You are on page 1of 162




{; 1*
.l :.-:,

*b&- .,.l ql.

- a

h. S.;.

Edwin IvL. Dyer, III


During the final years of World War 2, Japan

was desperate for technical advances to
combat the high-altitude B-29 bombers and
planned invasion of the homeland. As cities
were flattened and antiquated aircraft such
as the infamous Zero fighter bravely fought on
only to be shot out of the skies by superior
Allied aircraft, the Japanese military, like their
German counterparts, sought innovative
designs as time and resources drew to a
Japanese Secret Projecls looks in detail at
the multiplicity of aeronautical designs and
concepts that the Japanese developed to try
to counter the Allied onslaught. The book is
divided into two sections dedicated to the air
forces of the IJA and IJN with over 42 aircraft
examined in their historical and political
context, each with its variants, performance
and combat records laid out in an easy to
read fashion.
Beautifully illustrated by the world's leading
aviation artists, this book is complemented by
stunning colour renditions of aircraft in
combat including profiles of genuine
markings and camouflage schemes. Notable
emphasis is placed upon the transonic
rammer aircraft, strategic long-range
bombers, high-altitude fighters, Kamikaze
aircraft and the methods in which the
Japanese military adapted German
technology to their particular needs such as
the Messerschmitt 262 je| fighter and 163
Komet rocket interceptor. Secret Japanese
armaments are also covered in detail with
information on guided missiles, aerial rockets
and unique bombs.
With an incredible range of interesting and
bizarre aircraft designs and access to
previously unpublished information,
Japanese Secret Projects is a gripping and
educational read for aviation and military
enthusiasts around the world.

Front couer illustration: "Slice & Dice" - Having

slashed their way through the ramjet-equipped
P-5 I D Mustang escorts, a pair of Japanese Manshu
Ki-98 fighter dive on a formation of B-35 flying wing
bombers, high over Japan. Although originally
conceived for the ground attack role, the Ki-98 was
one of many fighters developed towards war's end
with the purpose to intercept high-flying allied heary
bombers. The Ki-98 in the cover illustration carries
the markings of the 4th Sentai's 3rd Chutai. During
the war the 4th Sentai achieved successes against
B-29 bombers, while fllng their twin-engined Ki-45
"Nick" heavy fighters. They utilised both conventional
and ramming techniques to bring down their victims.

This specially commissioned artuork was created by

Ronnie Olsthoorn - wuw.skyraider3d,com


An imprint of
lan Allan Publishing
THE IJA AND IJN 1939.r-945

Edwin 11 Dyer
Japanese Secret Projects: Experimental Aircraft
of the IJA and IJN 1939-1945

First published 2009

Reprinted 2010

ISBN 978 1857803 174

All rights reserved. No part of this book may be

reproduced or transmitted in any form or by any
means, electronic or mechanical, including
photocopflng, recording or by any information
storage and retrieval system, without permission
from the Publisher in writing.

@ Edwin M. Dyer III

Midland Publishing is an imprint of Ian Allan

Publishing Ltd, Hersham, Suney KTl2 4RG.
Printed by Ian Allan Printing Ltd,
Hersham, Suney KTl2 4RG.

Visit the IanAllan Publishing website at

www.ianallanpublishing. com

Distributed in the United States of America and

Canada by BookMasters Distribution Services Inc.

Illegal copying and selling of publications deprives
authors, publishers and booksellers of income,
without which there would be no investment
in new publications. Unauthorised versions of
publications are also likely to be inferior in quality
and contain incorrect information. You can help
by reporting copyright infringements and acts of
piracy to the Publisher or the UK Copyright

At ui*.d sources
w,ts.r! adm,sl{a{tlIa
F -L 6 rs. h;rrhdHld*i

Half-title: Profrle art of the ManshuKl-98. PeterAllen

Title page: ManshtrKi-98 in action. Ronnie Olsthoom
Preface .........6 Other Aircraft
Acknowledgments... ..........7 TheJapaneseRammer. ....... 123
AbbreviationsandGlossary ......8
TheKamikazeAirplane .......125
MitsubishiT.K.4Type0...... .........126
NakajimaAT27 ..... .........127
KawasakiKi-64.. .......9 S-31 Kurowashi ..... .........129
KawasakiKi-88.. ......12 T.K.19. .......131
KawasakiKi-91 .. ......14
KayabaKatsuodori .....17 Weapon Systems
KokusaiTa-Go. ........20 Japanese Missile and Guided Munitions
MaedaKu-6 ....22
Projects. 132
ManshtKi-g8.. ........24 Specialised Bombs 137
MitsubishiKi-73.. ......27 Specialised Bombs: Fu-Go. 139
NakajimaKi-87.. ......28
NakajimaKi-I15Tsurugi............... 31
NakajimaKi-201 Karyu .........34
RikugunKi-93... .......37 German Technical Exchange with Japan:
RikugunKi-202.. .......40 ABriefOverview.... .......142
RikugunKogikenseries . .......43 Bibliography. .........150
TachikawaKi-74.. .....50 ArtistBiographies ... ......... 151
TachikawaKi-g4-l . .....53 Ted Nomura Hlpothetical
TachikawaKi-g4-ll .....55 AircraftProfiles .....155
Tachikawa Ki-162
and other IJA jet projects . . . . . . . . . . . . 59 Index

Personalities .........158
Aircraft ....... 158
KawanishiBaika. ......61 Engines. ...... 160
KawanishiHllKSoku. .........63 WeaponSystems. .....160
KawanishiK-200. ......65
KugishOHighSpeedProiects...... .....67
KngishoMxY6. ........69
KugishdMXYTOka .....70
Kugisho MXYS/9 Akigusa/Shuka . . . . . . . .77
KugishoR2YKeiun .....78
KrlgishoTenga .........81
KyushuJTWShinden ..........84
MitsubishiGTMTaizan .........91
MitsubishiJ4MSenden .........93
MitsubishiJ8MSyusui ..........96
Mitsubishi-Payen Pa.4OO/Suzukaze 20 . . 101
MizunoShinryull...... .......104
NakajimaFugaku .....108
NakajimaJlNGekk6. .........112
NakajimaKitsuka .....114
If you browse any major book seller, you tend local hobby shop or major book retailers did Because of the constraints on the number
to see a good manyworks on the experimen- not turn up anything above and beyond what of pages, there had to be a process of select-
tal aircraft developed by Germany both I already knew. I found a rather large gap in ing aircraft for inclusion. The planes selected
before and, particularly, during World War 2. the online data pool on Japanese X-planes, at for this volume have been chosen based on
Also, you'd find a llne selection of books on least in English, and so I sought about cor- several factors. The first was the nature of the
the topic of American experimental planes. recting that. plane in terms of being a conventional or a
From time to time, you could find mention of In 1998, I began to assemble a website more radical design. Thus, while the Kugisho
such 'X-planes' of other nations amongst the inspired by Dan Johnson's Luft '46 which D3Y Mydjo may be a relatively obscure plane
text describing more well known aircraft. But started in 1997 as a one-stop site about Ger- of which only two were built, it was a very
you did not often see, if at all, books dedi- man X-planes. In 1999, my site, Hikoki: 1946, conventional aircraft in terms of design. The
cated to Japanese experimental aircraft. Usu- went live to the world. During its first few same applied to the Mitsubishi Ki-83. There-
ally, one had to visit specialty book dealers, years, Hikoki: 1946 expanded to encompass fore, these more conventional designs or pro-
hobbv shops, or be fortunate enough to be 31 Japanese experimental aircraft and sec- totypes received a lower selection priority
able to read another language in order to find tions on engine specifications, German air- over more advanced concepts. Another fac-
books on the subject of Japanese X-planes. craft the Japanese were interested in or tor concerned aircraft which were derivatives
I was first exposed to the world of Japanese bought, missiles and more. Support for the site of established production planes in the
experimental aircraft in 1988 through the was great. Such people as artist Ted Nomura, Japanese arsenal. As such, designs such as
classic book Japanese Aircraft of the Pacific Polish author Tadeusz Januszewski and J-Air- the Ki-l 16, which was derived from the Naka-
War by Ren6 J. Francillon. I found the book contributors Mike Goodwin, George jima Ki-84 Hayate, are also excluded. A third
on the shelf in your typical mall bookstore. Elephtheriou and D. Karacay helped the site by factor revolved around the pool of informa-
Sure, before then, I knew about the classic providing both artwork and data on some of tion available for a certain design. The more
Japanese planes such as the Mitsubishi A6M the planes presented. By 2002, I felt that I'd obscure or unknown the design was, the
Reisen and the Nakajima Ki-43 Hayabusa. But exhausted what there was on the subject and higher it was considered over other planes.
Francillon's book brought to me such planes the site entered a state of finality with no fur- For example, the scope of the Rikugun
as the Nakajima Kitsuka, Mitsubishi J8M ther updates having been done since. Kogiken designs were of far more interest
Syusui and the Tachikawa Ki-94. Fast forward to the fall of 2007.Jay Slater of and of a lesser known nature than the proto-
My interest in military technology sat on the Ian Allan Publishing e-mailed me to discuss types of the Nakajima G8N Renzan or the
kerb through my college years but afterwards the prospect of writing a book on Japanese Aichi S1A Denko of which more information
it I
slowly ramped back up. found that I experimental aircraft. This was not the first is readily available. Finally, X-planes that
focused my reading on the military machine time someone had approached me to do so. were purely research aircraft such as the
of Germany and the sheer breadth of techno- But unlike the others, Jay had a well knonm Kawasaki Ki-78, those experimental planes
logical innovation their engineers and scien- publisher behind him who had a number of constructed prior to the start of the war, and
tists churned out. Small arms, armour, X-plane books in print, many of which I had in most of the non-combat aircraft (transports,
artillery, missiles, submarines, aircraft, rock- my own library. It seemed natural to him that gliders and the like) were generally excluded
etry and much more - no stone was left a book on Japanese X-planes would be a wel- from contention. Perhaps in a future publica-
unturned by Germany's scientists. It was dur- come complement to their existing titles as tion, those designs that did not make the cut
ing my studies of German aviation that I well as providing the English aviation histo- for this book will get their chance.
would see the Japanese pop up from time to rian or enthusiast with a ready source of ded- It may appear that few aircraft remained
time. Most often, it was the acquisition of Ger- icated information on Japanese X-planes. I with such pruning but it still left a significant
man technology for development in Japan, or certainly agreed. number of planes to choose from, from the
German plane designs offered for the Japan- The work you hold is not simply my Hikoki: historically important Nakajima Kitsuka and
ese to purchase. This piqued my interest in 1946 website in book form. Yes, some of the Mitsubishi J8M Sy0sui, to more unknown
learning about what the Japanese had brew- aircraft in these pages can be found on the types such as the Krlgish6 Tenga and Kawan-
ing in their aviation cauldron. site but the information here has been further ishiK-200.
By this time, the World Wide Web was researched and revised. This means the data In so far as the book layout, aircraft are sep-
becoming the engine of information that it is in these pages is far more up to date than the arated by service (lJA and IJN) and then
today. While I was able to find bits of infor- site. And for sure, the outstanding artwork alphabetised by manufacturer. Those aircraft
mation regarding Japanese X-planes, it was provided makes this a spectacular publica- that were not of either service (or were joint
never anything substantive. Stops into the tion and investment for your library. projects) are listed last. Missiles and a selec-

tion of some of the more interesting aircraft can be viewed as another means by which accuracy in the information provided in this
munitions that were deployed or were in Japanese aviation technology was boosted book. Even as the writing of the book was
development are included along with a brief through a wartime ally, but more often than underway, I was acquiring additional sources
discourse on German technical exchange not it was an expediency to rapidly increase and checking and rechecking data to make
with Japan before and during World War 2. A the capability of Japanese aircraft in the face sure nothing was amiss. Of course, at some
feature in each aircraft chapter is the inclu- of aworsening war situation and ever improv- point I had to 'let it go'. If I held on to chapters
sion of a 'Contemporaries' section. The pur- ing Allied fighters and bombers. It is hoped waiting on the next titbit of information to
pose of this is to illustrate to the reader that the information in this section will be a cata- appear, the book would never get finished
designs didn't occur in a vacuum and similar lyst to learn more about the aircraft presented and you wouldn't be holding it in your hands.
concepts could be found in other Allied to expand one's knor,vledge of aviation by Thus, invariably, there is the risk of omitting
nations as well as Axis ones. This section other nations. Also keep in mind that this sec- something, interpreting a translation or
should not in any way be construed as point- tion does not list each and every plane that source incorrectly, or iust plain making an
ing to the Japanese as simply copying the could be considered a contemporary. error. To that end, corrections, new informa-
work of other nations. While it is true that the Instead, I have picked the more interesting tion and any and all comments can be
Japanese air forces prior to the war were very and have intentionally listed only the aircraft directed to the author at the e-mail address
keen on obtaining as much information on name(s) in order to not take away from the below.
aviation technology as possible (and, in some main topic at hand. The reader will also find I hope you, the reader, enloy the book and
cases, built and flew versions of foreign air- with certain aircraft a section called 'Sur- find it a worthwhile addition to your library as
craft), once hostilities began Japan knew she vivors'. Listed here are those aircraft that sur- a ready resource on some of the most inter-
could no longer rely on outside assistance for vived the war and what their fates were, esting Japanese airplanes of the war.
their aircraft industry and ensured it could either being scrapped or escaping the cutting
stand on its own. This it did, producing many torch. Where known, the Hepburn Romani-
successful aircraft that were indigenous. The sation system is utilised for Japanese words. Edwin M. Dyer III
influx of German technology during the war Every attempt has been made to ensure

This book would not be what it is without the ion. To Tim Hortman who graciously pro- Tadeusz Januszewski, James Long, Robert C.
assistance and support of a good number of vided his photographs of the Kyushu J7W1 Mikesh, Nicholas Millman, Ronnie Olsthoorn,
individuals and I would like to recognise Shinden, Nakajima Ki-l 15 Ko Tsurugi and the Masafumi Sawa and Akio Takahashi. As some
them here. Nakajima Kitsuka currently in store at the of the works used in researching the book
First and foremost, Jay Slater and his team National Air and Space Museum and which were in Japanese, Ryuki Arceno, Nanae
at lan Allan Publishing. It was Jay who are no longer available to the public. Konno, Lara Law and Tekla Munobe pro-
reached out to me and first proposed this During the information gathering phase of vided translations. For those works in Polish,
book and through him it became a reality. the book, research support and material was Michal Sporzyriski was the key translator.
To the artists whose work you will see in provided by several individuals. Their assis- Last but certainly not least, my parents,
this publication: PeterAllen, Kelcey Faulkner, tance helped to confirm or deny data, pro- Edwin and Margaret, for their support and
Muneo Hosaka, Gino Marcomini, Ted vided a sounding board for theories, offered encouragement. Also to Gail Lashley for
Nomura, Ronnie Olsthoorn and Daniel Uhr. comments on the information or brought always making sure I had my nose to the
Through their hard work, the aircraft within new information to the table. These folks are grindstone.
these pages come to life in spectacular fash- Shorzoe Abe, David Aiken, Paul Deweer,
Abbreviations and Glossary
AOAMC Atlantic Overseas Air X{aterial Centre in Nelvatk, Doi-lchi Kaigun Kaku Gijutsu-sho I st Naval Air Technical ' Rihugun Kohugijutsu Kenkyujo Air Technical Research
Nerv Jersey. Arsenal housed at Yokosuka, Japan. This eroup Institute. This unit conducted research, design
FE Foreign Equipment. FE numbers were used by rvas responsible for aircraft design fot the IJN. and development for the IJA. The facilities were
the USMF to mark captured aircraft for tracking Dai-lchi KaigunKoftusho I st Naval Air Arsenal located at located at Tachikawa.
and documentation purposes. Kasumigaura. Shi Beginning in 1931, when the IJN issued a specification
Ha Abbreviation for the Japanese rvord for engine, Dai-Juichi Kaigun Ko,ftusho This was the I 1th Naval Air for an aircraft, a shi number was assigned to it,
hatsudohi. For example, the Nakajima IHa-351 Arsenal located at the Hiro Naral Arsenal in based on the year of the Showa Japanese
25 l4-cvlinder radial engine. This was used as a Kure, Hiroshima,',vhere seaplanes, flying boats calendar, For example, the \'litsubishi G7M
designator for all engines used by the IJA while and engines for the IJN were produced. Taizan was designed to meet the l6-5hi Attack
the IJN used their own system. Eventually, the Hei Literally'grade C'. Hei is most often used in the Bomber specification. 16 stood for Showa 16 or
two systems were combined. In the example book to describe the third specification, such as 1 941.
above, the '3'was the engine type (14-cylinder, l8+hi Hei. Shimpu The rvord is an alternate reading of the
double+orv radial), the'5' the bore/stloke Hihohi Aeroplane in Japanese. Japanese franl foi'divine wind' and is used in
(130mm/150mm) while the '25' rvas the model Jidosha Japaneseforcar/automobile. this text to describe those aircraft for use by and
number, taken from the IJN system. Koi From time to time, Japanese aircraft i'vould for suicide unit missions. Shimprl Tokubetsu
IJA Imperial Japanese Army (Dai-Nippon Teikoku have Kai appended to their designation rvhen KOgekitai is the IJN designation for suicide
Rikugun). In this book, IJA refers to the Imperial the design r,as heavily modified but not to a units, meaning Divine Wind SpecialAttack
Japanese Army Air Sen'ice (Teikoku Rikugun point it required a nerv designation. Kai meant Force. The IJA would use Shimbu Tokubetsu
Kokltai). 'improled'. An example $ould be the Nakajima Kogekitai, shimbu meaning 'band of heroic
IJN Imperial Japanese Naq, (Dai-Nippon Teikoku Ki-84JI Hayate Kai. ltarriors' or'brandishing
a slvord'. Kamikaze is

Kaigun). In this book, IJN refers to the Imperial Kaigun Koku Hombu Imperial Japanese Nary Aviation not used by the Japanese as a description for
Japanese Na'"y Air Service (Dai-Nippon Teikoku Bureau of the l\'linistry of the Naty of Japan, suicide units but the word is popular in the
Kaiqun Koku Hombu). Among a number of duties of this bureau, the West.
K.K. Stands for Kabushiki Kaisha, which translates pertinent one for this book ivas the Technical Tochihaua Dai-lchi Rihugun Kohusho First Army Air
as Company, Limited. For example, Fuji Kokuki Depaltment that oversarv the design of nerv Arsenal. IJA air depot that repaired, modified
K,K. would be FujiAirctaft Co Ltd. aircraft and equipment. and distributed aircraft to IJA air units. It also
Ken Abbrel'iationforftensan,meaning'research', Ko Literally'grade A'. Ko is most often used in the produced the Ki-30, Ki-43 and Ki-51 in addition
Ki The IJA utilised Ki numbers for aircraft. Ki rvas book to describe the first model of a plane such to being one venue for experimental aircraft
the abbreviation for Arlar'which means as the Ki-l 15 Ko or the first specification, such development.
'airframe'. The numbet folloling the as l8-shi Ko, Tail Number For the purposes of this book, this refers to
abbreviation was the project number for the Koku Hombu Army Air Headquarters. This division of the the US Naly method of marking captured
aircraft under development regardless of the Imperial Japanese Army rvas responsible for aircraft. It is considered that the number
manufacturer or aircraft tlpe. For example, the research and development of aircraft, aircraft applied corresponded to shipping allocation
N'litsubishi Ki-73. engines and other aviation equipment. As such, numbers for the aircraft when they were
Ku Abbreviation forftakku, meaning'to glide'. specifications for aircraft nould be issued fiom prepared for transport to the US.
MAMA \'liddleton Air ['laterial Area at Olmstead Field in this division.
tr'liddleton, Pennsylvania. This rvas lvhere Kokuhi Aircraft in Japanese.
captured Japanese aircraft lvere shipped by the Klgisho Aircraft designed b1'the Dai-lchi Kaigun KOktl
t]SMF. Cijutsu-sho rvere given the KDgisho designation,
NA Not Available. In regards to specifications, NA a contraction of KOk[ Gijutsu-sho. One may see
means that at the time of the book printing, the KugishO aircraft called by the Yokosuka
jnformation on the spec rvas unknown and not designation but this rvould be incorrect.
available. Mitsubishi Jnhag'o Kabushihi Kctlsha The full name for
NAS Naval Air Station. N'litsubishi Healy Industries Co Ltd.
NASM National Air and Space N'luseum. Otsu Literallv'grade B'.Otsu is most often used in
Ne Abbreviation for A'ensfto Rocftello, meaning the book to describe the second model of a
'burning rocket' in Japanese. Ne r'vas used on plane sur h as lhe Ki-93 Otsu or lhe second
Japanese jet engine designations. For example, :pecificalion. such as l8-chi 0tsu.
the Ne 330. Parh Ridge This rvas the location of No. 803 Special Depot
TAIC Technical Air Intelligence Center located in in Park Ridge, Illinois. Captured aircraft selected
Washington DC at NAS Anacostia. This was for inclusion in the NASN'l museum rvere stored
rvhere data on enemy aircraft was collected here until the collection lvas culled during the
and maintained. Captured aircraftu'ere Korean War as the US Air Force needed the
evaluated at NAS Patuxent River, N,laryland. space.
USAAF United States Army Air Force. Poul. E. Garber Presenation, Restoratlon. and Storage
Idcili4 Aircraft that sun ived the culling at
Park Ridge rvere moved to this NASN'l facility
located in Suitland-Silvet Hill, I'laryland.

Imperial Japanese Army


Kawasaki Ki-64 Ki-60/Ki-61 Hein. The Ki-45 was proving to tures which were perceived as rather
be a horribly troublesome aircraft. Problems unorthodox in comparison to other more
Designers at times relish the freedom to let with the landing gear, concerns with drag, conventional types. Since Koi's aircraft was
theirvisions develop and flow from the drafting engine difficulties and more were proving purely his own and not created to meet any
board to the tarmac, ready to take to the skies. a thorn in Kawasaki's side. lt was Doi who sort of specification, Koi would be disap-
Takeo Doi, working for Kawasaki, was just stepped in to solve the issues plaguing the pointed when the IJA forbade Kawasaki giv-
such a designer. Despite the very real work initial Ki-45 design. In addition to working ing any further time to the design.
developing, testing and producing combat air- to fix the floundering Ki-45, by 1940, Doi Doi's disappointment then turned to joy
craft for the IJA, Doi had a concept that he, on was involved with the preliminary develop- when, in October 1940, the IJA authorised
his own, brought to the fore. Initially, the IJA ment of the Ki-60 heavy fighter and the Ki-61 Kawasaki to proceed with developing the
would not hear of the design but later, when Hein. Ki-64 to meet a requirement for a fighter
the Ki-64 took shape, their mind would change. Even with such responsibilities, Doi had a capable of a marimum speed of 700km/h at
In 1939, Doi was involved with two design of his own for a high-speed fighter. His 5,000m (435mph at 16,405ft) and a 5 minute
Kawasaki programs: the Ki-45 Toryu and the concept incorporated a number of novel fea- climb to that height.

Il,tpERral Jnpafrese ARN4y

To power the Ki-64, Doi would enlist the two coolant tanks holding 83.3 litres (22 gal- was then crated and shipped to the United
help of fellow engineers employed by the lons) and 117.4 litres (31 gallons) respec- States for study at Wright Field in Dayton,
Akashi engine plant. The decision was made tively. The smaller serviced the front engine, Ohio.
to use the Kawasaki Ha-201 ([Ha-72] 11) the larger the rear engine. Given the relatively long development time
engine. This was actually a combination of Doi tested the Ha-201 and the cooling sys- of the Ki-64, US intelligence had already
two Ha-40 engines, the Ha-40 being the tem to a considerable degree. In 1942, one of become aware of the design. Thus, expecting
licence-built version of the Daimler-Benz the Ki-61 Hein fighters was specially modified the Ki-64 would see service, the airplane was
DB601A. Each Ha-40 was a 12-cylinder, liq- to test the cooling method and its trial flights assigned the codename RoLr.
uid-cooled powerplant and the Ha-201 was began in October 1942 and ran through to the
formed through having one engine in front of end of 1943. Because ofthis thorough testing,
the cockpit and the second behind it. The rear the construction of the Ki-64was delayed and Kawasaki Ki-64 - data
engine drove the first of the contra-rotating so did not reach completion until December
propellers and it had a variable pitch' The 1943. Nevertheless, Doi was satisfied that the
Republic XP-69 (US), Heinkel P.1076 (Germanv), Caproni CA l83bis
front engine drove the second propeller, cooling system would grant an additional (ltal.v)

which was of a fixed pitch. Both propellers 41km/h (25mph) to the Ki-64 and battle dam-
were three-bladed. All told, the Ha-201 was age would not significantly reduce the effec- Tlpe Fighter

expected to produce 2,350hP. tiveness of it and as such the exhaustive Crew 0ne

The use of the Ha-201 was not the most testing was worth the effort.
novel feature of the Ki-64. Itwas the means of With the Ki-64 complete, flight testing com- Powerplant

menced in December 1943. Four test flights (lHal2l I l), 24-cvlinder. inverted'V. liquid'
cooling the engine that was notable' The One Kauasaki Ha-201

Ha-201 was to be cooled by a steam con- were successfully made without mishap. cooled engine developing 2,350hp at take'off, 2,200hp at 12,795it, driving

tlro, metal, 3-bladed contra-Iotatins pr0pellers

densing system. A tank of 15.4 gallons of However, on the fifth, the rear engine caught
water was fitted into each wing and the outer fire. The test pilot was able to land the aircraft
wings and flaps seled as the cooling surface, and the fire was quenched. Mechanics Span 13.48m 44.2:t
totalling 23.99m' (258.3ft'). The port wing ser- stripped the engine out of the Ki-64 and Length ll.00m 36.lft
viced the front engine while the starboard returned it to the Akashi plant for a full repair. Height 4.24m 13.9ft

wing provided coolant to the rear engine. As The Ki-64 itself was returned to the city of Gifu Wing area 28m' 30l,3ft'

the water coolant turned to steam, it was which, at the time of World War 2, was a lVing loading 3.471b/m' 37,31b/ft

pumped out into the wings where the steam major industrial centre. Pouer loading 2,17ksihp 4.8lbihp

would condense back into water which in Plans were made to improve the Ki-64 by
turn was pumped back into the engine. The replacing the 2,350hp Ha-201 '"vith an Weights
enhanced model that could generate up to Empt) 1,050kg
main benefit of this system was lower drag on t00kg 1,244Ib
Loaded 5, 1

the airframe as the need for air inlets was 2,800hp. The propellers '"vould be replaced
removed or minimised. Each laminar flow with two, constant-speed contra-rotating pro-
wing housed the fuel tanks and one 20mm pellers that were electrically operated. It was \'lar speed 690.3km/h 429mph

Ho-5 cannon while two more were fitted in envisioned that with these modifications, the at 5,000m at l6,40ift

the fuselage deck. One drawback to the Ki-64 Kai would be able to attain a top speed Range l.000km 62lmiles

wings being packed with the cooling system of 800km/h (497mph). Climb 5 min 30 sec to 5,000m (16,405ft)

apparatus was that it left little room for the Unfortunately for the Ki-64, repair work lan- Ceiling 12,000m 39,370ft

fuel tanks and consequently the Ki-64's oper- guished, especially in the face of more press- Fuel capacit,v 618 liires 136 gallons

ational range suffered. The majority of the ing needs for the Japanese arms industry and
fuel was carried in the fuselage, the tanks the production of more conventional aircraft'
Tn'o 2Omm Ho-5 cannons in the fuselage deck, one 20mm Ho-5 cannon
being placed in and around the space left As a result, both the Ki-64 and the Ki-64 Kai
in each ning
available by the Ha-201 engine. The main fuel were cancelled. After Japan's surrender, the
tank was situated in front of the cockpit, hold- Ha-201 was still at Akashi, its repairs incom- Deplolment
ing 306.7 litres (81 gallons). Under it was the plete. When US forces arrived in Gifu, the None. Onlv one Ki.6.l rras completed and flolvn before lhe end of lhe

oil reservoir, holding 79.5litres (21 gallons) of Ki-64 was found and technical teams gutted
oil for the engines. Behind the cockpit were the airplane of its cooling system. The system

l0 hpnnEss Stcner PRo;tcrs: ExpsRrN,reNrar- AIRcnant oF rHE IJA AND IJN 1939-1945

I[,tpsRrer- JnpanEsE Anuy l1

Kawasaki Ki-88


In August of 1942, the IJA saw a need for a air- nose. To get around this, Takeo placed the how best to accommodate a large calibre
craft that, in its primary role, would selve as 1,500hp Ha-140 liquid-cooled, turbo- (or cannon in a single engine aircraft.
an interceptor flying sorties to defend instal- super-) charged engine in the fuselage, In June of 1943, Takeo finalised his design
lations, airfields and other assets that were behind the cockpit. The three-bladed pro- for the Ki-88. Apart from the use of a 37mm
considered important and therefore subject peller was driven using an extension shaft cannon and the engine placement, the Ki-88
to enemy attack. Design specifications were that ran from the engine to a gearbox con- bore no further resemblance to the P-39. The
drafted by the IJA for the aircraft and it was nected to the propeller. In essence, Takeo Ki-88 had a deep fuselage to accommodate
Kawasaki who put forth what they felt was built the aircraft around the Ho-203. the Ha-l40 engine that was situated below
the answer: the Ki-88. The main advantage of placing the engine and to the rear of the cockpit. The air scoop
Prior to drafting their design specifications, in the fuselage was that it allowed the cannon for ihe Ha-140's radiator was mounted on the
the IJA had sifted through a number of ideas to fire through the propeller hub, producing a bottom of the fuselage, iust forward of the
before settling on a plane that had to have a more stable firing platform that resulted in wing roots. The radiator itself was positioned
hear,y armament to ensure it could inflict sig- improved accuracy. Another advantage was back from the scoop on the bottom interior of
nificant damage to enemy aircraft, especially that it allorved a skilful designer to make the the fuselage, almost directly underneath the
bombers, and also good handling character- nose more streamlined, enhancing speed pllot's seat. Jutting out on the left side of
istics to make it not only a stable gun plat- performance. the fuselage, iust above the trailing edge of
form, but also to avoid a steep learning cule A good number of references infer that the wing, was the scoop to provide air to the
for new pilots. Takeo's design rvas inspired by the Bell P-39 turbo- or super-charger of the Ha-140' The
To this end, Tsuchii Takeo, a designer for Airacobra. While there is no evidence that landing gear was conventional and the main
Kawasaki, began work on what would specifically states that Takeo simply copied gear retracted into the wings, while the tail
become the Ki-88. Takeo selected a 37mm the American fighter, the P-39 was in limited wheel was fixed. A fuel tank was provided in
cannon as the primaryweapon, supported by operational use by the time design work each wing, mounted behind the wheel wells'
two 20mm cannons. lt is probable that the began on the Ki-88 in 1942, notably seeing On either side and to the bottom of the
37mm Ho-203 cannon and two Ho-S 20mm action in the Battle of Guadalcanal. Thus, the Ho-203 cannon were the Ho-5 cannons.
cannons would serve as the armament fit Japanese were aware of the design. Whether With the final design comPlete, work
with all three weapons fitted in the nose' an example was ever captured for analysis is began on a full scale mock-up of the Ki-88 and
Given the size of the Ho-203 (which was a lit- unknown but certainly intelligence was avail- this was completed sometime in 1943' In
tle over 1.53m (5ft) in length, weighing 88.9kg able on the plane. Or, it may be that Takeo addition, work had already begun on con-
( 1 96 lb)), this presented a problem in squeez' arrived at the same conclusion as did H'M. struction of the fuselage and wings for a pro-
ing them, along with the engine, into the Poyer, designer of the P-39, when looking at totype and it was expected that by October

12 hpaNlesn Sscner PnotEcrs: ExpERIMeN-taL AtRcRant op rHE IJA aNo IJN 1939-1945
1943 the Ki-88 would be nearing completion.
The IJA, however, had other plans for the ...€*'.
Ki-88. After inspecting the mock-up and in
Tiyb irrt"
*)\ E
reviewing the projected performance data of z
the plane, it was seen that it offered no real
,:4 u
advantage over other designs then in opera- ra"e . ffi
tional use, notably the Kawasaki Ki-61 Hien. t t{i'W r-
Thus, Kawasaki was ordered to terminate all .'' ,h
"ib"'" ',1
work on the Ki-88. 'J
"" "ry i"T
Kawasaki Ki-88 - data
Bell P.39 Airacobra (l.S), Bell XFL-l Airabonita (trS). Bell P.63 King

Cobra INATO codename lrcd in Russian senice) (LS), Cudkov Gu-l


Tlpe lnterceptor

Crew 0nc

Onc Kau,asaki Ha-110, liquid-cooled. turbo, or supercharged engine

devcloping l.j00hp diiving a three.bladed propeller

Span 12.37m 40.6it
Length l0.l8m 33,4ft

Height 1.1{m r3.6ft

\\iing area 27.19m' 26911

Emptv 2,9.19k9 6,5031b

Loaded 3,899ks 8.598Ib

Performance (specifications arc estimations bl Kau,asaki)

\'lar speed at 19,6851t 600kmih 373mph
Range t,198km 715 niles
Climb 6 min 30 sec to 5.000m (16.{0.lft)

Ceiling I 1,000m 36,089ft

! :,;*.: ;\

fumament .i .,* *
Ss "*
One 37mm Ho.20ll cannon and tn'o 20mm Ho-5 cannons

None. The Ki-88 did nol progress past a mock.up and partlallv

completed prolot!pe.

