ltd..
Kwok, K. C. S. (1984), "Damping Increase in Building with Tuned Mass Damper",
Journal of Engineering Mechanics-Asce, 110(11), 1645-1649.
Kwok, K. C. S., and Samali, B. (1995), "Performance of Tuned Mass Dampers un-
der Wind Loads", Engineering Structures, 17(9), 655-667.
Nishihara, O., and Asami, T. (2002), "Closed-form solutions to the exact optimiza-
tions of dynamic vibration absorbers (mini-mizations of the maximum amplitude

www.samco.org Page 101 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

magnification factors)", Journal of Vibration and Acoustics-Transactions of the


Asme, 124(4), 576-582.
Ohtake, K., et al. (1992), "Full-Scale Measurements of Wind Actions on Chiba
Port Tower", Journal of Wind Engineering and Industrial Aerodynamics, 43(1-3),
22252236.
Ormondroyd, J., and Den Hartog, J. P. (1928), "The Theory of the Dynamic Vibra-
tion Absorber", Transactions of the ASME, APM-50-7, 9-22.
Petersen, C. (2001), Schwingungsdampfer im Ingenieurbau, Munchen: Fa.
Maurer und Sohne, GmbH & Co. KG.
Petersen, N. R. (1980), "Design of Large Scale Tuned Mass Dampers", Structural
control, June 4-7, 1979. Ontario, 581-596.
Soong, T. T., and Dargush, G. F. (1997), Passive Energy Dissipation Systems in
Structural Engineering, Chichester: John Wiley & Sons.
Ueda, T., et al. (1992), "Suppression of Wind-Induced Vibration by Dynamic
Dampers in Tower-Like Structures", Journal of Wind Engineering and Industrial
Aerodynamics, 43(1-3), 1907-1918.

www.samco.org Page 102 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

www.samco.org Page 103 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Warburton, G. B. (1982), "Optimum Absorber Parameters for Various Combina-


tions of Response and Excitation Parameters", Earthquake Engineering &
Structural Dynamics, 10(3), 381-401.

Description Unit
non-dimensional complex transfer function -
H„ norm of non-dimensional complex transfer function
-
H2 norm of non-dimensional complex transfer function
- dashpot constant frequency stiffness constant
mass
frequency ratio -
mass ratio -
circular frequency rad/s
Often used notations
Symbol
G(z)
IG(; )|_ |G(z )| 2
cfкm
ПM(
z
kg/s Hz kg/s2 kg
damping ratio -
Subscripts
p
rt
Description main structure relative
tuned mass damper

www.samco.org Page 104 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

A(z)
Appendix
F Up П2 - z2 2ntZtz
U0
F up -2ntZtz2 z(n2 - z2)
®p
UQ
F UP -z2(n2 - z2) -2ntZtz3
2
®p
UQ
F UQ z Q

Ug 2 (1 + |)n2 - z2 2(1 + |)ntZtz


®P
UP
Ug
Ug ®P -2(1 + |)nt Z tz2 z((1 + |)n2 - z2)
UP
Ug
Table 1: Non-dimensional transfer functions for various excitations and
response parameters.
Exci- Output tation parameter
B(z)
U p + UQ
-z2((1 + |i)n? - z2)
-2(1 + |)n Z tz3

www.samco.org Page 105 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

2
Q
1
g
®pur
g

Legend
A(z) + iB(z)
G(z) = : non-dimensional transfer function with coefficients A(z),
B(z),C(z) and
C(z) + iD(z)
D(z), where C(z) = z4 - ((1 + |)nt + pQnt + 1)z2 +n2 and
D(z) = 2z(((1 + |)Z t^t + Zp)z2 - (Z t + Z pHt)Ht)
F: excitation (force) applied to the primary system
ii : excitation (acceleration) applied to the base
uq : reference displacement ( uq = FQ/kp for harmonic excitation with
Feimt and
uq ^>/s7/kp for
z:
white noise excitation w(t) with autocorrelation Rww(t) = Sw8(t))
non-dimensional frequency z = ffl / fflp

www.samco.org Page 106 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Table 2: Approximate solutions of the H« optimization problem for primary systems with vanishing structural damping (Z p = 0).
Exci- Minimized Transfer „ tation parameter Ct,» mH|G(z)|| » ur(z)|» u0
function »
Fe1(t Up U 1 -1M ^ 3» J, 3 hi 17 4 1 + » 1 + 32 » if + 32 » JT5 »
0 8(1 +
»)
Fe1(

fi+2 ^8(1 +MJ


JMJi-
+ 7M 112 ц\ 1 + 23m 124
1
1+ »1
,,1 + — »

www.samco.org
Pa
SAMCO Final Report 2006
F05 Guidelines for Structural Control

8(1 + »)\ 24
p
°pUc
-JT5 » V 2 100^

5ГП
\M V1 + 23m 132
3m L 27
—. 1 + —m
8(1 + mH 32
1+ M12
1
Fe1(
p
1+m
>u0

www.samco.org
Pa
SAMCO Final Report
2006
F05 Guidelines for
Structural Control

4 1 + m л/Г5 m V1 -
4vm15 + 11m 110

2 u
°pUp uge
Ф-M12 1 + m g
* 1+m 8(1 + -4- ^ 1+3vm+62 m
M)V 48 V15 m v 5s* 75
1
1+m
^ Ittil
V15 m V 16^
3m
(pu p
uge
8(1 + m)
g

www.samco.org
Pa
SAMCO Final Report 2006
F05 Guidelines for Structural Control

li
4 (1 + ») г рг
1
1+m
-3* 1+i-m
8(1 + ») V 32
H1+32»
Up + uo
llp+ug
uge
g

Legend
Fe1(t: harmonic excitation applied to the primary system
uge1(t: harmonic excitation applied at base

www.samco.org
Pa
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Table 3: Closed form expressions of the non-dimensional H2 norm for various excitations and response parameters.

Exci- Response G(z) 2


tation parameter
F E[up]mpfflp 2nSf 1 (1 +1)2 ztn4 + (i + 4(1 + |)z2) cptf - (2 + |- 4(1 + |)z2 - 4^2 )nf + 4c?cpnt + zt 4 (1 +1)2z^4 + (i + 4(1 +
|)c?)с2л? - 2(1 - 2(1 + - 2ф ztcpnf + (i + 4ф^ч + ztzp
F E^Kfflp 2nSf 1 (1 + tn4 + (i + 4z2) - 2(1 - 2z2 - 2фп2 + 4c?cpnt + zt 4 (1 +1)2z^4 + (i + 4(1 + |K?)- 2(1 - 2(1 + - 2ф +
(i + + ztzp
F E[U 2]mp 1 N(|'n" zp,Q - 4(otc+c) 4(1 + |)2 ад + (i + 4(1 + |)c?) cpn? - 2(1 - 2(1 + |)z2 - 2ф ctcpn? + (i + p

2nSf rap N(|, nt, zp, zt) = (1 + 4(1 + |) zt) + (1 + 4(1 + 3| + 2|2(1 + ф) z2)ztn4 + (i-4(1 + |) zf + 16DC + 4(| + 4(1 +
2|) ф ф ^3 - (24(1 + i-i%2 + 4(1 - 4(1 + |) ф zp-1бф n2 + 4ф22(1 + 2| + 4z2)nt + W + 4ф
F E[u2]mprap 1 zpnt + zt
2nSf 4 (1 +1)2ztzpn4 + (i + 4(1 + !)Ф- 2(1 - 2(1 + - 2ф ctcpn? + (i + 4ф^Ч + ztzp
Ug E[u>p 2nSf 1 (1+ i)4 z tn4 + (1+ i)2 (i + 4 (1 + IK2 ) zpn? - (1+ i)2 (2 + i- 4(1 + iK2 - 4zp)n2 + (4 (1+ i)2 z2 +12) zpnt
+ zt 4 (1 +1)2 ztcpn4 + (i + 4 (1 + dc?) qn? - 2(1 - 2(1 + iK? - 2ф z tcpn? + (i + 4ф?2 n + ztzp
Ug Щрн 1 (1+i)3 z tn4+(1+i)(i+4(1+!)Ф zpn? - (1+i-2(1+i)2 z2 - 2zp)nr + 4 z2 zpnt+z t 4 (1 +1)2ztzpn4 + (i + 4(1
2nSf + !)Фc2n3 - 2(1 - 2(1 + DC - 2ф вд + (i + 4?r)?t2nt + ztzp
E[(U p + U g)2] 1(1 + i)(1 + i + 4cp)ctn4 + (i + 4(1 +i + 4ф^ + 4^)zpn3 -(2 + i + 4(1 -4фzp -4(1 + i + 4cr)c2)ctn2 + 4(1 +

www.samco.org
Pa
SAMCO Final Report 2006
F05 Guidelines for Structural Control

4фад + (1 + 4^
Ug 2nSf rap 4 (1 +1)2 ztcpn4 + (i + 4(1 + i)z2) c2n3 - 2(1 - 2(1 + i)z2 - 2q) z tzpn2 + (i + 4фс2 nt + ztzp
E[u2]mpK>p 2nSf
1 (1 +1)(1 + | + 4zp) gtn? + (| + 4(1 + |)z2 + 4^g)cpnt2 + (4 zp +i) сл + zp
4 (1 +1)2zIn4 + (1 + 4(1 + cpn3 - 2(1 - 2(1 + - 2ф ztcpnf + (i + 4cP)cfnt + ctc
g

random excitation applied to the primary system with


spectral density Sf random excitation at base with spectral
density Sf
Legend F:
g
E[up], E[Up], E[U2]:
E[ur2]:
mean of the square of the displacement up, velocity Up
and acceleration U p mean of the square of the relative
displacement ur

www.samco.org
Pa
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Table 4: Exact solutions of the H2 optimization problem for primary systems with
vanishing structural damping (Zp = 0).
Parameter optimized
E(u2)mpa>p 2nSf
Exci- Minimized pa- tation rameter
Zt,2
min G(z)
П2
2

F E[u2K®p up J- A 1 1Г1+» j3/2


p
2nSf ь+» \ 4 v»\ 1+» 21 M J
V1+»/2 1+»
4(1 + »)V
1 + 3»/4

www.samco.org
Pa
SAMCO Final Report 2006
F05 Guidelines for Structural Control

1 + »/2
1 1 + 3» / 4
1 (11» 2 ^ »
1 + 3»/4
вд2К
2nSf rap
V1 -»/2
(1 -» / 4)(1 + »)
4(1 + »)\(1 -» /2)(1 -m + »2)
-V (1 -» + M )(1 -»/4)
(1 -m + m2)
u
F
E[up]mprnp 2nSf
1
M-
p
F
mv1+»
2 [1J \ 1 -M/4

www.samco.org
Pa
SAMCO Final Report 2006
F05 Guidelines for Structural Control

E[u2] ^p 2nSf
i
V1 -»/2 1+»
1 -»/4
1 -»/4 1 + »
1Г1 + » ) + »/2 1
-(1+m)2
u
g
4(1 + »)\1 -»/2
2^»
1 r 1 (1 + »)2 r '
1 -»/2^1 -»/4
E[uplfflp
2nSf
1(^J
1

www.samco.org
Pa
SAMCO Final Report 2006
F05 Guidelines for Structural Control

1+»
(1 + »)3
g
4(1 + »)
E[(tip + iig^] 2nSf rap
V1 + »/2 1 + »
111+»J (1+»)21+3»/M 1
1 + 3»/4
1 1 + 3» / 4
M-
uipr + ug
4(1 + »)V
g
2^»
1 + »/2 \1 + 3»/4
1 + »/2
»v 1+»

Legend

www.samco.org
Pa
SAMCO Final Report 2006
F05 Guidelines for Structural Control

F: random excitation applied to the primary system with spectral density Sf


ug: random excitation at base with spectral density Sf
E[up], E[up], E[uip]: mean of the square of the displacement up, velocity up
and acceleration up
E[u2]: mean of the square of the relative displacement ur

www.samco.org
Pa
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Table 5: Approximate solutions of the H2 optimization problem for primary systems with
structural damping (Zp > Q).

Ex-
Z t,2
n
2
cita Minimized
tion parameter

E[up]mprnp 2nSf i
(1+ I/4) I17' rzp -
F
(1 + I)3/2(1 + I
17 I1/2(1 + I /4)
/2)1/2(1 + 3i/4)1'
" 2 (1 + I)3/2 (1 +1/ 2)1/2
(1 + 3I / 4)1

www.samco.org
Pa
SAMCO Final Report 2006
F05 Guidelines for Structural Control

32(1 + i)(1 + 3i/4)(1 + i /2)3


Z I(3 + 19I/4 + 33I2/16 +
13ц18/64) ■ p + 2(1 + I)3(1
+ I /2)3/2(1 + 3I/4)
(1 + 5i/4-15i2/64) i5/2 64(1
+ i)3/2(1 + i /2)5/2(1 +
3i/4)2

18 i1/2 I(1 -7I/8) 2 + 5 (12


+ I) Z3

www.samco.org
Pa
SAMCO Final Report 2006
F05 Guidelines for Structural Control

i1/2 z i(1 + 7i/8) 7 i1/2(1 + 25i/28) F


,3 (1+ i)+ (1 + i)3/2 zp + 4 (1 + i)2 (1+i)
zp 2nSf
^ ^ r 3 i1/2(1 + 19i/12) r2 21 (1 +
4(1 + I)1/^p 4 63i /64) Z3 + 2
-c3 p (1+i)3

E[U 2]mp 2nSf


rap
(1 - 4351/112)
(1-121I/28)
(1 -I/4)
zp + 2
F
(1 -I /2)(1 -I-3i2)
3 i1/2(1-
25i/12+3i2/4) 2
(1 -i/4)
I(1 -I/4)
(1 -I /2)(1 -I-3
2(1
1279ц/592)^J-If *
+ 2(1 + 119I/4 +
1281I2/4) V ^ (1 -
1459i/592) ^ 19
^ \

: n, Zp
„-(5/2
www.samco.org
Pa
SAMCO Final Report 2006
F05 Guidelines for Structural Control

E[u2]rap 2nSf g
(1+ I/4) i1/ I (11 + 83I /4 + 2Q1I2 /16 +
i 157I3 / 64), ' p + 8 (1 + I)2(1
rzp + + 3I/4)2(1 + I/2)3/2 '
(1 + I)3/2(1 + I/2)1/2(1 + 3i I3/ 2 (1 + 215I /64 + 1163I2 /
/4)1, 256 + 395I3 /128) (1 +
32(1 + i)(1 + 3i/4)(1 + i /2)3 I)3/2(1 + I /2)5/2(1 + 3i/4)2

E[(U p + U g^] 2nSf 64(1 + i)(1 + 3i / 4)(1 + i /2)3


rap g
11 (11 + 83i/4 + 4 (1 + i)2(1 + 3i /4)2(1 + i /
2Q1i2/16 + 157ц3/6 2)3'
i (1 + 215i /64 + 1163i2 / 256
rc - + 395i3 /128) i3' (1 +1)3/2
^ +T (1+ i / 2)5/2 (1 + 3i / 4)5/2

Legend F:
g
random excitation applied to the primary system with
spectral density Sf random excitation at base with
spectral density Sf

E^]:
E[up], E[Up], E[ii p]: mean of the square of the displacement up, velocity Up
and acceleration U p
mean of the square of the relative displacement ur

www.samco.org
Pa
SAMCO Final Report 2006
F05 Guidelines for Structural Control

www.samco.org
Pa
SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

2.1.1.9 Tuned liquid damper Theoretical background

Fig. 50: Principal parameters of sloshing and column TLD.


The basic principle of a tuned liquid damper (TLD) is similar to that of a tuned
mass damper (TMD). In particular, a moving secondary mass which is realized
by the fluid mass is introduced into the primary structural system. The gravity
acts as restoring force and energy dissipation is provided through viscous
action mainly in the boundary layers of the fluid. A simple realization of TLD
consists of rectangular or circular container with water inside (Fig. 49). A
horizontal motion of the container produces a sloshing motion of the fluid. A
variation of TLDs is the tuned liquid column damper (TLCD) which consists of
a U-shaped container partially filled with fluid (Fig. 50). The fluid column starts
oscillating as soon as the container is subjected to horizontal motion. The
dynamic vibration absorber for ship applications proposed by Frahm was
actually a forerunner of modern tuned liquid column damper (Den Hartog
(1997), Frahm (1909)). For civil engineering applications, the first
investigations started in the mid-1980s by the work of Bauer (1984).

mary structure.

www.samco.org Page 67 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Unlike a TMD, generally the response of a TLD is highly nonlinear. Liquid


sloshing or fluid flow through orifices is typically a highly nonlinear
phenomenon. The response
of the primary structure with a TLD is therefore amplitude dependent, even for
primary structures operating within the linear elastic regime. Considerable
research effort has been focused in understanding and quantifying the dy-
namic behaviour of nonlinear sloshing TLD. The most promising macroscopic
models are based on extensions of classical theory of shallow water gravity
waves with finite amplitude (Lepelletier and Raichlen (1988), Shimizu and Ha-
yama (1987), Sun et al. (1992)). These models contain special terms to
account for damping and fit reasonably well with test results. The steady state
response curves of nonlinear sloshing TLD show some similarity with the
steady state amplitude response curves of oscillating systems with nonlinear
hardening properties (Fujino et al. (1992)). The fundamental frequency of
sloshing TLD can be estimated from linearized small amplitude theory and is
given by

www.samco.org Page 68 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

The energy dissipation per cycle of a TLD depends strongly from the excitation
frequency. For frequencies in the neighbourhood of the natural frequency the
energy dissipation was found to be significantly higher than elsewhere.

f = ± П tanh у L
(1)
The dynamic behaviour of tuned liquid column dampers (TLCD) is more suited
for a description with simple one degree of freedom systems. Nevertheless,
one degree of freedom systems are still nonlinear for significant vibration
amplitudes of the fluid column (Hochrainer (2002), Saoka et al. (1988)). The
nonlinearity mainly involves the damping term because of the energy
dissipation due to turbulent flow. This term is usually modelled to be
proportional to the square of the velocity of the oscillating fluid column. The
natural frequency of a CTLD can be estimated using

(2)

Design issues
Because of the lack of reasonably simple models for describing the dynamic
behaviour of a TLD, no generally accepted design procedure exists so far. In
principle, because of the close similarity in the basic principle of operation
between TMD and TLD, the same approach as described for TMDs can be
applied for TLDs. In fact, a tuned mass damper analogy for the sloshing TLD was
proposed for design issues with frequency dependent virtual mass and dashpot
(Sun et al. (1995)). However, because the virtual mass and dashpot are
amplitude-dependent, these had to be determined by experiments and may

www.samco.org Page 69 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

therefore change due to implementation details. Because of the much lower mass
density of water with respect to steel, TLD needs much more space than TMD for
installation. In addition, sloshing TLDs are usually implemented as a rack of tanks
therefore requiring even more space. The fundamental frequency of sloshing
TLDs can be estimated using the linearized theory of shallow water waves. Much
more difficult is predicting the energy dissipation of a TLD. Therefore, laboratory
tests are mandatory to optimize the dynamic behaviour of TLD with respect to the
primary structure.
Because of the nonlinear hardening behaviour of sloshing TLD, its effectiveness is
less sensitive to a correct frequency tuning of the damper. Rectangular tanks are
applied for structures with different natural frequencies in the two principal
directions. For a given water depth, the frequencies are tuned by an adequate
selection of the plan dimensions of the tank. However, because the existing
theories are only tested for unidirectional excitation, special care should be taken
in designing sloshing TLD acting in two directions. For structures with nearly equal
natural frequencies in the two principal directions, a circular tank may be used.
Testing and validation
The testing and validation of TLDs follows the same procedures as described for
TMD.

