Discussion

Building Design Procedure

Paper by JOHN F. BAKOTA (4th Quarter 1977)

Discussion by A n a n d B. G o g a t e T h e discusser concurs with M r . Dakota's observation regarding the lack of systematic information on the design of mill buildings in text books. One of the main reasons for this situation is that Consulting Engineers, who have the expertise in this field, do not have the time to write textbooks. This discussion is aimed to add a few remarks based on the discusser's experience of about two decades of Consulting Engineering. A " m u s t " in the list of references on the design of mill buildings is the pioneering book written by the late Milo S. Ketchum^ in 1903. This book was the very bible of steel building design about sixty years ago. During the discusser's employment with one of the major steel fabricating firms in 1966, this old book proved to be very valuable; it provided more information on the design of mill buildings than any modern text book then available. A reasonable argument is given in Ketchum's book as to why a mill building constituting a single span roof truss with knee braces, and supported on hinged base columns, can be treated as a frame for lateral load resistance. T h i s method has been used ever since by designers with no regrets. Even a recent paper by Zweig^ has recommended full frame action in design of buildings with bottom chord extension of roof trusses into supporting columns. SPACING OF EXPANSION JOINTS AND THERMAL STRESSES T h e effect of thermal stresses becomes pronounced when thermal expansion is restrained. Expansion joints, when detailed and spaced adequately, minimize thermal expansion restraints. Consequently, there is less need to account for the negligible thermal stresses. Indeed this is very Anand B. Gogate is Senior Structural Engineer, Elgar Brown Consulting Engineers, Worthington, Ohio.

fortunate, since an accurate stress distribution in any three-dimensional enclosure subjected to temperature gradients is extremely complex. From the design standpoint, therefore, the discusser would like to recommend the upper limit on the spacing of expansion joints as 200 ft. With this limitation, the thermal stresses can be safely ignored. On the other hand, expansion joints are expensive and are a potential source of roof leakage, which must be considered. Incidentally, the location of true expansion joints also affects the design of wind bracing, and the discusser has found on many projects that this fact was not fully realized in the location and proportioning of the wall bracing and diaphragm bracing in the roof system. SIDESWAY AND COLUMN STABILITY In regard to effective length of framed columns and the benefits derived by rotating the axes of adjacent columns the discusser would like to refer to the papers by Yura^ and Zweig^ and the discussion of Yura's paper by Gogate."^ These papers conclude that for a safe lower bound the total buckling load of a rigid frame with an infinitely stiff horizontal member (a roof truss with bottom chord extensions very nearly fulfills this requirement)^ equals approximately the sum of the buckling loads of each frame column. In other words, two columns tied at the top with bottom chord extensions of the roof truss would act together in such a way that an overload of one column (less than the Euler Load) can be justified on the basis of a smaller load on the other column. Yura's^ inelastic G concept and an iterative procedure to obtain a modified AT-factor for such columns is also a major contribution towards economical design of column grids where loads (like crane loads) create an excess load at one or two adjacent columns at a given time only. Finally the discusser hopes that other Consultants will share their knowledge through papers such as M r . Bakota's in the coming years. REFERENCES 1. Ketchum, Milo S. General Specification for Steel Frame Mill Buildings 1903, {Retitled ''Steel Mill Buildings'' in 1932 and published by McGraw-Hill Co.)

75
SECOND QUARTER/1978

2. Zweigy Alfred Discussion of "Column Stability Under Elastic Supports" by T. R. Higgins AISC Engineering Journal, Vol. 2, No. 3 July 1965, pp. 105-106. 3. Yura, Joseph A. The Effective Length of Columns in Unbraced Frames AISC Engineering Journal, Vol. 8, No. 2, April 1971, pp. 37-42. 4. Gogate, Anand B. Discussion of "The Effective Length of Columns in Unbraced Frames" by Yura AISC Engineering Journal, Vol. 8, No. 3, July 1971, pp. 110-111.