IltptRra,r- .llpnrussE ARn'rv l3

Kawasaki Ki-91


1.s -r
222s .t

This story centres on the failure of a bomber DC-4E four-engine passenger aircraft was Nakajima submitted the Ki-68 version using
that inspired the development of another funded by five airlines and Douglas with either the Mitsubishi Ha-101 or Nakaiima
new type. The Nakajima Ki-68 and the United Airlines building and testing the one Ha-103 engines in place of the Nakaiima
Kawanishi Ki-85, both four-engine, long- prototype. While the DC-4E was impressive, NKTA Mamoru 11 units on the G5N1. Kawan-
range bomber designs, hinged on the success in terms of its operating costs it did not add ishi also submitted their Ki-85 which was to
of the IJN's Nakajima G5N Shinzan (Mountain up. The aircraft was complex and this use the Mitsubishi Ha-111M engines.
Recess). The G5N would prove to be a failure resulied in maintenance issues, which As it was, the GsNl proved to be a dismal
and in turn led to the termination of the Ki-68 increased the cost of using the plane. Support failure. The NKZA engines were problematic
and Ki-85 programs; therefore the IJAwas left for the DC-4E'"vas withdrawn and Douglas and underpowered and the aircraft was too
without a long-range bomber project. It was 'uvas asked to simplify the design. As a conse- hear'y and complex. These difficulties con-
Kawasaki who stepped in to fill the gap with quence, the DC-4 sarv operational use with tributed to the overall poor performance of the
their own design. the US Army as the Douglas C-54 Skymaster. G5N1. Despite the problems, three more G5N1
In 1938, the IJN was enamoured with the In earlv 1939, the sale of the DC-4E was aircraft were built followed by a further two air-
idea of a bomber that was capable of operat- completed and arrived in Japan to be craft that replaced the NKTA engines for four
ing up to 6,486km (4,030 miles) from its base. reassembled. By this time, the IJN had Mitsubishi Kasei 12 engines. The two addi-
In part, this was due to the initial desire to informed Naka.iima to be ready to study the tional aircraft were designated G5N2, but even
strike targets deep in Russia from Manchurian DC-4E to produce a suitable bomber devel- the Kasei 12 engines could not resuscitate the
bases. Later, when Japan went to war with opment from it. After having been flown sev- design and the problems remained. Due to its
the United States, a need to attack the US eral times, ihe DC-4E rvas then reported as complications, the G5N1 was never used as a
mainland was identified and it was recog- having'gone down in Tokyo Bay', but in real- bomber. Two G5N1 (using Kasei 12s) and two
nised that a two-engine design would not suf- ity had been handed over to Nakajima whose G5N2 aircraft were converted to transports and
fice - four engines would be required. On the engineers took it apart. Within a year, Naka- serued in this role until the end of the war. The
understanding that the Japanese aircraft jima had built the prototl,pe G5N1 which first Allies gave the G5N the codename lz.
industry had very little experience in building flew on 1 0 April 1941 . The G5N 1 used only the By May 1943, the cancellation of the G5N
such aircraft, the IJN used the Mitsui Trading landing gear layout, wing design and radial had also brought the demise of both the Ki-68
Company as a cover to acquire a Douglas engine fittings from the DC-4E coupled to a and the Ki-85 (of which Kawanishi had a
DC-4E four-engine airliner, ostensibly for use new fuselage, tail design and a bomb bay. mock-up constructed by November 1942),
by Japan Air Lines. The development of the The IJA planned to produce the G5N1 and leaving the IJA with no active four-engine

14 JapnNesn SEcRET PRo:Ecrs: ExprRrurural AtRcRnlr op rHE IJA a.Nl IJN 1939-1945

Kawasaki Ki-91 - data

Focke-Wulf T0 'Transozean' Projekt B [Germany), Focke.Wull Frv l9l C

(Germany), Focke-Wulf Fw300 (Germanl)

Type High-Alttude, Long-Range Bomber

Crew EEht

Four Milsubishi Ha-214 Ru, I 8-oiinder, supercharged radial engines
developing a mariimum of 2,500hp; each engine to drive a fouFbladed

Span 47.9m 157.4ft

Length 32,97m l08.2ft

Height 9.99m 32.8fr

Wing area 223.99m 2,41Lt lft:

Empiy 33,999k9 74,955 1b

Loaded 57,999k9 127,865Ib


Max speed ;80kn-/h 360mph

at 9,808m at 32,180ft

Range 9,000 to I 0,000km 5,592 to 6,2 1 3 miles

Climb 20 min 30 sec to 8,000m (26,246f1)

Ceiling 13.500rn 44,291ft

Tweh'e 20mm cannons, two cannons each in one offour turrets located
in the nose, underside of lhe fonvard fuselage and top and bottom 0l
the aft fuseiage respectively. A fifth tunet mounted the remaining four

cannons. Standard bomb load 4,000kg (8,816tb) (perhaps up to 8,000kg

(1 7,636 lb) for shorter ranged missions)

None. Prototlpe 60 per cent complele by the end ofthe war

bomber designs on the table. Kawasaki, see- bomber and engineer Jun Kitano would work ing B-29 Superfortress which was to be mass
ing the opportunity, immediately got to work with Doi to help turn the aircraft into reality. In produced in late 1943. Four MitsubishiHa-214
on designing a new bomber. The man behind June 1943, Doi and Kitano began their initial 18-cylinder radial engines were chosen to
the Ki-91 was Takeo Doi, an engineer research and by October, work on the first power the Ki-91. As the plane was expected
employed by Kawasaki. It was his goal to see design concept for the Ki-91 was underway. to operate at high-altitude, provisions were
the development of a successful four-engine The Ki-91 was slightly larger than the Boe- made to utilise superchargers with the

Il,IpEntaL JnpnrussE ARnrv l5

engines and the projected maximum speed surisation so as to avoid holding up develop- try. This would change by the close of 1944
was 580km/h (360mph). To provide for the ment and allow its flight characteristics to be when B-29s began to operate from the Mari-
anticipated 10,001km (6,214 mile) range, measured. Once the pressurised crew cabin ana Islands and by 1945 bombing raids were
each wing carried eight fuel tanks with a fur- for the Ki-91 was ready, subsequent aircraft far more frequent. In February 1945, a raid
ther two mounted in the fuselage above the were to have it installed. heavily damaged the factory in which the
bomb bay. For weapons, the Ki-91 was to In April 1944, a full-scale wooden mock-up Ki91 prototype was being constructed. The
carry a hea'uy armament of tr.t elve 20mm can- was completed and Kawasaki invited IJA offi- damage was extensive, ruining the tools and
nons. Five power-operated turrets were to be cials to come and review the Ki-91. Up until jigs. With the loss of equipment needed for
used; one in the nose, one on the underside this time, the project was a private venture by future production coupled with dwindling
of the forward fuselage, one above and below Kawasaki to which considerable company supplies of aluminium, the IJA decided that
the aft portion of the fuselage, and the last in resources has been allocated. lf the IJA did fighters to combat the marauding B-29s had
the tail. The bottom turrets were remotely not find the bomber to their liking, it would become a higher priority than bombers. Any
controlledwhile the remainder were have been a waste of time, effort and money. hope of utilising such bombers was at best
manned. The tail turret was to mount four Fortunately, the IJA saw potential in the Ki-91 slim. With the Ki-91 at 60 per cent comple-
cannons while the rest had two cannons and work continued. In May, the IJA tion, Kawasaki stopped further work on the
each. As far as bombs, a total payload of inspected the Ki-91 mock-up and immedi- bomber and the project was officially can-
4,000kg (8,8181b) was envisioned and the ately ordered production of the first proto- celled in February 1945.
Ki-91 was to have a tricycle landing gear with type. Kawasaki planned to construct the Had the Ki-91 achieved service, plans to
the nose gear using a single tyre and the main Ki-91 at a new plant in Miyakonojo in attack the US mainland were in place to oper-
landing gear using dual tyres. A semi- Miyazaki Prefecture. However, the IJAdid not ate the bomber from the Kurile lslands using
recessed tail wheel was also installed. want to wait for the construction of a new temporary bases, while another plan to strike
Another feature of the Ki-91 was to be the plant and directed Kawasaki to use their Hawaii was formulated using bases in the Mar-
use of a pressure cabin for the eight man established factory in Gifu Prefecture. By shall Islands. The second plan was rendered
crew. But the development of such a large June 1944, the construction of the prototype obsolete when the Japanese lost the Marshall
pressurised cabin for the Ki-91 was expected Ki-91 had begun at the Gifu factory, together Islands to the Allies in February 1944. As a note,
to take some time to implement, even using with the necessary tools and jigs to produce contemporary images sometimes show the
knowledge from another of Doi's designs, the further aircraft. Ki-91 as having a bomb bay battery of down-
Kawasaki Ki-108, a twin-engine high-altitude However, June would see the first B-29 ward firing cannons for a ground-attack role.
fighter fitted with a pressure cabin for the raids over Japan, but as the attacks were few While the Japanese were interested in such
pilot. Therefore, it was decided that the initial and far between, work on the Ki-91 continued concepts, there is no evidence that Kawasaki
Ki-91 prototype would be built without pres- despite the worsening situation for the coun- envisioned such a task for the Ki-91.


l6 JnpanesE SrcnEr Pno:Ecrs: ExpeRttteNrnt- AtRcRnrt op rHE IJA AND IJN 1939-1945
Kayaba Katsuodori


Kayaba envisioned that his design for a fast, designing such an engine. The main draw- The Katsuodori depicted here is shown in the
point defence interceptor would sweep back with a ramjet engine is that at subsonic colours and markings of the Tlst Sentai. It is
intercepting TolryGbound Nonhrop B-35 bombers
through the Allied bombers like the kat- speeds its performance is poor. Below
of the 44th Bomb Squadron, 40th Bomb Group
suodori bird hunts for fish. Impressed with the 6l2km/h (380mph), a ramjet suffers signifi- operating from Tinian.
prowess of the katsuodori, Kayaba named his cant loss in speed and becomes highly ineffi-
design after the bird. But as we will see, his cient in terms of fuel consumption. The been purchased from Germany. His interest
vision was to meet with a harsh reality. ramjet typically requires another power in aviation saw him produce four aircraft
The genesis of the Kayaba Katsuodori source to bring the aircraft up to the speed at designs: the Hino No. l, No. 2 and the No. 3
began as far back as 1937 with the Kayaba which the ramjet can operate efficiently. Typ- and No. 4 Kamikaze-go airplanes. However,
Ramjet Study Group, a collection of engineers ically, this speed is at least 966km/h each of these designs was a failure. Pressure
and scientists who sought to investigate ram- (600mph). Once the ramjet reaches that from his military superiors saw Hino give up
jet propulsion in Japan. The concept of the speed the engine is self-sufficient and, with- on aviation by 1912.
ramjet was actually patented in 1908 by out fuel injection moderation, would propel However, in 1937, Hino was inspired to cre-
French engineer Ren6 Lorin, but it was the the plane to speeds far in excess of the ate a tailless glider. The project was taken
Russian I. A. Merkulov who first built and design's ability to handle the high tempera- over by the Kayaba Seisakusho (Kayaba Man-
tested one, the GIRD-04 in 1933. A ramjet is a tures and Mach number. ufacturing Works) and then by Dr. Hidemasa
very basic engine with few moving parts. In The Kayaba Ramjet Study Croup saw the Kimura who worked for the Aeronautical
simple terms, it uses the high pressure air benefits of high speed with a relatively easy to Research Institute of the Tokyo Imperial Uni-
generated by the aircraft's forward motion manufacture engine. The group produced versity under Dr. Taichiro Ogawa. The result
and forces it through the inlet. The air is then two test models before the final product, the was the HK-l. The HK-l (standing for Hino
mixedwith combusted fuel-this heats the air Kayaba Model 1 ramjet, was realised. The Kumazo) was built by the lt6 Hikoki K.K. and
and is forced out of the rear of the engine, pro- Model I was projected to be able to offer was completed in February 1938. lt was
viding forward propulsion. Unlike pulsejets speeds of 900km/h (559mph). With the purely a research glider to test the tailless
(which were to be used on the Kawanishi engine complete, all thatwas neededwas the concept. Testing commenced in December
Baika, see Page 61), the fuel flow is continu- aircraft to fit it into. 1938 with ground towing at Kashima in
ous. Without getting into the specifics of a The airframe design began with Kumazo Ibaraki Prefecture and the first air released
ramiet, adjustments in the design of the inlet Hino. Hino was an officer in the IJA and had flights began in September 1939 at Tsu-
(to maximise the intake of air), combustor (to been the first Japanese to unofficially make a danuma in Chiba Prefecture. Because it
ensure effective operation during flight move- flight on 14 December 1910 when he acci- showed positive results, the IJA took an inter-
ments) and the outlet nozzle (to effect accel- dentally took to the air in a Hans Grade mono- est in the concept. The HK-l was purchased
eration increases) all come into play on plane while he was taxiing. This aircraft had by the Rikugun Kokugijutsu Kenkyujo in April

Itr,rpsnral JapaNssE AnN,ry 17

1940 for continued testing. However, a subse- In order to get the Katsuodori off the
quent test flight on 16 April by an IJA officer ground, Kayaba envisioned the use of four
pilot resulted in a hard landing that damaged rocket booster units. Secured to each side of
the HK-l beyond repair. In all, 182 flights had the fuselage under the wings were two rocket
been made in the HK-1. units and together all four could provide an
With the UA still interested, the Rikugun estimated 7,200kg (15,873Ib) of thrust. The
Kokugiiutsu Kenkyujo set aside 200,000 yen planned procedure for using the rockets was
to continue the project. Kimura, along with to have one on each side being fired first, and
Kayaba's chief development designer Dr. when these had burned out the next pair
Shigeki Nait6, set about the task of producing would be fired. Each rocket contained pro-
a new tailless aircraft, this time with a possi- pellant for five seconds ofthrust and, all told,
ble military application. The result was the the scheme would give the Katsuodori a total
Ku-2. The Ku-2 had no tail but rudders were of ten seconds of thrust with which to get the
fixed to the wing tips and the design was plane off the ground and the ramjet function-
tested extensively from November 1940 ing. Kayaba estimated that the Katsuodori
through to May 1941 making 270 flights in all would need to achieve 367km/h (228mph)
before it was damaged in a crash on l0 May. before the ramjet would operate and cer-
To further test the concept, Kimura (with the tainly the speed provided by the rocket units
aid of J6iiWashimi) produced the Ku-3which would have been sufficient for this to happen.
had no vertical control surfaces at all and fea- The rockets, once used, may or may not have
tured a cranked delta wing form to test vari- been releasable but the latter is likely in order
ous angles of sweep. The only control came to minimise drag.
from the flaps arranged along the wings. 65 With the ramjet operating, the estimated
flights were carried out with the single Ku-3 performance of the Katsuodori was a speed
before a crash in 1941 wrecked the glider. of 900km/h (559mph) and a climb rate of
The last design put forward by Kimura was three minutes to reach an altitude of 10,000
the Ku-4. At the request of the Rikugun (32,808f1). Fuel load was 1,500kg (3,306lb)
Kokugijutsu Kenkyujo, the Ku-4 was to be and with a fuel consumption of 50kg (1 l0 lb)
powered and a rear mounted 120hp de Hav- per minute would grant a combat endurance
illand Gipsy 4-cylinder, air-cooled, inline of thirty minutes. Once the fuel was
engine was selected, turning a two-bladed exhausted, the Katsuodori would use its glid-
propeller. Unfortunately for Kimura, the IJA ing properties to return to base.
lost interest in the entire concept. With the For weapons, Kayaba planned on mount-
loss of the Ku-2 and the Ku-3, the IJA can- ing two 30mm cannons externally, one under
celled the Ku-4 before it could be finished. each wing near the wing root. Kayaba did not
With no backing, Kayaba could not afford the wish to use existing 30mm cannon designs
100,000 yen to finish the Ku-4 alone. The IJA such as the Ho-155 preferring to produce a
had paid Kimura and Kayaba 17,000 yen out 30mm version of the 40mm Ho-301 cannon
of the 200,000 yen project money for costs which his manufacturing facilities were con-
associated with the Ku-2, Ku-3 and what was structing for use in the Ki-44-ll Hei Shoki
already paid into for the Ku-4. The remaining fighter. The Ho-301 used caseless ammuni-
funding was not released. Shiro Kayaba, how- tion with each round being, in effect, a rocket.
ever, still had hopes that the concept could be The propellant cavity was partially lined with
a potent weapon and from this came the Kat- a thin aluminium cap. When the primer was
suodori. struck, the propellant was ignited and the
The roots of the Katsuodori come from the pressure would build up until the cap burst,
Ku-2. Unlike the Ku-2, the wings were moved the exhaust gas being vented out the back of
higher on the fuselage and the wing form had the round to move the projectile forward. The
a rearward sweep. The Katsuodori retained main advantage of the weapon was its light
the vertical wingtip rudders used on the Ku-2. weight for such a heavy calibre.
The ramjet filled most of the fuselage which The design of the Katsuodori was nearly
meant there was no room for landing gear. compleie by 1943 and Kayaba anticipated
Instead, a main skid was incorporated on the that he could have had a flying prototype by
underside of the fuselage along with a small 1944. By this time, however, the IJA was
wheel mounted at the rear of the aircraft. already involved with the rocket powered
Without integral landing gear, the Kaisuodori Ki-200 (the IJA version of the IJN's Mitsubishi
rvas to be fitted with a simple, sprung set of J8Ml Syusui - see Page 96) and so paid no
landing gear that could be jettisoned when attention to the Katsuodori. Kayaba, in trying
the aircraft took to the air. The pilot sat to salvage the design, stated that he could
towards the front of the fuselage and was pro- adapt the Katsuodori to accept the Kugisho
vided with a one piece canopy that offered Ne20 turbojet or the KRl0 rocket motor as
respectable visibility to the front and sides. used in the Ki-200. And since his design was

l8 JepeNosn SEcRET PRoJECTs: ExpERTMENTAL ArRCRarr oF rHE IJA aruo IJN 1939-1945
nearly complete a protot,?e Katsuodori record for using such units. Improper place- Japanese, had an incredibly short range -
could be ready for testing before the Ki-200. ment of rocket boosters was the reason only 149.5m (490ft) since it had a muzzle
The advantages ofthe Katsuodori included behind the aborted second flight of the Naka- velocityof 241mlsec (79OfVsec). Coupling the
a ramjet that was far less complex to con- jima Kitsuka (Page 114), heavily damaging very short range of such a weapon with a high
struct than a turbojet. This would have been the aircraft. Attempts to use rocket boosters closure rate due to the speed of the Kat-
a critical asset in a Japanese war industry that on the Mitsubishi Ki-l09 to boost take-off and suodori against a slow bomber, the pilot
was devastated by US bombing. It could also climb met with such poor results that the would have had mere seconds or less to line
use standard aviation fuel, unlike the Ki-200 rockets were removed from the Kil09 devel- up the target, fire, and then bank to avoid col-
that required special fuels, and by extension, opment all together. A misfire or variation in lision. Since Kayaba did not proceed with a
could operate from any airfield without the the thrust output might result in the plane 30mm variant of the Ho-301, the muzzle
need ofspecial fuelling apparatus and proce- careering out of control. Like the Ki-200, once velocity for the round is unknown but it can-
dures. While the speed of the Katsuodoriwas fuel was exhausted the Katsuodori lost its not have been substantially more than the
on par with the Ki-200, the Katsuodori's com- speed advantage, and on the ground its Ho-301.
bat endurance was far superior to the Ki-200 recovery would take longer since the Kat- Despite the potential advantages over the
and the IJA's own planned rocket intercep- suodori could not move on its own without Ki-200, the Katsuodori would never see life
tor, the Rikugun Ki-202 Syusui-Kai (see Page means of wheeled apparatus. This made it outside plans on Kayaba's design board. The
40). r,ulnerable to intruder aircraft dedicated to IJA was looking to the Ki-200 and their own
However, the Katsuodori had several draw- airfield interdiction missions. Kayaba's elec- Ki-202 for their interceptor needs and thus
backs. The first was the use of the rocket tion to use a 30mmversicln of the Ho-301 can- ended Kayaba's dream of seeing his Kat-
boosters to get the plane up to speed. The non would have been a recipe for disaster. suodori taking to the skies to defend Japan.
Japanese did not have a successful track The 40mm version, as used in combat by the

z Ihyaba Katsuodori - data


L Handley Page H,P.i5 Mam (tiK), B0K-5 (Russia), Blohm undVoss
P.210/P.215 (Germany), Heinkel P.1078 (Germany), NorthropXP-56 Black

Bullet (US), Northrop XP-79 Flying Ram (US), Skoda-Kauba SK P,14.01

(Cermany), Lippisch Li P. I 3a (Germany), l'lesserschmitt Me P.l l0l L

(Germany), Heinkel He P,1080 (Germany), St0ckel Rammschussjiiger
(Cermany), Leduc Model 010 (France), Kostikov 302 (Russia)

Type Point Interceptor

Crew 0ne

One Kayaba Model I (or possibly later) ramiel producing 300kg (661 Ib)

ofthru$ at 367km,rir (228mph), 420k9 (925 Ib) of thrust at 490kn/h

(304mph), 550kg (1,212 lb) of thrust at 6i 2kny'h (380mph) and 750kg
(1,653 lb) ol thrust at 734knr/h through l,l03kffi (456mph through


Span 8.99m 29.5ft

Length 4.48m I 4.7fr

Height 1.85m 6.lfr

Wing area 12.57m' l35.4fr:

Wng su'eep 25.5"

Empty 850k9 l,873lb
Loaded 3,000kg 6,613 lb

Max speed 900knt/it 559mph
Landing speed l00krdl' 62mph

Range 400km 248miles

Climb 3 min to 10,000m (32,808ft)

Ceiling 15,000m 49,21211

Two 30mm cannon

None. The Katsuodori did nol advance beyond the drawing board.