www.samco.org Page 70 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Implementation
Tuned liquid dampers have been primarily used for mitigating wind-induced vibra-
tions of tall structures with very small natural frequencies. The first applications of
nonlinear sloshing TLD were implemented in Japan and include the Nagasaki
Airport Tower, the Yokohama Marine Tower and the Shin-Yokohama Prince Hotel
(Tamura et al. (1995), Tamura et al. (1996), Wakahara et al. (1992)). The
hardware implementation of sloshing TLDs is much simpler as that for TMD. Each
damper, generally, consists of a rack of fluid containers. The installation
requirements are minimal. Sloshing TLD may be easily added in existing buildings
and produce very little maintenance costs.
References
Bauer, H. F. (1984), "Oscillations of Immiscible Liquids in a Rectangular Container
- a New Damper for Excited Structures", Journal of Sound and Vibration, 93(1),
117133.
Den Hartog, J. P. (1947), Mechanical Vibrations. Third ed. 1947, New York and
London: McGraw-Hill Book Company.
Frahm, H. (1909), Device for Damped Vibrations of Bodies. U. S. Patent No.
989958, 1909.
Fujino, Y., Sun, L., Pacheco, B. M., and Chaiseri, P. (1992), "Tuned Liquid
Damper (Tld) for Suppressing Hori-zontal Motion of Structures", Journal of
Engineering Mechanics-Asce, 118(10), 2017-2030.
Hochrainer, M. (2002), Control of Vibrations of Civil Engineering Structures with
Special Emphasis on Tall Building, Institut fur Allgemeine Mechanik, Vienna
University of Technology, Wien.
Lepelletier, T. G., and Raichlen, F. (1988), "Nonlinear Oscillations in Rectangular
Tanks", Journal of Engineering Mechanics-Asce, 114(1), 1 -23.
Saoka, Y., Sakai, F., Takaeda, S., and Tamaki, T. (1988), "On the Suppression of
Vibrations by Tuned Liquid Column Dampers", in Annual Meeting of JSCE, Tokyo:
JSCE.

www.samco.org Page 71 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Shimizu, T., and Hayama, S. (1987), "Nonlinear Responses of Sloshing Based on


the Shallow-Water Wave Theory", Jsme International Journal, 30(263), 806-813.
Sun, L. M., Fujino, Y., Pacheco, B. M., and Chaiseri, P. (1992), "Modelling of
Tuned Liquid Damper (Tld)", Journal of Wind Engineering and Industrial
Aerodynamics, 43(1-3), 1883-1894.
Sun, L. M., Fujino, Y., Chaiseri, P., and Pacheco, B. M. (1995), "The Properties of
Tuned Liquid Dampers Using a Tmd Analogy", Earthquake Engineering &
Structural Dynamics, 24(7), 967-976.
Tamura, Y., Fujii, K., Ohtsuki, T., Wakahara, T., and Kohsaka, R. (1995), "Effec-
tiveness of Tuned Liquid Dampers under Wind Excitation", Engineering
Structures, 17(9), 609-621.
Tamura, Y., Kohsaka, R., Nakamura, O., Miyashita, K., and Modi, V. J. (1996),
"Wind-induced responses of an airport tower - efficiency of tuned liquid damper",
Journal of Wind Engineering and Industrial Aerodynamics, 65(1-3), 121-131.

www.samco.org Page 72 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Wakahara, T., Ohyama, T., and Fujii, K. (1992), "Suppression of Wind-


Induced Vibration of a Tall Building Using Tuned Liquid Damper", Journal of Wind
Engineering and Industrial Aerodynamics, 43(1 -3), 1895-1906.
Notations
Symbol Description Unit
A cross sectional area m2
L container length in the sloshing direction m
g gravity constant m/s2
h depth m
m mass kg
p mass density kg/m3
ц mass ratio -
Subscripts Description
f fluid

www.samco.org Page 73 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural
Control

2.1.2 Semi-active devices 2.1.2.1


Defintion of term "semi-active"

aassva
davioe
sam-a^-tva
da vie
fj у active
devioa

arty d ss uat va force active & dissipative no fo roe m tat о n


va ja s fo rue values s
tj fa ice I m tat о n s tj fa roe lim tat 'j n t ■ foica ad jslaa a
tj mate-a iiffihii tj a^tjat'j'dynamics in II and IV
(ела mple: I ива' v only "J a a native
Sta j s da mper) bice values

л.

о n у d a oative fo favCe Ivrttatians:


ice va jes matef ial it'angti n t

www.samco.org Page 74 of
155
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Tiata-a it ffne is maximum farce rritatknS-


да та rates oo а irritalon (rrin. fa 'се I st ffnest t« ™n of
л1 tn active fo n tat on тнд g Ые) davice Tjt nag g b le.
ices fastavtjat'jr dynam ii tla n a so active
поп n«ar darrp tj 'Л у d ii aat va for jb fa rca va jes (Tjt da
active S d ti aat v« va jet и ctad he ne)
ft*он va jat maiimmi 8, mi nimim foroe iBKinsa t ma
fa ice Tjt vary fait

ib 1

: operating range quadrant I. Ill : active ptrtrtr ihhi : desired force trajectory
quadrant 11, IV : (fesipalive ptrtrtr Я : raafeaUe ftiCa trajeCtciy
Fig. 51: Definitions and characteristics of damping devices often used in
structural control.
Semi-active devices are "actuators" that cannot produce power and therefore,
e.g., excite structures (Gavin and Alhan (2005), Spencer Jr. et al. (1997),
Spencer Jr. and Nagarajaiah (2003), Xu et al. (2003)). However, they are able
to dissipate structural vibration and - in contrast to passive dampers - the
amount of energy dissipation within the semi-active device may adjusted by
control of the device force. Thus, the operating range of semi-active devices
may be defined by the force, which is desired by the controller, and the device

www.samco.org Page 75 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

velocity, which is the velocity of the piston, rod, shaft etc depending on the
device design, as follows
0 • (f des ' X sa-device ) > 0
p j w des sa-device / (1)
sa-device |p • ( f ' )< 0
I des • vf des Xsa-device f 0

www.samco.org Page 76 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

The different operating ranges of passive dampers, semi-active devices such as


controllable dampers and fully active devices, usually called actuators, are
illustrated in Fig. 51. The operating range of many semi-active devices is not
only constraint to quadrants II and IV, which include dissipative force values
only (in view of the device), but further constraint by a maximum and minimum
device force (Fig. 51, bottom right). The force limitation by a maximum value is
nothing special since this property is common for any semi-active and fully
active control device. However, the additional restriction by a minimum device
force that is relatively large compared to the minimum force of actuators given
by friction of seals, bearings, and gears, plays an important issue in control point
of view. This fact implies that the desired force trajectory is not only bounded by
the maximum device force but also constraint by a considerable minimum force
value. Especially in the field of vibration mitigation, the minimum device force
leads to clamping effects at position of the semi-active device if the desired
damping force is far smaller than the minimum force of the semi-active device
because then minimum force is applied to mitigate the structure. Clamping the
structure at device position is equivalent with generating a nodal point of the
structure at that position. Then, vibration velocity at device position is zero. As a
result, structural vibration energy may not be dissipated anymore within the
semi-active device. Hence, structural vibrations decay as without external semi-
active damping device.
Consequently, not only the maximum force of semi-active devices but also the
minimum device force must be taken into account for the design of semi-active
devices.
In accordance with customers, engineers have to assume the maximum
magnitude of expected structural vibrations (assumption of worst case scenario)
in order to determine the maximum device force which is mandatory so that the
desired maximum control force may be tracked. In order to determine the
minimum force value of the semi-active device, engineers and customers
together have to define this vibration magnitude threshold below which the
semi-active device may clamp the structure and therefore cannot dampen
vibrations.
According to the values of maximum and minimum damping forces, the semi-
active device may be evaluated or designed and manufactured, respectively.
References

www.samco.org Page 77 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Gavin, H. P., and Alhan, C. (2005), "Guidelines for low-transmissibility semi-


active vibration isolation", Journal of Smart Materials and Structures, 14, 297-
306.
Spencer Jr., B. F., Dyke, S. J., Sain, M. K., and Charlson, J. D. (1997), "Phe-
nomenological Model of a Magnetorheological Damper", Journal of Engineering
Mechanics, ASCE, 123(3), 230-238.
Spencer Jr., B. F., and Nagarajaiah, S. (2003), "State of the Art of Structural
Control", Journal of Structural Engineering, 129(8), 845-856.
Xu, Z.-D., Shen, Y.-P., and Guo, Y.-Q. (2003), "Semi-active control of structures
incorporated with magnetorheological dampers using neural networks", Journal
of Smart Materials and Structures, 12(1), 80-87.

www.samco.org Page 78 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

2.1.2.2 Magnetorheological/electrorheological fluid damper Main working


principles
Magnetorheological and electrorheological fluid dampers (MR dampers, ER
dampers) represent a class of controllable fluid dampers where the shear force
of the fluid is controlled by an external magnetic and electric field, respectively
(Chen and Yang (2003), Gordaninejad et al. (2002), Han et al. (2002), Kornbluh
et al. (2004), Naka- mura et al. (2002), Oh and Onoda (2002), Spencer Jr. and
Nagarajaiah (2003)).

Fig. 52: Rotational MR damper, force range approximately [10, 320] N.


Fig. 53: Cylindrical MR damper with gas pocket (accumulator) for compensation
of piston volume; force range approximately [70, 1800] N.
Besides the control input variable (current/voltage), the main difference between
these two semi-active damping devices is that MR dampers produce far larger
damping forces relatively to eR dampers for the same damper size. That is why
MR dampers are often used for vibration mitigation in the field of large civil
structures whereas ER dampers are used for vibration control of smaller
structures. In the following, the working principles of these two damper types
shall be explained on behalf of MR dampers.
Characteristics of MR dampers
The MR fluid consists of oil, magnetizable particles and additives. The viscosity
of the oil defines the slope of the force-velocity trajectories at zero current (Fig.
54). The magnetizable particles may be polarized by the applied magnetic field
which ends up in a controllable shear force depending on the amount of
polarized particles. Since the shear force is controlled, the force trajectory in the

www.samco.org Page 79 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

force-velocity diagram approximately describes a Coulomb friction, superposed


by oil viscosity (Guglielmino et
Basically, MR/ER dampers consist of a housing including the MR/ER fluid and
the moving part of a disc or piston, depending if the dampers are built in
rotational or cylindrical form (Fig. 52, Fig. 53, Liao and Lai (2002), Weber et al.
(2002), Weber et al. (2005a), Weber et al. (2005c)). The MR/ER fluid is a
suspension of oil, additives and particles whose polarisation may be influenced
by an external magnetic/electric field. Hence, the control input variable of MR
and ER dampers is current and voltage, respectively. The amount of polarised
particles depends on the magnetic/electric field magnitude and influences the
shear stress of the MR/ER fluid. Therefore, according to the applied
current/voltage, the shear force of the MR/ER fluid and eventually the damper
force acting on the piston may be controlled.

www.samco.org Page 80 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

www.samco.org Page 81 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

6 AMCO

Fig. 56: Force displacement trajectories including transient measurement data


from preyield to postyield regions (MR damper of Maurer Sohne).
Fig. 55: Measured MR damper behaviour in pre- and postyield regions (MR
damper of Maurer Sohne).
Additives within the fluid should help to avoid agglomeration and sedimentation
of the magnetizbale particles. Sedimentation effects are not of high priority if MR
dampers are used to mitigate, e.g., cable vibrations due to wind loading
because such vibrations exist for fairly long time compared to the time of one
cycle. In this case, the MR fluid will be mixed up during the first couple of cycles.
After that time, the particles are
al. (2005), Zhou and Sun (2005)). Thus, MR dampers may be seen as
controllable Coulomb dampers where the slope of the Coulomb force
trajectories depends on the oil viscosity. Usually, the slope is of minor
importance or even negligible compared to the force value at a certain velocity.
At very small velocities, the force trajectories go into the zero-point tracking a
very large viscosity. The reason for that phenomenon is that MR fluids behave
more like rigid bodies (elastic) below yield stress and therefore at very small
velocity values, whereas the fluid only starts to flow above yield stress, thus
behaves plastic-viscous (Fig. 55, Fig. 56, Pignon et al. (1996), Powell (1995),
Weiss et al. (1994)). This ends up in the fact that force-displacement trajectories
or force-velocity trajectories of sinusoidal measurements include transient
measurement data when the fluid changes between these two regions (Jung et
al. (2003), Gor-

www.samco.org Page 82 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Fig. 54: Schematic steady-state damper behaviour of MR dampers.

www.samco.org Page 83 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

distributed homogeneously again. Then, the MR damper works as expected.


However, if dampers should optimally damp structural vibrations resulting from
earthquakes, they have to react according to the programmed control strategy.
Hence, dampers with sedimentation which changes the damper behaviour
instantaneously are not appropriate.
Modelling
Numbers of phenomenological MR damper models have been presented by
many researchers (Dominguez et al. (2004), Jimenez and Alvarez-Icaza (2005),
Ramallo et al. (2004), Rosenfeld and Wereley (2004), Sahasrabudhe and
Nagarajaiah (2005), Spencer Jr. et al. (1997), Wand and Liao (2005), Yang et
al. (2004)). Most models are based the so-called Bouc-Wen model. This
approach tries to model the MR damper behaviour by combining linear spring
elements, linear dash-pot elements, nonlinear damping elements, friction
elements with the Bouc-Wen element that describes the hysteresis
phenomenon of MR dampers. Exemplarily, the simple Bouc- Wen model
proposed by Spencer Jr. et al. (1997) is depicted in Fig. 57. The governing
equation of the MR damper force including the force offset f0 describing the fric-
tion due to seals or the so-called accumulator is
f -f0 = m-x + c(x) •x + k-x + a- z (1)
nonlinear
damping
coefficient
where z describes the evolutionary variable which is defined as follows
i.i II n-1 n | | n .
z = -Y • |x|-z-|z| -p- x • |z| + A-x (2)
The nonlinear damping coefficient c(x) describes the plastic, nonlinear damping
characteristics within the post-yield region. According to Yang et al. (2004), this
damping coefficient may be modelled as follows
c( x) = a1 • e "(a2-|x| )P (3)
A completely different modelling approach is the simple approach of fitting a
two- dimensional function using measurement data (Song et al. (2005), Weber
et al. (2002), Weber et al. (2005c)). The function is two-dimensional since the
MR damper force depends on the actual damper piston velocity and actual
damper current. The advantage of this "mapping" procedure is that the function
may be simply inverted in order to have an inverse model of the steady-state

www.samco.org Page 84 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

MR damper behaviour available which compensates for the stationary main


nonlinearities of MR dampers (Fig. 58).

Fig. 57: Simple Bouc-Wen modelling ap- Fig. 58: Fitted inverse MR damper
function. proach for MR-dampers.

Testing MR/ER dampers


The common way to determine the MR damper behaviour is to measure the MR
damper characteristics at constant current level for sinusoidal controlled damper
piston displacement (Oyadiji and Sarafianos (2000), Tse and Chang (2004),
Zhang et al. (2004)). The reasons for sinusoidal controlled displacement are:
usually, hydraulic and pneumatic aggregates are displacement controlled (Fig.
59).
sinusoidal displacement with defined amplitude and frequency produces the
velocity range of [0, |imax| ]. Hence, the dependency of the damper force on
the velocity may be measured over the entire desired velocity range. However, it
must be mentioned that this measurement data also includes transient data due
to the change of MR damper behaviour between preyield and po- styield regions
(Fig. 56 and Fig. 60). That is the reason why force-velocity diagrams do not only
show approximate Coulomb force trajectories at constant damper current but
also trajectories including non-dissipative force values and transient force
values, see Fig. 60. Depending on the viscosity of the MR fluid, the steady-state
parts of the force trajectories describe a Coulomb friction behaviour superposed
by a viscous part (Fig. 61).
The following points for damper measurements with high accuracy have to be
respected:

www.samco.org Page 85 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

The joints must not have play. Therefore, fitting bolts have to be used (Fig. 59).
Otherwise, knocking effects will falsify measurements.
If the force sensor is mounted between aggregate and damping device, the
sensor moves up and down according to the piston displacement. Due to the
acceleration of the force sensor, the sensor signal has to be compensated by
the acceleration term which results from the sensor's mass.
It must be ensured that the aggregate piston at high frequencies of the desired
displacement sinus really tracks the desired sinus signal. Depending on the
aggregate properties, the actual piston displacement tends to triangular shape
with increasing frequency of the desired displacement sinus.

www.samco.org Page 86 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Fig. 60: Transient measurement data due to MR damper behaviour in preyield


and postyield regions with force overshoot, fairly small MR fluid viscosity.
Fig. 61: Transient measurement data due to MR damper behaviour in preyield
and postyield regions, no force overshoot, fairly large MR fluid viscosity.
• The force response of MR/ER dampers may be measured if the desired dis-
placement is triangular. Then, the force response on velocity step may be
measured (Fig. 62, Fig. 63).

Fig. 59: Testing damping devices.

www.samco.org Page 87 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Fig. 62: MR damper force response on approxi- Fig. 63: MR damper force
response on approximate velocity step at 0 A. mate velocity step at 4 A.

Implementation
Since forces of MR dampers are larger than those of ER dampers for the same
housing dimensions, MR dampers have been installed on numbers of civil
structures for mitigation of vibrations caused by wind, combined wind-rain,
combined wind-snow, pedestrian, traffic, and earthquake loading. Wind and
earthquake loading may evoke unacceptable large vibration amplitudes of tall
buildings, especially skyscrapers. Large cable vibrations of stay cables of long
span bridges are mainly due to wind loading or wind loading combined with rain
or snow, which change the aerodynamic profile of cables. Pedestrian loading is
mainly reported in the case of footbridges. In this case vibration amplitudes
become too large if the first eigenfrequency of the footbridge is near the
frequency of walking, namely around two Hertz (Bachmann et al. (1995)).
Although e.g. bridge deck vibrations due to traffic loading exist, these vibrations
are of minor importance due to the large deck mass and deck stiffness
compared to the exciting forces caused by traffic.
MR dampers are primarily manufactured by LORD Corporation and Maurer
Sohne, Munich, Germany. MR dampers of LORD have been installed on the
"Dongting Lake Bridge" in China on all cables because vibrations due to
combined rain-wind effects were unacceptable large (Fig. 64, Ko et al. (2002)).
Each cable is equipped with two MR dampers in order to dampen in-plane and

www.samco.org Page 88 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

out-of-plane vibrations. In the case of cable amplitudes above a defined level,


these MR dampers operate in the so-called "passive-on" mode which means
that MR damper current is hold on a constant value different from zero (Lee and
Jeon (2002), Wan and Gordaninejad (2002), Weber et al. (2005a)). If vibration
amplitudes are below this defined threshold, MR damper current is zero. Then,
MR dampers work in the so-called "passive-off" mode.
Maurer Sohne had the opportunity to install one MR damper on the longest stay
cable of the "Eiland Bridge" nearby Kampen, The Netherlands, manly for long-
term field tests (Fig. 65, please see the following case study, Weber et al.
(2005b)). It must be mentioned that unacceptable large cable vibrations on this
cable-stayed bridge have not been observed so far. In collaboration with Empa,
the Swiss Federal Laboratories for Materials Testing and Research, this MR
damper has been model-based designed using cable properties, damper
position and assuming a worst-case scenario which has to be damped. When
the MR damper was installed in October 2004 on the bridge, decay
measurements of the cable without and with MR damper at different
current levels have been carried out. The target was to determine
experimentally the optimal current level depending on vibration amplitude for
maximum additional damping. Finally, an intelligent on/off control strategy was
implemented and runs since late October 2004.