76 ENGINEERING JOURNAL/AMERICAN INSTITUTE OF STEEL CONSTRUCTION

Discussion

Building Design Procedure

Paper presented by J O H N F. BAKOTA (4th Quarter 1977 issue)

Discussion by C. F. Diefenderfer Mr. Bakota has included in his references many of the universally accepted articles on this subject. However, the Association of Iron and Steel Engineers publication Specifications for the Design & Construction of Mill Buildings is not mentioned in the article or references. A discussion of Mr. Dakota's article follow^s, with emphasis that the AISE specification will provide the guidance he was seeking. Mr. Bakota is quite correct when he states, "The design of an industrial mill building involves myriad assumptions and parameters." The Association of Iron and Steel Engineers, recognizing the need to consolidate available information and to guide designers, contractors, and suppliers to the building requirements of the steel industry, established a subcommittee for the design and construction of mill buildings in the mid-sixties, under the chairmanship of the late John J. Murray. AISE Standard No. 13, Specifications for the Design and Construction of Mill Buildings, was issued by the Association in 1969. As stated in Sect. 1.1, ''This specification provides owners, engineers and contractors with a comprehensive and rational approach to the design and construction of mill buildings and other buildings or structures having related or similar usage." This specification guides the owner and designer through the questions and problems presented by Mr. Bakota. It suggests minimum floor loads and offers wind load factors for determining average normal pressures on walls and roofs, in accordance with ANSI and ASCE wind studies. It lists load combinations for the design of crane runways and supporting structures, including the foundations. C. F. Diefenderfer is Consulting Engineer, Technical Services, Bethlehem Steel Corp., Bethlehem, Pa., and Chairman, AISE Subcommittee on Design & Construction of Mill Buildings.

The specification requires a continuous bracing system in the plane of the bottom chords of the roof trusses. Transverse shears are to be distributed to the columns in proportion to their stiffness in that direction. Guidance is given for the combined stress evaluation in crane girders from simultaneous vertical and lateral loading. It indicates when fatigue provisions should be applied. Information is given for soil investigation, earthwork, and excavation requirements. General criteria and procedures for the design of mill building foundation components are included. Information on electrical grounding of the structural frame and general illumination levels for various operations is presented. In general, this specification provides the parameters and guidance for the design and construction of mill buildings that Mr. Bakota was seeking. It is an excellent reference for all engineers involved in industrial buildings. Mr. Bakota recommends independent action of the crane and building shafts for type (b) columns of his Fig. 1. There is limited data, both theoretical and experimental, that supports integral action of these shafts, provided there is a stiff connection between the top and bottom of the segments and adequate diaphragms between them to resist the shear flow (VQ/I). Using the integral assumption, more economical columns are achieved, especially where the transverse forces are larger than those in the example. The AISE standard recognizes such an analysis. With this assumption, one must include the eccentric load of the crane girder reaction about the strong axis of the combined shaft. This adds to the moment forces being analyzed. In the design of the crane girder, the moment from lateral force should be 21.4 kip-ft, using the AISC recommended value for lateral force on crane runways. This lateral force is one of three criteria for side force given in the AISE standard. Carrying that correction through, a designer might accept a W24xl20 as being adequate, even though the sum of the combined stresses will exceed 24 ksi by about 2%. However, this section, selected utilizing Gaylord's liberal assumption, would not be in conformance with the AISE standard. In heavy duty operation, it is desirable to have a relatively stiff horizontal resistance at the crane

143
FOURTH QUARTER / 1978

girder level to maintain gage of the crane rails and minimize wear on the rail and wheels. Thus, the designer familiar with heavy duty service requirements will select a section that satisfies the more general requirement oiftx/Fbx + fby/Fby ^ 1Some mention should be made in the example concerning the fatigue life of the crane girder and its support. In this particular case, the indicated stress range of the W24xl20 plain beam without any attachments should provide a life

of 2 X 10^ cycles of loading. This is generally considered adequate. However, a caution against welding any attachments to the girder in the tension region should be made in order to maintain the fatigue life of the chosen section in this high stress range. Mr. Bakota's conclusions are the best advice to all designers. The designer must determine if the assumptions of any analysis are valid for his particular structure before he utilizes them.

144
ENGINEERING JOURNAL / AMERICAN INSTITUTE OF STEEL CONSTRUCTION