IMprRr.q.L hpnNEse Annv l9

Kokusai Ta-Go


In 1943, the Allied island-hopping campaign addition to war factories being razed to the
was underway and in 1944 the Japanese ground, Japan was being starved of materials
would see their island outposts, bases and needed to sustain weapon production. Alu-
strongholds destroyed and lost to them for- minium was a key material in aircraft pro-
ever. In 1945, the Japanese lost their holdings duction and it was estimated that by
in Burma, Borneo, Iwo Jima and Okinawa. December 1945, even with strict control, the
Japanese military planners had no doubts supply of this metal would be exhausted.
that the Allies would continue their progress Consequently, wood was to become the
and land forces on the main islands of Japan. main material for aircraft construction,
The Allies did indeed have such a plan knorvn regardless of the type of aircraft concerned.
as Operation Dovmfall. The Japanese, to Examples included the Tachikawa Ki-l06 (a
defend against the invasion they felt was wooden version of the Nakaiima Ki-84 Hay-
coming, put into motion Ketsugd Sakusen or ate) and the Kugish6 D3Y MyOio (which was
Operation Decision. the wood derivative of the Aichi D3A Vcl).
Operation Decision's main component With the loss of heavy industrial machinery, it
was the use of shimpu and shimbu missions fell more and more to smaller workshops to
targeting the Allied naval force, specifically produce components and sub-assemblies for
landing craft, troop ships and support ships. aircraft. Often the labour force was not as
To repel the invaders, all manner of craft skilled as before and working with wood was
were assembled for the Japanese defender of easier as it did not require the sophisticated
kokutai, the national polity of self sacrifice. tools and jigs necessary for construction of
Midget submarines such as the Kairyl, Koryl more conventional aircraft using metal com-
and the crude U-Kanamono, the Kaiten ponents.
human torpedo, small explosive laden Captain Yoshiyuka Mizuyama, an officer in
powerboats like the Maru-Ni (lJA) and Shinyo the IJA's aviation equipment section, was the
(lJN), and even frogmen (the Fukuryu) were man behind the Ta-Go (Ta being short for
prepared for the final showdown. Even the take-yari, or bamboo spear). It was his desire
best tanks, the Type 3 Chi-Nu and Type 4 Chi- to design and build a plane that was simple in
To, were held in Japan to counter Allied construction, used the bare minimum of war
arrnour. Aircraft would also play a significant critical materials and could be produced
role in the defence of Japan. Itwas estimated rapidly. By doing so, the Ta-Go could quickly
that 10,000 aircraft of every type would be populate the aircraft pool available to units
available to throw at the Allied invasion fleet. destined for shimbu missions and also
It was thought that the mass wave tactics replenish losses in short order. He hoped
would result in a tremendous loss of aircraft that the Ta-Go could be used to defend
which the Japanese industry in 1945 would the seaside cities of Osaka and Kdbe. In an
be unable to keep pace with unless steps effort to help realise the Ta-Go, Mizuyama
were taken to remedy such a situation. The approached the Tachikawa Hik6ki K.K. with
Ta-Go was one such remedY. his concept. Despite Mizuyama being an IJA
By 1945, Japanese industries were under officer, Tachikawa refused to assist him as his
The profile depicts the Kokusai Ta-Go prototype
in the colours it actually sponed. regular bombardment from US airpower. In plan had no official sanction and was not

20 Jnpmlrse SEcnrr PRorEcrs: ExpeRturNteL Alncnanr on rHE IJA ano IJN 1939-1945
approved by the Koku Hombu. As such, In addition to the size reduction, steps were tilities in August 1945, the Kokusai Ta-Go
Tachikawa could not spare the capacity to taken to simplify the Ta-Go even more. Gone never entered production. The close of the
develop the aircraft. was the canopy and the pilot sat in a open war also saw the end of two Kokusai devel-
Undeterred, Mizuyama discovered a small cockpit with only a small acrylic glass wind- opments of the Ta-Go, known as the Gi-Go
shop in the city of Tachikawa within which he screen as protection from the elements. For and Tsu-Go. Both remain shrouded in mys-
and his fellow men went about the task of instrumentation, only the absolute basics terybecause no information on them has sur-
designing and constructing the first proto- were used consisting of a speedometer, faced to date.
type. Once the concept was completed, work altimeter, compass and the essential engine Ironically, Tachikawa would return to the
began on building the Ta-Go. Using wood related gauges such as fuel and oil. The fuse- Ta-Go when the Gunjushd (Ministry of Muni-
lathes to construct the fuselage and other lage was slab sided and box shaped. While tions) authorised development of
components, the aircraft was made from ply- this granted easier construction, it was not Mizuyama's initial design following the com-
wood while fabric was used for some of the the most aerodynamic design. Much of the pletion of the Kokusai Ta-Go prototype. The
skinning and coverings for the control sur- fuselage used wood sparring and structure end of the war would find the Tachikawa
faces. The pilot was given a simple acrylic while the skinning was of plywood. The Ta-Go prototype incomplete. As a note, with
glass canopy. Instrumentation was kept to the wings were low mounted with squared wing the acceptance of the Ta-Go by the Koku
bare minimum. The landing gear was fixed. tips and they were hinged just outside of the Hombu, a project number (meaning a Ki
For a motor, a Hitachi Ha- 1 3 Ko 9-cylinder, air- landing gear to enable them to fold upwards number) was assigned to the Ta-Go - Ki-128.
cooled radial engine developing 450hp was to allow the aircraft to be hidden in caves as It has not yet been confirmed whether this Ki
selected, the cowling for it being made from well as facilitate their construction within the number applied to the Kokusai Ta-Go, the
plain sheet steel. The only armament was a confines of caves or small manufacturing Tachikawa Ta-Go or both.
single 500kg (1,102 lb) bomb. In February lines. Both the vertical stabiliser and the hori-
1945, the Ta-Go prototype was nearly com- zontal stabilisers were rectangular in shape.
plete when Tachikawa was subjected to a The landing gear was fixed, being made of
bombing raid. In the ensuing attack, the shop steel tubing and fitted with rubber wheels, Ta-Go - data
was burnt to the ground and the Ta-Go inside each gear supported by a single strut. To pro-
destroyed. vide a modicum of streamlining the tubing Contemporaries
Despite the setback Mizuyama forged that made up the landing gearwas faired over Messerschmitt P. I 104 Sprengst0fftrager (Cermany)

ahead, going to Nippon Kokusai Kogyo K.K. using aluminium. The only measure of shock Specificalions are for the Kohusai Ta-Go,
(Japan International Air Industries Co. Ltd.) to absorption came from the tlres and the tail
pitch his Ta-Go. In the end the proiect was skid, the latter also being built from steel tub- Type Special Attack Aircrafl

accepted and in part this may have been due ing with a portion rubberised. Crew 0ne
to Kokusai's experience with light aircraft The [Ha-47] 1l engine was fitted with an
such as the Ki-76 (known as S/e114 by the angular plywood cowling, the engine driving Powerplant

Allies) and the Ki-86 Ko (codenamed a fixed-pitch, two-bladed wooden propeller. One Hitachi IHa-47] I 1, 4-c.vlinder, air-cooled inline engine developing

lor take-off driving lvooden, lwo.bladed propeller 7.lft in diameter

Cypress), the latter of which Kokusai had A metal engine mount was used while the I I Ohp a

as the prototype all-wood Ki-86 Otsu. Of fuel tank was situated on top of the engine
course, Kokusai was not as heavily taxed by and used a gravity feed system. Behind the
Span 8.90m 29.211
wartime demands from either the IJA or the tank and in front of the windscreen was a sim- Length 7.40m 24.311
IJN and could thus allocate some assets to the ple oil cooler, mounted flush in the fuselage. Height 3.87m 12.711

development of the Ta-Go. Despite Kokusai Given the much smaller dimensions of the Mng area 5.10m' 54.9ft:

taking on the Ta-Go project, it still remained revised Ta-Go, it was no longer able to carry Wing loading 34.66k9/m' i,l lbiit'

an unofficial design and thus bore no Ki the 500kg (1,102 lb) bomb Mizuyama's origi-
number. nal version was designed for. Instead, it could Weights

Mizuyama's designfor Kokusai differed only carry a 100kg (220 lb) bomb. The bomb Empty 345,5k9 761 lb
from the one he proposed to Tachikawa was fitted to the underside of the fuselage Loaded 585.5kg

because the new version was significantly and once in place could not be released by
scaled down and much smaller. In so doing, the pilot. l2lmph
)!1ar speed l95kn/h
this reduced the amount of assemblies Mizuyama, with the assistance of his onn Cruise speed I Tgkntiit illmph
needed to produce the aircraft which, by men and Kokusai, had completed the first Range l50km 93 miles
extension, lowered the man-hours required prototype of the new Ta-Go by the middle of Ceiiing 4,600m I 5,091 [r

to build it. Fewer assemblies meant less use June 1945 and it was made ready for flight. On
of construction materials. With the resizing, 25 June, the Ta-Go took to the air for the first fumanent
the Ha-I3 radial became too large for the pro- time with a Kokusai test pilot at the controls. One l00kg (220|b) bomb

posed airframe and so the Hitachi [Ha-47] 1l Not surprisingly, the pilot reported handling
inline engine, rated at 110hp, was selected as concerns. After a number of additional test
None. A total ofthree Ta-Go aircralt were constructed: \'lizuyama's own
a replacement. This same engine was used in flights, revisions were made to the design.
protot!?e aircraft thatwas destroyed by fire prior to flight, the one built
the TdkyO Koku Ki-107 all-wood two-seat Once complete, Kokusai created a complete
and flown at Kokusai, and the Tachikawa Ta-Go which remained
trainer which was to be the replacement for set of working blueprints for the production incomplete at the end ofthe !var.
the Ki-86 had the former made it into service. version. However, with the cessation of hos-

IupERreL JepnruEsE Anl,rv 21

Maeda Ku-6

Interest in airborne forces can be traced as far M1A1 .30cal carbine and the Japanese Type 2 striking power of their paratroopers and one
back as 1917 because they can provide sev- Paratroop rifle. Despite such weapons, air- such plan was a tank borne into battle on
eral tactical advantages. Being air dropped, borne forces were deficient in one critical wings.
parachute troops can be deployed into areas area: armoured vehicles. The ability to pro- ln 1943, the IJA set the wheels in motion to
not easily accessible by ground forces as well vide airborne troops with armoured support investigate a flying tank. The tumy Head Avi-
as bypassing defences meant to hinder or such as tanks was one sought by all the war- ation Office in league with the Fourth Army
repel attacks from specific avenues of ring powers and tank designs did emerge. Research Department drafted the initial con-
approach. Also, the ability to place troops The key problem was how to send in the cept for the weapon. The aviation research
anyr,vhere on the battlefield requires the tanks with the troops during an operation. section of Maeda was tasked with producing
enemy to use assets to protect against such One of the first solutions was the glider tank. the wings that would form the glider portion
operations, thereby spreading defending The Japanese would create and utilise air- of the weapon and the Army Head Aviation
forces thinner. Such advantages come at a borne forces during World War 2. The IJA Office assigned the designation Ku-6 to the
cost, however. Airborne forces tlpically do called their forces the Teishin Dan (Raiding glider. The tank was to be designed and built
not have the firepower of comparable ground Brigades) while the IJN had the Rikusentai. by Mitsubishi and called the So-Ra (or Sora-
forces nor the ability to remain independent Both would be used first in 1942 during the Sha, literally'sky tank'). To ensure there was
for long before outside support must be fighting in the Dutch East Indies. Unlike the no confusion, the Army Head Aviation Office
obtained. Germans, British and Americans, the Japan- called the entire combination the Kuro-Sha
Airborne troops were used by all of the ese did not provide their paratroopers with a (taking the 'Ku' from Ku-6 with 'ro' meaning 6
major waning powers in World War 2 and significant amount of specialised healy and the 'Sha' for tank).
special equipment and weapons were cre- weapons. ln part, this may have been due to Mitsubishi's So-Ra was, due to the purpose
ated for use by these units in an attempt to the fact that the Japanese parachute forces for which it was intended, a tankette design.
provide them with heavier firepower. Artillery would rarely be used in their designated role. With a crew of two (driver/pilot and the com-
such as the US fumy M1A1 75mm pack how- Instead, much of their fighting would be done mander/gunner), the So-Ra was to weigh
itzer and the German 7.5cm LG 40 recoilless as light infantry (much like the German 2,812kg (3.1 tons). The turret was set behind
gun were air-droppable and the troops used Fallschirmjiigers). Nevertheless, the IJA and the driver/pilot compartment and was pro-
modified or special small arms such as the US IJN were considering ways to improve the vided with three large, hinged ports to allow

v U^/ lltu


22 JaparuEsE SEcnnr PRoLEcrs: ExprRIr'rrltaL AIRcRnrr op rHE IJA axo IJN 1939-1945
some measure of vision for landing. Armour Maeda Ku-6 - data
was likely very light and was certainly less
than the 6mm-12mm armour protection of Contemporaries
AntonovA-40 (or KT for Kryl'ya Tanka, flying tank) (Russia), Raoul
the Type 95 Ha-Go light tank then being used
Hafnels Rotabuggy and Rotatank (UK), Baynes Bat (UK), John Waller
by Japanese airborne forces. Three weapon
Christie's M1932 (US)
fits were proposed for the So-Ra. The first was
a 37mm cannon (such as the 37mm Type 94 Specifications are based on the second t)atialion of lhe Kuro'Sha,

used in the Ha-Go), the second consisted ofa Lith the So-Ra beneath the uing,

machine gun armament (either a light

Type Glider (Ku-6)
weapon like the 7.7mm Type 97 machine gun
or a heavier calibre) and the third was a
flamethrower. Power was to come from an
air-cooled engine producing 50hp that was Dimensions
estimated to give the So-Ra a maximum road Span 21.97m 72.lft

speed of 42kmlh (26mph). Length 14.96m 4l.gfr

The glider portion, the Maeda Ku-6, has Height 2.98m 9.8ft

been interpreted in at least two ways since Wing area 59,99m' 645.83ft'

the original design is not known, the docu-

ments either having not survived the war or
Loaded (wjth the So-Ra) 4,200kg
have yet to be discovered. One version shows
the wings secured to the So-Ra at the hull, on Performance
either side of the turret, with a tail boom fitted Mar glide speed l74km/h 108mph
to the rear of the tank. A horizontal stabiliser
sat on top of the vertical stabiliser. The dri-
ver/pilot moved the control surfaces via wires
that ran into the tank. On the hull front was
Tankette (So-Ra)
the tow cable attachment point. The second Type
version has the So-Ra fitted with struts on the Crew

hull sides. Atop the struts was the main wing

to which twin tail booms were fitted with a
One 4-cylinder, air-cooled, gasoline engine developing 50hp at 2,400rp
low mounted horizontal stabiliser connecting
the vertical stabilisers. ln essence, the So-Ra Dinensions
would hang below the wing. On landing, the widrh 1.43m 4.nt
tank would shed the wings and move into Length 4.05m I 3.3f1

action with the paratrooPers. Height 1,88m 6.2fr

By 1945, the Ku-6 had been completed and

Mitsubishi had produced a full scale mock-up Weights

of the So-Ra. Flight testing was conducted for Loaded 2,90Okg

a brief period and it is likely that the mock-up

was used, suitably weighted to simulate the 26mph
Nla\ speed
2,812kg (3.1 tons) of an operational So-Ra.
The So-Ra was to be towed by a Mitsubishi Armament
Ki-21 bomber. Tests soon showed the diffi- One 37mm cannon, machine gun ot flamethrower
culty of the concept. The Kuro-Sha suffered
from in-flight control problems, the driver/
pilot had poor vision and landing was
extremely difficult. Another concern was that Deployment
None. Only one prototlpe built and flown.
the So-Ra could not stand up to heavier and
more powerful tanks. Any usefulness the
Ku-6 may have possessed was minimised
with the advent of the Kokusai Ku-7 Man-
azuru (meaning 'Crane') glider that began
development in 1942. First flown in August
1944, the Ku-7 was able to carry a 7,257kg (8
ton) tank within its fuselage which was more
than enough to hold the 6,7l3kg (7.4 tons) of
the Ha-Go light tank. With the Kuro-Sha's
problems evident, the IJA terminated any fur-
therworkon the Kuro-Shafavouring the Ku-7'

IMpsRrar- hpnNesE AnuY 23

Manshfl Ki-98

In late i942, the Koku Hombu was looking for advanced trainer. Manshu's main plant was
a number of new aircraft tlpes as improve- located in Harbin in the Japanese puppet
ments on those in seruice. These included a state of Mdnzhdugu6. On learning of the Koku
hear,y fighter capable of conducting ground Hombu's desire for new aircraft, Manshu
attack operations and a high-altitude fighter. sought to put together a proposal to meet the
Nakajima and Tachikawa were tasked with fighter requirement. The company assigned
the latter, coming up with designs that would their two best men to the project, engineers
later result in the Ki-87 and Ki-94 respectively Noda and Hayashi, and what resulted was an
(see Pages 28 and 53). For the former, aircraft that was far from the conventional
Kawasaki attracted the interest of the Koku types Manshl had worked on in the past.
Hombu with their multi-role Ki-i02. However, The aircraft was a single-engine fighter
Kawasaki's design was not to go uncontested with a pusher, twin-boom configuration. The
and the competition would come from a rela- heart of the plane was to be a Mitsubishi
tively small aviation company. Ha-211-lll 18-cylinder, air-cooled radial
Manshlkoku HikOki Seizo K.K. - the engine fitted within the fuselage and behind
Manchurian Aeroplane Manufacturing Com- the cockpit. The four-bladed propeller, situ-
pany Ltd., and better kno'uvn as Manshu, a ated at the very rear of the fuselage, was dri-
contraction of the kanji 'Man' in Manshukoku ven by a 2m (6.5ft) long extension shaft. In
and 'Hi' in Hikdki-was founded in 1938. Man- order to maintain a well streamlined airframe
shr-r was a subsidiary of Nakajima Hikdki K.K. no air scoops were used; instead, flush inlets
and produced the Nakajima Ki-27 (code- were fitted along the top of the fuselage
named Nate by the Allies) and the Nakajima behind the canopv. To increase the flow ofair
Ki-84 (Frank) for the company. Manshu to the engine, a fan driven by the engine was
would produce few of their own designs installed. Flush outlets forward of the pro-
and only one ever saw service, the Ki-79 peller completed the air circuit across the

24 JnpenEse Sncnsr PnoLecrs: ExpeRnrnxteL AIRcnapr oF rHE IJA,qNo IJN 1939-1945

engine. The thin wings were mounted low did, sending Manshu suggestions for design Manshu Ki-98 - data
and on each wing was a boom that ended in changes to the Ki-98 to make it suitable for the
an ovoid vertical stabiliser. A single, high new role. On receiving the news Manshrl had Contemporaries

mounted horizontal stabiliser connected the tukhangelskiy BSh (Russia), Saab 2l (Sweden), Vultee V.78 (US),
to substantially alter its initial design to meet
Bell XP-S2 (US)
two tails. the new demands. With strained manpower
A tricycle landing gear system was used, the and resources, the mandated changes set the Pertornance specificatians arc estimotes based on Manshi's prqections

nose gear retracting backwards into a wheel Ki-98 program further back and scuppered
well that ran underneath the cockpit. Each of plans to build the prototlpe. Type HighAltitude Fighter

the two main wheels retracted into their One of the most important changes was the Crew One

respective tail booms. As the aircraft sat very need to fit an engine with a turbosupercharger
high offthe ground, the pilot had to access the resulting with the Ha-211-lll being replaced by
One l'litsubishi Ha-21 I Ru l8-cylinder, air-cooled radial engine uith a
cockpit via a hatch in the nose wheel well. If the Mitsubishi Ha-211 Ru which incorporated
turbosupercharger developing 2,200hp [or take.oif, l,960hp at
the pilot had to bail out, he had two choices. this feature. As the turbosupercharger was 2.000m/6,56 I ft and I,750hp at 8,500m/27,887[t driring a four-bladed,
He could leave in a conventional fashion, but exhaust driven itrequired the appropriate metal, 3,6n/l l.8lt diameter propeller
had to contend with both the twin tails and additional piping, which, of course, was not
horizontal stabiliser along with the propeller. originally included. The new engine was Dimensions
Manshu recommended that the pilot egress therefore larger than the original and this Span ll,24m 36.9f1

through the hatch out of the bottom of the air- made it necessary for the fuselage to be Length (total) 11,39m 3i.4ft

craft. This method allowed the pilot to avoid lengthened and slightly widened. As the new Boom length 8.26m 27.lll
being dashed on the tail but still had to con- propeller had a larger diameter, the twin Height 4.29m 14.1ft

tend with the propeller. Nevertheless, the booms had to be moved further apart to
Mng area 23.99m 258.3ft'
Wing loading 187.48k9tm' 38.41b/fl
chances of lowering the nose gear, sliding accommodate the blades and, by extension,
Polver loading 2,72k{hp 6lb/hp
down and out through the hatch in a stricken the wings had to be reworked as well. Finally,
plane were slim and Manshtr were aware of the airframe had to be strengthened to support Weights
this flaw in the design. The canopy was a bub- the heavier weight. Another alteration was to Empty 3,500kg 7,i16 rb
ble type that afforded an excellent field of offer the pilot a more suitable way to bail out Loaded 4,500kg 9,9201b

view. Forweapons, two Ho-5 20mm cannons of the aircraft. Given the extreme difficulty in
and one Ho-204 37mm cannon were installed having to drop the nose wheel to gain access Performance

in the nose. Due to the short length of the fuse- to the well hatch, the revised Ki-98 incorpo- Nlar speed T3lkrni h 454mph

lage, the barrels for the cannons, especially rated explosive bolts that shed the tail unit to at 10,000m at32,81Oft
Climb min 30 sec to 5,000m (16,404ft)
the Ho-204, protruded out from the nose. allow the pilot to exit more conventionally. 5

Once the preliminary design for the fighter

Range 1,249km 776miles
The weapon fit remained unchanged.
Endurance 2 houn l5 min at 499kn/h (310mph)
had been completed, Manshl submitted it to With the new specifications in hand, the Ceiling 10,000m 32,808ft
the Koku Hombu. Despite the unorthodox Ki-98 design was reworked and redrafted but
approach, it was accepted as the Ki-98 and it would not be until October 1944 that the Armament
work was allowed to proceed. Interestingly, redesign was completed to be followed by a One Ho-204 37mm cannon and two Ho-5 20mm cannons
the Koku Hombu reiected Tachikawa's mock-up of its revised fuselage. Manshlr
Ki-94-l that was similar in concept to the expected to have the first prototype finished Deployment

Ki-98. With approval in hand, the draft for the and ready for flight testing by early 1945. None. The only prototlpe was never compleled and was destrol'ed to
prevent capture.
Ki-98 was finalised by July 1943. Work then These plans were dashed following a US
commenced on a wooden mock-up that was bombing raid on Manshu's Harbin factory on
completed in December. Design work con- 7 December 1944. lt was not until January
tinued into the beginning of 1944 further refin- 1945 when work commenced on the Ki-98.
ing the Ki-98. A scale model of the aircraft Despite Manshu attempts to increase the
were constructed and sent to Japan for wind pace of construction work, progress still
tunnel testing at Rikugun Kokugijutsu lagged.
Kenkyujo. Unfortunately for Manshd, the At the start of August 1945, the fuselage,
worsening war situation saw some of theii wings and the tail booms were completed
personnel called into service or shifted to and were ready to be assembled. However,
other departments and this, coupled with a on 8 August 1945, the Soviet Union declared
plethora of design revisions, saw work on the war on Japan and initiated its invasion of
Ki-98 slow down. Nevertheless, wind tunnel Mdnzh6ugu6 the next day. With the
tests showed excellent results and Manshu Mdnzhdugu6 Imperial fumy and the Japan-
began to make the preparations to construct ese Kwantung Army unable to stem the tide
the first prototlpe. ofSoviet forces, Manshl ordered all relevant
In the spring of 1944, the Koku Hombu documenta.tion including models, mock-ups,
instructed Rikugun Kokugijutsu Kenkyujo to jigs, tools and the incomplete Ki-98 to be
tell Manshu that the Ki-98 should be adapted destroyed to prevent the aircraft and informa-
to serve as a high-altitude fighter. This they tion on it being captured by the Soviets.

lupEnral JaparuEse ARuv 25


JnpaNEse SscRsr PRoJECTs: Exprntl,lerutnL AtRcnLnr op rHE IJA nNn IJN 1939-1945
Mitsubishi Ki-73

In 1943, the Koku Hombu issued a specifica- izontal-H design the engine experienced near
tion for a fighter capable of operating for long constant problems during its development.
distances in order to act as an escort for Ultimately Mitsubishi was unable to over-
bomber formations. Despite the defensive come these difficulties and abandoned the
weapons Japanese bombers carried, they Ha-203-ll. Due to the delays and eventual can-
were still vulnerable to interception. If a cellation of the engine, Kubo's Ki-73 design
fighter had the extended range, it would be was abandoned even before he and his team
able to protect the bomber formations by could produce a mock-up, let alone a proto-
being able to engage enemy interceptors and tlpe. Even though the Ki-73 went nowhere,
allow as many bombers as possible to sulive Allied intelligence was aware of this new pro-
and deliver their bomb loads. It was this ject. Information obtained from various
desire that fuelled the Koku Hombu to issue sources, including captured documents, led
their specification and from which Mitsubishi intelligence officers to conclude that the Ki-73
would build the aircraft to meet it. would see selice. As such, in 1944, the Ki-73
Mitsubishi's Tomio Kubo, along with engi- was assigned the codename Sfeue. As it was,
neers Kato, Sugiyama and Mizuno, began the noAllied pilotwould ever encounter the Ki-73
investigation on how best to meet the specifi- in any form.
cations. They settled on using a single engine What Allied pilots might have encountered
design and the heart of it would be the Mit- had the war gone on would have been the Mit-
subishi Ha-20311 engine. This was a 24-cylin- subishi Ki-83. Not discouraged by the Ki-73's
der, horizontal-H, liquid-cooled engine that demise, Kubo would go on to design the twin-
was projected to generate 2,600hp. The engine Ki-83 to meet Koku Hombu's specifi-
Ha-203-ll was chosen due to its horizontal-H cation. The result was a highly capable aircraft
Mitsubishi Ki-73 configuration - in essence, two flat engines thatwould have provided a challenge toAllied
- data placed one on top of the other and geared air power. However, only four Ki-83 proto-
Contemporaries together (a flat engine is one in which the pis- types were built before the end of the war.
Consolidated Vultee XP-8 i 'Silver Bullet' (tJS), North American P-5 1 D tons move horizontally). Each flat engine had Very little is known to show what the Ki-73
'Mustang' (US). Larochkin La-l I (NATO codename fang) (Rusia), its own crankshaft. Although horizontal-H looked like. The artwork depicted for the
Westland \Itvern (UK) engines have a poor power-to-weight ratio, Ki-73 in this book is based on an interpreta-
they offer the advantage of being more com- tion of the Ki-73 printed in Richard Bueschel's
pact, which made the Ha-203-ll the ideal 1966 bookJopanese Code Names. The illus-
Outside of the intended engine and the aircraft's role, specifications on
choice for the aircraft, now designated the tration was based on the Ki-83 on the
the Ki-73 are unknolvn
Ki-73. assumption that Tomio Kubo would have
Unfortunately, Mitsubishi was having a very used aspects of the Ki-73 in the Ki-83. Sfeue is
None. The Ki-73 never advanced past the concept stase difficult time with the Ha-203-ll. In fact, shown here in the markings and colours of
because of the relative complexity of the hor- the l0lstSentai.



lnrpsnraL JapeNesE ARI'ry 27

Nakajima Ki-87



. .::rrrr.r..i..::.,:l.i::ilai:1@

t**,.{ffi$.;s . ::.,

Altitude is a maior factor in an engine's per- ter flight time in addition to a half hour of having a turbine and a compressor linked
formance and, by extension, the aircraft as a combat flight time up to 800kmih (497 miles) together via a shared axle. Engine exhaust
whole. Known as density altitude, the higher from the airfield that the aircraft operated spins the turbine which in turn spins the com-
the altitude, the less dense the air. The effects from. Finally, a hear,y armament requirement pressor which draws in outside air, com-
called for two 30mm cannons and two 20mm presses it and then directs the air to the intake
of this manifest themsleves in lower wing lift,
a reduction in propeller efficiency and cannons. manifold of the engine. This compressed air,
reduced horsepower output from the engine. With these new specifications, Nakajima delivered at high pressure, results in more air
As such, a plane that was not designed to 'uvas contracted in November 1942 to produce reaching the cylinders for combustion. The
operate in such conditions suffers accord- three prototypes and seven pre-production net effect of this is that at higher altitudes
ingly. The Koku Hombu sought an answer to aircraft for the IJA. The proiot)'pes were to be where the air is thinner, the turbosuper-
the problem and Nakaiima looked to provide completed betrveen November 1944 and Jan- charger allows the engine to function as if it
the solution. The result was the Ki-87. uary 1945 rvith the pre-production planes fin- was at a lower altitude where the air is heav-
In mid-1942, the Koku Hombu drew uP a ished between February and April 1945. The ier and thus engine performance is not
set of specifications for a high-altitude fighter. design of the Ki-87 rvas headed by Kunihiro adversely affected. A benefit of this is that
These called for a plane capable of operating Aoki. because the air is thinner at higher altitudes'
at high altitude, heavily armed with a maxi- Nakajima initially selected the Nakajima there is less drag on the aircraft and since the
mum range of 3,000km (1,864 miles) and [Ha-44] 1l l8-cylinder radial engine as the turbosupercharger preserues the horsepower
capable of 800km/h (497mph). Examination heartof the Ki-87. The IHa-44] 21 (known also of the engine, overall speed is improved'
of the specifications called into question the as the Ha-219 Ru) was also considered but A sizable portion of the aircraft's forward
viability of meeting such performance expec- the [Ha-44] 11 would be the engine used in fuselage was taken up by the [Ha-44] 11
tations. After deliberation, they were revised. the first prototype. Both engines were rated at engine assembly and the large turbosuper-
The role remained the same but the speed 2,400hp and each used a turbosupercharger charger',vas fitted to the starboard side of the
requirement was dropped entirely to the that would maintain and enhance the fuselage, just ahead of the cockpit. To cool
point that no mention was made at all for a engine's power output at altitude. A turbosu- the engine, a sixteen-bladed fan was mated
minimum or marimum speed. The range percharger is an air compressor used to force to the four-bladed, constant speed propeller,
requirement was adjusted to one hour of loi- air induction to the engine. It does this by turning at 1 50 per cent of the propeller speed.