Fig. 64: Cable-stayed "Dongting Lake Bridge" in China, equipped with MR


dampers of LORD Corporation.

www.samco.org Page 89 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Fig. 65: Cable-stayed "Eiland Bridge" nearby Kampen, The Netherlands,


equipped with one MR damper of Maurer Sohne for long-term field tests and
damping measurements.
The "Dubrovnik Bridge" nearby Dubrovnik, Croatia, is an example of a cable-
stayed bridge, where cable vibration amplitudes during combined wind-snow
events were that large that severe material damage resulted (Fig. 66). The
damage was caused by contact friction between cables and bridge deck
lightings. Again in collaboration with Empa, the Swiss Federal Laboratories for
Materials Testing and Research, MR dampers were model-based designed and
manufactured for cables no. 1-6, no. 11- 13 and for the backstays. These MR
dampers were installed in March 2006. Maurer and Empa will perform field tests
with forced cable vibrations in order to measure the additional damping provided
by feedback controlled MR dampers. Each MR damper will be equipped with
one displacement sensor between damper piston and damper cylinder. Based
on the measured relative damper displacement at 100 Hz sampling frequency,
two control concepts will be tested: velocity feedback and controlled friction
force level. It is planned that all MR dampers will be real-time controlled by the
friction force control.

www.samco.org Page 90 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 66: Cable-stayed „Dubrovnik Bridge" nearby Dubrovnik, Croatia.

MR dampers have been also installed as tuneable damping element in tuned


mass dampers in footbridges to mitigate vertical bridge deck vibrations due to
pedestrian's impact from walking (Seiler et al. (2002)). In the field of earthquake
engineering, MR dampers have been installed in skyscrapers using two different
damping concepts. One idea is to increase the overall building damping by
introducing damping forces that decrease the relative displacement between
adjacent floors. Hence, MR dampers are installed on each floor and are
connected with the adjacent floor. The other concept is base isolation. Here,
dampers shall isolate the sensitive system - in this case the tall building - from
the impact of the earthquake. Then, the MR damper pistons are, e.g., connected
to the ceiling of the ground floor which is built as a soft story, whereas the MR
damper cylinder must be fixed on the ground outside of the building. Hence, the
whole soft story including the MR dampers represent the isolating system.
Case Studies
The case study described here is the implementation of one feedback controlled
MR damper of Maurer Sohne on the "Eiland Bridge" nearby Kampen, The
Netherlands (Weber et al. (2005b)). The entire procedure of implementation
comprises the following steps:
1. MR damper design based on known cable properties, intended damper posi-
tion on the cable and assumed worst case vibration scenario which still shall be
possible to be damped.

6 AMCO

www.samco.org Page 91 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

MR damper fabrication and measurement of the force-displacement charac-


teristics at constant damper current.
Measurements of the decay rate of the free cable and of the cable with con-
nected MR damper operating at different settings.
Evaluation of measurement data in order to determine the optimum controller
tuning that controls the MR damper force.
Controller design and implementation on the bridge.
1) Model-based MR damper design
The basic design variables of MR dampers are:
Maximum damper force for given damper piston velocity:
f MR-max (xMR 'IMR-max )
Minimum damper force at zero current for given damper piston velocity:
fMR-min (xMR 'IMR-min = 0)
Maximum damper piston displacement: xMR-max
According to these design variables, the MR fluid may be evaluated. Based on
that evaluation, the MR damper geometry may be designed.
In order to be able to determine the above listed design variables, a worst-case
stay cable amplitude of the expected predominant mode has to be assumed for
the cable with connected MR damper, thus for the damped cable (Fig. 67).
Based on that assumption, the maximum damper force and maximum damper
piston displacement may be model-based estimated. The basic idea is to
simulate the vibrating cable with linear viscous damper that is optimally tuned
for the target mode (Fig. 67). Although the linear viscous damper is
characterized by a different force trajectory than the one of MR dampers
operating at constant current, the simulated maximum damper force and
damper displacement values may be transformed to their corresponding values
of a Coulomb friction damper whose behaviour is close to the one of MR
dampers operating at constant current. The reason to simulate a fairly simple
linear cable model with a linear viscous damper is that the connected system is
again linear. Thus, computing time is small and the correction of the disturbance
force amplitude may be done also linearly (Fig. 67).
One run of the simulation is completed when steady-state conditions occur.
Then, error e and correction factor A may be determined using the steady-state
results. The whole design procedure stops when the error of the actual
amplitude is smaller than a defined maximum error. Then, the steady-state

www.samco.org Page 92 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

simulation results deliver the wanted values of maximum damper force and
maximum damper displacement of the linear viscous damper that is optimally
tuned for the vibrating mode according to (Fig. 68, Krenk (2000))

www.samco.org Page 93 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 67: Flow chart of damper design procedure.

T
,opt
n
(4)

The distribution of the disturbance force fw accords to the shape of the mode to be
excited (Fig. 68). This guarantees that other modes are not excited. The amplitude A
of the external disturbance force is chosen the way that the steady-state amplitude of
the cable with linear viscous damper accords to the worst-case amplitude (Fig. 67).
^-■гггПТЛТГ.

a |±|c
7777"
L
Fig. 68: Cable/damper system with distributed disturbance force according to the
shape of the excited mode (here shown for the first vibration mode).

The maximum force of the linear viscous damper may be transformed to its counter-
part of the energy equivalent Coulomb friction damper provided that the maximum
displacements of both damper types are equal (Fig. 69)

www.samco.org Page 94 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

n
ff- = — ■ f (5)
J fri-max 4 J vis-max \f
with: Xfri-max ^ Xvis-max (6)

www.samco.org Page 95 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 69: Characteristics of linear viscous and friction dampers.

Since MR dampers working at constant current almost produce Coulomb force trajec-
tories, the transformed values are the wanted values of maximum MR damper force
and maximum MR damper displacement
/MR-max ~ //ri-max (7)
X MR-max ~ X/ri-max (8)
The minimum friction force cannot be smaller than the sum of sealing friction and
shear force of MR fluids at zero current. If the target is to produce MR dampers with
maximum large operating range, then the friction force at zero current will be chosen
to be equal to the minimum friction force mentioned above. If the target is to manu-
facture MR dampers that produce fairly large friction forces at zero current in order to
enhance the fail safe behaviour against vibrations of large amplitudes (Weber and
Feltrin (2003)), then the friction force at zero current may be increased by additional
sealing and different type of MR fluids.
J /mm = Min(seal friction + MR fluid friction @ 0 A): maximumoperating range MR-mm
[ > /min : enhanced /ail sa/e behaviour
(9)
Summarizing, the presented design procedure yields the maximum values of force
and displacement of MR dampers operating at constant current for maximum damping
of the target mode with defined, steady-state worst-case amplitude. The minimum MR

www.samco.org Page 96 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

damper force at zero current is bounded by MR fluid characteristics and damper


construction. The minimum MR damper force may be chosen to be larger than that
minimum value but not smaller.
The test MR damper had to be designed for the longest, 163 meter long stay cable of
the "Eiland Bridge". The worst-case assumptions for that cable with MR damper were:

www.samco.org Page 97 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

0.40 meter amplitude at mid span, and


only the first mode vibrates.
Using the known cable data listed in Table 6, the first eigenfrequency was estimated
using the formula of a linear cable
f»=i=2, mL do)

Fig. 70: Force displacement trajectories of MR damper of the "Eiland Bridge".


Fig. 71: Force current relation of MR damper of the "Eiland Bridge".
2Vm•L
Table 6: Properties of longest stay cable of "Eiland Bridge".
L m T

[m] [kg/m] [kN] [m]


163.7 66.55 5082 7.88

The estimated first eigenfrequency became 0.844 Hz. Applying Eq. (4), the optimum
viscosity of a linear viscous damper located at 7.88 meter was determined as 121.6e3
kg/s. Using this optimum linear viscous damper, the results of the model- based
damper design were the following required maximum MR damper values:
fMR-max = 40 kN and
x MR-max = °.°35 m.
Further desired design properties were:
The MR damper force at zero current shall be as small as possible in order to produce
a maximum large operating range of the MR damper.

www.samco.org Page 98 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

The dependency of the MR damper force at constant current on the damper piston
velocity shall be as small as possible. Then, the MR damper force is only a function of
current. Hence, the inversion of the steady-state damper function for control purposes
becomes simple and the actual velocity does not have to be measured.

www.samco.org Page 99 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

2) Measured MR damper characteristics


Based on the choice of MR fluid, the geometry of the MR damper was designed by
Maurer Sohne GmbH & Co. KG, Munich, Germany, which also manufactured the MR
damper. It was tested at the "Lehrstuhl fur Konstruktiven Ingenieurbau der Universitat
der Bundeswehr" in Neubiberg, nearby Munich. As can be readily seen from Fig. 70
and Fig. 71, the required MR damper characteristics could be fulfilled.
3) Damping measurements on the bridge
In order to know the amount of additional damping provided by the MR damper of
Maurer Sohne on the "Eiland Bridge" (Fig. 72, Fig. 73), the decay rate of both the free
cable and the cable with MR damper at different constant current levels were
measured.
The stay cable was excited by man power with a perpendicularly connected rope at
the frequency of the first vibration mode using a metronome (Fig. 74). The decay rate
of the cable was evaluated at 12% relative cable length. The accelerometers on the
top and bottom of the MR damper and the displacement sensor as well detected
clamping of the cable due to the MR damper force which remained approximately
constant during decay time as a result of constant current. The MR damper clamps
the cable in its equilibrium position as soon as the cable force component in damper
direction is smaller than the damper force (Fig. 75, Weber et al. (2005d))
\fMR (i) < T(sin(a2) - sin(a)) ^ xMR Ф 0 ^ P Ф 0 : no clamping

www.samco.org Page 100 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 72: Instrumentation of MR damper on the Fig. 73: Taking decay measurements
on the
"Eiland Bridge" nearby Kampen, The Netherlands. "Eiland Bridge" nearby Kampen,
The Netherlands.

www.samco.org Page 101 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 76: Measured vibration decay at 12% cable Fig. 77: Measured vibration decay at
12% cable length and damper position with MR damper at 0 length and damper
position with MR damper at 0.4

A.

Fig. 74: Sketch of decay measurements on the "Eiland Bridge" nearby Kampen, The
Netherlands.
Fig. 75: Clamping of cables.
A.

www.samco.org Page 102 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

www.samco.org Page 103 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

0 50 100 150^.200 250 300 0 50 100 1SQt[s]200 2b J 300


Fig. 78: Measured vibration decay at 12% cable Fig. 79: Measured vibration decay at
12% cable length and damper position with MR damper at length and damper position
of free cable. 2.0 A.

Due to the approximate Coulomb friction behaviour of the MR damper under consid-
eration, vibrations at 12% relative cable length decay approximately linearly as long as
the MR damper works. If the MR damper works, the MR damper relative displacement
is larger than a certain band value which results from out-of-plane vibrations of the
damper piston (Fig. 76, Fig. 77). If clamping occurs, the relative MR damper
displacement becomes zero or is smaller than the band value, respectively. With the
start of clamping, the decay envelope changes to an approximate exponential
function. This indicates that the damping is dominated by the approximate viscous
material damping of the cable only. If the damper current is set to 2 A, the MR damper
fully clamps the cable at damper position. Hence, the decay envelope has an
exponential shape and the decay rate is that of the free cable (Fig. 78, Fig. 79). At
such high current level, the MR damper behaves like an almost completely stiff bear-
ing.
4) Evaluation procedure of measurement data
Although the measured acceleration at 12% relative cable length decays almost with
linear envelope during unclamped conditions (Fig. 76, Fig. 77), the damping was

www.samco.org Page 104 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

evaluated applying the approach of the logarithmic decrement that basically assumes
linear viscous damping (Bachmann et al. (1995)). In that case, the ratio of two follow-
ing maxima is constant

www.samco.org Page 105 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

(12)

X (t,)
= constant
X (t, + Td)
where 2п/Td denotes the radial eigenfrequency of the damped structure. The loga-
rithmic decrement becomes

f
X (t,)
S =1 • ln
n
n
Л
= 1 • ln(X(t,))-ln(X(t, + nTd)) (13)
X (t, + nTd) J

which leads to the equivalent damping ratio as follows


(14)
S ^ S = л/ 4n2 +S2 2n

Fig. 80: Exponential approximation of linear decay within the amplitude range
from 95% to 50%.

www.samco.org Page 106 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

In order to keep the evaluation error due to the approximately linear decay
envelope fairly small, the local maxima within the amplitude range of 95% to
50% of the maximum value were fitted by an exponential function in order to
determine the logarithmic decrement and the equivalent damping ratio,
respectively (Fig. 80). Moreover, this way, the decay rate is evaluated within
the amplitude range of large values which could cause material damage.
Hence, if the MR damper current is chosen for maximum logarithmic
decrement for that amplitude range, the MR damper force is tuned in order to
prevent from possible material damage.
Due to some small modulation of the decay measurements by higher modes,
the evaluation procedure of the logarithmic decrement is modified as follows
(Weber et al. (2005d)):
Band pass filtering of the measured signals for the excited mode, in that case
mode number 1.
Determination of the local maxima within the amplitude range of 95% - 50%.
Linear fitting of the natural logarithm of the selected local maxima in order to
derive an estimate of the logarithmic decrement with small error, Eqs.(16)-
(18).

www.samco.org Page 107 of 155


SAMCO Final
Report 2006
F05 Guidelines
for Structural
Control

lnfc (ti)}, i = [1, k]


linear fitting
-> fit = a + b ■ t
^ = b ■(tk -ti) k -1

(15)-(17)
For the cable amplitudes within the range of 95% - 50% of the maximum mid
span amplitude of approximately 0.075 m, 0.4 A turned out to be the optimum
value of constant damper current for minimum decay time within the chosen
amplitude (Fig. 81). The logarithmic decrement of the free stay cable was
measured as 0.96% which is a typical value for lightly damped steel structures.
By attaching the MR damper operating in the passive-off mode, thus 0 A, the
overall damping could be increased by an approximate factor of four (Fig. 81). If
the MR damper current is optimally tuned to the actual vibration amplitude, the
overall damping was approximately eight times larger than the damping of the
free cable. If the MR damper clamped the cable at damper position due to a far
too large current, the overall damping was the same as for the free cable (see
logarithmic decrement for 2 A in Fig. 81).

Fig. 81: Evaluated mean logarithmic decrement.

www.samco.or
g Page
108 of 155
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 82: Schematic controller structure.


5) Controller
The basic controller structure consists of analogue/digital converter, the band
pass filter, the control law, and the digital/analogue converter (Fig. 82). The input
is the measured relative damper displacement and the controller output is a
voltage signal [-10, 10] V, which is the command signal of the power module that
generates the desired MR damper current (Fig. 83). The controller parameters
such as sample frequency, cut-off frequencies and control law parameters may
be defined using a lab- top computer connected to the controller by USB
interface.
In the case of the "Eiland Bridge", a fairly simple control approach has been
implemented. Basically, it is an on/off strategy that applies 0.4 A to the MR
damper if the actual relative damper displacement amplitude is above a defined
threshold for a certain time and 0 A for relative damper displacement amplitude
below the threshold for a certain time.

www.samco.org Page 109 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

www.samco.org Page 110 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

The memory function is taken into consideration in order to prevent from


numerous on-off switching. The definition of a displacement amplitude threshold
results in:
Low coupling of cable to bridge deck in the case of relative displacement
amplitude below the threshold. This guarantees that the MR damper does not
operate in passive-on mode due to small, traffic induced bridge deck vibrations.
MR damper energy consumption is small.
The required power in the passive-off mode counts for 0.25 W with a current of
21 mA and in the passive-on operating mode at 0.4 A is 2.6 W with a current of
220 mA. The required power is supplied by a solar panel und a battery 24Ah /
12V, made lead/gel technology, representing the necessary energy buffer (Fig.
84).

Fig. 83: Displacement sensor and MR Fig. 84: Solar panel and controller box
mounted on the
damper on the "Eiland Bridge" nearby lightning pylon on the "Eiland Bridge"
nearby Kampen,
Kampen, The Netherlands. The Netherlands.

The control law may be extended to a gain scheduling approach where different
current levels correspond to different relative damper displacement amplitude
ranges. Due to the fairly small mid span amplitude that could have been excited
by the presented man powered tests, only one optimum damper current value
for one amplitude range could have been determined. Therefore, the control law
consists of only one "gain".

www.samco.org Page 111 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

The physical reason for the gain scheduling approach depending on the actual
relative damper displacement amplitude may be explained in the following.
Depending on the MR fluid viscosity, the force trajectory of Mr dampers
describes more or less a Coulomb force trajectory (Fig. 54). Hence, the force of
MR dampers operating at constant current is independent of the actual
collocated cable velocity or displacement, respectively. As a result, during the
decay of cable vibrations, the MR damper force remains constant and finally
clamps the cable at damper position as soon as the MR damper force is larger
than the cable force in damper direction (Fig. 75). In contrast, linear viscous
dampers will not clamp the cable during a decay phase due

to the proportionality between force and velocity. Concluding, if MR dampers


shall be used for cable vibration mitigation, control of their friction force is
mandatory. The control law may be derived from the energy equivalent,
optimum linear viscous damper as follows (Fig. 69)
W =n- f x = W = 4• f x (18)
vis J vis-max vis-max fri J fri-max fri-max \'
Assuming sinusoidal displacement at damper position, the velocity amplitude
may be expressed in terms of displacement amplitude and frequency
W = W =ncop'x x =ncop'x 2 а (19)
vis fri n vis-max vis-max n vis-max n
The optimum viscosity given by Eq. (4) may be substituted in Eq. (19). For the
same maximum damper displacements, the force level of the energy equivalent
friction damper becomes
жT
fequiv = x • — •— (20)
J fri-max fri-max * \'
4 xd
Hence, if friction force dampers shall dissipate the same amount of vibration
energy as an optimally tuned linear viscous damper, then the actual friction
damper force has to be controlled proportionally to the actual damper
displacement amplitude. The absolute force value of friction dampers remains
constant during one cycle of steady state vibrations in contrast to the force of
linear viscous dampers which changes si- nusoidally during one cycle of steady

www.samco.org Page 112 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

state vibrations. For MR dampers with force trajectories at constant current with
fairly small slope, the equations derived for the friction damper may be taken
also for MR dampers working at constant current.
Provided that damper displacement amplitude may be measured directly, the
feedback gain of energy equivalent friction dampers becomes
жT
8T =V — (21)
4 xd
Linear viscous dampers represent the case of an ideal damping element where
the damping coefficient is a constant independent variable. The force of an
optimally tuned linear viscous damper is the product of viscosity and collocated
velocity
fOP = cOP' • Xvis (22)
If such an optimum linear viscous damper has to be realized using an ideal
actuator (Fig. 51), then the optimum viscosity represents the optimum feedback
gain if the collocated velocity is measured directly (Marathe et al. (2004),
Preumont (2002))
goP = С (23)

www.samco.org Page 113 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 85: Emulation of desired damping characteris- Fig. 86: Characteristics of


real "linear viscous tics using feedback controlled actuators. dampers".

Generating the force trajectory of a, e.g., linear viscous damper by feedback


control of actuators is called "emulating" linear viscous dampers (Fig. 85,
Nakamura and Nakazawa (2002), Marathe et al. (2004), Stelzer et al. (2003)). If
a feedback controlled actuator emulates a linear viscous damper, the control
strategy is called "velocity feedback". There are several reasons why emulation
of, e.g., linear viscous dampers using feedback controlled actuators may be
sensible:
the wanted linear viscous damper with the desired optimum viscosity is not
available on the market,
the offered "linear viscous dampers" often do not have a pure linear force tra-
jectory, especially below a certain velocity limit ilim due to the sealing friction
which ends up in a highly nonlinear force trajectory below ilim (Fig. 86), and
the viscosity of a linear viscous damper cannot be tuned on site to the real
structural properties in contrast to feedback controlled actuators, where the
feedback gain allows for a fine tuning.
References
Bachmann, H., et al. (1995), Vibration Problems in Structures: Practical Guide-
lines, Birkhauser Verlag Basel, ISBN 3-7643-5148-9.

6 AMCO

Chen, S. H., and Yang, G. (2003), "A method for determining locations of
electro- rheological dampers in structures", Journal of Smart Materials and
Structures, 12(2), 164-169.
Dominguez, A., Sedaghati, R., and Stiharu, I. (2004), "Modelling the hysteresis
phenomenon of magnetorheological dampers ", Journal of Smart Materials and
Structures, 13, 1351-1361.
Gordaninejad, F., Saiidi, M., Hansen, B. C., Ericksen, E. O., and Chang, F.-K.
(2002), "Magneto-Rheological Fluid Dampers for Control of Bridges", Journal of
Intelligent Material Systems and Structures, 13(2/3), 167-180.
Guglielmino, E., Stammers, C. W., Edge, K. A., Sireteanu, T., and Stancioiu, D.
(2005), "Damp-By-Wire: Magnetorheological Vs. Friction Dampers",

www.samco.org Page 114 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Proceedings of the 16th IFAC World Congress, Prague, Czech Republic, 4-8
July 2005.
Han, Y. M., Ham, M. H., Han, S. S., Lee, H. G., and Choi, S. B. (2002),
"Vibration Control Evaluation of Commercial Vehicle Featuring MR Seat
Damper", Journal of Intelligent Material Systems and Structures, 13(9), 575-579.
Jimenez, R., and Alvarez-Icaza, L. (2005), "LuGre friction model for a magnetor-
heological damper", Journal of Structural Control and Health Monitoring, 12, 91-
116.
Jung, H.-J., Spencer Jr., B. F., M.ASCE; and Lee, I.-W., M.ASCE (2003),
"Control of Seismically Excited Cable-Stayed Bridge Employing
Magnetorheological Fluid Dampers", Journal of Structural Engineering, 129(7),
873-883.
Ko, J. M., Zheng, G., Chen, Z. Q., and Ni, Y. Q. (2002), "Field vibration tests of
bridge stay cables incorporated with magneto-rheological (MR) dampers",
Proceedings of the International Conference on Smart Structures and Materials
2002: Smart Systems of Bridges, Structures, and Highways, S.-C. Liu and
Darryll J. Pines (eds.), Proceedings of SPIE (publ.), 4696, 30-40.
Kornbluh, R., Prahlad, H., Pelrine, R., Stanford, S., Rosenthal, M., and von Gug-
genberg, P. (2004), "Rubber to rigid, clamped to undamped: Toward composite
materials with wide-range controllable stiffness and damping", Proceedings of
the International Conference on Smart Structures and Materials 2004: Industrial
and Commercial Applications of Structures and Technologies, E. H. Anderson
(ed.), Proceedings of SPIE (publ.), Vol. 5388, 372-386.
Krenk, S. (2000), "Vibrations of a Taut Cable With an External Damper", Journal
of Applied Mechanics, 67, 772-776.
Lee, Y., and Jeon, D. (2002), "A Study on The Vibration Attenuation of a Driver
Seat Using an MR Damper", Journal of Intelligent Material Systems and
Structures, 13(7/8), 575-579.
Liao, W. H., and Lai, C. Y. (2002), "Harmonic Analysis of a Magnetorheological
Damper for Vibration Control", Journal of Smart Materials and Structures, 11 (2),
288296.
Marathe, S. S., Wang, K. W., and Gandhi, F. (2004), "Feedback linearization
control of magnetorheological fluid damper based systems with model
uncertainties", Journal of Smart Materials and Structures, 13(5), 1006-1016.

www.samco.org Page 115 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Nakamura, T., Saga, N., and Nakazawa, M. (2002), "Impedance Control of a


Single Shaft-type Clutch Using Homogeneous Electrorheological Fluid", Journal
of Intelligent Material Systems and Structures, 13(7/8), 465-469.

www.samco.org Page 116 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Oh, H.-U., and Onoda, J. (2002), "An Experimental Study of a Semiactive Mag-
neto-Rheological Fluid Variable Damper for Vibration Suppression of Truss
Structures", Journal of Smart Materials and Structures, 11(1), 156-162.
Oyadiji, S. O., and Sarafianos, P. (2000), "Comparing the Dynamic Properties
Conventional and Electro-Rheological Fluid Shock Absorbers", Proceedings of
the 7th International Conference on Recent Advances in Structural Dynamics,
Southampton, England, July 2000, 579-590.
Pignon, F., Magnin, A., and Piau, J.-M. (1996), "Thixotropic colloidal
suspensions and flow curves with minimum: Identification of flow regimes and
rheometric consequences", Journal of Rheology, 40(4), 573-587.
Powell, J.A. (1995), "Application of a nonlinear phenomenological model to the
oscillatory behavior of ER materials", Journal of Rheology, 39(5), 1075-1094.
Preumont, A. (2002), Vibration Control of Active Structures, Kluwer Academic
Publishers, Dordrecht.
Ramallo, J. C., Yoshioka, H., and Spencer Jr, B. F. (2004), "A two-step
identification technique for semi-active control systems", Journal of Structural
Control and Health Monitoring, 11(4), 273-289.
Rosenfeld, N. C., and Wereley, M. (2004), "Volume-constrained optimization of
magnetorheological and electrorheological valves and dampers ", Journal of
Smart Materials and Structures, 13, 1303-1313.
Sahasrabudhe, S., and Nagarajaiah, S. (2005), "Experimental Study of Sliding
Base-Isolated Buildings with Magnetorheological Dampers in Near-Fault Earth-
quakes", Journal of Structural Engineering, 131(7), 1025-1034.
Seiler, C., Fischer, O., and Huber, P. (2002), "Semi-active MR dampers in
TMD's for vibration control of footbridges, Part 2: Numerical analysis and
practical realisation", Proceedings of the International Conference on
Footbridge, Paris, France, 2022 November 2002, AFGC - OTUA (eds.), on CD.
Song, X., Ahmadian, M., and Southward, S. C. (2005), "Modeling Magnetor-
heological Dampers with Application of Nonparametric Approach", Journal of
Intelligent Material Systems and Structures, 16(5), 421-432.
Spencer Jr., B. F., Dyke, S. J., Sain, M. K., and Charlson, J. D. (1997), "Phe-
nomenological Model of a Magnetorheological Damper", Journal of Engineering
Mechanics, ASCE, 123(3), 230-238.
Spencer Jr., B. F., and Nagarajaiah, S. (2003), "State of the Art of Structural
Control", Journal of Structural Engineering, 129(8), 845-856.

www.samco.org Page 117 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Stelzer, G., J., Schulz, M., J., Kim, J., and Allemang, R., J. (2003), "A Magnetor-
heological Semi-active Isolator to Reduce Noise and Vibration Transmissibility in
Atomobiles", Journal of Intelligent Materials Systems and Structures, 14(12),
743765.
Tse, T., and Chang, C. C. (2004), "Shear-Mode Rotary Magnetorheological
Damper for Small-Scale Structural Control Experiments", Journal of Structural
Engineering, 130(6), 904-911.
Wan, X., and Gordaninejad, F. (2002), "Lyapunov-Based Control of a Bridge Us-
ing Magneto-Rehological Fluid Dampers", Journal of Intelligent Material Systems
and Structures, 13(7/8), 471-478.
Wand, D. H., and Liao, W. H. (2005), "Modeling and control of
magnetorheological fluid dampers using neural networks", Journal of Smart
Materials and Structures, 14, 111-126.
Weber, F., Feltrin, G., Motavalli, M., and Aalderink, B. J. (2002), "Cable Vibration
Mitigation Using Controlled Magnetorheological Fluid Dampers: A Theoretical
and Experimental Investigation", Proceedings of the International Conference on
Footbridge, Paris, France, November 20-22 2002, AFGC - OTUA (eds.), on CD.
Weber, F., and Feltrin, G. (2003), "Influence of the offset friction of rheological
fluid dampers on the vibration mitigation performance", Proceedings of Fifth
International Symposium on Cable Dynamics, Santa Margherita Ligure (Italy),
September 15-18 2003, 427-435.
Weber, F., Feltrin G., and Motavalli, M. (2005a), "Passive damping of cables
with MR dampers", Journal of Materials and Structures, 38(279), 568-577.
Weber, F., Distl, H., and Nutzel, O. (2005b), "Versuchsweiser Einbau eines
adap- tiven Seildampfers in eine Schragseilbrucke", Beton- und Stahlbetonbau
100(2005), Heft 7, 582-589.
Weber, F., Feltrin, G., and Motavalli, M. (2005c), "Measured LQG Controlled
Damping", Journal of Smart Materials and Structures, 14, 1172-1183.
Weber, F., Distl, H., Feltrin, F., and Motavalli, M. (2005d), "Evaluation Procedure
of Decay Measurements of a Cable with passive-on operating MR Damper",
Proceedings of Sixth International Symposium on Cable Dynamics, Charleston,
SC (USA), September 19-22 2005, 143-150.
Weiss, K. D., Carlson, J. D., and Nixon, D. A. (1994), "Viscoelastic properties of
magneto- and electro-rheological fluids", Journal of Intelligent Material Systems
and Structures, 5(11), 772-775.

www.samco.org Page 118 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Williams, K., Chiu, G., and Berhard, R. (2002), "Adaptive-passive absorbers


using shape-memory alloys", Journal of Sound and Vibration, 249(5), 835-848.
Yang, G., Ramallo, J.C., Spencer, B.F., Jr., Carson, J.D., and Sain, M.K. (2000),
"Large-scale MR fluid dampers: Dynamic performance considerations",
Proceedings of the International Conference on Advances in Structure
Dynamics, Hong Kong, 341-348.
Yang, G. (2001), Large-scale magnetorheological fluid damper for vibration
mitigation: Modeling, testing and control, PhD dissertation, University of Notre
Dame, Notre Dame, Indiana.
Yang, G., Spencer Jr., B. F., Jung, H.-J., and Carlson, J. D. (2004), "Dynamic
Modeling of Large-Scale Magnetorheological Damper Systems for Civil
Engineering Applications", Journal of Engineering Mechanics, 130(9), 1107-
1114.
Zhang, T., Jiang, C., Zhang, H., and Xu, H. (2004), "Giant magnetostrictive
actuators for active vibration control", Journal of Smart Materials and Structures,
13(3), 473-477.
Zhou, H. and Sun, L. (2005), "A full-scale cable vibration mitigation experiment
using MR damper", Proceedings of 6th International Conference on Cable
Dynamics, Charleston, SC (USA), September 19 - 22, 2005, 167-174.
Notations
Symbol Description
I current A
L cable length m
P power W
T cable force N
W energy, work J
X amplitude of x m
c viscosity kg/s
f force, frequency N, Hz
g gain -
k stiffness kg/s2
m mass; cable mass per unit length kg; kg/m
x displacement m
a angle rad
S logarithmic decrement -
£ error -

www.samco.org Page 119 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

(0 radial frequency rad/s


z damping ratio -

Superscripts Description
opt optimal
x mean value of x
Subscripts Description
MR MR damper
act actual
d damper
des desired
fri friction damper
max maximum
min minimum
sa semi-active
u control input
vis linear viscous damper
w disturbance

2.1.2.3 Controlled shape memory alloy Theoretical background


Shape memory alloys (SMAs) are mostly known for their so-called "shape
memory effect", where SMAs revert to their initial shape upon heating after
having been deformed at low temperatures (Duerig (1990), Funakubo (1987),
Humbeeck (2001), Janke et al. (2005), Otsuka and Wayman (1999), Otsuka and
Kakeshita (2002)). This effect is also called pseudoplasticity. If the deformation
recovery is restrained, mechanical stress results within the SMA. The stress
may be used to introduce external forces into structures.
Another effect occurs when austenite phase transforms to martensite phase due
to external mechanical load at constant temperature. This is called super- or
pseudoe- lasticity, respectively. The alloy retransforms automatically to
austenitic state if the SMA is unloaded. The stress-strain curve shows a
hysteresis built by the two paths of transformation and retransformation.
The "two way SMA effect" describes the behaviour of some SMAs that may
remember two different shapes depending on different temperatures.

www.samco.org Page 120 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Besides the transformation processes briefly described above, SMAs are


characterized by high damping capacity due to internal friction which is based on
mainly two different mechanisms. If martensite is the stable phase, martensite
variants are reoriented within the SMA upon loading above yield stress. This
effect produces large hysteresis areas if the loading is sinusoidal around zero. If
the SMA is heated to a higher temperature, the hysteresis loop changes its form
until it reaches the shape similar to the superelastic behaviour of SMAs.
However, due to the complicated shape of the hysteresis trajectories between
ambient temperature and high temperatures (Otsuka and Wayman (1999)), this
"adaptive" material behaviour is not used to produce controllable damping
devices.
The second mechanism is based on the superelasticity of SMAs (Fig. 87). The
relative displacement between austenite and martensite interfaces and between
martensite variants, respectively, produces friction and therefore damping. The
hysteresis area during loading and unloading in the superelastic state is
equivalent to the dissipated energy. Since the hysteresis curve appears within
the stress-strain map, this kind of damping is structural damping. If the
temperature of the SMA is increased, the hysteresis does change its area, form
and location within the stress/strain map (Fig. 87). For small temperature
variations, the hysteresis curve does hardly change its area and form, but
moves slightly to higher stress/strain values (compare the thick solid line and
thick dashed line in Fig. 87).
Adaptive shape memory alloy
In the field of structural control, the superelastic behaviour may be used to
produce a damping device with variable damping characteristics. The crucial
points of such a controllable damping device may be:
Heating up the entire SMA is energy consuming and a constant, homogeneous
temperature of the entire SMA is difficult to achieve and maintain.
Fast cooling of SMAs is a challenging issue.
Both, heating and cooling are rather slow processes due to the thermal capacity
of the SMA material which results in a very small bandwidth of the semi-active
damping device.

www.samco.org Page 121 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

www.samco.org Page 122 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

• The working area of the semi-actively working SMA is small compared to the
entire area of the stress/strain area. Consequently, in many cases, the desired
optimum control force cannot be tracked. The resulting suboptimal force leads to
mediocre vibration mitigation performance.
Due to the difficulties listed above, the potential of SMAs working in the
superelastic mode at different temperature is not a current issue. However,
researchers have investigated numerically the damping potential of SMAs in
superelastic mode at constant temperature (Li et al. (2004)).
A different way of using SMAs as an adaptive material is proposed by, e.g.,
Rustighi et al. (2005) and Williams et al. (2002). Here, the SMA is used as a
spring element with two different stiffness values depending on the temperature.
At ambient temperature, the SMA is completely in the martensite phase (Fig.
88). At temperature "austenite finish", the SMA is completely in the austenite
phase. The two phases are characterized by two different moduli of elasticity
within the elastic stress/strain region. By combining several SMA elements
which are either in martensite or austenite state, a tuneable spring can be
designed. Such an adaptive spring may be used in order to tune the frequency
of a tuned mass damper to the actual structural target frequency (Rustighi et al.
(2005), Williams et al. (2002)). It must be mentioned that the SMA stiffness does
not change smoothly with increasing temperature. Therefore, each SMA
element of the overall spring is either in full martensite or full austenite state
(Williams et al. (2002)). That is the reason why the number of SMA elements
has to be fairly high in order to be able to produce a spring element with almost
continuously tuneable stiffness, bounded by the low stiffness of martensite and
the high stiffness of austenite.

www.samco.org Page 123 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural
Control

strain
Fig. 88: Different moduli of
elasticity for martensite and
austenite.
5 tress

Fig. 87: Stress-strain curves for


the superelastic behaviour of
SMAs at different temperatures.

References
Duerig, T. W. (1990), Engineering Aspects of Shape Memory Alloys,
Butterworth- Heinemann, London.
Funakubo, H. (1987), "Precision Machinery and Robotics, Vol. 1 - Shape
Memory Alloys", Gordon and Breach.
Humbeeck, J. V. (2001), "Shape memory alloys: A material and a technology",
Adv. Eng. Mater, 3(11), 837-850.

www.samco.org Page 124 of


155
SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

Janke, L., Czaderski, C., Motavalli, M., and Ruth, J. (2005), "Applications of
shape memory alloys in civil engineering structures - Overview, limits and new
ideas", Journal of Materials and Structures, 38(279), 578-592.
Li, H., Liu, M., and Oh, J. (2004), "Vibration mitigation of a stay cable with one
shape memory alloy damper", Journal of Structural Control and Health
Monitoring, 11(1), 21-36.
Otsuka, K., and Wayman, C. M. (1999), Shape Memory Materials, Cambridge
University Press.
Otsuka, K., and Kakeshita, T. (2002), "Science and technology of shape-
memory alloys. New developments", Mrs Bulletin 27(2), 91-100.
Rustighi, E., Brennan, M. J., and Mace, B. R. (2005), "A shape memory alloy
adaptive tuned vibration absorber: design and implementation", Journal of
Smart Materials and Structures, 14, 19-28.
Williams, K., Chiu, G., and Berhard, R. (2002), "Adaptive-passive absorbers
using shape-memory alloys", Journal of Sound and Vibration, 249(5), 835-848.
Notations
Symbol Description Unit
E elasticity modulus N/m2
T temperature K

www.samco.org Page 125 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

www.samco.org Page 126 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

'-^^Тгчцг «
О АМСО
2.1.3 Actuators
2.1.3.1 Actuators for increase of damping
In the field of structural control, increasing structural damping is the main target
because, in general, both frequency and spatial distribution of disturbance
forces due to, e.g., wind loading, are unknown. If the structural damping c shall
be augmented by the external control force fu, the control force has to be
controlled proportionally
to the collocated structural velocity, which is shown for the simple case of a
single degree of freedom system (SDOFS), see Eqs. (1) and (2).

fu =-CUX
(2)

The overall damping of the SDOFS


becomes

Ctot =(c + Cu )
The control concept described by Eq. (2) is called "velocity feedback" and will
be discussed in detail in Chapter 2.2. According to Eq. (2), the control forces
required are dissipative. Consequently, controllable dampers are sufficient to
track the desired control force fu. However, there are mainly three reasons, why
actuators that can apply active forces to structures may be mandatory in the
field of structural control:
Besides increasing structural damping, structural control comprises also, e.g.,
displacement and shape control of structures.
If the control force shall compensate for disturbance forces that influence the
structure only at well known positions of the structure (Achkire et al. (1998),
Caruso et al. (2003), Raja et al. (2002)).

www.samco.org Page 127 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

There exist many other control concepts used to mitigate structural vibrations
which require dissipative and active force values (d'Azzo and Houpis (1995)).
In the special case of "sky hook" damping requires dissipative and active force
values (Lee and Jeon (2002), Preumont (2002), Preumont (2004), Stel- zer et
al. (2003)).
From the wide range of actuator types, two of them are described in the
following which are fairly often used in the field of structural control, namely
hydraulic aggregates and piezo actuators.
References
Achkire, Y., Bossens, F., and Preumont, A. (1998), "Active damping and flutter
control of cable-stayed bridges", Journal of Wind Engineering and Industrial
Aerodynamics, 74-76, 913-921.
(3)
d'Azzo, J. J., and Houpis, C. H. (1995), Linear Control System Analyzes and
Design, Conventional and Modern, Fourth Edition, McGraw-Hill (eds.), ISBN 0-
07016321-9.