2g Jap.qxpsE StcREr PRorEcrs: EtpERruENral AtRcRapr ol rnE IJA lr'rp IJN 1939-1945
The engine reduction gear ratio was set at
0.578. As the Ki-87 was designed for high alti- I
tude operation, the pilot was to be provided
with a pressurised cockpit (though the proto- z
t!?e was not equipped with one).
For weapons, Nakajima kept to the specifi-
cations mounting a 20mm Ho-5 cannon in
each wing root, synchronised to fire through
the propeller, and a 30mm Ho-155 cannon in
each wing to the outside of the main landing
gear wheel wells. Ammunition was stored in
the inner wing near the fuselage. Hydraulic
pressure was used to load the cannons and
they were fired by electrical triggers. If
required, provision was made to carry a
250k9 (551 lb) bomb or a drop tank along the
centreline. Because of the heavy weapon fit
and to ensure enough room for the self-seal-
ing wing fuel tanks, Nakajima designed a
landing gear arrangement that was rare in
Japanese aircraft development - the main
landing gear struts would retract backwards
and the wheels would rotate 90" to fit flush
into the wheel wells.
Given the task the Ki-87 had to perform, @*
Nakajima provided a degree of protection for
the pilot in the form of 66mm thick, bullet
proofglass in the front of the canopy and back
protection via armour plate 16mm thick. To
extend the range of the Ki-87, two 300 litre (79
gallon) drop tanks could be fitted under each
wing beside the landing gear wells. The pilot
could jettison them via electrically controlled
releases and these could be used in conjunc-
tion with centreline payloads.
As work progressed on the Ki-87, the IJA
saw fit to change the design by insisting that
the turbosupercharger be placed in the rear
of the fuselage beginning with the third pre-
production Ki-87. Nakajima protested against

Iupnnrlr. .llplnssn ARuv 29

Nakajima Ki-87 - data the change but could do little to sway the IJA reported good handling characteristics and ii
on the matter. In addition, the third prototype was thought that the Ki-87 was superior in
Contemp0raries SukhoiSu-l(Rusia) Ki-87 would have a reduction gear ratio set at comparison to the Nakajima Ki-84 Hayate
0.43i and the seventh pre-production Ki-87 (Gale).
Be.,use the Ki-E7 u)as nat flou)n b its fulllbilit!, Ihe pelomance
\/u/r'tl ' 0rr ('/.lltdl, s ma4e bt \ohtjinu was to feature a cooling fan that spun faster to Even as testing of the Ki-87 was underway
facilitate enhanced engine cooling. with work continuing to meet the IJA turbo-
T}?e Hish-Altitude Interceptor Despite the worsening war situation, Naka- supercharger position requirement, Naka-
Crerv 0ne jima was able to complete the first prototype, jima designers developed the Ki-87-ll.
c/n 8701, by February 1945 rolling it out from Replacing the IHa-44] 11 would be the Naka-
Poserplant One trakajima lHa.lll I l, l8'olindeL. ail'cooled their Ota Plant. Problems wiih the electrical jima [Ha-46] 11 (known also as the Ha-219)
developing 2,100hp lor take-ofl 2.20|)hp at l ,r00m' l -020ft, 2.050hp
rad ial
system that operated the landing gear and dif- that could provide 3,000hp. The turbosuper-
at 6.000m1 9.685|t and ,8i0hp at 0.i00m ,i1, l50il and diiYing a constant
charger was situated in the belly of the fuse-
I 1 1

ficulties with the turbosupercharger delayed

speed, l.bladed propellei
flight testing. It was not until April 1945 that lage as demanded by the IJA. Performance
the Ki-87 was able to take to the air. Due to the estimates showed a 4 per cent increase in
Span l3.llm .t l.l.)fr issues wiih the landing gear, Nakajima for- speed compared to the Ki-87.
Length I L79nr 38.7ir bade the test pilot from retracting the main Ultimately, the Ki-87's design team failed to
Height 1.18m ll.itr gear lest it fail in the up position, thereby dam- overcome the problems with its engine.
\\iing area 2i.99m' 2i9.8ft, aging or destroying the Ki-87 vuith the resul- Because they were unable to solve difficulties
\\iing loading 2l6.29ksrm 11.3 lbrfr
tant belly landing. This, however, prevented with both the turbosupercharger and the
Po*'er loading 2.35kq'hp 5.2 lbrhp
any chance of a thorough evaluation of the IHa-44] 1 1 aswell as the temperamental land-
Ki-87's top speed and iull manoeuvrability. ing gear system, the Ki-87 would make no
Consequently, there was no attempt to mon- more test flights. When hostilities ceased the
Empt\ l,387kg 9,62Ib
itor and collect performance data. During the other two prototypes remained incomplete
Loaded i.632kg I 2.116

\larimrtm 6.100k11 13,1J8tb

five flights the prototype did make, the pilot and the Ki-87-ll rvas still on the drawing board.


\lar specd i0Tkmrh l39mph =

at I l,i]il()nt al ll6.090li

Endurance 2 hoLrs =

Climb 1l min l2 sec to 10,000rn (32.8J0it)

Ceilinq l2,85rn 12,liii1

Tu o 3()nm Ho-i ii cannons, Tu o 20mm Ho.5 cannons and ptorision ior

one i5l lb bomb

fone.0nlvone Ki-87t\as completed and test fl01\nuith h\0 Othels

incomplete beiore the trarended.

\akarinra Ki-8i (FE-1iii)
This n as the onh Ki-87 to flr. havinq the sctial 8i0 l. CaptuLtd at the I.lA aiL

base at Chofu, the Ki-87 (nicknaned 'Big Bot br the nlen \\ h0 sas the

larqe aircraft) nas crated and shipped to the ll5, appearinq on l0 \latch
l916 at \lA\lA. Under iestoration fot the museurn. lhe Ki'8i sas s00n

rnovcd lo Park Ridqe. Hotret er. ailer I \lav l9J9 (lhe last nriiten report

docunleniing the airciaft) all trace oithe Ki"87 disappcared, a likehvlctim

oithe cutters torch.

Nakajima Ki-87 (FE.l55)

It has been surmised that FE-15; \\as, in fact, altpogiaphical enot nlade
0n a Iater rep0il concerning Ft-157 ftec belo$j, On thc flip side, it mai

be that the FE-155 enhr \\as a cOrrection and that FE-l;7 as listed on the

earlierLeport nas designated in enor. In either case,Onlr hl0 ofthe Ki'8i

aircraft reached the L'S.

Nakajinra Ki-8i (FE-lri)

FE-157 u,as, most likeh, the sccond 0lthe two remaining Ki'87 protottpes

found incomplete u'hcn the trai endtd. Listed as FE'l 57 on l0 trlarch

1916 at \1A[1A. lhc plane uould lati'r reappear 0n a ] August l916 report

as FE.155 and uas located at lhe A0A\1C in Netraik. \en Jetsev.

\o lurthcr trace of this Ki-s7 e)iists aiter the AuqLrst report and the aircraft

nas most piobablr sctapped.

30 I.rparuesr SEcirEr Pno.lEcrs: EtpEnlntElltel AtRcnanr oF THE IJA AND IJN 1939-1945
Nakajima Ki-l 15 Tsurugi z

By 1945, Japan was reeling from one defeat structure with steel panelling and centre sec- models of the Ki-l15 were to be fitted with
after another in the face of the Allied advance. tions with tin used for the engine cowling. two solid-fuel rockets, one under each wing.
With the possibility of an Allied invasion The tail was made of wood with fabric cover- The purpose of the rockets was to boost the
looming in the minds of Japanese military ing while the slightly swept wings were of speed of the airclaft during the final, terminal
leaders and planners, several means to repel metal with stressed skinning on the outer dive on the target. With the Ki-l 15 Ko deemed
the invaders were considered, investigated, wing surfaces. The Ki-I15 Ko could accept a acceptable, Nakajima began production of
and in some cases, allowed to proceed variety of radial engines and to simplify the the Tsurugi at both their lwate and Ota plants.
towards a finalisation. One of the ideas devel- installation only four bolts were used to The IJA anticipated that 8,000 aircraft per
oped was to use aircraft for shimbu (suicide) secure the engine to the fuselage. The Naka- month would be assembled from production
missions against the invasion fleet. Airworthy jima [Ha-35] 23 (Ha-25) radial engine was lines scattered throughout Japan.
aircraft of any type were to be thrown against used on the prototype Ki-l15 Ko and would Even with production underway, steps
Allied shipping. In order for the missions to be found on the subsequent production air- were being taken to further simplify the
succeed, wave attacks were envisioned, craft. The pilot was provided with an open Ki-115. To save on precious metals, the wings
involving scores of aircraft. Sheer numbers cockpit with simple instruments and con- of the Ki-115 Ko would be replaced with
would ensure successful hits on naval ships trols. A crude aiming sight was provided as wooden versions and the wing area
and landing craft even in the face of heary well. The landing gear could be jettisoned increased. To better address pilot vision, the
anti-aircraft fire and c.ombat air patrols. Even after take-off, had no suspension outside of cockpit would be moved forwards. The ver-
one aircraft that struck a ship had the poten- the balloon tyres and was made out of pipes. sion of the Tsurugi was to be designated the
tial to cause significant damage. Such mass For weapons, the Ki- 1 1 5 Ko carried only a sin- Ki-l15 Otsu. A variation of the Ki-l15 Ko was
attacks, however, led to the conclusion that gle bomb and this was held in a recess under the Ki-115-lll (also known as the Ki-l15 Hei).
the available pool of aircraft would quickly be the fuselage between the wings. The heaviest The only two modifications was the provision
depleted. Thus, it was clear that an airplane bomb that could be carried weighed 800kg of a bomb release and cockpit being moved
had to be designed that could be built rapidly (1,764Ib) and the bomb had no provision for even further forwards. But even these models
to swell the number of aircraft available for release from the cockpit. would not be the end because the Ki-230, a
these shimptr missions. It was Nakajima that In March 1945, the prototype of the Ki-l 15 further development of the Ki-l 15, was also
would provide one answer. Ko, called the Tsurugi (which means'sword' investigated.
On 20 January 1945, the IJA issued specifi- or 'sabre'), was rolled out and flight testing The IJN, having learned of the new plane,
cations for an aircraft that could be built by commenced. As soon as the trials had started became interested in the Ki-l 15 and sought to
semi-skilled labour, would use very few war problems began to surface. The landing gear produce it for themselves. To facilitate this,
critical materials, had the ability to accept any contributed to poor ground handling and this Nakajima provided Showa Hikdki K.K.
radial engine with a 800hp to l,300hp rating, was compounded by the poor view afforded (Showa Aeroplane Co Ltd) with two Ki-115
was easy to maintain in the field, was able to the pilot. Once in the air, the flight character- Ko aircraft. In IJN service, the aircraft was to
carryat least one bomb and had a maximum istics of the Ki- 1 1 5 Ko were anything but stel- be called the Toka, meaning Wisteria. Showa
speed of at least 340kmih (2llmph) with lar and even skilled test pilots had some was to adapt the design to accept any num-
landing gear and 51skm/h (320mph) without difficulty in flying the aircraft, let alone a pilot ber of IJN radial engines from older, refur-
landing gear. Nakajima was tasked with mak- with minimal training. Nevertheless, given bished motors to ones then in current seruice.
ing the specifications a reality and engineer the mission of the Ki-1 15, flight trials contin- By the time ihe war ended, Nakajima had
Aori Kunihiro was assigned the project. Kuni- ued while modifications were investigated to only been able to produce 104 of the Ki-l15
hiro would have assistance from the Mitaka improve the aircraft. By June 1945, the initial Ko (22 from the Iwate plant and 82 from the
Kenkyujo (Mitaka Research Institute) and Ota flight testing was completed. Two further Ota plant) and none would be used in anger.
Seisakusho K.K. (Ota Manufacturing Co Ltd). changes were made to the Ki-l l5 Ko and this Neither the Ki-115 Otsu, Ki-115-lll or the
Because semi-skilled workers would be involved adding suspension to the landing Ki-230 would be constructed, remaining for-
used to build Kunihiro's aircraft, the Ki- 1 I 5 Ko gear and including auxiliary flaps to the ever as design board projects. Likel,r,ise,
was simplicity itself. The fuselage used a steel inboard trailing edge of the wings. Production Showa had no time to produce the Toka.

In'lpEntel JapaxEss ARlrv 31

Japa:resE Slcrrrt Pno:Ecls: Etprnlttlt"rt,tl Arncnalr on rHe IJA lln IJN 1939-1945

tr Nakajima Ki-l 15 Tsurugi - data

E Contemporaries Survivors
Bell XP-77 (US), Blohm und Voss BV40 (German.-v), Zeppelin 'Fliegende Nakajima Ki-i 15 Ko Tsurugi (FE-156)
cF Panzerfaust' (Germanrl, \'lessenchmitt N'le328 (Germanv). One offour caplured at Nakajima's No.l plant in Ota. Gunma Prelecture,
E ['leserschmitt P 1104 (Germany) this Ki-l i 5 Ko Gerial I 002) uas listed on the ['1Al{A I August 1946 report

as being in storage and was moved to Park Ridge in September 1919.

Spe' ilkutioni in purpnthc:es ate lor lhe Ki-l l5 lt:u Ki-t ljlll and aty
Luckl'enough to sunive the scrap heap, the Ki-1 15 Ko is cunentl.-v in
estimates onlj. The Toha uauld harc had simi[at dimensions though the
storage, unrestored and in poor condition at the Paul. E. Garber iacility
ueights and speeduould hate uaried dependingonthe engine used.
in Suitland-Silver Hill, i\4anland (pictured lelt).

T}?e Special AitackAucrall Nakajima Ki-ll5 Ko

Crew 0ne Apparentlv another surviving Ki-l 15 Ko is beinq restored in Japan but

there are felv, ifan,v details. on luho is restoring the aircraft nor the
Powerplant One Nakajima IHa-351 23, 14-cllinder, air-cooled history of the Ki-1 I5 involved.
radial engine developing 1,1 30hp for lake-off and 980hp al 6,000nr/ in 1991, tlo other Ki'l l5 aircraft irere reporled to be found in.lapan,

l9.o85ll (.ame engine rrar to be u.ed in rhe Ki-l l;0r.u Ki-l l5-l1l one in Kanda and the other in Koganei. \\ho has them and in u'hat
condition is neither knoun nor conllrmed.
Span 8.59m 28.2ft

(0tsuilll) 9.69m 31.8ft

Length 8.53m 28.0ft

(0tsu/lll) 8,53m 28.0ft Nakajima Ki-230 - data (estimated)

Height 3.29m t0.8ft
(Otsu/lll) 3.29m 10.8ft T}?e Suic;de Attack Aircraft
\Ving area 12.39m' 133.4ft. Crew One
(0bu/ilr) 11.49m' 156.0f1

lVing loading 207.99kgim' 42.6lbrft' Powerplant

(0tsu/lll) 181.13k9/m' 37.1 lbrfl' One l4-cvlinder, air-cooled radial engine developing at Ieast 1.100hp.
Polver Ioading 2.3lkghp 5.1 lbihp
(0tsuilll) 2.3lko'hp 5.1 lbihp Dimensions
Span 39,69m 1.8ft

Weights Lenglh 8.47m '27.811

Emptv 1,640k9 3,61 5 lb Height 3.29m 10.8ft

(Otsuilll) l,69{)kg 3,2;lb \\ring area 13.09m' I 4 t.Ofr

Loaded 2,580k9 s,6881b

(0lsurlll) 2,630kg ;,7981b Weights

l'larimum 2,880k9 6,319 1b Empt! l,700kg 3,747tb

Loaded 2,400kg 5,291 lb

trlal speed 5;lknlh 312mph Performance
at 2,800m at 9,i86it \'lai speed 557knl'h 346mph
620knih 385mph at 2.800m at 9,185ft

(0tsurlll) at5,800m at 19,028it Range I.l99km 745 miles

Cruise speed 300kni/h 186mph Ceiling 6.500m 21.325fr

Range l,l99km 715 miles

(Otsu/lll) l,l99km 745 miles Armament

Ceiling 6,500m 21,32511 One 250kg (5i1 lb) or 500kg (l.l 02 lb) bomb
(Otsu/lll) 6,500m 21,32;ft

Armam€nt One 250kg (55 I lb), 500kg ( I, I 02 lb) or 800kg None, A total oi 105 Ki-1 I5 Ko aircralt rvere built but none saw combat
(1,i64 lb) bomb Game ior the Ki-l I 5 Otsu/Ki-l l;-llD No Ki-l l5 Otsu. Ki-i l5lll, Ki-230 or Toka aircrafl $erc construcled.


lupERrnr- .lapallesE ARl,lv JJ

Nakajima Ki-201 Karyu


As a result of the development of the Naka- of the Me 262 and requested reports and pro-
jima Kitsuka for the IJN, Japan's first turbojet- jections for the production of 100 and 500 air-
powered aircraft to fly (see Page 114), craft a month. It was known that the wartime
Nakajima rvas in the position of being the manufacturing capability of Japan could not
leader in the fledgling iet aircraft field. Seek- produce an exact copy of the Me262 and
ing to expand on that position, Nakajima took adaptations '"vould have to be made to
it upon themselves to offer a jet that rvould be accommodate .lapanese capabilities. Naka-
superior to the Kitsuka. This rvas to make the jima sought to provide that answer.
most of rvhat little data was received from The genesis of the Ki-201 took place on 12
Germany on the N'lesserschmitt N4e 262. With January 1945 rvith the formation of the design
the Kitsuka under development for the IJN' team led by Nakaiima engineer Iwao
Nakajimaprovided the IJAwith their proposal Shibuya. Unlike the Kitsuka proiect, from the
for rvhat was to be ihe definitive Japanese outset Shibuya designed the Karyu as a
version of the Me262, the Ki-201 Karyu, the fighter. ln addition, Shibuya realised that the
Fire Dragon. aerodynamics of the Me 262had been tested
Depending on the source, the IJA was or and felt assured that by applying as much of
was not interested in developing its own jet the design of the N4e 262 into the Karytt as r,vas
aircraft. However, evidence supports the fact possible rvould result in an aircraft that would
that the IJA wished to have its own jet-pow- need minimal testing before production was
ered fighter or was looking to have an option staited. This idea was shown to good effect
should the Ki-202 and fighter variant of the in the development of the Mitsubishi J8M
Kiisuka not meet their expectations. In Octo- Syusui.
The Ki-201 depicted here sports the colours of the
ber 1944, ihe Japanese embassy informed the Shibuya had the same access to the Me 262
244th Sentai, one of the more successful Japanese
home defence air units. Germans that the IJA would be the producer information as the IJN. It consisted of sketches

34 JApANESE SrcnEr Puo:Ecrs: ExpEnr:uenrll AIncRnnr or rna IJA AND IJN 1939-1945

IlrprnrnI- JnpnNEse ARlrv 35

Nakajima Ki-201 Karyu- data and drawings of the Me 262A-1 and little else' with the Ne 130s was a maximum 852km/h
Whereas the Kitsuka only bore a superficial (529mph).
Contemp0raries Messerschmitt \1e 262A'la (Germany), Avia S'92 resemblance to the Me262, Shibuya's design For armament, the Karyu was fitted with
Turbina (Czechoslovakia) would seek to match the Me 262 as much as two Ho-l55-ll 30mm cannons and two Ho-5
possible. Shibuya and his team may have had 20mm cannons. On the chance that the IJN
Type Fighter
little, if any, contact with the Kitsuka develop- might acquire the Ki-201, provision was also
Crew One
ers despite being in the same company. The made to use two Type 5 30mm cannons and
first draft of the Karyu nearly matched the two Type 99 20mm cannons. More notable
Trvo Ne 230 arial-florv turbojets rated at 885kg (1,951 lb) of static thrusl
dimensions of the Me 262. However, it fea- was that the Karyrlwas slated to be equipped
tured a straight wing as opposed to the swept with the Ta-Ki 15 airborne intercept radar.
each; later, t1v0 Ne 130 arial-florv turbojets rated at 908kg (2,002 lb) of

static thrust each wing of the German jet. This was quickly Used in coniunction with the Ta-Chi 13
changed and the revised Karyuwas largerand ground control radar, the Karyl could be
Dimensions heavier than the Me262, but replaced the guided to its targets by ground controllers
Span 13.68m 44.9fr
straight wing with a gently swept wing' with a 153km (95 mile) radius. Such a system
Length I L49m 37.7fr
Initial design work, including wind tunnel would have been a benefit in lowJight, night
Heighl 4.05m 13.3ft
testing, was completed in June 1945' For all or poor flying weather interceptions. In addi-
Winq area 23.96m 258|t:
intents and purposes, the Ki-201 was a larger tion to the cannon fits, the Karyu was to be
derivative of the Me 262. That it was bigger capable of carrying a 800kg (1,7631b) or
and heavier than the German jet may point to 500kg (1,102Ib) bomb.
Empt] 4,495k9 9,9l0tb

Loaded 7,021kg 15,478 lb adaptations the Japanese had to make in With the initial progress of the IJN's J8M
Overload 8,492k9 18,722Ib order to produce the Karyrr. For example, the Syusui program, which would provide the IJA
Japanese did not have the experienced fabri- with the Ki-200 and the subsequent IJA Ki-202
Performance cators to make the thin, sheet steel used in Syusui-kai project, IJA interest in the Ki-201
I'lax speed 812kmih 504mph the nose of the Me 262. The result was that the looked to have waned. The result was delays
(Ne 230) at 10,000m at 32,808ft
Karyr-r's nose had to make do with duralumin in further developing the Karyu. Nakajima
845km/h 525mph wanted to have the final design of the Ki-201
',vhich was heavier. In addition, it is certain
(Ne 130) at 10,000m a|32,8081t
that the Karyr1 incorporated simplifications to completed by July 1945 with more advanced
Landing speed l6lkm/h lOOmph testing underway by August. The first proto-
1,006kn1,h 625mph
accommodate production by semi-skilled
Mar dive speed
labour and construction using less critical type of the Karyu was to be completed and
Take-off distance 945m 3,l00it loaded
1,588m 5.209ft in overload war materials. The latter was borne out by the ready for flight trails by December 1945, and in
Range at 60110 thrust 987km 613 miles intense interest by the Japanese in obtaining addition, a further 18 examples of the Ki-201
at 7,995m at 26,2301t the German process for making plywood were to be built and delivered by March 1946.
Fuel capacitv 2.200 to 2,590 litres 560 to 684 sallons (and likely the bonding glues as well) which Despite the delays, work commenced on
Climb (Ne 230) 6 min 54 sec to 6,000m (19,685it) the Germans used in their aviation industry the prototl,pe. Nakajima's Mitaka plant,
(xe 130) 6 min l7 sec to 6,000m (19,685ft)
because, although Japan was lacking in avia- which was located on the western edge of
(Ne230) l4 min 56 sec to 10,000m (32,808f1)
tion metals by the close of the war, they had Toky6, was the facility for the prototlpe
l3 min l5 sec ro 10,000m (32,808ft)
(Ne 130)
ample access to wood. Ki-201's construction. Regular production of
Ceiling 12,000m 39,370ft
the Ki-201 was intended to be carried out at
The Karlu was initially slated to be fitted
with Ne 230 a-rial-flow turbojet engines each the Kurosawaiiri Research Works No.2l situ-
Tlro Ho-l55ll 30mm cannons and tivo Ho-5 2Omm cannonsi one rated at 85kg (1,951 lb) thrust. These were cal- ated near Kitakami, in Iwate Prefecture, in
1,763 lb bomb or one I,I02 lb bomb; proposed \a\ version to be fi tted culated to push the Kary-r at a maximum Honshu. The fuselage for ihe Karyu was
rvith hlo Tlpe 5 30mm cannons and hvo Tlpe 99 20mm cannons speed of 812km/h (504mph). However, itwas nearly complete when Japan surrendered on
also planned that once they became avail- 15 August 1945. With the surrender, work on
Deployment able, the Ne 230 engines would be switched the Ki-201 ceased. Itwould be nearly 30years
NOne. The protot\pe Ki-201 rvas incomplete by the close of the war'
for the improved Ne 130 axial-flow turboiets. before the next Japanese designed and built
Projected to produce 908kg (2,001 Ib) of iet fighter would fly, this being the Mitsubishi
thrust each, the calculated speed of the Karyu F-1 which first flew on 3 June 1975.

36 J.qpeNEse Secnsr Pnorscrs: ExpenrN,rlNral AIRcnapt oF rHE IJA AND IJN 1939-1945
Rikugun Ki-93


7t Ja
-L t.,l,

At the time the Ki-93 was conceived the war ered at first to power the Ki-93 but both mod-
situation for Japan was dire. The mainland els were ultimately powered by two Mit-
was suffering from near daily B-29 raids and subishi Ha-214 18-cylinder, air-cooled radial
looming on the horizon was the anticipated engines, each providing a maximum of
US invasion of Japan. A means to counter the 2,400hp. In order to give the aircraft a mea-
B-29s as well as to attackAllied invasion ships sure of survivability in the face of enemy fire,
was needed. The resulting Ki-93 would be a armour plating was used. The pilot was pro-
first and a last for Rikugun and Japan. vided with five armour plates, each 12mm
When Rikugun Kokugijutsu Kenkyujo thick. Two plates were placed just forward of
began the design research for the Ki-93, the the cockpit in the nose, one on each side of
goal was to provide an aircraft that could pro- the pilot and the fifth would protect his back.
vide a platform for anti-bomber operations The front glazing was composed of 70mm
and anti-shipping missions. In both cases the thick bullet proof glass. The rear gunner was
aircraft had to be able to absorb damage also protected by a 12mm armour plate, offer-
when flying in the face of interceptors, the ing defence from rounds being fired at the
defensive machine guns of the bombers, and Ki-93 from behind. Likewise, the fuselage fuel
the anti-aircraft weapons of ships. tanks were given a measure of protection
Two versions of the all-metal Ki-93 were to from incoming fire via an 8mm thick armour
be constructed. The first, the Ki-93-l Ko, was plate. Each engine was also provided with
the heary fighter that would combat armour plating in the nacelles. Should the
bombers. The second was the Ki-93-l Otsu armour protecting the fuel tanks be pene-
and this was the anti-shipping model. The trated, each tank was self-sealing and, to
Mitsubishi Ha-211 radial engine was consid- prevent fuel fires, had an automatic fire

IupERtar- JapaNesE AnN,ry JI




w z,ffi

\ ffi

38 .lapar-EsE SrcnEr Pno.lncrs: Ext'ERIntEllal AtncRnn oF rHE IJA ,qNn IJN 1939-1945
Rikugun Ki-93 - data extinguishing system. Finally, a defensive Mitsubishi Ki-109 (flown by the 107th Sentai).
armament, consisting of a single 72.7mm The Type 88 had to be hand loaded by the
Contemporaries Ho-103 machine gun was fitted in a rear firing second crewman. In addition to the cannon
Henschel Hs I29B-3.,\\ a (GemanlJ, \'lessenchmitt ['le 110A-]rLl{ position to be operated by the second crew- the Ki-93-l Otsu would carry two 250k9
(Cermanr'). Tupoler, ANT-46 (Russia), North American B.25G Mitchell
man. (551Ib) bombs.
(tS), Bell YFlll Airacuda (LrS), Curtiss XP-71 (ltS). de Havilland
The differencc in the two versions rvas in Rikugun had Dai-lchi Rihugun Kokusho,
\'losquilo FB ['lk.X\rlll (tJK)
the offensive weapon fits, both mounted in located in Tachikawa (r'vhich is about 24
Tlpe Heal'Fighter (Ki.93l Ko) and
ventral gondolas. The Ki-93-l Ko was miles from the centre of Tdkyd), construct the
Ground Atlack Aircraft [Ki-93-l Ohu) equipped with a powerful 57mm Ho-401 can- Ki-93. The first prototype in the Ki-93-l Ko con-
Tuo non and this was backed up by two 20mm figurationwas completed byApril 1945. In the
Ho-5 cannons (although one initial design did same month the aircraft successfully took to
Po$'€rplant away with the two Ho-5 cannons and used a thc air making it the first Rikugun aircraft to be
Tu,o tr{itsubishi Ha-214, l8-olinder, air.cooled radials, dereloping single 37mm cannon with 40 rounds of built and flown. However, further flight testing
2.100hp for take-ofi, l.9i0hp at 1,500mi 1,92011 and 1,73l.)hp at
ammunition). It was anticipated that the was hampered by the war situation, so much
8.1i2mi27,72-qit: each engine drole a ti-bladed. metal propeller
Ho-401 cannon would inflict enough damage so that the test program was never com-
with a single hit to cripple or shoot down a pleted. Despite the lvorsening conditions in
B-29. The Ho-401 could fire 90 rounds per Japan and delays with the flights of the first
Span 18.98m 62.3iI

Length l4.20nr {6.6fl

minute with a muzzle velocity of 518.2mlsec prototype the second aircraft in the Ki-93-l
Height 1.81m 15.9ft
(1,700f1/sec). For the Ki-93-t Otsu, the large Otsu configuration was completed. However,
\\ring area 5.1.74nr 589.3ft, 75mm Type 88 cannon was fitted. The it'"vould never fly.
Wing loading 18.l.80kgim 39.9 lbiir weapon was an adaptation of the Type 88 With the surrender of Japan, the Ki-93
Pouer loading 2,22kq,'hp 4.9lbrhp anti-aircraft gun that had been modified for would become the last hear,y fighter and
use on aircraft. Besides the Ki-93, this ground attack aircraft to be built during the
weapon rvas also used operationally in the war.
Emptv 7,686k9 16,94;lb

Loaded 10,660kg 23,501 lb

Performance c
l\'1a:i speed 625kmrh
at 8,300nr
at 27.230ft
Cruise speed 350kmih 2liniph
Range 3.000km 1.861 miles

Endurance 6 houn
Clirnb 4 min l8 sec to 3.000m (9,810f1)
9 min 3 sec to 6,1)00m (l9,fi8Sfl)

Ceiling I 2.049m 39.;30tt

One 57mm Ho-101 cannon with 20 rounds 0f ammunition, hro 20mm
Ho-5 cannons n ith 300 rounds ol ammunilion per gun and one 1 2. Tmnr

Ho-1 03 machine gun iljth 400 rounds of ammunition (Ki-93.1 Ko); One
75mm lipe 88 cannon, one l2,imn Ho-103 machine gun and tuo
250kg (551 1b.) bombs (Ki-931 Otsu)

\one. Tuo Ki-93 prototipes (one of each lersion) rrere produced but
did not enler producti0n belore the end 0f lhe \rar.

Rikugun Ki-93.1 Otsu (FE-152)

The second protollpe that had not flou'n bv the end oi thc narrvas

taken at uhat is present dal Takahagi in lbaraki Prelecture. The aircrail

was craled and arriled at I'lA['lA being listed on ]0 \'{arch 19{6. It was tcr

be restored for displav and this began on September 1!116. Moied to

Park RiJge on ld 5eplember lu 10. All lrar * ut lh" Lirr:i-l 0l'u .topped
x+-G €E
at Park Ridge in 1949.

IupEnrnl Japa\sse AR\,rv 39

Rikugun K-202 Syusui- Kai

JapaNEsr SEcnEr PRotscrs: ExpEntnEIltnL AtRcnapr on rHs IJA eNo IJN 1 939-l 945
The IJA was not satisfied with the Ki-200 (the then the KR20 was most likely the Japanese Rikugun Ki-202 Syusui-Kai - data
IJA designation for the Mitsubishi J8M1 Syusui development of the HWK 509C motor and the
- See Page 96). They felt that the IJN's plans to answer the IJA was looking for in extending Contemporaries
Messerschmitti\'le l63C-1a (Cermanr)
adopt and adapt the Messerschmitt Me l638 the range of the Ki-202. Contemporary illustra-
as the J8Ml would amount to the same, if not tions of the Ki-202 clearly show some form of
Type Interceptori Fighter
more, effort and development compared to a secondary means of thrust. As a stop-gap
Crew 0ne
creating a new design based on, but not a measure, the Ki-202 could have accepted the
direct adaptation of, the Me 1638. Although the KR10 motor if problems arose with the devel- Powerplant (planned)
IJA attempted to make the IJN see their point- opment and production of the KR20 and thus 0ne Toku-R0.3 (KR20) bi-iuel rocket motor producing 2,000kg (4,4091b)
of-view, the IJN pushed aside such plans, forg- any delays in flight testing could have been 0f thrust $ith supplemenlary r0cket 0r cruise chamber producing 100kg

ing ahead with their J8M program. Thus, the avoided. (880Ib) olthrust

IJA took it upon themselves to design the bet- Although the Ki-202 was larger than the
ter aircraft they had wanted from the outset. Ki-200, no attempt was made to include a Dimensions

The IJA saw a main flaw in the Ki-200 that landing gear system. Like the Ki-200, the Span 9,72m 31,9ft

resulted in aspects of the plane's performance Ki-202 retained a central landing skid, tail Length 7.68m 25.2f1

that they found unacceptable: the limited fuel wheel and would use the Heighl 2.74m gft
iettisonable Wing area 18.39m: l98ll:
capacity. Because of this, combat endurance wheeled dolly for take-off and ground han- Wing loading 272.43k9i11' 55.8ib/ft'
was reduced and fuel was rapidly consumed dling. No provision for catapult launching is
by the KRl0 (Toku-Ro 2) engine. Even with the known to have been considered as a means to Weights
IJN's proposed J8M2, which removed a Tlpe 5 conserve fuel that would have been con- Empty l,6lgkg 3.5691b

30mm cannon to make way for more fuel, the sumed during normal take-off procedures. Loaded 3.384k9 7,4601b

IJA felt that the endurance was still insufficient. For weapons, the Ki-202 was slated to use Nlaximum loaded 5.01 5kg I l,057lb

Starting in 1945, Rikugun Kokugijitsu Kenkyujo two Ho-155 30mm cannons, one mounted in
began the process of developing the IJA's own Performance (estimated by Rikugun)
the each wing root, the same as the Ki-200.
rocket aircraft using the Me 1638 as a template. Insofar as the larger size and motor, the Mal speed 900kmih 559mph

This development was in secret and the desig-

at 10,000m at32.808i1
Ki-202 was estimated to have an endurance of
Landing speed l32krD-h 82mph
nation given to the aircraft was the Ki-202 10 minutes and 28 seconds, whereby the Range i0 min 28 sec ofendurance
Syusui-Kai which meant 'Autumn Water - Ki-202 was calculated to achieve 5 minutes Climb I min 21 sec to 2.000m (6.561|t)
Improved'. and 30 seconds. With a near doubling of the 2 min 0 sec to 4,000m (l3,l23ft)
The IJA took the obvious route and endurance time, this would have allowed the 2 min 34 sec to 6,000m (19,685it)

increased the fuel capacity by stretching the Ki-202 to remain in combat for a longer period 3 min 2 sec to 8.000m (26,246ft)

fuselage to make room for larger fuel tanks. or, at the least, extend its operational radius. It 3 min 26 sec to 10,000m (32,8081t)

They also planned to use an improved motor, was projected that the final design of the Ceiling 12,000m 39,370ft

but exactly what power plant depends on Ki-202 would be completed by February 1945
which source is referred to. Two main options with construction of the first prototype com-
Tn'o Ho-i55 30mm cannon
appear. One was the KR10 as used in the mencing shortly afterwards. The first test flight
Ki-200 that developed 1,500kg (3,3061b) of was scheduled forAugust 1945.
thrust, but with a secondary rocket added pro- As it was, the Ki-202 design would remain None. The Ki-202 did not advance berond the design board.
viding a further 400kg (881 lb) of thrust. The just that, a design. When the war ended, no
other motor was the KR20, which may also be metal had been cut on the Ki202 prototy?e
known as the Mitsubishi Toku-Ro.3. The KR20 nor was a mock-up even constructed. In part,
promised 2,000kg (4,409 lb) of total thrust and the Ki-202 program may have hinged on the
may have been fitted with a cruise chamber. success or failure of the J8Ml. The technical
This is a secondary combustion chamber issues in producing the KRl0 in a reliable form
which was typically mounted above or below most likely stymied work on the KR20, which
the main combustion chamber. The purpose was to be the main powerplant for the Ki-202.
of having two such chambers is that the main The problems with the KR10 delayed flight
one (or both if necessary) can be used for full testing of the J8M1 until July 1945 and even
power needs such as take-off and rapid then, a fuel system failure caused the crash of
ascent, while the cruise chamber has a lower the Syusui during its maiden flight. This set
thrust output and can be employed for normal back the J8M1 further still and although the
cruise speeds once the plane is aloft and the fuel system problem was corrected, the war
main chamber shut off. The benefit of this is ended before any further flights could be
the conservation of fuel, allowing the plane to made. Had the J8M1 succeeded and the IJA
remain airborne and in action longer. version, the Ki-200, entered service, it is likely
Wartime Allied intelligence reports stated that development of the Ki-202 would have rapidly
the Germans had provided data to the Japan- proceeded and had it succeeded, the IJA
ese on the Walther HWK 509C rocket motor would have offered it to the IJN. If accepted,
which used a cruise chamber. lf this was so, the designation would have been the J8M3.

Il,tpERrnr- Japaruesr Anuv 41


42 J.qp,qr,lEsE SEcnor PRorEcrs: ExpnRrunrutnL Atncnnrr on rns IJA ar'"o IJN 1939-1945
Rikugun Kogiken Series

Kogiken Plan I - data (estimated) Without doubt, war often provides for rapid incorporated into designs in the other four
advancements in military technology. The classes. In addition to Kogiken, the IJA's two
Contemporaries Arado P.530 light bomber (Germany), BeJI P-39
key driver of this is the need for a weapon that biggest aircraft providers, Tachikawa and
Airacobra fighter (US), Caudron C.670 light bomber (France), Farman
is superior to those used by the opposition. In Kawasaki, were also invited to participate in
N.C.223 heary bomber (France), Messerschmitt Me 210 and 410

fighter (Germany), Piaggio P.l

the majority of cases, improvements in some of the studies and tender their own
19 fighter (ltaly), Potez 63 heavy fighter
(France), Yakovlev Yak-2 light bomber (Russia).
weapons come about because one partici- designs.
pant fields a weapon that the other partici- A central theme in all of the Kogiken air-
Type Tlpe A Heary Fighter
pants have no answer for or cannot craft was the use of only a select group of
Crew 0ne adequately counter with currently available engines. The two most prominent were the
weapons. This, then, spurs development of Nakajima Ha-45 and the Mitsubishi Ha-211-ll,
Powerplant One Nakajima Ha-45 l8.cylinder, air-cooled radial an equal or better weapon and the cycle both of which were l8-cylinder, air-cooled
engine developing l,480hp for lake-offand l,460hp at 5,800m (19,028ft), repeats - an arms race! In some cases, devel- radial engines. At the time of the Kogiken
driving a four-bladed metal propeller
opment of a weapon does not result from study, these engines were still in develop-
actually encountering new developments on ment but were expected to be operational in
the battlefield. Instead, study of what could the very near future. Another factor concern-
Span 9.32m 30.6fr

Length 8.71m
be encountered in the future, coupled with ing the engines was that, in some cases, the

Height 3,29m 10.8ft

what can reasonably be advanced in terms of designs were built around the ability to
Wing area 14.59m? l5i.t ft?
technology, provides the required emphasis replace the radial power unit with an inline
for development - in essence, an attempt to engine without significant modifications. The
Weights second-guess likely developments so that if primary inline engines were the Daimler-
Empty 2,29lkg 5,0501b and when they manifest themselves the Benz D8601 and DB605. The former was
Loaded 3,l40kg 6,922Ib answer will already be in place. It was the lat- licence built in Japan as the Kawasaki Ha-40.
ter that prompted the IJA's Rikugun Kokugi- By the close of September the designs
Performance jutsu Kenkyujo to initiate a series of studies were nearing completion. Since there were
Ma\ speed 699kn/h 434mph
and plans for a host of aircraft suitable to no performance requirements or specifica-
at 5,800m at 19,028ft
Mar range 599km
meet different roles. tions applied to any of the classes of aircraft,
3i2 miies plus I hour
In the summer of 1941, Kogiken (a con- this essentially left the engineers and
Armament traction of Kokugijutsu Kenkyujo) formed a designers with free rein to come up with
One 37mm Ho-203 cannon or one 20mm Ho-5 cannon and four l2.7mm group headed by aeronautical engineer Lieu- aircraft they thought would be suitable for
Ho-103 machine guns tenant Commander Ando Sheigo. The task the roles. In many cases this led to aircraft
put to Sheigo and his group of engineers was concepts that featured, at least for the
to study Japanese aviation technology in Japanese, unconventional mechanisms and
terms of what was possible at present and in approaches. Each aircraft was designated as
the near future. In addition, some effort was a plan with each plan grouped by their
to be spent on reviewing the aircraft technol- respective class.
Type Type B Heary Fighter
ogy of other countries. From the results of The first class, the hear,y fighter, had four
Crew 0ne these studies, the group was to assemble and main designs, all Kogiken creations. Two
draft proposals for aircraft to fill various roles were for a single-engine aircraft while the
Powerplant One Nakajima Ha.45 lS.cylinder, air.cooled radial that could take advantage of future innova- other two were twin-engine concepts. The
engine developing l,480hp for take-off and l,460hp at 5,800ni/l 9,028f1, tion. In all there were four roles or classes of first was the Kogiken Plan I Type A heavy
driving a four-bladed metal propeller aircraft the Kogiken group had to produce fighter. Outwardly it was a fairly conventional
designs for: heavy fighter, light bomber, aircraft. It used low mounted, thin laminar
heavy bomber and reconnaissance. There flow wings but in order to reduce drag the
Span 9,32m 30.6ft
was a fifth class, that of high speed, but meth- Ha-45 engine was placed within the fuselage,
Length 8.86m 29. lfr
ods to achieve high speed aircraft were often along the aircraft's centre of gravity. This, in
Height 3.68m l 2.1 ft
Winq area 14.59m'? 157.lfl!

Empty 2,295kn1,t' 5,0591b

Loaded 3,205ks i,065lb

Mar speed 699kn/h 434mph
at5,800m at 19,028ft
Ma{ range 599km 372 miles plus I hour
Armament Kogiken Plan I Type A heavy fighter (in the colours J

One 20mm Ho-5 cannon, four l2,7rnm Ho.l03 machine guns and hvo of the 2nd Chutai, 48th Sentai)
7.7mm Type 89 Model 2 machine quns

IuprRrnr- JnpeNEsE Anruy 43

KogikenPlanl-data z
Kogiken Plan I Type A Long Range Heavy Fighter
Tlpe: Trpe A Long Range Hearl Fightet (in the colours of the lst Chutai, l8th Sentai).
Crelv: Trvo

Powerplant: Trvo ['litsubishi Ha-2 II ll I 8.ct ]inder, air-cooled

radial engines developing 2.100hp for take-off, each driving a four

bladed metal propeller

Span: 15,97m i2.1ft
Lengih: I 1.97m 39.3ft

Height: N/A

Wrng Area: 44.99m' 181.3f1

essencer put the motor approximately in the 1.45m(4.7f|) extension shaft, and the cockpit
Empt.v: N/A
middle of the aircraft. The cockpit was situ- was moved to the back behind the wings.
Loaded: 9,100k9 20,231b
ated ahead of the engine, the four-bladed Gone was the tricycle landing gear, replaced
propeller being driven via a 2.5m (8.2ft) long with a more conventional retractable main
\'lax Speed: 699krn'h 434mph extension shaft. For weapons, either a 37mm gear and tail wheel arrangement. The healy
\'la\ Rang€: 1,000km 2,{85miles Ho-203 cannon or a 20mm Ho-S cannon was armament of the Type A was modified for the
Climb: 8 min to 6,000m (19,685i1) to be fitted, firing through the propeller hub Type B with one Ho-5 cannon firing through
Ceiling: ll.500m 37.729f1 with two 12.7mm Ho-103 machine guns in the the hub, four Ho-103 machine guns (two in
nose and one in each wing. Perhaps as an the nose on either side of the engine, one in
oversight, the design of the fighter made no each wing) and two 7.7mm Type 89 Model 2
One 20mm Ho-S cannon and thlee 12.imm Ho-103 machine guns
provision for cooling the internal engine. machine guns situated just forward of the
Mention was made that the fighter would cockpit. A Type 3 radio was also to be
have been equipped with a Type 3 radio sys- included for the pilot.
tem and provision was made to use the antic- The second pair of plans fell into the role of
ipated Nakajima 2,000hp Ha-145 l8-cylinder long range heavy fighters. In both cases, the
radial engine. Finally, the aircraft was to use a aircraft bore a striking resemblance to the
retractable tricycle landing gear. German Messerschmitt Bf 110 although the
The Kogiken Plan I Type B Hear,y Fighter influence may have come from closer to
plan used the same fuselage shape and wings home in the civilian Mitsubishi Otori
as the Type A but this time the Ha-45 engine (Phoenix) long range communication aircraft
was moved towards the nose, needing only a built in 1936 for the Asahi Shimbun newspa-

Kogiken Plan II - data Kogiken Plan III - data Kogiken PIan III - data
Tlpe Light Bombei T}?e Light Bomber Type Revised Light Bomber

Crew Three Crew Three Crerv Three

Powerplant Two Nakajima Ha-39 l8-cylinder, watercooled Powerplant Tno Xakajima Ha-15 18-olindet, air.cooled radial Powerplant

radial engines, each developing l,760hp at 3,000mi 9,812ft ddving a engines, each developing l.160hp at 5,800nrt19,028fi driting a lour Tno Nakajima Ha-45 l8.o'linder, air-cooled radial engines,

lour'bladed metal propeller bladed metai propeller each developing l,480hp driving a four-bladed metal propeller

Dimensions Dimemions Dimensions

Span i4.59m 47.9fr Span l,1,l4m l6,4tt Span 14.14m 46.4it

Lenglh 10.69m 35. I fr Lengih 10.85m 35.6fr Length 9.96m 32.711

Height 3.77m I 2.4ft Height 3.77m i2.4fr Height 3.77m 12.411

Wing area 31,99m' 344.4ft. Wing area 26.59m' 286.3ft, Wing area 26.59m' 286.3ft

Weights Weights Weights

Emptv 5,303kg 1 1,691 lb Emptl' 1,6i Okg 10.163Ib EmptI 4.480kg 9,8761b

Loaded i,5l(]kg 16.556Ih Loaded 6,6tOkg l4.52lb Loaded 6,480kg 14,2851b

Performance Speed SPeed

N{ar speed 649km/h 403mph \'lax speed 7l0km h 44lmph l\'lax speed 7l0kn-!-h 441mph

at 3,000m at 9,812ft at 5,800m at 19.02Sft 5.800m

at at 19,028f1

['la\ range t,400km 869 miles plus J hour ['1a: range l,40Okm 869 miles plus 1.5 hours \'lar tange l,400km 869milesplusl.5hours

fumament Armament fumament

Tno 7.7mm Trpe 89 machine guns; normal bomb load of 300kg (661 lb) Two 7.?mm Trpe 89 machine guns: normal bomb load of 300kg (661 lb) Tn o 7,7mm Trpe 89 machine guns; normal bomb load of 300k9 (661 lb)

to a marimum ol400kg (881 Ib) to a ma'iimum oi400kg (881 lb) to a marimum ol.100kg (881 lb)

44 JepaxEsE Secnsr Pno;ecrs: Expenll,lnNrar- AIRcnlpr oF rHE IJA AND IJN 1939-1945
Kogiken PlanVI - data
i Kogiken Plan V Revised Light Bomber
(in the colours of the lst Chutai, 2lst Sentai). Tlpe Hear--r Bomber

Crew Forrr

&{!;e"e"- Four liitsubishi Ha-21 ]ll 18.cllinder. air-cooled radial engines, each

developing 2,i00hp driling a iouFbladed metal propeller

Span 22,09m 72.51t

Length 17.09m 56.lft

Height 5.12m 16.8tt

Winc area 61.00m' 656,5f1'

per. Bothwere twin-engine designs and fitted evolve from the Ho-S. A variation of the Type
with the Ha-211-ll radial although the Ha-145 B, the Plan II Long Range Fighter, had a
Empt\, 9,0r3kg 19,958lb

Loaded l5,l98kg 33,505Ib

and the Kawasaki Ha-l40 12-cylinder, liquid slightly redesigned wing that reduced the
cooled inline engine were also considered. At wing area from 44.99m' (484.3ft') down to Performance
this time, the latter was a projected develop- 41.99m' (452f1'). A final variation, the Plan III \'lar speed i20kmih 447mph
ment of the Ha-40 engine, which as noted Long Range Fighter, reduced the crew glaz- at 6.000m at l9,68ill
was a licence-built version of the German ing and fitted a remote controlled Ho-103 \'lax range 3,000km 1,864 miles plus 2 hours

Daimler-Benz DB60lA. They each carried a machine gun in the tail.

crew of two. The Kogiken Plan I Type A Long The light bomber plans would feature an Armament
One 7.7mm Tlpe 89 machine gun, one J 2.7mm Ho-103 machine gun
Range Hear,y Fighter carried an armament of approach that would not be repeated in sub-
and fourteen 50kg (l l0lb) bombs, eight l00kg (220lb) bombs, two
one 20mm Ho-5 cannon and two 12.7mm sequent Japanese designs. The Kogiken Plan
250kg (551 lb) bombs or one 500k9 (1,102 lb) bomb
Ho-103 machine guns in the nose with a rear II Light Bomber used two airframes con-
firing Ho-103 for the second crewman. The nected by a wing centre section. The vertical
Kogiken Plan I Type B Long Range Heary stabilisers were connected by a single hori-
Fighter was identical to the Type A but fea- zontal stabiliser. The pilot and crewman sat in
tured a slightly different weapon fit. It a cockpit on the left airframe while a third
retained the two forward firing Ho-103 crewman had a station in the right. The main Kogiken Plan I Type A - data
machine guns but would either add a second landing gear retracted outwards into the
Ho-5 in the nose or swap both Ho-5 weapons outer wing panels. Buried in each airframe T}?e High Speed Hean Bombcr
for two 30mm cannons, perhaps based on was a Nakajima Ha-39, l8-cylinder, water Crew Four

the 30mm Ho-155 that would eventually cooled radial engine that developed l,760hp
Four Mitsubishi Ha.21 IJI lS-c.vlinder, air'cooled radial engines,
KogikenPlanV-data Kogiken PlanV- data each developing 2.100hp driling a fourbladed melal propeller

Type Light Bomber Tlpe Revised Light Bomber Dimensions

Crew Three Crew Trvo Span 826.33m 6,4ft

Length 19.78m 64,9ft

Powerplant Powerplant Height 5,09m i6.7fl

Two Nakajima Ha.45 l8-cylinder, air-cooled radral engines, Trvo tr'litsubishi Ha-21 l-ll l8-cilinder, air-cooled radial engines, Wing area 88.00m' 947,2ft'

each developing 1,480hp drir,ing a fourbladed metal propeller each developing 2,l00hp driling a lour-bladed metal propelier Wing loading 2l9.7kgim' 45lbril'
Poiver loading 2.76k9hp 6.1 lbihp
Dimensions Dimensions
Span l4.l4m 46.4fr Span 14,23m 46.7|t Weights
Length 10.48m 34.4fr Length i0.63m 31.9ft Emptl 12.510k9 27,645lb

Height 3,77m 12.4fi Height 3.71m 12.2t1 Loaded 19.310kg 42,637]b

Wing area 26.59m' 286.3ft, Wing area 26.59m' 286.3ft. Useful load 6,800k9 14,991 lb

Weights Weights Performance

Empty 4,480m 9,8761b Emptr 4,138k99,7841b \'lar speed 684kmft 125mph

Loaded 6,18Oks 14,285lb Loaded 6.445kg/14,208 1b at 8.500m at 27,887it

Cruise speed 5l5knh 320mph

Performance Performance at 8,500m at 27,887f1

Mar speed 71Oknh 44Jmph \'lal speed 746krn,h 463mph Climb 3 min I sec to 3,500m (l l.l82lt)
at 6,000m at 19,685ft at 6,000m at 19.685ft 6 min 3 sec to 5,000m (l6.40llt)
['lar range l.400km 869 miles plus 1,5 hours \'la,r range l,400km 869 miles plus 1,5 hours ['lar range 3,000km/1,861 miles

Armament fumament Armament

Trvo 7.7mm Tlpe 89 machine guns, 300kg (661 lb) in bombs One 7.7mm Tlpe 89 machine gun and 250kg (551 lb) in bombs Two 7.7mm Trpe 89 machine guns. one l2.7mm Ho-103 machine gun

and up to l,500kg (3.306Ib) ofbornbs

IupERrar- JapaNEse Anuv 45

Kogiken Plan I Type B - data and drove a four-bladed propeller via an war Japanese light bombers, the armament
extension shaft. It was proposed that a sur- was light consisting of two 7.7mm Type 89
Type High Speed Heary Bomber face evaporation cooling system be used machine guns. For bombs, a regular payload
Crew Four
which consisted of a network of piping that of 300kg (661 lb) could be carried with a max-
would take the steam produced by the imum of 400kg (881 lb).
Powerplant four Mitsubishi Ha-21 l-ll l8-cylinder, air-cooled radial
The Kogiken Plan III Light Bomber was
engines as the water circulated through them
engines, each developing 2,l00hp driving a four-bladed metal propeller
out into the wings where the cooler airwould basically identical to Plan II. The main change
condense the steam back into water that was was the replacement of the Ha-39 engines
Span 28.19m 92.5ft
then recycled through the engines. The main with the Ha-45 and the latter were situated
Length 20.66m 6i.8ft benefit from this method of cooling was that deeper into the fuselages requiring a .83m
Height 5,18m I 7fr it allowed for a more streamlined fuselage (2.7ft) long extension shaft. Without the com-
Wing area 99.00m' 1,065.6ft' which increased air speed. Typical of early plex cooling system of Plan II, Plan III was
Wing loading 221.66kg/m' 45.4 rb/fr,

Power loading 3.l2kg/hp 6.9tb,hp

Empty l3,0sOkg 28,858Ib

Loaded 2t,985kg 48,468Ib

Useful load 8,895k9 l9,6l0tb

Max speed 660krnh 4lOmph

at8,500m at27,88ift
Cruise speed 495km,/ir 307mph

at8,500m at27,887ft

Climb 4 min 8 sec ro 3,000m (9,842ft)

8 min 0 sec to 5,000m (16,404ft)

Max range 4,000km 2,485 miles

Kogiken Plan III Revked Light Bomber.

fumament Two 7.imm Type 89 machine guns, one l2.7mm

Ho-103 machine gun and up to l,500kg (3,306|b) ofbombs

Kogiken Plan II - data

Type High Speed Heary Bomber

Crew Five

Powerplant Four Mitsubishi Ha-2 I l-ll I 8-cylinder, ait-cooled radiai

engines, each developing 2,l00hp driving a four-bladed metal propeller

Span 27.09m 88,9ft

Length 20.54m 67.4f1

Height 5.33m l7.5ft

Wng area 98,00m' 968.7ft1

Wng loading 219.70kg/m: 4slb/tf

Power loading 2.8lkg,/trp 6,2lb,hp

Empty 12,805k9 28,2301b

Loaded 21,985k9 48,468tb

Uselul load 6,975kg l5,3ntb

Max speed 680krnh 422mph

at 8,500m at 27,887ft

Cruise speed 509krilh 3l6mph

at 8,500m at 27,887ft

Climb 4 min 0 sec to 3,000m (9,842ft)

6 min 4 sec ro 5,000m (16,404ft)

Max range 3,000km 1,864 miles

Armament One i.imm Type 89 machine gun, three l2.7mm

to l,500kg (3,306 lb) of bombs
Ho-103 machine guns and up J

46 JapeNssr SecRnr PRo.lecrs: ExpERruEnrrl ArncRnnt oF rHE IJA AND IJN 1939-1945
smaller, lighter and faster. In addition, the In the heary bomber class, only one stan- TachikawaPlanl-data
canopy for the third crewman was made dard healy bomber design was completed by
T}?e High Speed Hearv Bomber
flush with the fuselage and was provided with Kogiken and this was the Kogiken Plan VI
Crew Six
a ventral fairing to facilitate bombing or Heavy Bomber. The fuselage was conven-
reconnaissance duties. The propellers were tional and was reminiscent of the solid nosed
Powerplant Four Nakajima Ha'145 l8-cllinder, air'cooled radial
fitted with ducted spinners to help cool the Junkers Ju88 bombers although the Plan VI engines, each developing 2,000hp driving a four-bladed metal propeller
engines. The Kogiken Plan III Revised Light had longer glazing over the compartment for
Bomber was a version of the Plan lll but with the crew of four. The similarities ended there. Dimensions
the Ha-45 radials placed in a conventional Set into each wing was a nacelle that housed Span 24.47n 80.3f1

fashion in the nose - the ducted spinners two engines driving two propellers in a push- Length 1i.67m 58lt

were left out. Another light bomber, the pull configuration. Originally Ha-45 radial Height 4,45m 14.6ft

Kogiken Plan V Light Bomber, shared a simi- engines were planned but the Ha-211-ll was Wing area 85.00m' 9l4.9ft'

the engine of choice. The design could have \\iing loading 223.6lkgim: ,15.8lb/fl'
larity to the Plan III Revised but whereas the
Porver loading 3.19kgihp 7,71b/hp
twin airframes were of the same length, been adapted to take the forthcoming Mit-
the Plan V had the right airframe shorter to subishi Ha-214 18-cylinder air cooled radial or
the point that the propeller spun behind the the Kawasaki Ha-201 which was made from Emptt' 9,540k9 21,032 lb

one on the left airframe. The canopy for the two Ha-40 12-cylinderinline engines mounted Loaded 19,000k9 41,887 lb

pilot and second crewman was also length- in tandem and was under development at the Useful load 9,460k9 20.8551b

ened. The Kogiken Plan V Revised Light time. Defensive armament was light consist-
Bomber broke away from the previous ing of a single 7.7mm Type 89 machine gun Performance

designs. The twin airframe scheme was and a 12.7mm Ho-103 machine gun. A vari- l\'lai speed ijS4knL'h 425mph

dropped and an engine was placed in a large able bomb load could be carried depending at9,300m al30,51lft

nacelle in each wing. The Ha-211-ll engine on the mission. A slight variation of the Plan VI Cruise speed 480krn'h 298mph

at 9,300m ai 30,51 lft

was specified but the Ha-45 could also be increased the wing area to 69.99m': (753.4f1').
Climb N/A
used. The crew was reduced to two and the There was far more activity in the high
\'lari range 4,899km 3,041 miles
armament was dropped to a single 7.7mm speed healy bomber category. Here,
Type 89 machine gun while the bomb pay- Tachikawa and Kawasaki made proposals as Armament
load was lowered to 250kg (551 lb). well as Kogiken. The Kogiken Plan I Type A Trqo 7.7mm Ttpe 89 machine guns, one 2Omm Ho-5 cannon and up to
High Speed Healy Bomber was similar to the l,500kg (3,306lb) of bombs

Plan lV Healy Bomber but was larger. Instead

of a conventional empennage, the Plan I Type
Kawasaki - data
A used a horizontal stabiliser ending in
Type High Speed Healy Bombet rounded vertical stabilisers. The glazing for
Crew Four to Five the four man crew was longer and the fuse-
lagewas more streamlined. The same engine Tachikawa Plan II - data
Powerplant and engine arrangement was used but the
Tne High Speed Hean Bomber
Tu'o Kalvasaki Ha-110 24-cvlinder, liquid-cooled inverted V engines, nacelles had a slightly improved shape. For
Crew N/A
each developing l.350hp driving a iourbladed metal propeller
defence, two 7.7mm Type 89 machine guns
and one 12.7mm Ho-l03 machine gun were Powerplant Four Nakajima Ha-115 l8'cllinder, air'cooled radial
fitted and a maximum bomb load of 1,500kg
Span l9.l7m 62.91t engines, each developing 2,000hp driving a four-bladed metal propeller
(3,306Ib) could be carried. The Kogiken Plan
Length 14.87m 48.8ft

Height N/A I Tlpe B High Speed Heavy Bomberwas basi- Dimensions

Wing area 48.00m: 5l6.6ft: cally identical save it was larger and had a Span 28.49m 93,5ft