www.samco.org Page 128 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

Caruso, G., Galeani, S., and Menini, L. (2003), "Active Vibration Control of an
Elastic Plate Using Piezoelectric Sensors and Actuators", Journal of Simulation
Modelling Practice and Theory, 11 (5-6), 403-419.
Lee, Y., and Jeon, D. (2002), "A Study on The Vibration Attenuation of a Driver
Seat Using an MR Damper", Journal of Intelligent Material Systems and
Structures, 13(7/8), 575-579.
Preumont, A. (2002), Vibration Control of Active Structures, Chapter 12, Kluwer
Academic Publishers, Dordrecht.
Preumont, A. (2004), "Semi-active sky hook, does it work?", EUROMECH
Colloquium 455 on Semi-Active Vibration Suppression, M. Valasek and A.
Preumont (Eds.), CTU in Prague, Czech Republic, July 5-7 2004, on CD.
Raja, S., Prathap, G., and Sinha, P. K. (2002), "Active vibration control of
composite sandwich beams with piezoelectric extension-bending and shear
actuators", Journal of Smart Materials and Structures, 11(6), 63-71.
Stelzer, G., J., Schulz, M., J., Kim, J., and Allemang, R., J. (2003), "A Magnetor-
heological Semi-active Isolator to Reduce Noise and Vibration Transmissibility
in Atomobiles", Journal of Intelligent Materials Systems and Structures, 14(12),
743765.
Notations
Symbol Description Unit
c viscous damping coefficient kg/s
f force N
k stiffness kg/s2
m mass kg
x displacementm
Subscripts
total
control
disturbance

www.samco.org Page 129 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

tot
u
w

www.samco.org Page 130 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

2.1.3.2 Hydraulic aggregate Theoretical background


(1)

fa- dyn-n
, ^X
a-dyn-max p-max \ p p ,
Fig. 89: Hydraulic aggregate, mounted on a steel frame.
(2)
te • Xp)
The force of hydraulic aggregates is produced by the oil pressure within the
aggregate cylinder
fa-sta = P • Ap
where P is the oil pressure and AP the cross sectional area of the piston.
The force level is controlled by servo valves (Fig. 89). Usually, hydraulic
aggregates are equipped with both displacement and force sensor. This allows
for controlling the hydraulic aggregate in displacement control mode or force
control mode. The most often used control mode is displacement control and

www.samco.org Page 131 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

the reaction force between tested device and aggregate piston is measured by
an additional force sensor or by the force sensor of the aggregate.
The maximum available displacement amplitude depends on the frequency
assuming sinusoidal displacement because the dynamic force of the hydraulic
aggregate is limited by a maximum tolerable value in order to avoid damage of
the aggregate (Eq. (2)). The common way is that data sheets give information
about the maximum values of frequency cop and amplitude Xp of a sinusoidal
displacement.

www.samco.org Page 132 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

www.samco.org Page 133 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

The maximum value of the static aggregate force is usually approximately 20%
higher than the maximum value of the dynamic aggregate force. However, since
aggregates are used in dynamic testing of structures, the maximum dynamic
force is relevant.
Implementation
Hydraulic aggregates are mainly used to test novel materials in the laboratory.
Usually, the test material or test device is displacement controlled deformed and
the reaction force as system output is measured (Krenk et al. (1996)). The
results are the well known stress-strain or force-displacement curves,
respectively. Another application of hydraulic aggregates is the forced excitation
of structures such as buildings, bridges, and towers in order to identify structural
parameters based on forced vibrations (Irwin and Stoyanoff (2005), Occhiuzzi et
al. (2003), Stoyanoff et al. (2005), Weber and Huth (2005), Yang et al. (2004)).
A third variant is simply to control the displacement or shape of large civil
structures installing hydraulic aggregates between stories or between the first
floor and ground.
As a result of the ability to produce active forces, the following points should be
included in the displacement or force control loop of such actuators in order to
avoid unintentional material or actuator damage:
A maximum force value should be defined above which the controller shuts off.
A maximum displacement value should be defined above which the controller
shuts off.
The first test should be run using a dummy.
Design and testing
Since hydraulic aggregates are commercially available, they are designed,
manufactured and tested by the corresponding company. Operating maps of the
aggregate are delivered together with the aggregate.
References
Irwin, P., and Stoyanoff, S. (2005), "Use of vibration testing as a diagnostic
tool", Tutorial of 6th International Conference on Cable Dynamics, Charleston,
SC (USA), September 19 - 22, 2005.
Krenk, S., Jonsson, J., and Hansen, L. P. (1996), "Fatigue Analysis and Testing
of Adhesive Joints", Journal of Engineering Fracture Mechanics, 53(6), 859-872.

www.samco.org Page 134 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Occhiuzzi, A., Spizzuoco, M., and Serino, G. (2003), "Experimental analysis of


magnetorheological dampers for structural control", Journal of Smart Materials
and Structures, 12(5), 703-711.
Stoyanoff, S., Theryo, T., and Garcia, P. (2005), "Full dynamic test of the stay
cables of Sunshine Skyway Bridge", Proceedings of 6th International
Conference on Cable Dynamics, Charleston, SC (USA), September 19 - 22,
2005, 191-198.
Weber, F., and Huth, O. (2005), "Charakterisierung der dynamischen
Eigenschaf- ten eines Einfamilienhauses in Monthey / Wallis", Empa Bericht Nr.
841173, Swiss Federal Laboratories for Materials Testing and Reserach,
Dubendorf, Switzerland.
Yang, J., N., Lei, Y., Lin, S., and Huang, N. (2004), "Identification of Natural Fre-
quencies and Dampings of In Situ Tall Buildings Using Ambient Wind Vibration
Data", Journal of Engineering Mechanics, 130(5), 570-577.
Notations
Symbol Description Unit
Л cross sectional area m2
X displacement amplitude m
f force N
p pressure Pa
x displacement m
( radial frequency rad/s

Subscripts
a actuator
dyn dynamic
max maximum
p piston
sta static
2.1.3.3 Piezo actuator
Besides books, information about piezo actuators may be found in the internet,
e.g., on the following pages:
http://de.wikipedia.org/wiki/Piezo
http://www.efunda.com/Materials/piezo/general_info/gen_info_index.cfm

www.samco.org Page 135 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

http://www.marco.de/pb.html
http://www.physikinstrumente.com/tutorial/index.html
http://www.piezo.com/ (Piezo Systems, Inc., MA 02139 USA)
http://www.piezo-kinetics.com/ (Piezo Kinetics, Inc. Bellefonte, PA 16823, USA)
http://www.piezomechanik.com/ (Piezomechanik GmbH, D-81673 Munich,
Germany)
Theoretical background
Piezo actuators like stacks, benders, tubes, and rings make use of the
deformation of electroactive PZT-ceramics (PZT: lead (Pb), zirconia (Zr),
Titanate (Ti)) when they are exposed to electrical fields. This deformation can
be used to produce motions or forces if the deformation of the piezo element is
constraint.
The above effect is the complementary effect to piezo electricity, where
electrical charges are produced upon application of mechanical stress to the
ceramics. As an analogy, the term "piezo mechanics" was introduced in the
early 80's of the past century by L. Pickelmann to describe the conversion of
electricity into a mechanical action by piezo materials.
For piezo mechanical conversion in the simplest case a single PZT layer is
used. Such a PZT monolayer structure as shown in Fig. 90 is a capacitive
element. It consists of two thin conductive electrode coatings enclosing the
piezo ceramic as dielectric. Applying a voltage to this "piezo capacitor", the
capacitor is charged and deformation results. Hence, PZT actuators may be
seen as "actuating" capacitors.
If the maximum deformation of PZT actuators shall be increased, PZT actuators
consist of several layers. These piezo stack actuators and stacked piezo rings
make use of the increase of the ceramic thickness in direction of the applied
electrical field (Fig. 91). This is called "d33 effect". Stacking of several layers
towards a multilayer structure increases equivalently the total stroke. In practice,
axial strain rates up to 2%o of stack's length can be achieved under certain
conditions.

www.samco.org Page 136 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Fig. 90: Schematic of a piezo-electric single layer element Fig. 91: Schematic of
an axially acting (http://www.piezomechanik.com/). multilayer piezo stack
(http://www.piezomechanik.com/).

Similar to elastic deformation of solid bodies, the thickness expansion of a PZT


layer is accompanied by an in-plane shrinking (Fig. 90). This is called the "d31
effect", being complementary in motion and showing roughly half linear strain
compared to the d33 effect. The d31 effect is mainly used for bending
structures.
Design
Piezo actuators were used preferentially in the past for quasistatic precision
positioning tasks, but find now increasing interest in completely new fields of
application like dynamically actuated mechanics, e.g., valves, fuel injection
devices, or adaptive smart structures such as shape tuning, vibration generation
and cancellation, and mode tuning (Caruso et al. (2003), Herold et al. (2004),
Ma (2002), Raja (2002), Sa- dri et al. (2002)). Of course, such a broad variety of
applications cannot be covered by one general actuator type. The main
parameters to adapt a piezo actuator to a distinct application are:
Selection of proper PZT material defining achievable strain, stroke, energy
balance, temperature range etc.
Preparation of a highly reliable and efficient stack structure which is, e.g., shock
and vibration resistant.
Sealing for corrosion resistance.
Packaging of the ceramic stack.
Actuator system design, e.g., systems able to push and pull.
Performance and reliability of actuator systems depend further on the electrical
driving characteristics like voltage / charge / current control operating strategies
and unipolar / semibipolar / bipolar operation.

www.samco.org Page 137 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

www.samco.org Page 138 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Some of the currently used basic stack designs are:


Bare stacks: The mounting and motion transfer is always done via end faces.
Bare stack must not be hold by sideway clamping.
Ring stacks with centre hole (Fig. 92): Ring stacks are needed if an accessible
system axis is needed for transmissive optical set-ups or the feed through of
mechanical parts is required. Ring stacks may be also required in order to
increase the bending stiffness by diameter enlargement of the stack without the
need of increase of the operated ceramic volume (e.g. for long-stroke
elements). Ring actuators provide further a fairly high cooling performance due
to possible access of the inner and outer surface by cooling media.
Cased stacks with internal preload mechanism (Fig. 93): The incorporation of
piezo stacks into a metallic casing generally improves reliability and stability
against mechanical impact and deteriorating environmental influences. The
implementation of a preload mechanism compensates for tensile stress. In
contrast, ceramics are extremely vulnerable to such impacts.
Actuator characterization

Fig. 92: Schematic of a stacked ring actuator


(http://www.piezomechanik.com/).

www.samco.org Page 139 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 93: Schematic of an actuator with prestress


(http://www.piezomechanik.com/).
Like any other electrical actuator, piezo actuators convert electrical energy into
motion and force, respectively, depending on the actuator stroke constraints.
The generated motion or force is coupled to an external device (Fig. 94). In the
simplest case, this device shall be shifted from the actual position to the desired
one. More complex applications may be, e.g., position control of valves or
controlled piezo mechanical elements incorporated in smart structures. In all
these cases the interaction between the actuator and the driven mechanism
must be analyzed. The mechanical interaction is defined by the stiffness values
of both systems, namely of the piezo actuator and of the actuated mechanical
system.

www.samco.org Page 140 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 95: Schematic voltage/stroke diagram of a stack actuator


(http://www.piezomechanik.com/).
Two basic experiments are carried out in order to determine the actuator
characteristics:
actuator stroke generation as a function of the applied voltage (Fig. 95), and
actuator force generation as a function of the applied voltage (Fig. 96).
The condition for determination of the stroke voltage relation is that the stiffness
of the actuated mechanical system is zero or in other words that the piezo
actuator displacement is not constraint. Then, the piezo actuator generates
maximum stroke and minimum force.
The necessary condition for determination of the maximum actuator force, also
called "blocking" force, is infinite large stiffness of the actuated mechanical
system or zero piezo actuator displacement, respectively. The force response
shows remarkably lower hysteresis than the stroke response. It must be noted
that the hysteresis depends on preload conditions.
In practice, piezo actuators interact always with mechanical systems showing
an intermediate stiffness value between the two theoretical limits 0 and ^. Then,
the piezo actuator distributes the electrical energy partially into generation of
stroke and partially into generation of force (Fig. 97). The ratio between stroke
and force generation depends on the quantitative relation of piezo actuator
stiffness and mechanical system stiffness. The achievable force-stroke relations
of a real system can be derived in the following way:
Draw a line from maximum stroke to maximum blocking force.
Draw a line from origin with a slope according the stiffness of the coupled me-
chanical system.
The intersection point A of these two lines describes the achievable stroke and
force at maximum voltage of the piezo actuator. Notice that the schematic
depicted in Fig. 97 represents a linearized actuator response. It does not take
into account stroke enhancement effects.

www.samco.org Page 141 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 94: Schematic of a piezo actuated system


(http://www.piezomechanik.com/).
If the stiffness values of the piezo actuator and of the coupled mechanical
system are equal, the achievable displacement is 50% of its maximum value
and the achievable force variation is 50% of the blocking force. Under this
condition, the mechanical energy transfer efficiency from the piezo actuator to
the mechanical system is maximized.

www.samco.org Page 142 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 97: Maximum piezo actuator stroke versus maximum (blocking) piezo
actuator force (http://www.piezomechanik.com/).
Piezo actuators convert electrical energy into a mechanical energy, often called
piezo
mechanical response. The electrical energy content W of an electrically charged
piezo actuator is
C ■ U2
W = CU (1)
2
where C is the actuator capacitance and U the applied voltage. The piezo me-
chanical response parameters are:
stroke generation Al,
force generation AF , and
mechanical energy W = Al ■ AF.
Control of piezo actuators
The user of an actuator is mainly interested in the displacement of the actuator
and, perhaps, in the ultra fine positioning capability down to the sub nano meter
range. A variation of force or generation of mechanical energy during the action
of the piezo actuator is not of high priority. However, piezo actuators then do not
operate energy efficient because only a small part of the actuator energy
content is needed to be transferred to the coupled mechanics.
Optimizing piezo actuators for dynamic positioning purposes means minimizing
the electrical energy input for given desired displacement. This is equivalent
with minimum actuator capacitance in relation to the driving voltage. Then, the
electrical power requirements for a distinct dynamic response, e.g., oscillating
arrangements, are minimal. Consequently, a PZT material with low dielectric
constant £ together with a high piezoelectric constant d33 is required.

www.samco.org Page 143 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 96: Schematic voltage force relation of a piezo stack


(http://www.piezomechanik.com/).
Position control of piezo actuators is the classical control target. However, in the
case of smart structures of high stiffness, which shall be highly deformed by
integrated piezo actuators, large displacements together with high force
generation to achieve high mechanical energy transfer are required. Examples
are adaptive frame struc-

www.samco.org Page 144 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

www.samco.org Page 145 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

tures of machines, car bodies, wings of air planes for active vibration excitation
and cancellation or shape optimization (Caruso et al. (2003), Herold et al.
(2004), Ma (2002), Raja (2002), Sadri et al. (2002)). To provide this large
mechanical energy output, an increased electrical energy input is required via a
reasonable large electrical capacitance of the actuator to get a high mechanical
energy density within the ceramics. Therefore, PZT materials are requested
showing an elevated dielectric constant £ combined with a very high strain and
force generation rate. Notice that materials with large dielectric constant £ are
widely offered but often do not show the wanted large piezo mechanical
response.

Fig. 98: Approximately linear relation between stroke and charge content of
piezo actuators (http://www.piezomechanik.com/).
Fig. 99: Creep of piezo actuators (http://www.piezomechanik.com/).

According the simple capacitance equation


Q=U■C (2)
the voltage U at capacitor's contacts is proportional to the stored electrical
charge content Q. For commercially available capacitors, the capacitance C is
constant. Piezo actuators represent also electrical capacitors. In contrast to
normal capacitors, the capacitance of piezo actuators is not constant but
depends to some extent on driving conditions like voltage, load, temperature
etc. Therefore, the position, velocity, and acceleration of piezo actuator do not
depend linearly on the applied voltage U (Fig. 95). However, position, velocity,
and acceleration are proportional to the charge content Q (Fig. 98).

www.samco.org Page 146 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Consequently, if the nonlinear relation between voltage and displacement shall


not be compensated, piezo actuators should be "charge"-controlled (Preumont
(2002)). Summarizing, the controlled variables of the piezo actuator are:
position, which is proportional to the electrical charge content
x « Q (3)
velocity, which is proportional to time derivative of the electrical charge content
dQ / dt = I (4)
thus the velocity is proportional to current:
x«I (5)

www.samco.org Page 147 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

acceleration, which is proportional to the second time derivative of the electri-


cal charge content (d2Q/dt2 = dl/dt), thus the acceleration is proportional to
the change of current (x <x I).
As explained above, "charge"-control of piezo actuators within an open loop
system since sensors are not needed results in linear position control. Besides
this very important issue, charge control leads also to the following benefits:
increase of actuator stiffness, and
increasing actuator's reliability under very high dynamic operating conditions.
The slight nonlinear relation between stroke and charge shown in Fig. 98 may
be compensated by the inverse function of that relation. The most important
point is that hysteresis does almost not exist (approximately 1%) as in the
case of voltage control (Fig. 95). The main advantage of linearization via the
charge philosophy is the pure mode excitation. A sinusoidal electrical current
will be converted into a sinusoidal oscillation of the piezo actuator velocity
without any modulations due as it is the case for voltage controlled stroke due
to nonlinearities and hysteresis.
Piezo actuators show small positive drift in expansion over a distinct time
when a voltage step is applied (Fig. 99). This is called creep of piezo
actuators. It is based on a "ferroelectric" effect, where the polarization state of
the ceramic alters as long as an electrical charge flows that is delivered by the
voltage supply. Hence, creep may be immediately stopped, when the charge
content of piezo actuators is kept constant.
Sensing using piezo actuators is also possible (Law et al. (2003)). Then, the
displacement of the connected mechanical system produces a charge and
therefore a voltage within the piezo actuator.
References
Caruso, G., Galeani, S., and Menini, L. (2003), "Active Vibration Control of an
Elastic Plate Using Piezoelectric Sensors and Actuators", Journal of
Simulation Modelling Practice and Theory, 11 (5-6), 403-419.
Herold, S., Mayer, D., and Hanselka, H. (2004), "Transient Simulation of
Adaptive Structures", Journal of Intelligent Material Systems and Structures,
15(3), 215-224.
Law, W. W., Liao, W.-H., and Huang, J. (2003), "Vibration control of structures
with self-sensing piezoelectric actuators incorporating adaptive mechanisms",
Journal of Smart Materials and Structures, 12(5), 720-730.

www.samco.org Page 148 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Ma, K. (2002), "Vibration control of smart structures with bonded PZT patches:
novel adaptive filtering algorithm and hybrid control scheme", Journal of Smart
Materials and Structures, 12(3), 473-482.
Preumont, A. (2002), Vibration Control of Active Structures, Chapter 3, Kluwer
Academic Publishers, Dordrecht.
Raja, S., Prathap, G., and Sinha, P. K. (2002), "Active vibration control of
composite sandwich beams with piezoelectric extension-bending and shear
actuators", Journal of Smart Materials and Structures, 11 (6), 63-71.
Sadri, A. M., Wright, J. R., and Wynne, R. J. (2002), "LQG control design for
panel flutter suppression using piezoelectric actuators", Journal of Smart
Materials and Structures, 11(6), 834-839.
http://de.wikipedia.org/wiki/Piezo
http://www.efunda.com/Materials/piezo/general_info/gen_info_index.cfm

6 AMCO

http://www.marco.de/pb.html
http://www.physikinstrumente.com/tutorial/index.html
http://www.piezo.com/ (Piezo Systems, Inc., MA 02139 USA)
http://www.piezo-kinetics.com/ (Piezo Kinetics, Inc. Bellefonte, PA 16823,
USA)
http://www.piezomechanik.com/ (Piezomechanik GmbH, D-81673 Munich,
Germany)
Notations
Symbol Description Unit
C capacitance F
F force N
I electrical current A
Q charge C
U voltage V
W work, energy J
l stroke, displacement m
t time s
X displacement m
А difference -
е dielectric constant -

www.samco.org Page 149 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

2.2 Control algorithms


The three basic control strategies, such as feedback, feed forward, and
passive control are described in the first subchapter. The following
subchapters present different control approaches often used in the field of
structural control. Active damping using the Root Locus method is
documented in the second subchapter. It is extended with a short introduction
into PID control. The third subchapter describes the approach of optimal
control. This chapter also introduces the state space representation. The
fourth subchapter summarizes other linear control approaches often used in
the field of structural control.
2.2.1 Control strategies
Basically, there exist three ways how to control actuators connected to
dynamic systems:
feedback control,
feed forward control, and
passive control.
Within the following subchapters, these three control strategies are explained.
In the case of structural control, controlled actuators may also be controllable
damping devices since the main target of structural control in the field of civil
engineering is damping increase. Actuators and dampers will be called
"control device", the controlled dynamic systems such as tall buildings, slender
bridges, long cables, and other civil structures will be called "plant".
2.2.1.1 Feedback control
When the main disturbance w of the plant GP (s) is unknown or not
measurable (Fig. 100), feedback control may be applied. The basic idea is to
measure the system response у , which includes measurement noise v, then
to compare the variable у to the desired value of у , the so-called reference
signal r, and feed back the resulting error e through the controller GC (s) to the
plant GP (s) (d'Azzo and Houpis
(1995), Geering (1990), Preumont (2002), Soong (1990)). The controller is
designed in order to minimize the error signal e with the constraint that the
closed-loop system must not become unstable.

www.samco.org Page 150 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

Gc(s uc GA( u G GS
) >. s) p(s) (s)
Fig. 101: Closed-loop structure of feedback
control including actuator and sensor
dynamics.