Wing Ioading 199.69k9m' 40.9lb/fr longer range. The final design from Kogiken, Length N1A

Polver Ioading 3.40k9hp i,Slb/hp the Plan Il High Speed Heavy Bomber, was Height N/A

slightly smaller and lighter than the Tlpe B but Wing area 80.00m' 861.1ft'

Wing Ioading 223.61kg/m) 45.8lb/ftr

Weights added a fifth crew member and beefed up
Polrer loading 3.3lkgihp 7.31b/hp
Empty 6,1 70kg 13,6021b
the defensive armament to three Ho-103
Loaded 9,590k9 21,142tb
machine guns and one Type 89 machine gun. Weights
lxeiul load 3,420k9 7,5391b
The Kawasaki High Speed Healy Bomber Empty N/A

was the smallest of all the designs in the class' Loaded 17,900k9

Mar speed 580kn-/h 360mph Consequently, its performance was less and Lseiul load N/A

at 6,000m at 19,685ft carried a lighter bomb load and defensive

6l Okn h 379mph armament. Instead of the Ha-21 1-ll engines it Performance

at 7,500m at 24,606ft used two Ha- 1 40 24-cylinder, inverted V inline Max speed 694knr,t 43lmph
at 9,300m llt
Cruise speed 400krdh 248mph engines, which again were two Ha-40 at 30.51

Cruise speed 490kmih 304mph

al 7,500m at 21,606f1
engines placed in tandem. It was also unique
N/A at9,300m at30,51lft
in using the 7.92mm Type 98 machine gun in N/A
['lax range 3,000m 1,864 miles
its defensive armament fit. Tachikawa's Nlari range 5,049km 3,137 miles
designs, the Plan I, Plan II and Plan III High
Armament ltuo 7.92rn6 T1,pe g8 machine guns, one l2.7mm

Ho-103 machine gun and up to l,000kg (2,204 tb) of bombs

Speed Healy Bombers, were all variations on Up lo l,000kg (2,204Ib) olbombs
the same theme with minor differences

IupERIel JnpaNEss ARIaY 47

Kogiken Plan IV- data between them. The key changes from the Reconnaissance Plane was essentially the
Kogiken plans were the engines used. The Kogiken Plan III Revised Light Bomber
Type Reconnaissance Plane Plan I and Plan II bombers used four Ha-145 adapted for the reconnaissance role. Not sur-
Crew Two radials while the Plan III used only two of the prisingly, the Kogiken Plan IV Revised Recon-
Mitsubishi Ha-21 IMB 18-cylinder, air-cooled naissance Plane was derived from the
Powerplant Two Nakajima Ha-45 l8"cylinder, air-cooled radial
radial engines. Of all the designs, the Kogiken Plan V Revised Light Bomber. The
endnes, each developing l,480hp driving a four-bladed metal propellet
Tachikawa Plan I had the largest crew com- Kogiken Plan MII High Speed Reconrnis-
pliment at six men. sance Plane was based on the Kogiken Pbn
Span 14.47m 47.'ft In the reconnaissance class, three primary M Heavy Bomber but had a highly strearrr
Lenglh 10.33m 33.gft designs emerged, all by Kogiken. Each of lined fuselage with nearly flush glazing orrcr
Height 3.74m 12.3fr them simply followed trends set in the fighter the crew compartment. As was typical of
Wng area 30.00m' 301.3[r, and bomber classes. The Kogiken Plan IV early Japanese reconnaissance aircraft tlre
Wing loading N/A

Power loading N/A

Empty 4,59lkg l0,l2l lb
Loaded 7,367kg 16,241 lb

Useful load N/A

Max speed 705km/h 438mph

at6,000m at 19,685ft

Cruise speed N/A

Climb NiA
Max range 3,000km 1,864 miles plus 2 hours

One 7.7mm Type 89 machine gun

Kogiken Plan IV Revised - data

Type Reconnaissance Plane

Crew Two

Powerplant Two Nakajima Ha-45 l8-cylinder, air.cooled radial

engines, each developing l,480hp driving a four-bladed metal propellet

Span 14.47m 4l.5lt
Length 10.69m 35.lft
Height 3.59m 11.8ft

Wing area 30.00m'? 30l.3ft'

Mng loading N/A

Power loadinq N/A

Empty 4,426Yq 9,7571b

Loaded 7,202W 15,8771b

llseful load N/A Koglken Plan II High

Speed Heary Bomber
Performance (in the colours of the
Max speed 7l5km/h 444mph lst Chutal, 83rd Sental).
at6,000m at 19,685ft

Cruise speed N/A

Climb N/A

Max range 3,000km 1,864 miles plus 2 hours

One 7.7mm Type 89 machine gun

48 JApeNEss SEcnrr PRo;Ecrs: ExpERttrlEntnL Atncnnpr oF rHE IJA AND IJN 1939-1945
defensive armament was extremely light for sume that the information provided valuable Tachikawa Plan III - data
all of these planes. data and worthwhile methods for subse-
None of the Kogiken aircraft would be con- quent IJA aircraft. For example, the Kawasaki Type High Speed Heary Bomber
structed. In part, this was due to the designs Ki-64 would use the Ha-201 coupled engine Crew

using technology that was either in develop- with a surface evaporation cooling system.
ment or not yet available. Another factor was Another example was the Kawasaki Ki-88
Tlro l'litsubishi Ha-21 iNlB l8-c,llinder, air-cooled radial engines,
the unorthodox nature of many of the aircraft which was to use a fuselage-buried Ha-I40 each developine 2,200hp drivine a fourbladed metal propeller
drafted - one might liken the Kogiken aircraft engine that drove the propeller via an exten-
to the results of a brainstorming with minimal sion shaft. The Mitsubishi Ki-46-lll (code- Dimensions
restrictions on what might be deemed possi- named Dinah by the Allies) may have also Span 24.38m 80ft
ble. Nevertheless, the studywas not awasted benefitted from some of the aerodynamic Length l7.58m 57.71t

effort and it is certainly reasonable to pre- streamlining studies done by Kogiken. Height 4,81m 15,8ft

Wing area 66.00m' 710.4ftr

z Wing Ioading 219.70k9/m: 45lb/ftj

- Polver loading 9.1 lbihp

Empty 8,075k9 17,8021b

Loaded 14,500k9 31.967 lb

Useful ioad 6,234kg 13,743 lb

['lar speed 64lknth 398mph

at 8,500m at 27,887f1

Cruise speed 499km/h 3lOmph

at 8.500m at 27,8871t
Kogiken Plan VIII High Climb N/A
Speed Reconnaissance
tr'lar range 4,999km 3,106 miles
Plane (in the colours of the
.:ti 2nd Chutai, Slst Sentai).
Armament Tlvo 7,7mm Type 89 machine guns, one l2.7mm
Ho-103 machine gun and up to 1.500kg (3,306 lb) of bombs

Kogiken Plan IV Revised - data

Type Reconnaissance Plane

Crew Trvo

Four Mitsubishi Ha-21 l-ll l8-cylinder, air-cooled radial engines, each

developing 2,l00hp driving a four-bladed metal propeller

Span 17.67m 58ft

Length l3.l0m 43ft

Height 3.56m I l.7ft

Wing area 61,00m' 656,5ft

Wing loading N/A

Pon'er loading N/A

Emptv 7,420k9 16,3581b

Loaded 9,9i0kg 21,8471b

llsefrrl load N/A

Mar speed 775kn/h 48imph
at 6,000m at l9,685ft
Cruise speed N/A

CIimb N1A

Nlar range 3.000km 1.864 miles plus 2 hours

Deployment \one olthe Rikugun Kogiken plans progressed

be\'ond the planninq staqe.

IuprRrnr- J.rparussr ARl,ry 49

Tachikawa K-74


i ii1,-

, ..I,:rtt,:::_:t::t:,1-;ila

In 1939, a specification was drawn up that Koku Hombu's specifications, drawn up by conditions like hypoxia, barotrauma and alti-
called for an aircraft capable of conducting the design team led by Dr. H. Kimura. The ini- tude/decompression sickness. Two designs
long range reconnaissance and it was tial design for the Ki-74 was drafted in 1939. were built that would test pressure cabin con-
Tachikawa that answered the call to provide To achieve the required performance, cepts. The first was the Tachikawa Ki-77.
such a plane. However, the resulting Ki-74 Kimura selected two Mitsubishi Ha-214M Conceived as a plane to make a non-stop
would find itself both a victim of development radial engines. Each engine developed flight from Toky6 to New York, work on the
delays and the changing fortunes of war. 2,400hp and would drive a six-bladed pro- Ki-77 began in 1940. For the Ki-77, Dr. Kimura
In the spring of 1939, the Koku Hombu issued peller. It was the pressure cabin for the Ki74 utilised a sealed crew cabin but one which
a specification for a long range reconnais- that would be the stumbling block to finalis- was not pressurised. The belief was that the
sance aircraft that could muster a range of ing the design. cabin would keep in the oxygen but in testing
5,000km (3,107 miles) and a cruise speed of Even before the Ki-74 was conceived, work it failed to meet expectations and the crew
at least 280mph. The reason for this request on developing a pressure cabin for use in had to wear their oxygen masks constantly.
was to provide a plane capable of operating high-altitude aircraft was being conducted by With the poor showing of the Ki-77 cabin,
from bases in Manchuria and flying to the Tachikawa. The purpose of such a cabin was Tachikawa tested a fully pressurised cabin in
west of Lake Baikal. The lake, the deepest in to maintain air pressure for the crew when the Tachikawa SS-1. The SS-1 was a modifi-
the world, is located north of Mongolia, near operating at altitudes in which the outside air cation of the Army Type LO transport which
the southern Siberian city of Irkutsk. Certainly pressure is much lower and the air thinner. itself was a licence built version of the Lock-
the intent was to monitor Russian and their Cabin pressurisation is desired for aircraft fly- heed Model i4 Super Electra (codenamed
allied Mongolian forces, especially in the face ing higher than 3,048m (10,000ft) and doing Thelma by the Allies). Work on the SS-1 also
of Japanese defeats at the Battle of Lake so provided the crewwith a much more com- began in 1940 and the one and only example
Khasan in 1938 and the Battle of Khalkhin Gol fortable working environment that did not was completed in May 1943 with subsequent
in 1939, both at the hands of the Soviets. require the use of oxygen and flight gear to testing providing excellent data on pres-
Tachikawa submitted a proposal to meet the protect against the cold. It also prevented srrrised crew cabins.

50 Jep,qNtEss Sscntr PRo:Ects: ExprRtluENral AIRcRlpt or rHe IJA,qNo IJN 1939-1945


The research into pressure cabins and the

construction of the Ki-77 and SS-1 delayed the
Ki-74 project so badly that work on it was
halted; once it became obvious that the cabin
for the new plane would not be ready the
entire project was shelved. No prototype was
However, towards the close of 1941, the
Ki-74 project was resurrected in order to
investigate the possibility of adapting the
design to suit the role of a long range, high alti-
tude bomber and reconnaissance platform.
One such bombing mission envisioned was
against the United States. To meet such a
requirement, adjustments had to be made to
the initial reconnaissance-only Ki-74 design.
Armour was incorporated along with self-
sealing fuel tanks.In addition, the appropriate
apparatus for bombing was introduced.
Finally, the initial Ha-214M radial engines
were replaced by two Mitsubishi Ha-211-l
radial engines, each producing 2,200hp.
Once the redesign was completed it was pre-
sented to the Koku Hombu who approved it
in September 1942 ordering three prototypes.
It was not until March 1944 that the first
Ki-74 prototype was completed, but the other
two aircraft were ready soon afterwards. The
latter two aircraft differed from the first Ki-74
only in the fact that they used the Mitsubishi
Ha-211-l Ru engines that incorporated turbo-
superchargers. With the completion of the
authorised batch ofthree aircraft, flight testing
of the Ki-74 began. Handling was considered
acceptable but the Ha-211-l and the Ha-211-l
Ru engines were proving to be temperamen-
tal and prone to mechanical problems. A fur-
ther thirteen pre-production aircraft were
ordered, but due to the severe problems with
the engines it was decided to replace them
with Mitsubishi Ha-104 Ru radials. While the
Ha-104 Ruwas more reliable italso developed
less horsepower, in fact only 2,000hp.

IvprRrar- JapalesE Ann'rv 5l

Tachikawa Ki-74 - data The Ki-74 as built was a mid-wing mono- ficient to fly from Tokyo to Berlin. In addition,
plane. The pressure cabin made up the all non-essential equipment was removed
Contemporaries Dornier Do217P-0 (German,v), Dornier Do3l7 maiority of the front of the aircraft with the and the flight deck was reallocated to the
(Germanv). Focke-Wulf Fw I 9 I (Cermany), Junkers Ju 86P and Ju 86R
bomb bay underneath. The use of the pres- middle of the fuselage and lengthened. The
(Cermanv), Junken Ju 388 Sttirtebeker (Germanv), BOK'l I (Russia),
sure cabin necessitated smaller window modified Ki-74 would never make such a
De Harilland l\'losquito B,Nlk XVI (UK)
arrangements (as opposed to more conven- flight as Germany surrendered to the Allies
tional glazing) and the flight deck was offset before it could make the first attempt. The fif-
Type Hilh-Altitude, Long-Range Reconnaissance/Bombet
to the port side of the fuselage. The aircraft teen Ki-74s were never to see combat, in part
Crew Five
had a crew of five which consisted of the because the flight testing of the twelve
Powerplant I First Prototlpe I Trvo l'litsubishi Ha'2 I I J, I 8- bombardier, pilot, co-pilot, navigator/radio remaining pre-production aircraft was not
cvlinder, air-cooled radial engines rated at 2,200hp ior take'off, 2,070hp operator and gunner. All were housed within completed before the end of Japan's part in
at 1,000nr3,280ft and 1,93Ohp at 5,000d16,405ft1 ISecond and Third the cabin with the bombardier in the nose, World War 2.
Prototlpes and Fourlh Pre-production Ki-741 Two N'lilsubishi Ha'21 ll Ru the pilot and co-pilot on the flight deck and Prior to the end of hostilities, two further
l8.cvlinder, air-cooled radial engines rated at 2,200hp for take-ofi, the navigator/radio operator and gunner in variants of the Ki-74 were proposed. The first
2,070hp at 1,000rry3,280ft and l,72Ohp at 9,500n13 l, I 70ft; I Remaining
stations behind the flight deck. Because the adapted the Ki-74 to a transport role but this
aircraftl Tn'o l{itsubishi Ha-101 Ru l8-cvlinder, air-cooled radial engines
Ki-74 was expected to operate at high- was soon discarded. The second was the
ratcd at 2.000hp lor take-off, l,900hp at 2.000n'6,560ft and l,750hp at
altitude where interception would be diffi- Ki-74-ll which would have been a dedicated
6,00Onr/19,68ift; all engines spun lout"bladed propellers
cult, the plane carried a very minimal arma- bombing platform. As such, the Ki-74-ll was a
ment consisting of a tail mounted l2.7mm redesignwhich showed some big differences
Span 26.97m 88,5fr Ho-103 machine gun that was fired by the to the Ki-74. The foremost alteration was the
Length I /.b4m 57.9ft gunnervia remote control. Although the Ki-74 pressure cabin which was smaller and kept
Height 5.09m l6.7ft was not a dedicated bomber, it carried a forward of the wings within the fuselage. This
\\jing area 80.00m' 86l.llfr, payload of 1,000kg (2,205 lb) of bombs which allowed for a deeper bomb bay that was
lVing loading 242,65k9m 49,7lbifr:
was comparable to IJA bombers then in needed to carry the planned 2,000kg
Porver loading 4.39kghp 9.7lb/hp
service. (4,4101b) bomb load. Due to the heavier
As construction commenced on the pre- weight, the operating range was estimated to
production Ki-74s, plans were formulated to be 7,144km (4,439 miles) (in comparison to
the 8,000km/4,971 miles of the Ki-74).
Empty 10.200k9 22,1871b

Loaded 19,400kg 42,770tb

put them to use. When a number of Ki-74s
had been built, they were to be assembled Because of the heavier bomb load, the
Performance into shotai (a flight comprised of three air- Ki-74-ll was to be supported on a twin tyred
Mar speed iT0knxh 351mph craft) and massed to conduct bombing mis- front landing gear. The wings for the Ki-74-ll
at 8.500m at 27,890it sions against the US airbases in Saipan from were to be more slender than the Ki-74 for
Cruise speed 101kmi'h 249mph which B-29 Superfortresses operated. high-altitude operation and instead of the sin-
al 8,000m at 26,21iit
While the bulk of the Ki-74s were con- gle tail mounted machine gun, two machine
Range 8,000km 4.971 miles
structed as bombers and reconnaissance guns or cannons were to be used, Two Mit-
l7 min to 8,000m (26.245f1)
platforms, another task was proposed. With subishi Ha-l04-Ru engines would power the
Ceiling 12,000m 39,370fr
much of the communication between Japan bomber, but like the transport concept, the
and Germany severed by the Allies, especially Ki-74-ll was abandoned and did not advance
Armament One remote-controlled l2,7mm H0'103 machine
qun in the taili l,000kg (2,205Ib) ofbombs by sea, a plan was made to use a Ki-74 for beyond the design stage.
non-stop flights to and from Germany. In Thanks to its long development history the
Deployment Did not see senice. 16 buill (3 prototlpes. 1 long 1944, the fourth pre-production Ki-74 was US was aware of the Ki-74. The main thing
range transc0nlinental aircrait. l2 pre-production aircrall), removed from the line and underwent modi- they did not know was the role. Thinking that
fications to allow it to accomplish such a feat. the Ki-74 was a fighter it was assigned the
Survivors Higher powered Ha-211-l-Ru engines were codename Pdf. It was not until May 1945 that
Tachikawa Ki-74 (FE-2206)
utilised and fuelwas carried to enable the air- the true role of the Ki-74 was discovered and
This luas one olfour total K-74 b0mbers captured al the end 0f lhe r|ar,
craft to fly up to 12,000km (7,456 miles) - suf- so the codename was changed Io Patsy.
Two rvere taken at Tachikala's facton lvhile the remainderrvere

obtained from the IJA's test centre at Tama Airfield (nowYokota Air
Base in Fussa, Japan). All lour Gee belolv) rvere listed on the l0 \'larch
1916 report. FE-2206 rvas listed as being a! AOAMC on I August 1946 but
lvas later relegated for disposal.

Tachikawa Ki-74 (FE-2207)

This example was at [']A\'lA in storage as listed on the I Augusl 1916

manifest. It $as later slated fol transfel to the Park Ridge storage facilit\
in September 1946 but no furlher trace oi FE-2207 remains and rvas

most likelv scrapped.

Tachikarva Ki-74 (FE-2208)

Like FE-2206, this Ki-74 uas housed at AOA\'IC 0n I August 1946 and

mei the same fate.

Tachikau'a Ki-74 (FEJ209) I

F8"2209 rvas the third Ki-74 kept at AOAMC and joined the previous hro
aircraft on the scrapheap. z

52 Jepnlrlsr SrcnEr PRo:Ects: ExpERIuENtaL ArRcRanr oF THE IJA AND IJN 1939-1945
Tachikawa Ki-94-I


rq# l

In the summer of 1942, the Koku Hombuwas lage in a push-pull configuration. The key into each wing while a 37mm Ho-2044 can-
formulating specifications for a fighter that advantage this offered was a reduction in non was situated directly below the Ho-155,
supported a healy armament and could drag over a more conventional, wing installed in the tail boom. If required, the
operate at high altitude. After settling on the mounted engine arrangement. Secondly, the Ki-94-l would be capable of carrying up to
specifications, the Koku Hombu approached centreline thrust symmetry of the aircraft 500kg (1,102 lb) of bombs. The Ki-941 was to
Tachikawa and Nakajima and asked each would be maintained in case of engine failure use a pressurised cockpit for the pilot and fea-
firm to produce a design to meet these which, in turn, allowed for nearly no loss of tured a tricycle landing gear.
requirements. control. Each engine was to drive a four- Once the Ki-94-l was finalised, construction
The specifications for the fighter included a bladed propeller. A twin-boom arrangement began on a wooden mock-up to be com-
maximum speed of 800km/h (497mph) and a was mated to the low mounted wings. For pleted in October 1943. Tachikawa then
range of 3,000km (1,864 miles). The Koku weapons, a 30mm Ho-l55 cannon was fitted invited representatives from the Koku
Hombu knew the demands were high, per-
haps even impossible to obtain, and so asked
Tachikawa and Nakajima to put forward pro-
posals to meet the demands. But they handi-
capped Tachikawa by allowing Nakajima to
ignore the range requirement. As such, Naka-
jima could concern themselves with making
their design fast without worrying about how
far it could operate. Undeterred, Tachikawa's
designers set about the task of coming up
with a concept that would achieve what the
Koku Hombu asked for. What resulted was a
departure from the conventional.
It was decided that two Mitsubishi Ha-211 E
Ru (Ha-43 Ru), 18-cylinder, air-cooled radial v
engines should be used placed in the fuse-

Il,rpsnrnL JapaNssu ARN,rv 53

Tachikawa Ki-94-l - data
Coniemporaries c
,lunkers EF I l 2 [Germanv), upolev A\T-23 (Russia)
l z
Because tlte Ki-911 uas nercr buill, the specificltions arc based ot1

Tadtihaua's final rlesign plans and estinlated perfonnance,

T}?e Healr Fighter

Crew One

Porverplant tu'o llitsubishi Ha-21 I Ru (Ha.43 Ru) l8-cvlindcr,

air-coolcd radial enginesi rated at 2,700hp at l.r00mll.921ft. 2,800hp at

2.800mi9,186ft and 1,7;0hp at 10,;00m'34.448f1; each engine drole a

lourbladed. VDM propeller. lhc fronl hal ing a 3.3m (10.8) diametet, the

iear a 3.4m (l 1.1 it) diameter

Span 12.U6m 42,211

Lenqth 1;l.04ni 12.8ft

Heisht 3.8{m l2.6fl

\\ring area 36.99m 398 2ft'

\\'rng loadinl 237,7ik{'m {8.ilbrlt

Pouer loading 2.4!lkq'hp 5.5 lbrhP

Empt"v 6.500ke 11,330lb

Loaded 8.800kg 19,400lb

\'1a\ speed 7Slknr h 185mph

at 10.000rn at 32,810ft

Range i.r20km 9ll miles

Mar range 2,Jl9km l,r6irmiles

Endurancc 2.i hours
l\'1ar cndurance 5 hours

Climb 9 min 56 sec to 10 000m (i12,80811)

Ceiling 11,0001im lr.93lft

Tuo 57mm Ho-101 cannons, t\\'0 37mm Ho-201 cannons, hro 30mm Hombu to visit and inspect the Ki-94-1. On gun Kogiken Ki-104. To bcost the armament,
Ho-15 cannons and up to 500kg (1.102]b) ofbombs inspection and review, Tachikawa was to be two 57mm Ho-401 cannons were added.
disappointed when the design was rejected Unfortunately, this design was also rejected.
outright. The Koku Hombu inspectors found With the rejection of the high altitude fighter
None. Did not advance past a mock-up
the Ki-94-l to be too unorthodox, too complex and the subsequent heavy fighter revision,
to build and that Tachikawa's performance Tachikawa finally abandoned the Ki-94-1.
estimates were optimistic. Tachikawa did not wish to let Nakajima's
Tachikawa, however, did not give up on design against the high altitude fighter speci-
the Ki-94-l and reworked the aircraft into a fications, the Ki-87, go unchallenged. The
heaw fighter that was designated the Riku- result was the Ki-94-ll.


54 Japar'-Esl Sectte't PRo.tscts: ExpnRtl,tgxral AtncRnpt oF THE IJA AND IJN 1939-1945
T'archilkarva Kit- $4- [n

Tachikawa Ki-94-ll - data In October 1943, the Koku Hombu rejected on the underside of the fuselage. Originally, it
Tachikawa's radical, twin engine, twin- was planned that a six-bladed propeller
Contemporaries boom, push-pull fighter, the company's should be used. However, testing showed
Focke-WullFw190Vl8tUl (Germany), Focke-Wulf Tal52H (Germany), answer to the Koku Hombu's request for a that when in operation a blur was created by
l\,'likoyan-Gurevich MiG-3D (Rusia), l\'liko)an-Gutevich l-220 and I-230
high-altitude fighter. Tachikawa was told the the spinning prop, obscuring the pilot's for-
series (Russia),Yakovlev I-28 (Russia), Polikarpov ITP(l\'l-2) (Russia),
Ki-94-l was unconventional, complex and its ward vision. Therefore, a four-bladed pro-
Yakovlev Yak-3PD and Yak-3TK (Rusia)
estimated performance specifications were peller was selected. For weapons the aircraft
Because the Ki-9411 uas neDet floun, the spedfications oe based on optimistic. The rejeciion left Nakajima's Ki-87 had a similar cannon armament as the
Tachihaua's estimated puformance, the sole contender for the specification, Ki-94-1, but this was downgraded to two
something Tachikawa was not going to let Ho-l55 30mm cannons and two Ho-5 20mm
Type High-Aliitude Fighter
happen. cannons with one of each calibre cannon fit-
Crew 0ne
Soon after the rejection Tachikawa utilised ted into each wing. The wings were a typical
Powerplant One Nakajima IHa-441 13 (Ha-219), l&cylinder the same specifications given to Nakajima to laminar flow-type with the cockpit situated
radial engine, developing 2.450hp for take-off, fitted with a Ru-204
plan a revised design. Whereas Tachikawa behind the trailing edges of the wings. As the
turb0supercharger, dd\'ing a c0nstant speed, f0ur-bladed metal propeller had to meet the Koku Hombu's 3,000m (1,864 dimensions show, the new design was fairly
mile) range requirement and a 800km/h large by Japanese standards, even supersed-
Dimensions (497mph) maximum speed, Nakajima only ing Nakajima's Ki-87 in size.
Span 13.99m 45.9ft had to contend with meeting the speed max- With the drafts completed, Tachikawa
Length I 1.97m 39.3ft
imum. Unbridled by the range issue, presented them to the Koku Hombu. After
Height 4.60m l5.tfr
Tachikawa engineers went about the task of review, the design was accepted as the
Wng area 27.99m] 301.3fr,
producing a design capable of matching and Ki-g4-ll and Tachikawa received the order to
Wing loading 230.45k9/m' 47.2tbllt,
exceeding the Ki-87. produce a static test airframe, three proto-
Porver loading 2.63kg/trp 5.8lb,hp
Knowing that the Koku Hombu would types and 18 pre-production aircraft. As soon
likely reject anything out of the ordinary in as approval was received for the Ki-g4-ll,
Empty 4,690k9 10,3401b concept, a more conventional approach was work began on the construction of the first
Loaded 6,450ks 14,220 lb taken. The man behind the revised aircraft two prototlpes with the first to be completed
was Tatsuo Hasegawa. Retained from the by 20 July 1945. However, this date would not
Performance Ki94-l was the pressurised cockpit for the be met as delays in production caused the
['lax speed 720kn/h 447mph pilot but everything else was redesigned. The program to fall behind schedule which, in
at 10,000m at 32,808ft
new project used a standard configuration for turn, delayed the final completion of the first
Range 2,200km 1,367miles
a single-engine, all-metal fighter to ensure Ki-94-il.
Climb I7 min 38 sec ro 10,000m (32,8081r)
acceptance by the Koku Hombu. The power- Tachikawa scheduled the first flight for 18
Senice ceiling 14,250m 46,751ft
plant selected for the plane was the Nakajima August 1 945. However, on I 5 August 1 945, the
Armament IHa-441 l3 (Ha-219), 18-cylinder radial engine war ended forJapan and the Ki-94-ll was pre-
Two 30mm Ho-155 cannons, h\'0 20mm Ho-5 cannons and either hto that was rated at 2,450hp. The engine was vented from making any flights. The end of
30kg (66 lb) air{o"air rockets or one 500kg (1,102 lb) bomb equipped with a fan cooled, exhaust driven the war meant that the second prototype was
Ru-204 turbosupercharger that was situated never completed.
None. One prototlpe was completed with a second under construction

lvhen the war ended,

Tachikarva Ki-9111 (FE-150)

This was the first protot1pe and lvas surrendered at Tachikalva's facility
nearTokyo. Delivered to trlAl'lA lor storage and evenlual senicing, it

was listed on a i0 March I 946 manifest of aircraft available for aviation

industry eraluation. 0n I August 1946, the Ki-9411 was still at [,1A['lA but
n0 eiforlu'as made t0 restore it although it lvas designated for display al

NAS['I, September 1946 sarv the start ol restOration rvork ior displav

pueoses and the project lvas to be completed by 18 Septembet I946.

The last kno$'n whereabouh oithe Ki-94-ll rvas in Patk Ridge in 1949.