Fig. 100: Closed-loop structure of feedback


control.
v
r
r
w

v
y
y

In reality, the closed-loop structure additionally consists of the control device


dynamics and of the sensor dynamics, represented by the corresponding transfer
functions Ga (s) and GS (s) in Fig. 101 (Dyke and Spencer Jr. (1997), Yeo et al.
(2002)). The

www.samco.org Page 151 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

controller output uC, e.g., a voltage signal, is the input signal of the actuator
which
eventually produces the desired control signal u , e.g., the desired force.
Usually, it is assumed that both the dynamics of the control device and the
dynamics of the sensor as well are much faster than the dynamics of the plant
and therefore negligible for model-based controller design. However, the
dynamics of the control device may limit the controller band width, especially in
the case of vibration control of high frequencies. The sensors have to be
chosen the way that the plant state variables of interest may be measured
without any aliasing.
2.2.1.2 Feed forward control
Feed forward control can be applied when a reference signal is available, which
is correlated to the disturbance acting on the plant (Fig. 102). The measured
reference signal is passed to an adaptive filter GF (s) whose output signal
influences the structure as a second disturbance. The criterion for the
adaptation of the filter coefficients is the minimization of the measured error e,
e.g., minimizing the displacement at a certain location of the structure (Ma
(2002)). Thus, the basic idea of adaptive filtering is producing such an additional
disturbance that the effect of the primary disturbance on the plant is
compensated. However, the full compensation is only possible at sensor
location.

Fig. 102: Structure of feed forward control. Fig. 103: Open-loop structure of
passive control.
e
r

Passive control

www.samco.org Page 152 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

In the field of structural control, passive control (Fig. 103) has been
implemented in many cases thanks to its simplicity and robustness
(Gordaninejad et al. (2002), Ko et al. (2002), Xu and Yu (1998a/b):
Passive control does not require any PC including control soft- and hardware.
Stability is not an issue because the loop is not closed.
Usually, the optimal value of the constant input signal uC of the control device is
determined experimentally. The criterion for the optimal value uC is minimum
error e, e.g., the minimum of the measured velocities at certain positions of the
plant.
References
d'Azzo, J. J., and Houpis, C. H. (1995), Linear Control System Analyzes and
Design, Conventional and Modern, Fouth Edition, McGraw-Hill (eds.), ISBN 0-
07016321-9.
Dyke, S. J., and Spencer Jr., B. F. (1997), "A Comparison of Semi-Active
Control Strategies for the MR Damper", Proceedings of the IASTED
International Conference, Intelligent Information Systems, The Bahamas, 8-10
December 1997, 580-584.

www.samco.org Page 153 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

О AMCO
Geering, H. P. (1990), Mess- und Regelungstechnik, 2. Auflage, Springer-
Verlag Berlin Heidelberg.
Gordaninejad, F., Saiidi, M., Hansen, B. C., Ericksen, E. O., and Chang, F.-K.
(2002), "Magneto-Rheological Fluid Dampers for Control of Bridges", Journal of
Intelligent Material Systems and Structures, 13(2/3), 167-180.
Ko, J. M., Zheng, G., Chen, Z. Q., and Ni, Y. Q. (2002), "Field vibration tests of
bridge stay cables incorporated with magneto-rheological (MR) dampers",
Proceedings of the International Conference on Smart Structures and Materials
2002: Smart Systems of Bridges, Structures, and Highways, S.-C. Liu and
Darryll J. Pines (eds.), Proceedings of SPIE (publ.), 4696, 30-40.
Ma, K. (2002), "Vibration control of smart structures with bonded PZT patches:
novel adaptive filtering algorithm and hybrid control scheme", Journal of Smart
Materials and Structures, 12(3), 473-482.
Preumont, A. (2002), Vibration Control of Active Structures, Kluwer Academic
Publishers, Dordrecht.
Soong, T. T. (1990), Active structural control: theory and practice, Cheng, W.F.
(advisory editor), Longman Scientific & Technical, ISBN 0-582-01782-3.
Xu, Y. L., and Yu, Z. (1998a), "Mitigation of three dimensional vibrations of in-
clined sag cable using discrete oil dampers - I. formulation", Journal of Sound
and Vibration, 214(4), 659-673.
Xu, Y. L., and Yu, Z. (1998b), "Mitigation of three dimensional vibrations of in-
clined sag cable using discrete oil dampers - II. application", Journal of Sound
and Vibration, 214(4), 675-693.
Yeo, M. S., Lee, H. G., and Kim, M. C. (2002), "A Study on the Performance
Estimation of Semi-active Suspension System Considering the Response Time
of Elec- tro-rheological Fluid", Journal of Intelligent Material Systems and
Structures, 13(7/8), 485-489.

Description Unit

www.samco.org Page 154 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

transfer function -
error -
desired value / desired trajectory -
complex frequency variable -
controlled input variable of the plant -
measurement noise - system noise / disturbance input
variable of the plant -
plant output variable -
Description
actuator
2.2.1.5 Notations Symbol
G (s)
e
r
s
u
v
w
У
Subscripts
AC
controller

www.samco.org Page 155 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural
Control

6 АМСО

www.samco.org Page 156


of 155
SAMCO Final Report 2006
F05 Guidelines for Structural
Control

F filter
P plant
S sensor

www.samco.org Page 157


of 155
SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

2.2.2 Active damping with collocated pairs


The following chapter describes active vibration suppression using collocated
pairs of actuators and sensors (Achkire et al. (1998), d'Azzo and Houpis (1995),
Preumont (2002), Soong (1990)). Actuator and sensor pairs are denoted as
collocated if they are physically located at the same place and energetically
conjugated, e.g., force and velocity or torque and angle. One main property of
such collocated pairs is that the closed-loop structure of each pair is independent
from the others. This leads to as many independent single input single output
(SISO) loops as collocated pairs.
In order to explain the different control schemes of active damping, first, linear
compensators will be introduced. Within this subchapter, the terms of "loop
shaping", "Nyquist Diagram", "Bode Diagram", and "Root Locus" will be described
(d'Azzo and Houpis (1995), Geering (1990), Preumont (2002)). Then, the different
closed-loop structures using different combinations of collocated actuator sensor
pairs will be presented.
2.2.2.1 Linear compensators PID control elements
The transfer function of a parallel PID controller (Fig. 104, Fig. 105) consists of
the proportional gain KP, integral gain Kt, and derivative gain KD (d'Azzo and
Houpis (1995), Geering (1990), Preumont (2002), Soong (1990))
K (1 ^
GPId (s) = Kp +-L + Kd ■ s = Kp ■ 1 + - + Tv ■ s (1)
■s
s
TN ■s ' J
The proportional gain is responsible for the "basic work" of the controller. The
proportional gain Kp amplifies the error e. This forces the measured plant output
у to follow the desired variable r (у tracks r). The integral gain makes it possible to
reach a tracking error equal to zero during steady-state conditions. The derivative
gain adds damping to the closed-loop structure which makes controlled plant re-
sponse fast to changes of r.
Bode Diagram kp/Tn kp/Tv 1/T

www.samco.org Page 158 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

100i

012345678 10 10 10 10 10 10 10 10 10 Frequency (rad/sec)


Fig. 104: Structure of parallel PID controller with Fig. 105: Bode diagram of P, I, D
elements, and inverse plant model NL-1. of low pass filter of 2nd order.

However, large values of KD may lead to unstable closed-loop behaviour.

www.samco.org Page 159 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Since PID control assumes a linear behaviour of the plant, the main non-linearity
(NL) between u and у must be compensated within the control algorithm by an in-
verse model of the nonlinearity (NL-1). Besides non-linearities of the plant itself,
the nonlinear relation between actuator input (voltage, current) and actuator
output (force, strain, displacement,...) has to be considered as well. In order to
suppress measurement noise, a low pass filter of order n is added in series to the
PID controller. The low pass filter is responsible for the so called "roll-off"
behaviour of the controller, which suppresses measurement and system noise
(Fig. 106). A first order filter produces -10 dB/dec "roll-off", which is not seen as
sufficient for measurement and system noise suppression. Usually, filters of 2nd
order (-20 dB/dec) or higher are applied.

M
f
1
1
(n > 1)
1+■
+ Tv ■ s
(T ■ s + l)n
TN ■ S
GPID-rf (S) = 1
KP

(2)
1
(aT)la
40

www.samco.org Page 160 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

30-
SB 20-
■O
V 10-
■O
M0
§-10- 2-20- (T ea d = 1/100 И)
(oTL
Bode Diagram

Bode Diagram
10 10 10 10 Frequency (rad/sec)
Fig. 106: Bode diagram of a parallel PID
control- Fig. 107: Characteristics of Lead
and Lag ele- ler with low pass filter of 2nd
order for a good ments shown in the Bode
Diagram. "roll-off".
0
-180
0
2
7
10
10
10
10
80
70 -
— 60- CQ
S 50- f 40 & С 30- I 20 10 - 0
90

lead

www.samco.org Page 161 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

a
a
_ a^=2 _
ai,g=10
aiag|

-30- (Tag = 1/100 [s]) -40 90


S 45
i0
i
' -45-90 10
123 10 10 10 Frequency (rad/sec)
a,, =10C
ag
8
3
10

Lead / Lag element


The Lead and Lag elements allow for adjusting the transfer function of the
controller in the frequency domain (d'Azzo and Houpis (1995), Preumont
(2002)). Since these elements change the shape of the transfer function in the
"Nyquist Diagram", where the transfer function is a trajectory in the real-
imaginary-plane and therefore appears as a line or "loop", this procedure is
called "loop shaping". The Lead element may increase the phase margin,
whereas the Lag element may increase the gain margin (Fig. 107).

GLead (S)
=
:a<1
(3)

www.samco.org Page 162 of 155


SAMCO
Final
Report
2006
F05
Guidelines
for
Structural
Control

T ■ s +1 a
■ T ■ s +1

GLag (s)
:a>1
(4)
T ■ s +1 a
■ T ■ s +1

www.samc
o.org
Pag
e 163 of
155
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Controller tuning
Besides the trial and error method for tuning the parameters of PID controllers, the
most common way of model-based controller parameter design is presented here
(Geering (1990)):
Given the transfer function of the plant together with a single P controller, the
value of KP may be determined where the closed-loop structure just becomes
unstable. This value of KP is called KP_rrit Then, the real parts of the closed-loop
poles are zero. The imaginary parts of the closed-loop poles are identical to the
critical frequency in radiant per second (^ Trrit). For higher
values of KP, the real parts of the closed-loop poles get positive and the closed-
loop structure becomes unstable.
KP_cnt, Tcnt = Inherit (5), (6)
Based on the values of KP_rrit and Trrit, the PID controller parameters may be
tuned according to Ziegler/Nichols:
P controller: Kp = 0.5 ■ KP_rrit (7)
PI controller: Kv = 0.45 ■ KP_mt, TN = 0.85 ■ Tcnt (8), (9)
PID controller: Kp = 0.6 ■ KP_mti , Tn = 0.5■ Tmtl, Tv = 0.12 ■ Tcrit
(10)-(12)
Please notice that the determination of KP_crit requires a model of the plant. For
this reason, the controller design may be called model-based.
Root Locus method
In the following, the determination of the values of KP_rrit and Tcrit shall be
demonstrated. It is assumed that the plant has the only real poles -1, -2, and -3
(Fig. 108, Geering (1990)). The transfer function of the closed-loop system with
the P controller only and negative unity feedback becomes
G (s) = YM. = Gc(s) ■ Gp(S) = KP
rl s U(s) 1 + GC(s)■ GP(s) s3 + 6s2 + 11s + (6 + KP)
(13)
The poles of GC are the zeros of the denominator of Gcl. The poles of the closed-
loop system are plotted in the real-imaginary-plane (Nyquist Diagram) with
parameter KP (Fig. 109). The resulting line or trajectory is called "Root Locus" of

www.samco.org Page 164 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

the closed- loop. For increasing values of KP, the closed-loop inclines more and
more to vibrate. For real parts of the closed-loop poles equal to zero, the closed-
loop system does not have any damping. Then, the closed-loop system is on the
border of instability. For the example presented, the critical values are:

www.samco.org Page 165 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

(14), (15)

KP.cnt = 60 [-] , Tcnt = 2n!4Tl [J]


If KP is further increased, the closed-loop system gets unstable or the damping of
the closed-loop system becomes "negative", respectively.

Root Locus

-4 -3 Ra(s)
Fig. 109: Root Locus of 3rd order
plant with feedback proportional
gain KP only.
5 4 32 1 -
f
-1 -2 -3 -4 -5

Gc(s)
Gp(s)
r
y
Fig. 108: Plant of 3rd order with
proportional gain only.

2.2.2.2 Active damping with collocated actuator sensor pairs Main idea

www.samco.org Page 166 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

The equation of motion of a system excited by the disturbance force w and con-
trolled by the force f is
Mz + Cz + Kz = w + f (16)
When the control target is increasing the damping of the plant, then the control
force f must be proportional to the structural velocity at actuator location in order
to increase apparently the plants damping
Mz + (Cz-f)+ Kz = w ^ f ~-z (17), (18)
This leads to a control force, which acts like an external viscous damper. The
collocated sensor may be, e.g., a velocity sensor. The matrix notation of the
equations above points out that the damping of the closed-loop structure may be
increased using numbers of collocated actuator sensor pairs (Fig. 110). Then, a
decentralized controller may be realized by a proportional negative feedback
f =-1 ■ [gi g2 ... gn]T ■ [z2 ... z„]T (19)
The controller is called decentralized because every sensor builds a closed-loop
only with its collocated actuator. Due to the positive definite values of gt (similar
to P element), this control law (negative feedback) guarantees power dissipation.

www.samco.org Page 167 of 155


SAMCO Final
Report 2006
F05 Guidelines
for Structural
Control

Fig. 110: Closed-


loop structure of
direct velocity
feedback for
collocated actuator
sensor pairs.
,"D О
:шо

GF
r
У

The locations of the velocity sensors must be chosen the way that the
vibration velocity of the mode of interest is not zero. If this is the case, the
collocated actuator sensor pair is located at the structure where the mode of
interest has a nodal point. In order to guarantee a proper working behaviour of
the proportional controller, a low pass filter should be added, which attenuates
the measurement noise. The resulting closed-loop structure is called direct
velocity feedback (DVF, Nakamura et al. (2002), Preumont (2002),
Subramanian (2002)).
Tuning the feedback gain for maximum additional damping
Given the four mass system depicted in Fig. 111 with identical values for
mass, stiffness, and damping elements, the equation of motion with the
external control forces f becomes

www.samco.or
g Page
168 of 155
SAMCO Final Report 2006
F05 Guidelines for Structural Control

z2+
z3
_z
4_
- с 2с - с 0
с
0 0 - с 2с
2k
-k 2k -k

zi fl
z2 f2
Z3 f3
z4 f4
k 2k -k
0 0 - k 2k
k0
000

с00
00
m
0m0000m0000m
z2z3
+
0
с 2с
0
4
(20)

www.samco.org Page 169 of 155


SA
M
CO
Fin
al
Re
por
t
20
06
F0
5
Gu
ide
lin
es
for
Str
uct
ura
l
Co
ntr
ol

сь
c
c
c
c
c
1
z

z
2
3
z

ww
w.s
am
co.
org

Pa
ge
17
0
of
15
5
SAMCO Final Report 2006
F05 Guidelines for Structural Control

z
4
Fig. 111: Four mass system.
For the following investigation, one force actuator and one velocity sensor at the
first mass are assumed. Then, as seen in the previous chapter, negative
feedback with proportional gain KP is able to damp structural vibrations. The
measurement equation becomes

www.samco.org Page 171 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

www.samco.org Page 172 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

y = [0 0 0 0 1 0 0 0] ■ [ z1 z 2 z3 z 4 z1 z2 z3 z4]T
(21)
Due to the low structural damping and the collocation, the transfer function
shows the alternating resonance/anti-resonance pattern (Fig. 112, Preumont
(2002)). The resonances are the poles of the plant, where the velocity z1 (output)
is infinite large for the external force f1 (input) near to zero. The anti-resonances
are the plant's zeros, where the velocity z1 becomes zero for any external force.
The alternating pole/zero pattern is also visible in the real-imaginary plane (Fig.
113 and Fig. 115).
The control target shall be maximum damping of the structural mode with the
lowest frequency, because this mode usually produces the largest displacement
(the modes in Fig. 113 are numbered with increasing frequency). The goal is to
find the optimum feedback gain g for maximum additional damping. Obviously,
there exist two cases for g that cannot be optimal:
g = 0 : The feedback gain is zero. Hence, the closed-loop system behaves like
the open-loop system. Then, the pole of the open-loop system is identical with
the pole of the closed-loop system (Fig. 113).
g = ™ : The feedback gain is infinite large. Then, the control force is also infinite
large, which results in a nodal point of the structure at actuator position. The
actuator force behaves like a completely stiff bearing. The closed-loop pole is
identical with the open-loop zero due to vanishing velocity and displacement,
respectively, at actuator position (Fig. 113).
closed-loop pole of mode i moving from open-loop pole (Kp=0)

Frequency (rad/sec)
Fig. 112: Transfer function of four mass system Fig. 113: Placement of closed-
loop pole for maxi- with force input and velocity output at mass no.1. mum
additional damping of mode 1 (direct velocity

www.samco.org Page 173 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

feedback).

As explained in the preceding chapter, the root locus is the trajectory of all
closed- loop poles for g increasing from zero to infinity. Considering the following
characteristics of the closed-loop pole
_ real(poled_. )=£& , \polecl_\ = at (22), (23)

www.samco.org Page 174 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

the closed-loop structure with maximum additional damping shows its pole on the
root locus where the angle between real and imaginary parts reaches its
maximum (Preumont (2002))

- real (рЫесЫ)
a
maximum damping of mode i: Zi-mca ^ max-

(24)

This fact also points out that large damping may only be achieved for
trajectories going far into the left half plane in the Nyquist Diagram. This is
only possible for poles and zeros being well separated. The transfer function
of the plant with optimally damped mode 1 is shown in the Bode Diagram
depicted in Fig. 114. The resulting pole locations of all four structural modes
are depicted in Fig. 115. This figure clearly demonstrates that single actuator
sensor pairs may only damp one mode optimally.
60 40 f 20 V 0
■O
5 -20 S-40
03
5 -60 -80 -100
90 L

g = 413: optimal for maximum damping


of mode 1
plant
controlled plant

www.samco.org Page 175 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

proportional gain: g = 413 -


™ 45
0
Bode Diagram 5 Root | Locus
# closed- m—^
loop poles
of modes
1-4 for g =
413
* open- / maximum И
loop poles: additional
closed- damping for
loop poles mode 1
of modes
1-4 for g =
0
о open-
loop zeros:
closed-
loop poles
of modes
1-4 for g =
8
321
I 0"1 -1 - -2-
4

45
90

www.samco.org Page 176 of 155


SAMCO Final Report
2006
F05 Guidelines for
Structural Control

10
10 10 Frequency
(rad/sec)
-0.5 0
-4.5 -4 -3.5 -3 -2.5 -2 -1.5
-1 Re(s)

Fig. 114: Comparison between very Fig. 115: Location of closed-loop


low damped plant and plant with poles of all four modes for maximum
optimally damped mode 1. damping of mode 1.