56 .lnpnNess SrcRrl PRo:scrs: ExpenlntsNrnl AIncRnrt oF THE IJA AND IJN 1939-1945
InrpERral hp,qn EsE ARr'tv 57
Tachikawa Ki-162 and Other IJAJet Projects

It was only after January 1945 that the Japan- The IJA's Captain Hayashi and Captain Yasuda
ese were given access to the very latest Ger- assembled and led two teams to research the
man jet technology. Prior to that, access concept and begin design. The one restriction
came only after German equipment had imposed was the requirememt for at least a
achieved operational status. Thus, in the half hour's endurance. At this time, only the
beginning of 1944, the Japanese were made Ne l2 series of turbojets was available. The
aware of the MesserschmittMe262 iet fighter Ne12, a development of an earlier project
and Me 163 rocket-powered interceptor long called the Ne 10, was found to be too healy. A
after thev were flying. Consequently, it was lighter model, the Ne 128, was produced in
not long before official requests for data limited numbers until the Ne 20 turbolet was
came from Japanese representatives in first tested in March 1945. This engine proved
Berlin. In the subsequent negotiations the to be superior to the Ne 128.
Germans were told that only the IJAwould be The problem facing both Hayashi and
building the Japanese version of the Me 262, Yasudawas that performance and endurance
the Nakajima Ki-201 Karyu, the development on a par with a twin engine design had to be
of which began in January 1945. But even achieved using only a single power unit. With
before this, the IJA began to formulate jet the Ne 128 unacceptable, both of these
designs of its own, spurred by the release of looked to the projected developments of the
technical information by the Germans. Ne 20. These included the Ishikawajima
In late 1944, the IJA initiated a study for a sin- Ne 130, Nakaiima Ne230 and the N4itsubishi
The depiction of the 'Tachikawa Ki-I62' shown
here is displayed in the colours of the 23rd gle engine jet fighter and the task was given to Ne 330 turbojets because theywere expected
Independent Chutai, operating in Okinawa, 1945. the Tachikawa Dai-lchi Rikugun Kokusho. to produce improved thrust over the Ne 20.

,t ,


IntpERral JlpnNpss Anl,rv 59

With at least the basic dimensions of the tur- In January 1945, Erprobungskommando Whether Captain Yasuda and his team
bojet in mind, each team got to work drafting 162 was created to field test the He 162 and on used the He 162 as their influence or arrived
their designs for the jet fighter. Two concepts February 6 I/JG 1 was ordered to convert to at a similar design by coincidence may never
emerged from both teams towards the mid- the jet fighter. The He 162 was an unforgiving be known. But the IJA would reject Yasuda's
dle of 1945. aircraft but in the hands of a skilled pilot it was fighter and instead selected Captain
Hayashi's team took the turbojet and an exceptional dogfighter. Pilots were told to Hayashi's design for continued work. Per-
placed it inside the fuselage near the tail. The avoid combat with Allied aircraft, however in haps it was felt that with the acquisition of the
fuselage was to be fairly slender in shape with late April and early May, I/JG 1 scored a hand- He 162, Yasuda's design was redundant. With
the cockpit forward of the low mounted wings ful of aerial victories. This was tempered by Hayashi's aircraft approved, he and his team
which were to be placed about mid-fuselage. the loss of 13 aircraft and 10 pilots, mostly due moved forward with refining the design and
The aircraft was to have a nose intake with the to accidents. The only other unit activated, constructing a prototype. However, by August
turbojet's exhaust being vented out through the 1.(Volkssturm)/JG 1 at Sagan-Kupper, 1945 when hostilities ceased, none of the
the tailpipe or rear of the fuselage. Yasuda's never received their He 162 fighters and this IJA's iet programs were ready to fly. Neither
group took the turbojet and placed it on top of was fortunate. The pilots of this unit were to Hayashi's jet nor the Japanese version of the
the fuselage. As opposed to Hayashi's more only receive training in the He 1625, a glider He 162 would progress past the initial design
slender aircraft, the Yasuda design had a version ofthe He 162 that had no engine, non- stage. Only the Nakajima Ki-201 ever made it
shorter and wider fuselage. As a result of the retractable landing gear and a rudimentary to the construction phase which, by the end
engine being on top, this eliminated the pos- second cockpit for the instructor. After a few of the war, amounted to a single incomplete
sibility of a conventional tail stabiliser arrange- gliding flights, the untrained pilot was fuselage.
ment and thus each tailplane ended in a expected to fly the jet powered He 162 and As a note, 'Ki-162' has been used by some
vertical stabiliser. Yasuda felt that with the the results would have been disastrous. A fly- as the designation for the Japanese produced
engine outside the fuselage, the nose could able two-seat He 162 known as the Doppel- He 162. There is no historical evidence to sup-
then be used to carry armament allowing for sitzer, fitted with a second cockpit for the port this although it is not without precedent,
improved and concentrated accuracy. instructor at the expense of the cannons and such as the case with the proposed Mitsubishi
Yasuda's design was very reminiscent of the ammunition, oxygen system and fuel capac- Ki-90 that was to be the Japanese version of
Heinkel He 162 Volksjdger. ity, was not completed by the end of the war the Junkers Ju90. In addition, one may see
At this juncture, the Volksjiiger will be dis- in Europe. the full name as the 'Tachikawa Ki-162'. It is
cussed briefly. The He 162 was designed as a On 15 April 1945, the Oberkommando der not unreasonable to believe that Tachikawa
fighter that could be built quickly with semi- Luftwaffe approved the release of specifica- may have played a part in assessing the
skilled or unskilled labour, using few war crit- tions and production data for the He 162 to He 162 data given that the IJA had the
ical materials and only a single turbojet. This the Japanese. A delegation of IJN officials Tachikawa Dai-lchi Rikugun Kokusho and
enabled the He 162 to be turned out more studied the Hel62 and visited Heinkel's the Rikagun Kokugijutsu Kenkyujo (also at
quickly than the more complex Me 262 then He 162 production line in Rostock, Germany. Tachikawa) at their disposal. Whether
in service. The He 162 began with the Volk- Impressed with the aircraft, the Japanese Tachikawa would have actually built the pro-
sjiiger competition, but as Heinkel had the quickly requested technical data on the duction Japanese He 162 is not known.
inside track with Deputy of the Reich Ministry fighter. Of course, by this stage of the war for
of Armament and War Production, Karl-Otto Germany there was no way that physical
Saur, who also happened to be the managing blueprints, production tools, jigs or a sample
director for Heinkel, the study for the He 162 He 162 could be shipped to Japan. Instead,
was already underway. By the time the other the only means available to rapidly send infor-
competing designs were submitted for the mation on the jet was via wireless transmis-
competition, the mock-up of the He 162 was sion - i.e., radio. To help facilitate the
already under construction. Little surprise the transmission of the data, CommanderYoshio
contract for the Volksjziger went to Heinkel in Nagamori used a datum line (a fixed, mea-
September 1944. On 6 December 1944, the surable line, used as a reference from which
He 162 Vl flew for the first time. angular or linear measurements are taken) to Tachikawa Ki-162 - data
The He 162 used a light metal monocoque measure the He 162 and transmitted the
fuselage with a moulded plywood nose. The resulting dimensions to Japan. Only a portion
Blohm und Voss P.22li0l (Germany), Republic F-84 Thunderjet (US),
high-mounted wing was one piece, made of of the data Nagamori sent was received in
Heinkel He 162 (Germanv), Arado 8.580 (Germanv), BMW Slrahljtiger I

wood and plywood skinning with metal wing Japan. (Germany), Heinkel P.1073 (Germany), Henschel Hs 132 (Germany),
tips. The two tailplanes sported vertical sta- Even though parts of the measurements for Anlonov SKh (Russia)
bilisers on their ends. Only four bolts secured the He 162 made it to the Japanese mainland,
the wings to the fuselage. A single BMW enough was received to begin making use of Specifirations
003A-l Sturm turbojet sat atop the fuselage the informaiion, filling in the gaps, revising Very little is known olthe jet designs created bvYasuda or Hayashi and

just behind the cockpit and to help him aban- the design to suit Japanese production capa- thus no specification inlonnation has surfaced. Likelvise, since there

don the aircraft in an emergency, the pilot bility and compiling the drafts needed to was little t;me to act on the He 162 data, there is no information on rvhat

was provided with an eiection seat. For realise the aircraft. Even though the IJN was the speciflcations \r,ould have been for the Japanese version though it
nould not be unreasonable t0 conclude the dimensions and
weapons, two MG 151 20mm cannons were the branch that obtained the He 162 data, it
periomance nould have been similar to lhe Cerman jet.
fitted with 120 rounds per gun. The top speed appeared that it would be the IJA who would
of the He 162 was 905km/h (562mph) at produce it. In addition to developing the air-
6,000m (19,690ft) with a 438km (272 mile) craft from the data, plans were made to begin None. \one of the jet fighters got past the design stage
radius at full power. production.

60 JnpnNEsE SEcRET PRo:ecrs: Expgnrnrsnr.qr- ArRcnapr or rus IJA aNo IJN 1939-1945
Imperial Japanese NaW
Kawanishi Baika much non-critical war materials as possible resulting gas can onlybe expelled through the
and be of simple design to allow for produc- exhaust and forward thrust is generated. This
The invasion of the Japanese home islands tion by unskilled or semi-skilled labour in cycle, or pulse, is repeated over and over up
was a genuine threat to Japanese military small, scattered workshops. These demands to 45 times per second in the case of the Argus
planners. Operation Downfall was the Allied were a result of the relentless Allied bombing Ar 109-014.
plan to launch the final blow against Japan. of Japanese industry and that Allied naval A pulsejet has four main benefits. Firstly,
This consisted of the capture of Kyrlshl forces had a stranglehold on imports of raw due to the simple nature of the engine it can
(Operation Olympic) that would provide the materials needed to sustain the Japanese mil- be easily built. Secondly, it can use low grade
jump-off point for the invasion of Honshl, itary. To meet this directive, Professors lchir0 fuels. Third, pulsejets offer reduced mainte-
near Tokyo (Operation Coronet). Given the Tani and Taichiro Ogawa, both of the Aero- nance. Finally, they have a lower cost per unit
geography of Japan, Japanese military lead- nautical Institute of the T6kyo lmperial Uni- when compared to other engines. However,
ership was able to narrow the likely avenues versity, began to study a means to achieve the the pulsejet does have three major flaws - it
of attack. To that end, Operation Ketsugdwas requirements of the task before them, sup- is not fuel efficient and, due to the operating
formulated. A critical component of the oper- ported by Kawanishi. nature of the engine, it is noisy and generates
ation was special attack units and they The engine for the Baika, the Maru Ka-10 significant vibration.
needed aircraft in mass numbers to succeed pulsejet, was derived from the German Argus The choice in using the pulseiet for the
in repelling the invasion. This provided the As 109-014 pulsejet, the technical plans for it Baika was clear. With the situation for Japan
spark for the Kawanishi Baika. having been delivered to Japan via subma- being what it was in late 1944 and into 1945,
On 2 July 1944, the Kaigun Koku Hombu rine in 1944. The Aeronautical lnstitute of the the Ka-10 offered a far less complex engine
issued a directive to Kawanishi Kokuki K.K. to T6ky0 Imperial University studied the design than a turbojet or piston engine. This meant it
produce a special attack aircraft. The design and in time developed the Ka-10. A pulsejet is could be built in greater numbers by unskilled
was to be a replacement for the Kugisho Oka a simplistic engine that operates by mixing air or semi-skilled labourers. Because it could
Model 11 and Model 22 as well as the special that is taken into the engine via a shuttered or use low grade gasoline it put less of a strain on
attack version of the Nakajima Kitsuka. The valve intake with fuel that is then ignited in the supply chain struggling to provide more
new aircraft, called the Baika (meaning Plum the combustion chamber. The force of the refined aviation fuel. Tani and Ogawa did find
Blossom), needed to be constructed from as explosion closes the intake and thus the that the Ka-10 suffered from having a short


InrpsRrer- Jeperupse Navv 6l

selvice life in regard to its fuelvalve. They also resulted in a fairing that enveloped a portion obtained in November
1943 and that one was
had concerns that the high noise of the of the front support strut for the engine. The 1944. These reports also suggest that the
engine would provide enough advance warn- second version, the Type II, moved the Ka-10 Japanese were very interested in the air
ing that countermeasures could be set or sent further back, eliminating the need for the fair- launching techniques for the weapon.
up to combat the Baika. Finally, it was ing. The final version, or Type III, placed the Another report indicates that the Japanese
realised that the vibration caused by the Ka-10 below the fuselage instead of above it. also knew ofthe Reichenberg project. A 1946
engine would put a strain on the aircraft, per- With the first two, the ad.iustment of the USMF report shows the Baika as a copy of
haps causing failures in the aircraft's struc- pulsejet may be due to maintain the centre of the Fi 103R. Finally, a manifest of cargo on the
ture. The benefits, however, outweighed the gravity on the aircraft. Facilitating pilot egress Japanese submarine I-29 (the same that car-
disadvantages and work on the Baika design was likely not the reason for the move. The ried data on the Messerschmitt Me 1638 and
proceeded. need for landing gear suggests that rail Me262; see the Mitsubishi J8M Slusui chapter
A meeting was called on 5 August 1945 to launching was not the only means to get the for more information) listed a single Fi 103
be held at the Aeronautical Institute of the Baika airborne. Towing or using rocket boost- fuselage as being onboard. The contempo-
Tdkyd Imperial University. It was attended by ers may have been considered or the use of rary illustrations of the Baika today (including
Admiral Wata and Admiral Katahira from the the landing gear may have simply been for the one here) are derived from the drawings
Kaigun Koku Hombu, professors Naganishi, training use only. It is probable that the land- of the aircraft made by Technical Comman-
Ogawa and Kihara of the Aeronautical Insti- ing gear could be iettisoned to improve aero- der Eiichi Iwaya in the 1953 Japanese book
tute, and Chairman Katachiro of Kawanishi. dynamics. The Type III, with the underslung Koku Gijutsu No Zenbo. Whether the Baika
In the discussions, the Baika was selected engine, would suggest that it was to be car- was truly inspired by the Fi l03R or was sim-
over the rival Krigisho Oka Model 43B, the lat- ried by a parent aircraft much in the manner ply an independent design may never be
ter being seen as too complex to build in of the Oka in addition to rail launching. It is known for certain.
numbers quickly, especially since it used the unknown what the two-seat trainer version
Ne 20 turbojet. Some revisions, specifically in would have looked like. It is not unreason-
simplifying the design, resulted in two ver- able to assume that the warhead would have
sions of the Baika being discussed. As the been removed and a second cockpit installed
Baika was to be used to defend the home with ballast simulating the warhead, much
island, launch rails were to be constructed like the Krlgisho oka Model 43 K-l Kai, the
that would use a solid fuel rocket to hurl the two-seat trainer for the Oka.
Baika into the air. The second version incor- Initially, the Baika featured 8mm of armour
porated a simple landing gear and it was protection for the pilot and although this only
decided that the Baika with the landing gear protected his back, it was a means to allow Kawanishi Baika - data
would be used for training pilots before they for some modicum of defence against inter-
converted to the rail launched Baika. After all, ceptors firing from behind rather than ground Contemporaries

given the mission of shimprl attacks on Allied fire. However, this was changed and the Fieseler Fi 103R Reichenberq (Germany)

invasion ships, the pilot would not be return- armour was removed, allowing for an
ing and would have no need of landing gear. increase in the warhead size up to 250kg Tlpe SpecialAttackAjrcraft

(551 lb). While this did result in a reduction of Crew One

At the conclusion of the meeting, Kawanishi
was given an order for one Baika prototype range, it was not seen as a detriment given
and ten two-seat trainers. The company was that the likely engagement distances would One I'lanr Ka-10 pulse jet; rated at 360kg (i91lb) olthrust
given a deadline of September to have the be rather less than 130km (80 miles).
finalised design completed as well as a pro- On 6 August 1945, another meeting on the Dimensions
duction plan finished. Mass production was Baika was called. Masayama Takeuchi of the Span 6.58m 21.6ft

to begin in October 1945. Kaigun Koku Hombu had concerns in regard Length 6.97m 22.91t

Kawanishi was given the following specifi- to its construction and sought to have the Height, unknown up to 3,99n/l3.llt by requirements

cations for the Baika: Baika built from as little war-critical material \Ving area 7.58mr 81.6ft1

length of no more than 8.5m (27.8ft) as possible, meaning that wood would be
height of no more than 4m (l3.1ft) used wherever possible. The Baika also had
Emptl 750k9 1,653 lb
width, with folded wings, of no more than to have passable handling characteristics in 3,152 lb
Loaded 1,430ks
3.6m (1 1.8ft) the air to give the pilot the best opportunity to Armament One 250kg (551 lb) rvarhead
speed, with the Ka-l0, must be at least strike his target. Beginning on 8 August 1945,
463kn/h (287mph) at sea level a team of 60 men was assembled at Kawan- Performance
a ceiling of 2,000m (6,561ft) ishi to oversee development and production I\'lrx speed 648kn/h 402mph

range of at least l30km (80 miles) of the Baika to be led by engineer Tamenobu. at 2.000m at 6,56lft

100kg (220Ib) explosive payload It all came too late for on 15 August hostilities Cruise speed 185kn,h 30lmph

came to a close and the Baika project came at 6,000m at l9,68iit

278km 172 miles
As the Baikawas developed, three versions of to an end before it had left the drawing board. Range

Climb 3 min 55 sec to 2,000m,'6.561|t

the aircraft emerged. Two were similar As anote, some sources make the case that
Ceiling 2,000m 6,56tfl
except for the cockpit placement. Despite the the German Fieseler Fi 103R Reichenberg, Fuel capacih 600 iitres 158 gallons
initial direction that the Baika would be rail the manned version of the Fi 103 (V-1), was
launched, all versions used landing gear. The an influence for the Baika. There are US intel- Deplolment
first version, or Type I, had the Ka-10 pulsejet ligence reports which indicate that the Japan- \one. Did not advance past the drawinq board,
directly above and behind the cockpit - this ese were well aware of the Fi 103 by October

62 JapaNsss StcnEr PRoJECTS: ExprRrrurENteL AIRcRnrr oF rHE IJA AND IJN 1939-1945
Kawanishi HllK Soku

By 1944, Japan was hard pressed to fuel her War 2. The company was also not lacking in
'uvar machine with the raw materials it des- cargo flying boat design having modified the
perately needed. Being an island, only two H6K to serve as a transport as the H6K2-L and
means were available for Japan to receive H6K4-L. Even the HSKwas adapted as a trans-
ore, fuel and other vital materials from what port, the H8K2-L Seikl (or Clear Sky). Kawan-
holdings Japan had left. The first was by sea ishi had also been working on the design of
and the second was by air. The sea route was the K-60, a long-range transport flying boat.
fraught with risk due to the US Nary and Allied With these credentials, Kawanishi was able
submarine and warship presence in the to capitalise on their knowledge to begin the
Pacific. Thus, delivering cargo by air, while design of the Hl1K Soku (Blue Sky) for the
also not without risk, appeared a better IJN.
option, despite the lower tonnage capacity in Kawanishi was instructed by the IJN to use
comparison to sea-going freighters. There- as muchwood as possible in the construction
fore, the Kaigun Koku Hombu asked Kawan- of the Soku since a flying boat of such size
ishi to design what would be one of only a would have consumed a large amount of pre-
handful of dedicated transport seaplanes of cious alloys needed for other aircraft such as
the war. fighters. Within Kawanishi, the Soku was
The IJN knew full well the impact the US called the I(X-8 and the initial design draft
Nar,y's unrestricted submarine warfare doc- was processed rapidly. The aircraft drew
trine was having, which made merchant heavily from the H8K being a high-wing, can-
ships the primary target. By the end of the tilever monoplane but overall, the Soku was
war, 1,200 Japanese merchant ships had much larger. The keel of the Soku was nearly
been sent to the bottom along with five mil- identical to the H8K. To power the flying boat
lion tons of cargo. In part this was due to poor four Mitsubishi MK4Q Kasei 22 (Ha-32-22)
convoy methods and protection by the IJN, radials, each developing 1,85Ohp, were
but also because the US had broken the selected with two perwing. As ordered by the
Japanese merchant marine cipher (the 'maru IJN, both the fuselage/hull and the wings
code'). Hence, moving cargo by sea would were to be built of wood and under each wing
more often than not result in the shipping would be a non-retractable float. The Soku
being located and sunk. ln 1944, shipping had two decks. The lower deck could accom-
losses were at their highest and by 1945, modate up to eighty fully equipped soldiers
while they did decline, it was because fewer including a number of vehicles or a compa-
ships were moving on the open ocean. rable amount of cargo. A smaller, upper deck
The desperation for fuel and other war housed quarters for the crew of five. The
materials was illustrated by the IJA's Kokusai main departure from the H8K transports was
Ki-105 Otori (meaning Phoenix), the twin- that the Soku utilised a split nose that was
engine version of the Kokusai Ku-7 Manazuru hinged to allow the two nose sections to be
(Crane) transport glider. The IJA planned to opened outward to each side of the fuselage,
use the Ki-l05 as a fuel tanker which would providing ready access to the lower deck.
fly from Japan to the Sumatra oil fields in This facilitated easier loading and unloading
Indonesia, load up with fuel and then return. increasing the speed and ease of these pro-
However, to make the return trip the Ki-l05 cedures. As a measure of protection the Soku
would consume some 80 per cent of the fuel was to be fitted with three 13mm Type 2
by the time the aircraft returned to Japan. The machine guns.
IJN looked for a better solution with a large Kawanishi presented the KX-8 to the IJN
transport capable of carrying a significant and the design was accepted. Authorisation
cargo load. Because a large expanse of ocean was given to construct a full scale wooden
had to be crossed in order to reach what few mock-up of the Soku now designaied the
territories the Japanese still held, the IJN Hl 1K1 for inspection before Kawanishi could
desired to have a seaplane to do the job. In proceed with the actual prototype. Construc-
January 1944, the IJN asked Kawanishi to tion of the mock-up commenced at the port
develop such an aircraft. of Komatsujima in the city of Komatsushima
Kawanishi was a leader in flying boat on the island of Shikoku (the smallest of the
development and had gained most of its four main islands making up Japan). This
experience in designing large sea going air- area was selected by Kawanishi because it
craft. Two of their most successful designs had access to the Seto Inland Sea which,
were the Kawanishi H6K (codenamed Mauis once the prototlpe was built, would be
by the Allies) and the H8K (Emily), with the needed to undertake sea and flight trials.
latter arguably the best flying boat of World Unfortunately for the Soku, the deteriorating

IupEnrnl JapaNesE Navv n.1

war picture saw delay after delay affect the
construction of the mock-up. To add to the
problem, Kawanishi was instructed by the IJN
in 1945 to reduce production of the H8K and
instead, concentrate on building the Kawan-
ishi NlK2-J Shiden-Kai fighter. Together,
these factors would see the mock-up
approaching its completion in April 1945,
well over a year after the design had been
On I April 1945, bombing raids conducted
on targets along the Seto Inland Sea saw the
nearly completed Soku mock-up destroyed.
With this loss, all further work on the Soku
design was shelved.

Kawanishi HllK Soku - data

Blohm und Voss BV222 Wking (Germany), Blohm und Vos BV238
(Germany), Dornier Do 214 (Germany), Boeing C-98/B-134 (US),

Consolidated PB2Y-3R Coronado (US), Martin JRM-l Mars (US), Martin

PBM{R Mariner (US), SikorskyVS"44 Lrcalibur (US), Convair R3Y
Tradewind (US), Short Sandringham (UK), Short Solent (UK)

Because the Hl I Kl uos not built, the specifications giDen arc estim1tes

made by KauanishL

Type Transport Flying Boat

Crew Five

Four Mitsubishi MK4Q Kasei 22 (Ha-32-22) l4-cylinder, air-cooled ndial
engines developing I,850hp for take-off, l,680hp at 6,886ft and l,550hp

at 5,500m/18,044ft; each engine drove a 4.3m (14, lft) diameter, four-

bladed, altematinq stroke propeller

Span 47.97m 157.4ft

Length 37,70m 123.7fi,

Height 12.55m 4l.2it

Wing area 289,95m'? 3,12lft?
Wng loading l56.i2lg/mr 32.1 lb/ft)

Power loading 6.l2k9hp l3.slb,ftp

Empty 26,405ks 58,2131b

Loaded 45,550kg 100,4201b

Useful load 19,095kg 42,0971b

Mar speed 4i0krnt' 292mph

at 5,000m at 16,404ft

Cruise speed 369km/h 229mph

Landing speed l44km/h 89mph

Range 3,890km 2,417 miles

Climb I I min 30 sec to 3,000m (9,842tt)

Ceiling N/A

Three l3mm Tlpe 2 machine guns with 200 rounds of ammunition per
z gun

U None. The Hl I Kl Soku did not advance past the mock-up stage

64 Jnpnnrsr SscRer Pno.lrcrs: ExptRrn,teNtaL AtRcRapr oF rHE IJA AND IJN 1939-1945
Kawanishi K-200

There are a few Japanese wartime aircraft the K-200 shows a hull not unlike the H6K but require at least 1 5,028kg (33,528 lb) of fuel for
such as the Kugisho Tenga jet bomber that deeper, though not to the extent of the H8K. A approximately one hour of operation at max-
remain shrouded in mystery to this day. The conventional tail akin to the H8K was used imum speed. Flying at a cruise speed would,
Kawanishi K-200 most certainly falls into this but the horizontal stabilisers were mounted of course, extend the operating range.
category, a design that had it proceeded halfway up the vertical stabiliser. The wings Options to attempt to save weight may have
would have resulted in the first turbojet pow- appeared very similar to the HSK and were fit- included removing any armour, stripping the
ered flying boat. ted to the hull in a like position, this being on defensive armament and/or constructing the
Very little is known of the genesis of the the top of the hull and, at least for the K-200, aircraft from wood as was the plan for the
K-200. Kawanishi may have been nearly central mounted on the hull. Interest- HllK Soku. Even with such measures the
approached by the IJN to initiate the project ingly, the K-200 was illustrated with fixed K-200 would have been hard pressed to
or Kawanishi may have undertaken the wing floats, which contrasted with the match, let alone exceed, the range of the H8K
design themselves to see if a flying boat could retractable floats used by the H8K3 as a or more conventional piston engine flying
be constructed using the new jet engines means to increase speed. Perhaps such a boats.
being designed following the success of the modification would have been considered It is not known how far Kawanishi studied
Ne 20 turbojet. Towards the close of the war, for the K-200 as well. the feasibility of the K-200, if at all. With
Kawanishi was developing two other large The K-200 is shown as having an armament resources allocated to the H8K, the HllK
flying boats: the Kawanishi K-60 and the layout similar to the H8K1. If this was the case, Soku and the K-60 among other projects,
Kawanishi Hl lK Soku. Both of these were at a Type 99 20mm cannon was fitted in a tail Kawanishi designers may have put the K-200
the behest of the IJN so it may not be unrea- turret and in the top mounted turret forward to one side pending availability of turboiets
sonable to assume that the IJN also asked of the wings. On either side of the forward sufficient to warrant the effort in developing
Kawanishi if they could add a jet powered fly- bow was a blister that would have been the flying boat. Aside from anything else, even
ing boat to the mix. Exactly when Kawanishi armed either with the Type 99 cannon or a a reliable turbojet such as the Ne 20 could
began to study the prospect of the K-200 is not Type 92 7.7mm machine gun. Finally, a Type only muster four to five hours of operation
known though 1945 is the likelyyear. 99 cannon would have been fitted in the bow. before it would suffer from problems. It may
Depending on the source, the K-200 was The K-200 was also probably able to carry a have been seen that preliminary perfor-
either to be the replacement for all IJN flying payload of bombs, depth charges or torpe- mance estimates fell short of expectations
boats in service or the K-200 was to be a car- does. and offered no significant advantage over
rier for a Japanese atomic weapon. The for- As far as propulsion, the K-200 was to use designs currently in use or projected to enter
mer assumption would likely have depended six turbojets. Theywere to be mounted on top service. Finally, a lack of materials necessary
on the performance of the K-200 had it been of the wings with each turbojet housed in a to construct the K-200 may have played a role
built. Certainly the prospect of the K-200 did separate nacelle. Grouped in sets of three, the in sidelining the design; the engine and con-
not deter other flying boat projects such as engines were fitted to each side of the hull on struction material issue saw the K-60 ground
the K-60 nor improvements of the H8K top of the wings. The reason for this was to to a halt and this mayverywell have extended
already in use. If the K-200 was to be such a minimise the amount of sea spray ingested by to the K-200. Regardless of the reasons, the
replacement for operational flying boats and the engines during use. If the K-200 was of K-200 would never be anything more than a
proved superior to them, it most likely would similar dimensions to the H8K, then the tur- concept.
not have entered widespread service until bojet engines would have to move something As a side note, following the war it was
1946. As far as the latter, the K-200 would have in the region of 24,948kg/55,000Ib of weight planned to construct a civilian version for use
needed capabilities that exceeded flying boat when the K-200 was fully loaded. If the Ne 330 by Japan Airlines.
designs then in service in order to serve as a turbojet was the engine of choice, all six
means to drop an atomic weapon on the US. would produce a combined thrust of
It has been suggested that the Nakajima 7,800kg/l 7,196Ib. This may have been suffi-
Fugaku was also devised to carry an atomic cient to give the K-200 a speed superior to the
weapon but there is no support for this H8K2, which topped out at 467km/h
notion. The same may be said for the K-200. (290mph).
One can speculate as to whether the K-200 Where the K-200 may have come up short
would have been any more successful in pen- is in terms of its range. Six turbojets would
etrating US coastal defences than a high flying have required a significant amount of fuel in
bomber. As we shall see, the K-200 may have order to give the flying boat a useful opera-
had a flaw that would have made any such tional radius. As an example, the Ne 20 turbo-
use all but impossible. jet consumed around 740kg (1,630Ib) of fuel Kawanishi K-200 - data
What the definitive shape of the K-200 was per hour. The Nakajima Kitsuka, which used
to be is open to conjecture. Certainly Kawan- two Ne 20 engines, carried a maximum of Beriev R-l (Russia), l'larlin P6\,1 Seama$er (US)
ishi would have utilised their successes with 1,447kg (3,190lb) of fuel (and without drop
the H8K and to a lesser degree with the tanks only 723.5k9/1,5951b) and therefore, at Specifications

Kawanishi H6K (codenamed Mouls) as a its cruise speed, could muster a 824km (512 There is n0 exact information available on the Kan'anishi K-200.

foundation for the K-200. As such, it is likely mile) operational range. If the Ne 330 con-
that the hull design would have followed a sumed approximately 2,535k9 (5,588Ib) of None. The K-200 existed only as a concept orpaper design.
similar pattern. One speculative illustration of fuel per hour at full thrust, then six would

IupnRrnr. .hplllEsE Nlvv 65

The K-200 depicted here is based
on a contemporaryJapanese
illustration of the flying boat whicb
itself is a conjecture on what the
K-200 may have looked like.