Active damping with force actuators


As seen in the example of direct velocity feedback, the idea is to feed the force
actuator with a signal being proportional to the actual vibration velocity at
actuator position. Negative feedback guarantees energy dissipation and thus
stability of the control loop. Basically, there are three ways how to measure the
actual vibration velocity. These three possibilities are described in the following.
a) Displacement measurement
Notice that measuring displacements requires the differentiation of these
signals in real time in order to get the actual velocities (Gandhi and Munsky
(2002)). Due to the transfer function of derivative elements, measurement noise
will be amplified. Therefore, when displacement sensors are used for active
damping, the derivative element must be extended by a low pass filter of nth
order (GF) for a sufficient good "roll-off" behaviour of the closed loop system

www.samco.org Page
177 of 155
SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

Of1
Gc(s) = { ( )n (25)
active damping 1s rf 3
with displacement 4 v /
sensor Gf
The filter order should be at least three (n = 3) in order to produce at least a
"roll-off" of -40 dB/dec (d'Azzo and Houpis (1995), Preumont (2002)). The filter
corner frequency is the frequency above which the signal parts are cut off. The
filter is designed the way that the steady-state amplification is one (GF (s = 0) =
1).
b) Velocity measurement
Although the signal does not have to be differentiated, again, it is recommended
to add a low pass filter to the proportional gain for measurement noise
suppression. The filter should be of 2nd order or higher in order to produce -
40dB/dec or more.
К
Gc (s) = g •т ^ (26)
{ (s + о )
active damping V rf '
with velocity 4 v '
sensor Gf
c) Acceleration measurement
Measuring the acceleration requires the integration of the signal in real time.
The integration helps already to suppress measurement noise, but only with a
"roll-off" of -20 dB/dec
Gc (s) = g •1 (27)
{s
active damping with accelerometer
In order to reach a "roll-off" behaviour of -40 dB/dec, a low pass filter of first
order may be added. It may help not destabilizing modes of high frequencies.

www.samco.org Page 178 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

The first order low pass filter together with the integration of the acceleration
signal gives the form of a 2nd order filter
Gc (s) = g • 2 Z Г (28)
s2 +2 rfF arFs + a>F
i ^^ j
1 'Gf
s
The tuning parameters of the second order filter, the filter frequency and the
filter damping ratio, allow for tuning the filter precisely to the mode of the plant
to be damped. According to Preumont (2002), the rule of thumb for the filter
tuning becomes (index F for filter, index P for plant, index i for mode number):
oF-i close to aP-i (rnF-i > rnF) (29)
ZF = 0.5 ... 0.7 (ZF T ^ increase of damping ratio)
(30)

6 AMCO

By designing several second order filters, several structural modes may be


damped. Here, the usual procedure is to determine the gains gi by experiment,
starting with the filter design for the mode with highest frequency. These n filters
targeted to damp n structural modes do not necessarily need n actuator sensor
pairs. The number of actuator sensor pairs may be smaller than n.
Active damping with strain actuators
Strain actuators may work in two principle ways:
If piezo elements deform freely, they produce a voltage signal (output) that is
proportional to the deformation of the piezo element.
If piezo elements are, e.g., glued to the structure surface, their deformation is
not free. Then, they apply a strain to the structure which is proportional to the
input voltage signal.
As a consequence of these two different working principles, either the
displacement of the structure or the force generated by the piezo element is
measured.
a) Displacement measurement
The output signal of the piezo element which acts as a strain sensor is
proportional to displacements. Then, active vibration mitigation is feasible by

www.samco.org Page 179 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

producing a stress being proportional to the displacement if the entire feedback


loop is positive (Preumont (2002))
Gc (s) = -g ■ 2 7 (31)
. —. s + 2ZFaFs + aF
active damping K 4гг г
with strain sensor
f
That is why this active control scheme is called Positive Position Feedback
(PPF). For a sufficient "roll-off", a second order filter should be designed
similarly to the filter design for DVF using accelerometers.
b) Force measurement
In the case of force measurements, piezo elements are collocated with force
transducers. If the force applied to the structure is measured, the measurement
signal must be integrated and fed back in a positive feedback loop in order to
produce a root locus being stable for all proportional gains. This active control
scheme is called Positive Force Feedback (PFF). In order to avoid the problem
of saturation for integral gains, the forgetting factor e should be introduced,
which moves the controller pole from the origin slightly to the negative real axis
Gc(s) = _ g ■— ■ Gf (32)
s+e
V
active damping with force sensor (including forgetting factor)

www.samco.org Page 180 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

References
Achkire, Y., Bossens, F., and Preumont, A. (1998), "Active damping and flutter
control of cable-stayed bridges", Journal of Wind Engineering and Industrial
Aerodynamics, 74-76, 913-921.
d'Azzo, J. J., and Houpis, C. H. (1995), Linear Control System Analyzes and
Design, Conventional and Modern, Fouth Edition, McGraw-Hill (eds.), ISBN 0-
07016321-9.
Gandhi, F., and Munsky, B. (2002), "Comparison of Damping Augmentation
Mechanisms with Position and Velocity Feedback in Active Constrained Layer
Treatments", Journal of Intelligent Material Systems and Structures, 13(5), 317-
326.
Geering, H. P. (1990), Mess- und Regelungstechnik, 2. Auflage, Springer-
Verlag Berlin Heidelberg.
Nakamura, T., Saga, N., and Nakazawa, M. (2002), "Impedance Control of a
Single Shaft-type Clutch Using Homogeneous Electrorheological Fluid", Journal
of Intelligent Material Systems and Structures, 13(7/8), 465-469.
Preumont, A. (2002), Vibration Control of Active Structures, Kluwer Academic
Publishers, Dordrecht.
Soong, T. T. (1990), Active structural control: theory and practice, Cheng, W.F.
(advisory editor), Longman Scientific & Technical, ISBN 0-582-01782-3.
Subramanian, P. (2002), "Vibration suppression of symmetric laminated
composite beams", Journal of Smart Materials and Structures, 11(6), 880-885.
Notations Symbol
G(s) K
M, C,K
T
e
fg
m, с, k n s t u
v, w
y

www.samco.org Page 181 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Description transfer function feedback gain


mass, damping, stiffness matrices
time constant
error
force
feedback gain
mass, viscosity, stiffness
order
complex frequency variable time
controlled input variable of the plant measurement noise, disturbance plant
output variable displacement, velocity, acceleration
Unit
s
N
kg, kg/s, kg/s2
s
m, m/s, m/s2
z, z, z

www.samco.org Page 182 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

www.samco.org Page 183 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

a tuning factors for Lead / Lag elements -


e forgetting factor -
circular frequency rad/s Z damping ratio -
Superscripts Description
T transposed
Subscripts Description
c controller
D derivative
F filter
integral
P plant, proportional
cl closed-loop
crit critical
f filter
i mode number
n number of collocated actuator sensor pairs
rf roll-off

www.samco.org Page 184 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

2.2.3 Optimal control


Optimal controllers are state-feedback controllers which try to equalize the
states to zero. The states are the state variables of the plant model that is
represented in the so-called state space form. Hence, first, the state space
representation of linear dynamic systems will be introduced. Then, the optimal
control approaches such as Linear Quadratic Regulator (LQR) and Linear
Quadratic Gaussian noise control algorithm (LQG) will be described.
2.2.3.1 State space representation
(1)
The representation of any dynamic system in state space representation shall
be explained on behalf of a single degree of freedom system (SDFS) subjected
to the external control force f and the external disturbance force w (d'Azzo and
Houpis (1995), Geering (1990)). The differential equation of the SDFS becomes
mz + cz + kz = f + w
The following substitutions

IkZ
c
u=f
(2)-(4)
О
2л/ mk
m

(5)
(6)
yield to the differential equation of a SDFS with resonance circular frequency o0
of the undamped SDFS and damping ratio Z
z + 2Z®0 Z + o02 z = —

www.samco.org Page 185 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

m
By introducing the state variables
" z
X1"
_X z
2_

the second order differential equation may be rewritten in state space


representation (SSR) including only first order differential equations

" " 0" ' 0"


X1" + •u+
_ у у
x2 _/ m
_ m_
1
- 2Z0>
w
"
X1"
_
x2
_
0
(7)
- 0)7,

In order to measure the vibration state of this SDFS, the following possibilities
exist:
Measuring the displacement (x1).
Measuring the velocity (x2).
Measuring the acceleration (X2).

www.samco.org Page 186 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

(8)
According to the sensor type, the measurement equation including
measurement noise v looks as follows
у = x1 = [1 0] • x + [0] • u + [0] • w + v

www.samco.org Page 187 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

у = x2 = [0 1]. x + [0] • u + [0] • w + v (9)


у = x&2 = [- «02 - 2Z®o] • x + [%J • u + [/m] • w + v (10)
The SSR of a any dynamic, linear, time-invariant system with several measured
output variables (MO: mixed output), one input variable (SI: single input), the
unknown process/system w, and the measurement noise v becomes in matrix
notation (Fig. 116)
X=A•x+B•u+G•w (11)
y=C•x+D•u+H•w+v (12)
The SSR is called time-invariant if the system matrices A , B , C, D, G, and H do
not vary in time. Usually, the measurement noise is assumed to be a zero mean
white noise signal. A white noise process is a mathematical idealization of a
stationary random process, whose elements are completely uncorrelated in
time. The matrix G describes the local distribution of the process/system noise
w. Besides modelling errors, it may include unknown, external forces and noise
in the vector of the input variable. Modelling errors may result from:
neglected nonlinearities of the plant,
neglected very fast dynamics of the plant,
neglected very slow dynamics of the plant, and
neglected nonlinearities and dynamics of actuators and sensors.
2.2.3.2 Linear quadratic regulator

Fig. 116: State space representation of the plant.


У
A full state feedback may be designed with an Optimal Linear Quadratic
Regulator (LQR, d'Azzo and Houpis (1995), Geering (1990), Preumont (2002),

www.samco.org Page 188 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Ribakov et al. (2003)). This feedback approach requires that all ( k ) state
variables are known and therefore measurable. The goal in structural control
often is to place the poles of the closed-loop system with LQ Regulator the way
that the damping of the closed-loop structure is increased compared to the
damping of the uncontrolled plant.
Fig. 117: Plant in SSR, all state variables measured, full state feedback with
stochastic LQ Regulator.

www.samco.org Page 189 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

(15)
(16)
One distinguishes between stochastic and deterministic LQ Regulators,
according to the fact if an unknown disturbance w (assumed as white noise) is
present or not. The stochastic LQ Regulator has to be designed for the following
plant subjected to the white noise excitation w with intensity W
x = A • x + B • u + G • w with Ј[w • wT ] = W
(13), (14)
Then, the full state feedback
u = -K • x
minimizes the performance index (cost functional in the case of deterministic
LQR)
J = Ј[xT • Q • x + uT • R • u]
qn 0
0 q22 0 •••
0
(17), (18)
Q=
r=1
O0
0q2j2j

This example assumes one actuator, so u and R are scalar. The input weight
factor R and the factors qii are design parameters of the LQ Regulator. For
large values of {qu/R), the control output u will be large and consequently
force the plant rapidly into its equilibrium position. Thus, the tracking error will
be small for large values of {qu/R), but at the expense of large control effort.
The state feedback, which is based on measurements, is not optimal due to
the measurement noise v

www.samco.org Page 190 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

-K • x = -K • y
(19), (20)
u
y=I•x+v^

(21)
Besides the regulator tuning parameters Q and R , the design of the regulator
matrix K requires the knowledge of the system matrices A and B. These
matrices are based on the physical model of the real plant. Hence, they
include modelling errors. In order to distinguish between the real plant
behaviour and the model behaviour, the model matrices used for model-based
controller design are indicated in the following with the index с
K = f (A с ,B с ,Q, R)
The resulting control system is a Single-Input-Mixed-Output System (SIMO,
Fig. 117), however, also MIMO (mixed input, mixed output), SISO (single
input, single output), and MISO (mixed input, single output) are conceivable.
As mentioned above, LQR design makes it absolutely necessary knowing all
state variables, but almost in all practical cases, the state variable vector is not
completely known. In this case, the state variables must be reconstructed
using a model of the plant (Zhang and Ro- schke (1999)). The procedure of
full state reconstruction will be explained in the next subchapter.

www.samco.org Page 191 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

2.2.3.3 Linear Quadratic Gaussian noise control Modelling the plant


The procedure how to design a Linear Quadratic Gaussian noise control
algorithm (LQG, d'Azzo and Houpis (1995), Christenson et al. (2001), Kim et
al. (2003), Preu- mont (2002)) will be explained on behalf of a beam model. It
is subjected to a white noise process w and the measured state variables
include the measurement noise v. The one-dimensional finite element model
(FE model) of a beam with displacement z perpendicular to the beam length
and control force f becomes
M feZ + C fe z + K fe Z = 0 f W + Л fJ (22)
The SSR of this differential equation may be written with the following state
variable
X [zn-l zn-2 "" zn-nn Zn-1 zn-2 "" zn-nn] (23)
This state vector x has twice as many elements as nodal points (nn) of the FE
model. For high spatial resolution, the order of the proposed SSR is fairly large
(dimensions of SSR matrices large). Thus, the simulation of such an SSR
would need large computing time. Moreover, vibrating modes with only a
couple of nodal points per wave length may lead to "noisy" or scattering
simulation results. In order to reduce simulation time and to avoid scattering
effects, the original FE model may be:
simulated with the Newmark integration method, which adds fictitious damping
to higher modes and this way helps to dampen "vibrations" or "scattering" pro-
duced by higher modes with insufficient spatial resolution, or
transformed from Cartesian coordinates into modal coordinates in order to
truncate higher modes with insufficient spatial resolution.
Since the Newmark integration algorithm is commercially available, the second
method of coordinate transformation shall be described hereafter.
The SSR in modal coordinates includes the displacements and velocities of
the first mm modes of the plant and not of all nn nodal points of the original FE

www.samco.org Page 192 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

model. One method of transforming the FE model from Cartesian coordinates


into modal coordinates is the Galerkin method (Ni et al. (2000), Raja et al.
(2002))
mm
z, (r, t) = ^ q} (t) ■Pj (r) (mm < nn) (24)
j=1
The shape functions may be approximated by sinusoidal functions
PPj (r ) = Sin{j ■^■{rjLbeam )} (25)
The resulting beam model in modal coordinates becomes
Mm&& + Cmq + K mq = 0m W + Лmf (26)
with the plant's matrices in modal coordinates
M m = Ф Г M fe Ф , Cm = Ф' C fe Ф , D m = Ф' D fe Ф , 0 „ = Ф' 0 fe , Лт = Л fe
(27)-(31)
and the transformation matrix

www.samco.org Page 193 of 155


SAMCO Final
Report 2006
F05 Guidelines
for Structural
Control

Pi(ri)
<Pl(rnn )
P (r)
mm1
P (r )
т mm ^ nn '
(32)
Ф

The beam model may take the first m modes into account. The according state
variable vector becomes
X [qm-1 qm-2 "" qm-mm Qm—l $m—2 "" tfm-mm]
(33)

Assuming that process noise w and measurement noise v are white noise proc-
esses with intensity matrices W > 0 and V > 0, and for the case that all nodal ac-
celerations are measured, then the state and measurement equations for one
actuator (u is scalar) become
x=A•x+B•u+G•wy=C•x+D•u+H•w+v
with
"0I" "0" "0"
A= - M -1K - M -1C , в = _M m , G = M -10
_ M m K m M m Cm Лm _ mm
_

(36)-(38)

C = [- ФTM-1C - ФTM-1K ] D = [-Ф T M -1Л L mm-1'


L mm m mJ ],

www.samco.or
g Page
194 of 155
SAMCO Final Report 2006
F05 Guidelines for Structural Control

H = [Ф M m 0 m ] (39)-(41)

Full state reconstruction


Since it would be too expensive or simply not feasible to measure all state
variables because they are not available, the idea is to reconstruct the full state
vector by a model of the plant (model matrices indicated by index c). The state
reconstruction of the plant with the process and measurement noise described
above is realized by the observer of the form
X = Ac • x + Bc • u + L•(y-Cc-red • x-Dc-red • u) (42)
Here, L is the observer gain matrix, which is chosen the way that the difference
between true state x and estimated state x converges to zero
e = x - x ^ 0 (43)
If L is chosen the way that the error covariance matrix is minimized, the designed
observer is called minimum variance observer or steady-state Kalman-Bucy
Filter (KBF, Geering (1990)).
In order to design the observer gain matrix L , model matrices must be available
and the intensities of process and measurement noise must be assumed. The
observer gain matrix is defined by the following input data

www.samco.org Page 195 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

www.samco.org Page 196 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

L = f (Aс,Gс,Cс-red ,Hс-red,E[w • wT],E[v • vT],E[w • vT])


(44)
Since only some of the nn nodal accelerations are measured, the system
matrices C, D, and H must be reduced to those columns where the sensors are
located
dimension(Cc) = (number of nodes)x(2 • number of modes)

^ dimension(Cc-red) = (number of sensors)x (2 • number of modes)

dimension(Dc) = (number of nodes)x1 (47)


^ dimension(Dc_red) = (number of sensors )x1 (48) dimension(Hc) = (number of
nodes)x 1 (49)
^ dimension(H c_red) = (number of sensors )x1 (50)
Process and measurement noise must be assumed. Hence, they may be seen
as observer design parameters (Kim et al. (2003), Sadri et al. (2002)). If
measurements of, e.g., the disturbance forces and some nodal velocities are
available, the intensities of process and measurement noise may be estimated
from the alternating signal parts (AC) of these measurements. In this example,
the measurement noise is assumed to count for 5% of the maximum value
E[w • wT ] - var(AC(w_)) (51)
E[v • vT ] - var(0.05 • AC(zmm)) (52)
Regulator design
The full state feedback (single actuator)
u = -K • x (53)
minimizes the performance index
J = E[xT • Q • x + uT • R • u] (54)
Besides the system matrices A c and B c, the regulator gain matrix requires the
state weight factor Q and the input weight factor R (single actuator) as regulator
design parameters
K = f (A с ,B с ,Q, R) (55)

www.samco.org Page 197 of 155


SAMCO Final
Report 2006
F05 Guidelines for
Structural Control
OAMCO

0
qn 0
Q=
R=1
0 q22

(56), (57)
The value of R may be set to 1 because the observer design only depends on
the ratio of qu / R.
When designing the regulator, the following rule of thumb should be respected
(Preumont (2002)): For every mode to be damped, the ratio of qu /R should be
chosen the way that the negative real part of the observer pole is
approximately 2-6 times larger than the negative real part of the corresponding
regulator pole. The reason for this rule is that regulator poles are the poles of
the closed-loop structure or, in other words, of the controlled plant (Fig. 118).
Considering that large negative real parts of observer poles imply that
estimation errors decay faster than the relevant dynamics of the plant, then,
the reconstructed state variables follow closer the actual true state variables.
Thus, small values of qu /R will produce a state estimate of high
accuracy but with the drawback of increased tracking error. The ratio qu / R is
the

www.samco.org
Page 198 of
155
SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

regulator tuning variable for the controller designer in order to find the desired
compromise between tracking and estimation error. The example of pole
placement depicted in Fig. 119 is the result of the following control
requirements:
The control target is to damp the first four structural modes.
The observer poles should be at least 4 times "faster" than the regulator
poles.
If only the first four modes (control target) were observed, the estimated vibra-
tion state would be falsified due to the influence of the higher modes (5-...) of
the real plant on the actual vibration state. However, a falsified state estimate
leads to a non-optimal or even wrong state feedback and therefore to a dete-
riorated control performance. Hence, higher modes than only the controlled
ones should be observed. The effect that unobserved modes falsify the state
estimation and evoke a false state feedback is called spillover instability. The
fact that the state estimation may only be based on a model of the plant is in-
dicated again by the index c of the system matrices (Fig. 118). In the example
shown here, the first eight modes are observed, although only the first four
modes are to be controlled (damped).
The matrix Q weights the modal state variables (Johnson et al. (1999)).

www.samco.org Page 199 of 155


SAMCO Final
Report 2006
F05 Guidelines
for Structural
Control

observer pole of mode


10
real measurements
dynamics of real plant
closed-loop pole of
mode 1 with increased
damping

H x=A cx+Bcu
-20 -40 -60 = -80 -100 - + L(y-Cc x-
120 -140 -160 Dc u)
D
observer
LQG
w
controller
C
*XЖX
=-y
*x
u
*
3
**
x
-10
x=Ax+Bu+Gw
Re(s)
1- x
-4 -2 0
K
control
1
target:
regul
increasing
ator
damping of

www.samco.or
g Page
200 of 155
SAMCO Final Report 2006
F05 Guidelines for Structural Control

the first 4 modes -20 -18 -16


□ poles of real plant X -14 -12
poles of plant model ~ ^
regulator poles о observer
poles (time continuous)

Fig. 118: Real plant with full state observer and regulator.
Discrete-time control algorithm
For the implementation of control algorithms, nowadays, it is common to
transform the algorithm into the discrete time domain. This enables to run the
algorithm a defined clock rate (controller frequency). The algorithm is written in
commercially available software which can be programmed on the host PC. The
use of a PC makes it possible to adjust controller parameters on site. For real-
time control, the software has to be downloaded on either a real-time controller
box or on a PC real-time card. This is the normal procedure as long as the
control algorithm is within the development process.
In the case of LQG control, the control algorithm consists of the following three
steps: The measurement update at the time k, the extrapolation step from time k
to (k +1), and the step of state feedback (discrete-time system matrices indicated
by index dt)
X[k I k] = X[k I k - 1] + M dt ■ (y[k] - Cc-red-dt X[k I k - 1] - D_d-dt ■ u[k])
(58)
X[k +11 k] = Ac-dt ■ X[k I k] + Bc-dt ■ u[k] (59)
u[k] = -Kdt ■X[k I k] (60)
Fig. 119: Observer and regulator design.
The estimated measurements become as follows
У c [k ] = C c_eed_dt ■ X [k I k ]+ D c_eed_dt ■u[k] (61)
The discretization may be realized using the Tustin approximation. Since the
measurements y[k = 0] at time k = 0 already exist, the observation starts with a
measurement update, followed alternating by an extrapolation step and

www.samco.org Page 201 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

measurement update (Geering (1990)).

www.samco.org Page 202 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Semi-active control
u
When controlling a structure using semi-active control devices such as, e.g.,
magnetorheological and electrorheological fluid dampers, the regulator should
feed its output to the damping device only if structural energy shall be dissipated.
Otherwise, if the state feedback desires active forces, the regulator output should
be set to zero (Ying et al. (2002)), because semi-active devices cannot produce
power (Fig. 120). This switching is called "clipping" of the control force which
ends up in high frequency excitation and the fact that the clipped control force is
not optimal anymore.

Fig. 120: Schematic operating range of magnetorheological fluid dampers.

The decision, if the regulator output is an active or passive force, may be based
on the measured or estimated structural velocity at the position of the semi-
active control device
K • x : {(-K • x • z„) < 0} or {(-K • x • z^) < 0} 0 : {(-K • x • 4-mea) * o} or {(-K • X •
zu-est) > 0}
(62)
The force of magnetorheological and electrorheological fluid dampers, which
represent often used semi-active control devices in the field of structural control,
is not only limited by the maximum damper force but also by the residual friction
at zero command input (Fig. 120). Hence, the optimum control force is not only

www.samco.org Page 203 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

constraint by the decision between active and dissipative forces but also by the
limited operating range of magnetorheological and electrorheological fluid
dampers. Not all desired forces, which are dissipative, may be tracked due to the
limited damper's operating range. This may be seen as a "secondary" clipping.
2.2.3.4 References
d'Azzo, J. J., and Houpis, C. H. (1995), Linear Control System Analyzes and De-
sign, Conventional and Modern, Fouth Edition, McGraw-Hill (eds.), ISBN 0-07-
016321-9.
Christenson, R. E., Spencer Jr., B. F., and Johnson, E. A. (2001), "Experimental
Verification of Semiactive Damping of Stay Cables", Proceedings of the 2001
American Control Conference, IEEE, Piscataway, NJ, USA, 6, 5058-5063.

www.samco.org Page 204 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

6 AMCO

Geering, H. P. (1990), Mess- und Regelungstechnik, 2. Auflage, Springer-Verlag


Berlin Heidelberg.
Johnson, E. A., Spencer Jr., B. F., and Fujino, Y. (1999), "Semiactive Damping
of Stay Cables: A Preliminary Study", Proceedings of the 1999 17th International
Modal Analysis Conference - IMAC, 1, 417-423.
Kim, D.-H., Han, J.-H., Yang, S.-M., Kim, D.-H., Lee, I., Kim, C.-G., and Hong,
C.- S. (2003), "Optimal Vibration Control of a Plate Using Optical Fiber Sensor
and PZT Actuator", Journal of Smart Materials and Structures, 12(4), 507-513.
Ni, Y. Q., Cao, D. Q., Ko, J. M., and Chen, Y. (2000), "Neuro-Control of Inclined
Sagged Cables Using Semi-Active MR Dampers", Proceedings of International
Conference on Advances in Structural Dynamics, J. M. Ko and Y. L. Xu (eds.),
Elsevier Science Ltd. (publ.), 2, 1373-1380.
Preumont, A. (2002), Vibration Control of Active Structures, Kluwer Academic
Publishers, Dordrecht.
Raja, S., Prathap, G., and Sinha, P. K. (2002), "Active vibration control of
composite sandwich beams with piezoelectric extension-bending and shear
actuators", Journal of Smart Materials and Structures, 11 (6), 63-71.
Ribakov, Y., and Reinhorn, A. M. (2003), "Design of Amplified Structural
Damping Using Optimal Considerations", Journal of Structural Engineering,
129(10), 14221427.
Sadri, A. M., Wright, J. R., and Wynne, R. J. (2002), "LQG control design for
panel flutter suppression using piezoelectric actuators", Journal of Smart
Materials and Structures, 11(6), 834-839.
Ying, Z. G., Ni, Y. Q., and Ko, J. M. (2002), "Non-clipping optimal control of ran-
domly excited nonlinear systems using semi-active ER/MR dampers",
Proceedings of SPIE, Smart Structures and Materials 2002: Smart Systems for
Bridges, Structures, and Highways, 4696, 209-218.
Zhang, J., and Roschke, P. N. (1999), "Active Control of a Tall Structure Excited
by Wind", Journal of Wind Engineering and Industrial Aerodynamics, 83(1-3),
209223.

www.samco.org Page 205 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

2.2.3.5 Notations
Symbol Description Unit
A, B, C, D, G, H matrices of SSR -
C damping matrix -
I, 0 identity matrix, matrix of zeros -
J performance index / cost functional -
K stiffness matrix, regulator gain matrix -
L observer gain matrix -
L length m
M mass matrix, diskrete-time observer gain matrix -
Q state weight matrix -

6 AMCO

R input weight factor -


V intensity matrices of measurement noise -
W intensity matrices of process noise -
Ф transformation matrix -
Л distribution vector of control forces -
0 distribution vector of disturbance forces -
c viscosity kg/s
e error vector -
f force; function N; -
k discrete-time variable; stiffness -; kg/s2
m mass kg
mm number of modes -
nn number of nodal points -
q only time-dependent vector of modes -
r Cartesian coordinate in the direction of the beam m
t time s
u controlled input variable of the plant N
var variance -
v vector of measurement noise -
w vector of disturbance -

www.samco.org Page 206 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

x vector of state variable -


x vector of estimated state variable -
y vector of plant output variable -
z displacementm
p mode shape -
g0 fundamental circular frequency of undamped structure
rad/s
Z damping ratio -
Superscripts Description
T transposed
Subscripts Description
c controller
dt discrete-time

6 АМСО

est estimated
fe finite element model
m mode, modal coordinates
mea measured
mm number of modes
n nodal point
nn number of nodal points
red reduced
sa semi-active
u controlled input variable of the plant
2.2.4 Other control algorithms
This chapter will give a brief introduction in fuzzy control and neural network
control representing linear control approaches that are used fairly often in the
field of structural control. Sometimes, they are used in combination with
control approaches described in the previous chapters.
2.2.4.1 Fuzzy control

www.samco.org Page 207 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fuzzy control represents an alternative control approach to the classical


control theory which relies on the mathematical formulation of the plant (Liu et
al. (2002)). The classical control strategies have been widely used for systems
that may be approximated by linear, time-invariant models. However, these
control strategies are rarely implemented for non-linear, time-variant, complex
systems including noisy measurements. Here, fuzzy control may have its
advantages.
The main idea of fuzzy control is to build a model of an expert operator who is
capable of controlling the plant without thinking in mathematical terms. The
person controls the plant by a set of linguistically expressed rules that result
from the a priori know-how of the physics of the plant. Some characteristics of
fuzzy logic may be listed here:
Fuzzy logic is conceptually easy to understand.
Fuzzy logic is tolerant of imprecise data.
Fuzzy logic can model nonlinear functions of arbitrary complexity.
Fuzzy logic is able to match any set of input-output data. This process is
made particularly easy by adaptive techniques.
Fuzzy logic can be built on top of the experience of experts.
In direct contrast to neural networks, which take training data and generate
opaque, impenetrable models, fuzzy logic relies on the experience of people
who already understand the plant. When fuzzy control is called adaptive fuzzy
control, then neural networks are used for the update of the implemented
rules.

www.samco.org Page 208 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fuzzy logic can be blended with conventional control techniques and it does
not necessarily replace conventional control methods. In many cases, fuzzy
control systems augment them and simplify their implementation.
Fuzzy logic is based on natural language.

www.samco.org Page 209 of 155


SAMCO Final
Report 2006
F05 Guidelines
for Structural
Control

Struct Sens
ure ors
Fuzzy Controller
Fuzzy Inference
Fuzzifier
u

w
L,

Fig. 121: Inverted pendulum on a


cart.
x
y
Actuator
— Defuzzifier
Fuzzy Rule Base Membership
Functions
Fig. 122: Structure of a fuzzy
controller.

www.samco.or
g Page
210 of 155
SAMCO Final Report 2006
F05 Guidelines for Structural Control

Developing a fuzzy controller for the stabilization of the pendulum in the upper
position using a motor-driven cart (Fig. 121) shall demonstrate how fuzzy control
may work. The linearized equations for small angles and small angular velocities
become

(1), (2)
m,
1 La = (m1 + m2 )ga- f, m1 x = f - m2ga

The fuzzy controller (Fig. 122) may be designed based on the measured angle
and angular velocity

(3)
У
= [a,a ]T

Assuming a person should stabilize the inverted pendulum, the rules for pushing
and pulling the cart may be similar to the fuzzy rules as following:
If the angle is "slightly to the left" (SL) and the angular velocity is "moderate to
the right", then do not move the cart, i.e., the force shall be zero (Z).
If the angle is "moderately to the left" (ML) and the angular velocity is "zero" (Z),
then move the cart moderately to the left, i.e., the force shall accelerate the cart
moderately to the left (ML).
All rules for balancing the pendulum in the upper position are listed in Table 7.
Each rule takes the two input variables, measured angle and measured angular
velocity, and produces the output variable force, which drives the cart. The rules
are IF-THEN rules, connected by logical operators such as AND, OR, and NOT.
The resulting five membership functions for the example described here are:

www.samco.org Page 211 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

"moderately to the left" (ML), "slightly to the left" (SL), "zero force" (Z), "slightly
to the right" (SR), and "moderately to the right" (MR).
Table 7: Possible fuzzy rules for inverted pendulum.
force a =ml a =SL a =Z a =SR a =mr
a =ML ML ML ML ML SL
a =SL ML ML SL Z Z
a =Z SL SL Z SR SR
a =SR Z Z SR MR MR
a =mr SR MR MR MR MR

www.samco.org Page 212 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Fig. 123: The steps of fuzzification, inference, and defuzzification.

The fuzzy controller may be divided into the following three steps:
Fuzzification: All measured variables are transformed to the degrees of mem-
bership ц (Fig. 123 left). For this step, the range of each measured variable
must be divided into sections.
Inference: According to the actual fuzzy variables, the corresponding rules are
applied using the fuzzy operations for the logic operators, such as AND, OR,
and NOT (Table 8). The implication method limits the membership function
according to the degree of truth (minimum method, Fig. 123 middle). The result
of the inference step is the aggregated membership function of the output
variable (maximum method), in this case, the cart force.
Defuzzification: The most common defuzzification method is the centroid cal-
culation, which returns the centre of area under the aggregated curve (Fig. 123
right). The position of the centre defines the value of the output variable, in this
case the value of the force applied to the cart.
Table 8: Fuzzy logic operations.

www.samco.org Page 213 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

AND OR NOT
min max additive
complement

2.2.4.2 Neural network control


Generally, neural networks are models of static and dynamic structures that are
not based on physical relations but consist on mathematical functions, whose
coefficients result from "learning" or "training" using measurement data (Fig.
124). Besides the varied coefficients, also the functions itself and the structure
of the neural network
'-^^Тгчцг «
О АМСО
may be varied. Once the structure and the functions are chosen, the neural
network is trained by adjusting the coefficients using measured input data as
long as the error between model output (output of the neural network) and the
target data (measured plant output) is smaller than the desired error (Fig. 124).
target data
(measured plant output)

Fig. 124: Black box identification by training of neural networks.


adjusting weights until error < error,
max

Neural networks consist of neurons working in parallel and of several layers


working in series (Fig. 125). The coefficients of the neurons are the weight
factor w and the bias factor b , which scale the input i. The result n passes the
transfer function f . Typical transfer functions (TF) may be:
Hard limit TF, symmetric hard limit TF.
Log sigmoid TF.
Positive linear TF, linear TF.
Saturation TF, symmetric saturation TF.

www.samco.org Page 214 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

Hyperbolic tangent sigmoid TF.


Triangular basis TF.
Several neurons together build one layer (Fig. 125). The number of neurons of
each layer depends on the design of the neural network. One layer includes the
same type of transfer function which must be assumed by the neural network
designer. The choice of the type of transfer functions depends on the a priori
know-how of the designer about the plant behaviour. Several layers together
build one neural network.
Due to the black-box characteristics of neural networks, they may be the
preferable modelling approach for non-linear, complex systems where
physically based models lack. Within the field of structural control, neural
networks are often used in order to replace a state estimator since state
estimators often are based on rather simple finite element modelling
approaches and therefore modelling errors may be too large (Xu et al. (2003a),
Xu and Shen (2003b)). Then, using vibration measurement data of the target
civil structure, a neural network may be trained in order to have a state
estimator of high accuracy available for structural control.

www.samco.org Page 215 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control
OAMCO

input first layer second layer kth layer output

Fig. 125: Neural network consisting of neurons (weights and transfer function)
and layers.

2.2.4.3 References
Liu, Y., Gordaninejad, F., Evrensel, C. A., Hitchcock, G., and Wang, X. (2002),
"Variable Structure System Based Fuzzy Logic Control of Bridge Vibration
Using FailSafe Magneto-Rheological Fluid Dampers", Proceedings of SPIE, S.-
C. Liu and D. J. Pines (eds.), 4696, 219-227.
Xu, Z.-D., Shen, Y.-P., and Guo, Y.-Q. (2003a), "Semi-active control of
structures incorporated with magnetorheological dampers using neural
networks", Journal of Smart Materials and Structures, 12(1), 80-87.
Xu, Z.-D., and Shen, Y.-P. (2003b), "Intelligent Bi-state Control of the Structure
with Magnetorheological Dampers", Journal of Intelligent Material Systems and
Structures, 14(1), 35-42.
2.2.4.4 Notations
Symbol Description Unit
L Length m
b bias factor -
c viscosity kg/s
f force; transfer function N; -
g gravitational acceleration m/s2
i layer input variable, layer output variable -
k stiffness kg/s2
Iw weight factor -
m mass kg

www.samco.org Page 216 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

n neuron output variable -o layer output variable -


u controlled input variable of the plant -
w disturbance N
y vector of plant output variables -
x displacementm
a angle rad
/ degree of membership -
Superscripts Description
T transposed
k number of layers
Subscripts Description
R number of input variables of the first layer
Sj number of neurons of layer j

www.samco.org Page 217 of 155


SAMCO Final Report 2006
F05 Guidelines for Structural Control

3 VIBRATION ISOLATION
The first subchapter presents passive isolators, whereas the second subchapter
describes the concept of active vibration isolation.
3.1 Passive vibration isolation
3.1.1 Theoretical background
The goal of vibration isolation is to mitigate vibrations by reducing the
mechanical interaction between the vibration source and the structure,
equipment etc. to be protected. The general approach is to insert secondary
mechanical components (anti- vibration mounts) between the vibration source
and the receiver with the objective to avoid any state of significant excitation in
the system. The concept of vibration isolation is illustrated in Fig. 126. A single
degree of freedom system whose mass represents a structure, an equipment
etc. is connected to a base support through an isolator. Generally, in vibration
isolation, two cases are considered. In the first, the isolator protects the base
support from the effects of dynamic force FS (t) acting directly on
the supported structure. The goal of vibration isolation is to reduce the force FT
(t) transmitted to the support. In the second case, the isolator protects the
structure from the motion us (t) of the base support. The goal is to reduce the
absolute or relative
displacement uT (t) respectively uT (t) — us (t) of the structure. The
performance of
an isolator is usually characterized by the transmissibility. It is defined by the
ratio of the magnitude of the item (displacement, velocity, force etc.) transmitted
by the isolator with respect to the magnitude of the exciting item.
A mechanical isolator is generally implemented as a system consisting of
masses, springs and dampers. A simple model of an isolator consists of a linear
spring and dashpot acting in parallel. For this isolator model, the transmissibility
of the force tf for harmonic force f(t) acting on the structure is given by

(1)

www.samco.org Page 218 of 155


SAMCO Final
Report 2006
F05 Guidelines for
Structural Control

Tf = Tu
1
Transmitted

Fig. 126:
Schematic
representation of
vibration isolation
systems. Left:
Isolation reducing
the effects of the
excitation force
acting on the
structure. Right:
Isolation reducing
the effects of a
forced excitation of
the support.
1 + (2Ј)'
(1 - 5 2)2 + (2Ј)'

www.samco.org
Page 219 of
155
SAMCO Final
Report 2006
F05 Guidelines for
Structural Control

Fig. 127: Force


l\ :: : :: - transmissibility for an
isolator consisting of a
linear spring and dashpot
acting in parallel for
different damping ratios.
— Zs = 0 ::::-- : :: - ^^v'*: " The absolute displacement
"i— Zs = transmissibility Tu for a har-
*
5% monic forced displacement
:.— Zs = uS (t) at the support is de-
10% Zs = scribed by exactly the same
20% function as TF. The force
.10' transmissibility TF is illus-
' 10" trated in Fig. 127 for
10 different damping ratios.
Frequency/ ratio f /f