JApANESE SrcREr PRo:Ects: ExpgntllENreL AtnCnnrr oF rHE IJA AND IJN 1939-1945
KugishO High-Speed Proj ects

Every aircraft creator seeks to reduce drag in Therefore, in order to realise higher air edge over the enemy. The Albatross D series
their designs. The definition of drag is the speeds, the designer must take steps to and the Roland 'Walfisch'would epitomise
force that resists movement through a fluid, reduce drag and thereby lower the amount of those efforts.
which, of course, includes air. The more drag, drag force slowing the aircraft down. Following World War 1, the resurgence in
the slower the aircraft moves through the air Before World War I some aircraft design- air racing such as the Schneider Trophy in
due to the resistance. Drag cannot be com- ers appreciated the need to reduce drag. This Europe and the National Air Races in America
pletely removed from a design, but even in often took the form of fuselages that had saw rapid advances in aerodynamics and
the earlyyears of aviation various methods for clean lines in an attempt to remove protru- drag reduction to produce fast flying racing
minimising drag were investigated and many sions and also to streamline propeller hubs to aircraft for competition. Aircraft such as the
different solutions were tried. Not surpris- help them cut through the airmore efficiently. Curtiss R2C-1 Na'"y Racer, the Adolphe
ingly, such applications were valued by those The best example would be the 1912 Deper- Bernard'Ferbois' (capturing the world speed
providing the military with aircraft and in dussin that won the Gordon Bennett race in record of 45lkm/h (280mph) in 1924), Gloster
Japan, prior to the outbreak of hostilities with Chicago, Illinois, which became the first air- III, Supermarine S.5, Kirkham-Williams Racer
the US, the Dai-lchi Kaigun Kok[ Gijutsu-sho plane to exceed 16lkm/h (l00mph) (in 1913 (which, unofficially, flew to a speed of
would study such efforts in an attempt to pro- a later model of the plane would achieve 51gkm/h (322mph) in 1927) and the Savoia-
duce fast flying aircraft. 205km/h (l27mph). Marchetti S.65 typified high performance race
Form drag is the component caused by the At the beginning of World War I , few of the aircraft. The benefits of these innovations
shape of the body moving through the air. major combat aircraft utilised significant drag were not lost on military aircraft designers.
Therefore, when designing an aircraft the reducing methods. Exhaust stacks, radiators, With the war clouds looming on the hori-
form and shape of the plane is one of the protruding machine guns, wire bracing, struts zon, the seeds planted by the air racers of the
most important factors a designer has to con- and engines only partially cowled predomi 1920s and early 1930s were germinating in
sider. The wider the cross section, the more nated. One of the few exceptions was the the aircraft used by the air forces of the major
drag is produced. Having significant form Morane-Saulnier N'Bullet'. Nevertheless, the powers. Designs by Curtiss for the US Army
drag results in lower speeds because the rapid pace of combat aircraft development Air Force were influenced by the Curtiss rac-
faster the aircraft moves through the air, the duringWorldWar 1 sawdesigners looking for ers while the retractable landing gear of the
more drag force is applied to the aircraft. ways to increase speed as a means to get the 1920 Dayton Wright RB racer would become

The DB 60lA engine aircraft,

The NKIB engine design, -

In'tprRrer- Jnpar{Ese Nnvy b/

a hallmark of Grumman aircraft such as the vortices that change how the air flows over the DB 601A equipped design, the wings were
F2F. In Great Britain, R. J. Mitchellwould draw the wings. This change results in less and less mounted mid-fuselage.
heavily from his experience designing lift which then requires a higher and higher Kugishd would not produce any direct pro-
Schneider Trophy racers to build the Super- angle of attack to compensate and, from this, totype aircraft from either concept. lnstead,
marine Type 300 which would eventually induced drag results. Elliptical wings offer the results of the various studies were likely
evolve into the Spitfire. In Italy, Mario Castoldi, less induced drag than more conventional kept available as reference for engineers to
lead designer for Macci, would turn his skills straight wings. However, low aspect ratio access as a means of obtaining data on the
in constructing racing aircraft to producing wings are more prone to larger vortices aerodynamic problem. Perhaps Krlgish6 in
fighters for the Regia Aeronautica with types because they cannot be spread out across a hindsight considered themselves fortunate to
such as the Macci C.202 Folgore. longer wing. Krlgisho's study on wing shapes not have expended additional expense and
Japan, like other countries, sought to pro- was the likely result of testing various airfoils effort in producing working prototypes given
duce racing aircraft and planes designed to in a wind tunnel to determine their effective- the failure of the IJA's Ki-78, a program that
beat world speed records. An early example ness and record the results. lingered on into 1944 and never met its design
was the Emi 16 Fuii-go built by Itoh Hikoki Another aspect Klgishd engineers reviewed goals.
Kenyusho (ltoh Aeroplane Research Studio), were the merits and flaws of using either an The DB 601A engine aircraft is shown in the
which from 1920 was used in Japanese com- inline or a radial engine and how each tlpe colours originally used on a Mitsubishi A6M3,
petitions, and the contemporary racing air- reduced the form drag. In both cases the engi- serial 3032, tail code V-190 of the Tainan
craft from Shirato Hikoki Kenkyusho. neers drew up two concept aircraft and each Kokutai. It was found on Buna Airfield on 27
Kawanishi was not far behind with the K-2 made use of streamlining. Streamlining is the December 1942 in disrepair. lt was a presen-
speed racer which, despite extreme mea- process of shaping an object, in this case, a tation aircraft donated by Sadahei, a civilian
sures to minimise drag, suffered from a drag- fuselage, to increase its speed by reducing volunteer group. The Hukuko number was
inducing radiator mounted on top of the the sources of drag. One concept used the 874. The NKIB engine design is painted in the
fuselage. The K-2 achieved an unofficial 1,159hp Daimler-Benz DB 601A, a 12-cylinder, standard training orange used on prototypes
speed of 258km/h (160mph) in a flight made inverted-V, liquid-cooled, inline engine. This and trainer aircraft.
on 31 July 192 1 . Other refinements in aerody- engine would be licence built for the IJN as
namics could be seen in the Kawasaki Ki-28 the Aichi AEl Atsuta (the 'A' stood for Aichi,
'l' for first liquid- Kugisho High-Speed Aircraft Project - data
of 1935 which, despite its advantages in 'E' for liquid-cooled and
speed, climb and acceleration, was not suc- cooled engine; Atsuta was a holy shrine in
cessful in attracting IJA contracts. Aichi Prefecture) and for the IJA as the Ha-40 Messerschmitt ['le 209 (Cermanv)
In 1 938, a group of designers sought to pro- before it was renamed the [Ha-60] 22. The
duce a high-speed aircraft to challenge the second concept aircraft used a 1,000hp Naka- Twe High.Speed Aircralt
world air speed record. Once war had broken iima NKlB Sakae 11 whichwas a 14-cylinder, Crew 0ne

out this aircraft, called the Ken lll, was soon air-cooled, radial engine. This engine was a
taken over by the IJA. Redesignated the Ki-78, licence version of the Gnome-Rhdne l4K Mis- Powerplant OneDaimler-BenzDB60lA,l2-c\'linder,inverted-V,

its development was continued under tral Major (in engine nomenclature, the 'N' liquid-cooled, inline engine developing l,l59hp or one Nakajima NKIB
I 14-c$inder, air-cooled, radial engine developing
Kawasaki. During this time, it may have been was for Nakaiima, 'K' for air-cooled, '1'as the Sakae I 1.000hp

the IJN who decided to conduct its own stud- first air-cooled engine, while the 'B'was for
ies of high speed aircraft with Kugisho the second version of the NKl; Sakae means
Span N/A
assigned the task of doing so. Whether the prosperity in Japanese).
Length (DB60lA) 6.91m 22.711

studies were initiated in response to the IJA's Kugisho would use the same basic air- (NKIB) 6.97m 22.91r

own high-speed aircraft project is unknown frame for the engine study. It consisted of a Height N/A

bui the prevalent aircraft design philosophy of well streamlined fuselage with the pilot Wing area N/A

both the IJN and the IJA prior to the war was mounted in a cockpit set behind the wing and Wing loading N/A

of speed, agility and range at the expense of just forward of the vertical stabiliser. This style Polrer loadinq ti/A

fi repower, durability and protection. was found in a number of racing aircraft such
Kugisho examined over half a dozen as the American GeeBee Rl and Geebee Z. Weights (approximate)

aspects of aerodynamics in order to produce Both used a standard tail-siiter configuration

Empty (DB601A) 1,600kg 3,527.31b

(NKlB) l,289kg 2,841.7 lb

data on what would be needed to realise an for the landing gear. The concept equipped
Loaded (DB60lA) l,900kg 4,188.7 lb
aircraft capable of significant speed. One with the DB 601A engine had a fuselage shape (NKtB) l,6s9kg 3.657.41b
area of research was the main wings. The that was not unlike the Kawasaki Ki-61 Hien Fuel&oiln'eight (DB60lA) 2lskg 473.9Ib
shape of a wing is one of the more critical (meaning Swallow; codenamed lony by the (NKIB) 270k9 595.2Ib

aspects of aircraft design. Factors such as Allies) which would appear in prototype form
wing loading, expected air speeds, angles of in December 1941 . The wings were mounted Performance

attack and the intended use of the aircraft all low on the fuselage. The fuselage appear- Mar speed \1A

influence how the wing is shaped. For high ance was due to the inverted-V engine which, Range \1A
speeds, a low aspect ratio wing is often con- by design, offered lower height, weight and Climb

Ceiling N/A
sidered. Typically, these are short span wings lengthwhen compared io more conventional
with the benefits of higher manoeuvrability motors. By contrast, the concept using the
and less drag. In addition, having a backward NKlB had a more ovoid fuselage shape, the
sweep to the wing also lowers drag. The drag result of the height of the radial engine. To Deplolment
most associated with wings is termed maintain the aerodynamic streamlining a None. Both Kngisho designs existed on paper only

induced drag, which is caused by wing tip large spinner was used. Also, in contrast to

68 JeplruEst SEcnsr PRo.lEcrs: Expnnnrlrvral ArRcRalr on rHe IJA nno IJN 1939-1945
Kugish6 M)ff6

The development of the KyushuJTWShinden ered MXY6, the validation of the canard
was an ambitious undertaking. Captain design provided the needed proof of concept
Masaoki Tsuruno, the man behind the Shin- and as such the IJN instructed Kyushu to pro-
den, needed to confirm the handling charac- ceed with the J7W Shinden.
teristics of a canard aircraft before
proceeding further with the plans and con-
struction of the J7W itself. To do this, he com-
missioned Klgish6 to design and build three
Kugisho MXYG - data
gliders that were based on his JTWI aircraft
plans. The result was the MXY6. Hamburger Ha l4l-0 (Germany), FGP 227 (Oermany), Goppingen Gti 9

Krlgisho drew up the design of the MXY6 (Germany), Horton Ho IllB and Ho lV (Germany), Berlin B 9 (Cermanv),
with the assistance of Captain Tsuruno. Con- Junkers Ju 49 (Germany), Lippisch DM- I (Gemany), DFS I 94 (German])

structed entirely of wood, the MXY6 featured

a slightly swept wing, vertical stabilisers fitted Type Proofof Concept Glider

inside of the wing ailerons and canards Crew One

mounted along the nose of the fuselage. The

braced tricycle landing gear was fixed and Powerplant Unpowered except lot one modified rvith a Nihon

Hainenki Semi I I ([Ha-90] i l) 4'cylinder, air-cooled engine developing

provided with suspension. Once the MXY6
22hp and driving a two-bladed, flred stroke wooden propeller
was finalised, construction was entrusted to
Chigasaki Seizo K.K. and they had completed Dimensions
the three gliders by the fall of 1943. Flight tri- Span ll,l2m 36.sft
als got under way soon thereafter and the Length 9,63m 31.6ft

MXY6 was found to have good handling char- Height 4.20m 13.8ft

acteristics which provided verification to the Wing area 20.49m' 220,6fr'

concept of the J7W.

For further testing, one of the three gliders
was modified by having a small engine l.oaded 640kg 1,410 lb

installed in the rear of the fuselage in the

same pusher configuration as the proposed Mar glide speed
J7W. The engine, a Nihon Hainenki Semi 11
([Ha-901 1l), allowed the handling under
The profile shown is based on one of the MXY6
power to be studied as opposed to unpow-
gliders found at Atsugi in September 1945.
The paint is training orange as normally used ered flight only. Following the conclusion of Deployment

on experimental and training aircraft, the testing of both the unpowered and porv- None, The N'IXY6 was purely a proof0fconcept glider



IupeRrer- Jap,qNEss Nnvv 69

Kugisho M)ff7 Oka



The Oka Model I I depicted here uses markings

commonly seen on deployed aircraft.

'No longer can we hope to sink the numeri- Kdgekitai. Their story, as well as that of the
cally superior enemy carriers through ordi- IJA's Shimbu Tokubetsu Kdgekitai, is beyond
nary attack methods. I urge the use of special the scope of this book (however, for those
attack units to crash dive their aircraft and I interested there is a wealth of material avail-
ask to be placed in command of them.' able on the subject such as David Brown's
These words by IJN Captain Eiichiro Jyo, Kamikaze and Earl Rice's Komikazes).
commander of the carrier Chiyoda, reflected The majority of the shimpu missions were
a mood he had observed in some of his pilots flovrm using types already in service. In addi-
Oka Model ll - data and men. Their feelings were that to carry on tion to the Reisen, the Kugish6 D4Y Suisei
with conventional tactics was doomed to fail- (meaning 'Comet'but known to the Allies as
Type Special Attacker ure. While death in combat was worthy, a Judy),Kawasaki Ki48 (l/y), Nakaiima Ki-49
Crew 0ne death that did no good was shameful and Donryu (meaning 'Storm Dragon'but called
would not serve the Emperor or Japan. Jyo's Helen by the Allies), Aichi D3A (Va1) and
Powerplant Three Tlpe 4 Mark I Model 20 solid fuel rockets,
words, written in a memo to Rear Admiral many others were modified, sometimes
each developing 267kg (588 lb) of thru$, for a total of 80lkg (1,764|b)
Soemu Obayashi and Vice Admiral Jisaburo heavily, and used against the Allies, but none
Ozawa, would be the catalyst for the forma- were specifically built from the ground up for
Span 5l2m I 6,8ft
tion of special attack units and from this a shimprl (suicide) operations. It would be IJN
Length 6.06m l9.gft new weapon would arise that would become Ensign Mitsuo Ota, a transport pilot flying with
Height Ll5m 3.8ft the only purpose-built special attack aircraft the 405th Kdkr.rtai, who put forward a design
Wing area 5,99m'' to see operational combat service during for a piloted glide bomb.
Wing loading 356.90kg/m' 73.1 lblt World War 2: the Krlgisho MXYT Oka. Ota's concept was not the only one that
Power loading 2.67kg,/t'p 5,9lb,t'p Vice Admiral Takijiro Onishi is most often called for a dedicated shimpu aircraft. Other
credited with officially forming and organis- ideas were considered such as the Showa
ing the special attack units, the first of which Toka (see elsewhere in this book on the
Empty 440kg 970rb
became operational in October 1944. A piv- Nakajima Ki-l15 Tsurugi for more informa-
Loaded 2,140.5k9 4,il8tb
Useful load 65Okg l,433lb
otal man in the formation of the IJN's tion), but what set Ota's idea apart was that
Rikusentai (airbome troops), Onishi was also he wanted to have the explosive payload car-
Perfornance eccentric which did not always endear him to ried internally as opposed to fitting an exter-
Max speed 649kn/h 403mph his superiors and so, prior to his assuming nal bomb. Also, the aircraft had to be carried
at 3,505m at 11,500ft. command of the IJN land air forces in the and released by a parent plane and rocket
Dive speed 927km,h 576mph Philippines, he served as a supply officer. boosters would be used to speed the
Cruise speed 462krl/it 287mph Speaking to the officers of the 201stAir Group, approach and terminal dive onto the target.
at 3,500m at ll,482ft
Onishi stated that because of the limited
nnge 37km 23 miles
resources only a Mitsubishi A6M Reisen with Ota did not have any aeronautical engineer-
Ceiling 8,250m 27,066ft
a 250k9 (551 lb) bomb that was crashed into ing experience and would not have been able
enemy ships would suffice in slowing the US to present a definitive plan for his aircraft. In
1,200k9 (2,646Ib) Tri-Nitroaminol explosive warhead
fleet. From this began the rise of the IJN spe- order to help his cause, Ota sought and
cial attack force, the Shimpu Tokubetsu received assistance from the Aeronautical

Z0 JepnxEse SEcnrr Pno;Ecrs: ExpsRrN4sNreL ArRcRepr oF rHE IJA nruo IJN 1939-1945
Research Institute of the University of Tokyo. ception and present a fast moving target to Oka Model 2l - data
Professor Taichiro Ogawa headed the study defending anti-aircraft gunners.
of Ota's conceptwhile Hidemasa Kimurapro- Miki and his team completed the design of Type Special Attacker

Crew 0ne
vided the basic design ofthe aircraft and even the MXYZ in weeks and by the end of Sep-
produced models that were wind tunnel tember 1944 ten MXYT had been completed
Powerplant Three Tlpe 4 I'lark I l'lodel 20 solid luel rockets,
tested. Within weeks, the proposal for Ota's and were ready for testing. The aircraft was
each developing 267k9 (588Ib) oflhrust, for a total of80lkg (1,764Ib)
design was drafted, the design illustrated then renamed the Oka Model I l, Oka mean-
and performance estimates presented along ing 'Cherry Blossom'. A 1,200kg (2,6461b) Dimensions
with the data obtained from the wind tunnel explosive charge was fitted into the nose and Span 4.llm l3.5lt
testing. five fuses were installed, one in the nose and Length 6.88m 22.6ft

In August 1 944, Ota brought his proposal to the remaining four on the rear plate of the Height l.l2m 3.7ft

the attention of Lieutenant Commander charge. The fuses were armed by the pilot Wing area 3.99m' 43ft:

Tadanao Miki. Miki was the department head from inside the cockpit and they could be set Wing loading 399.78kg/m' 8l.glbift'

of the aircraft design section of the Dai-lchi to explode on impact or the detonation could Polver loadrng l.99kg/hp 4.41b,t'p

Kaigun K6ku Gijutsu-sho. lt is said that when be delayed by up to 1.5 seconds to allow the
Miki reviewed Ota's concept he was taken Oka to penetrate the target (such as a ship
Empt".v 535kg I,l 79 lb
aback and shocked at the idea of putting men hull) and explode inside. The carrier for the Loaded l,60Okg 3,527Ib
into piloted bombs. However, by this time the Oka was the Mitsubishi G4M, known to the Usefui load 9lsks 2.01 7 lb
policy of shimprl tactics had been approved Allies as Beffy. A number of G4M2a Model 24B
and regardless of how Miki felt personally he and 24C bombers were modified by having Performance
could not deny the submission. Miki placed their bomb bay doors removed to be replaced N'lar speed 642kmih 399mph

the design before the Naval General Staff on 5 by the required shackles to hold the Oka. at4,000m al l3,l25ft

August 1944. Air Staff Officer Minuro Genda, These modified carriers were redesignated Cruise speed 413kryh 275mph

at 4,000m at l3,l25ft
after looking over Ota's plan, approved the G4M2e Model 24J. However, the Oka's loaded
[,lax range I llkm 69 miles
concept and instructed chief of staff Admiral weight of 2,140kg (4,718Ib) far exceeded the
Ceiling 8,500kg 27,8871t
Koshiro Oikawa to set the wheels in motion bomber's standard load of 1,000kg (2,205 lb)
for turning the design into reality. Perhaps it and as a consequence the G4M2e suffered 600kg (1.322 lb) explosive rvarhead
was ironic that the task of starting the devel- from poor handling and performance.
opment of the aircraft fell to Miki. Kugisho As the Oka did not take-off on its own nor
was the organisation that would develop the was it anticipated that it would fly at speeds
aircraft, which was given the initial designa- under 322kmlh (200mph), the wings were
tion MXY7. Miki assembled a team of engi- kept very short. For propulsion, three Type 4
neers led by three men, Masao Yamana, Mark I Model 20 solid fuel rockets were Oka Model 22 - data
Tadanao Mitsugi and Rokuro Hattori, and they installed in the tail of the fuselage. Each
began drafting and refining the MXYT design. rocket could produce up to 267k9 (588Ib) of Type Special Attacker

The MXYT was essentially a glider bomb thrust for a total of 801kg (1,764Ib). The pilot Crew 0ne

with a pilot providing the guidance. There could activate them as he saw fit and could
were several specific factors involved in the fire them one by one or all three at once. Total
One Tsu-1 I therrnojet developinq 200ke (440 lb) of thrust
MXY7, most of whichwere out of necessity. In burn time for each rocket was 8-10 seconds.
order to conserve war materials, the MXYZ Given that the Oka would have to fly through
was to be constructed using wood as well as significant anti-aircraft fire as it approached Span 4,llm i3.5fl
non-critical metals such as aluminium, if nec- its target as well as the possible aerial inter- Length 6.88m 22.6f1

essary. It was expected that pilots with mini- ception by Allied fighter cover, the pilot was Height Ll2m 3.71t

mal skill would be required to fly the machine afforded protection through armour plate. A Wing area 3.99m' 43ft'

and therefore the aircraft had to possess good 19mm strip of plating was fitted along the Wing loading 401.82k9m' 82.3lbiit:

handling and manoeuvrability to ensure a underside of the fuselage near to the pilot's Power loading 7,98ks/hp l7.6lbihp

successful strike. Not surprisingly, instrumen- feet while his bucket seat had between 8mm
tation for the MXYT was kept to the bare min- and 15mm of armour, the majority protecting
imum. The aircraft also had to be simple to his back.
EmpLv 545kg 1.201 lb

Loaded l,450kg 3,1971b

construct so as to allow rapid mass produc- As discussed above, the instrumentation IJsehrl load 965kg 2,1271b
tion by semi-skilled and unskilled labour. was kept to a minimum. The instrument
The MXYT's primary mission was anti-ship. panel contained the altimeter, compass, atti- Performance
The flight profile began with the MXYT being tude indicator (artificial horizon), airspeed I'lar speed 445knlrh 276mph

carried aloft by a modified Mitsubishi G4M indicator, arming handle for the fuses and the at 4.000m at l3,l25ft

bomber. At the point where it was within rocket motor ignition switches. Cruise speed 427knyh 265mph

range of the target, the G4M would release With the ten available MXYT prototypes, at3,500m at ll,482ft

the MXYZ which would then glide towards flight testing was to commence in October IIar range l60km 99 miles

the intended victim. During the approach the 1944. However, the IJN did not want to wait Ceiling 8,i00m 27,887ft

Fuel capacitl 290 litres 76.6 gailons

pilot would ignite the rocket motors in the for the results of the tests and in September,
Oil capacit-v l0 litres 2.6 gallons
rear of the plane to increase its speed and Rear Admiral Jiro Saba, director of the
close in to the target as quickly as possible. Krigisho Naval Aeronautical Research Labo- 600kq 0,322 lb) explosh'e rvarhead
This would minimise the chances of inter- ratory, went to Lieutenant CommanderYokei

Il,IprRIal, JnparvEse Nlvv 71

Oka Model 33 - data I\,latsurra at the Munitions Ministry to sort out Hyakurigahara Airfield on 1 October 1944
the arrangements for opening production of under the command of Commander Moto-
Type Special Attacker the Oka. Matsurra, who shared a similar dis- haru Okamura with Lieutenant Commander
Crew 0ne taste of the suicide concept to Miki, saw to it Goro Nonaka and Lieutenant Commander
that much of the production was handled by Kunihiro Iwaki as his operations officers. The
military contractors to maintain secrecy and unit was nicknamed the Jinrai Butai, translat-
One Ne 20 arial-florv turbolet developing 475kg (1,047 lb) of thrust or
not by the private aviation industry. As such, ing as 'Thunder God Corps'. Through October
one Ne I28 jet eneine developing 320k9 (705lb) oilhrust
Krigishd would build the Oka at Dai-lchi Kai- the unit received hundreds of volunteers.
gun K6kir Gijutsu-sho as well as at Dai-lchi Those who were too old, married or were
Span 4.99m 16,4fr Kaigun Kokusho, and two sub-contractors, only sons, or those with significant family
Length i.l9m 23.6ft Nippon Hikoki K.K. in Yokohama and Fuji responsibilities, were rejected for the Jinrai
Height 1.15m 3.8fr Hikdki K.K. in Kanegawa, would provide wing Butai, leaving 600 pilots to be accepted into
Wing area 5.99m' 64,5fr' and tail assemblies. It was expected that 100 the unit. The 721st Kikotai consisted of the
Wing loading 382.78k9/m: 78,4lb/fr,
Oka aircraft would be ready by November 708th Hikotai and the 711th Hikotai, eachwith
Power loading 4.76kgtrp lbi/t'p
I 0.5
1944. 18 G4M2e bombers. The 306th Hikotai and
The first unpowered flight tests of the Oka the 308th Hikotai were assigned the task of
Empty N/A
began at the Sagami fusenal located in escorting the Oka carrying bombers, each
Loaded 2,300k9
Sagamihara in Kanagawa Prefecture. To squadron maintaining 36 Mitsubishi A6M
Usefui load N/A begin with, unmanned, unpowered flights Reisen fighters. The unit's initial l0 Oka air-
were conducted to assess the Oka's flight craft were supplemented by some 40 Mit-
Performance characteristics and these were followed soon subishi A6M5 Reisens fitted with 250kg
Max speed 643knl/tt 399mph afterwards by unmanned, powered flight (551lb) bombs.
(Ne20) at 4,000m at 13,1251t tests. All of the Oka drops were made from Flight testing of the Oka continued through-
Cruise speed N/A
the C4M2e bombers with the Okas being out November. These tests showed that
\'lar range 132 miles
directed out into Sagami Bay. Flight testing when dropped from 5,944m (19,500ft) at a
Ceiling N/A
was then moved to Kashimi in Saga Prefec- downward glide angle of 5.5' the Oka could
Fue) capacity 250 litres 66 gallons
ture which was near the IJN base in Sasebo in achieve a range of 60km (37 miles) at a speed
0il capacity N/A
Nagasaki Prefecture. At Kashimi, the first of 317km/h (230mph). In anearlyvertical dive
Armamenl manned flight of an Oka took place on 31 it was clocked at over 966km/h (600mph).
800kg (1,763 lb) explosive warhead October 1944 with Lieutenant Kazutoshi However, under combat conditions the Oka
Nagano (other sources have his last name as could manage 25 to 29km (15 to 18 miles).
Nagoro) at the controls. The particular Oka Based on the tests and flight experience, a
Oka Model 43A- data (estimated) that Nagano was to fly was the prototype for mission profile was developed for the Oka's
the Oka K-1 trainer. In place of the warhead deployment. Flying at a height between
Type Special Attacker and the three rocket motors were tanks hold- 6,096m and 8,230m (20,000f1 and 27,000f1),
Crew 0ne ing water as ballast that simulated the com- the G4M2e would release the Oka when it
bat weight of the Oka. Since there was no was within 1 7 to 33km (10 to 20 miles) of the
room for a conventional landing gear, a cen- target. The pilot would then enter a shallow
One Ne axialflow turbojet developing 475kg (1,047Ib) oi thrust
tral landing skid was fitted to the underside of glide with an airspeed of between 371km/h
the fuselage and under each wing tip were and 45lkm/h (230mph and 280mph). At a
rounded skids to protect the wings and pre- point about 8 to 12km (5 to 7 miles) from the
Span 8.99m 29,5ft

Length 8.16m 26.8ft

vent them from digging into the ground on target, and from an altitude of approximately
Height 1.12m 3.7ft landing. Prior to landing, the water was to be 3,505m (l 1 ,500ft), the pilot would activate the
Wing area 12.99m: l39.9lt' jettisoned which slowed the landing speed to rocket boosters increasing the speed to
Wing loading 193.83kg/mr 39,ilb/it' 223knt/h (138mph). For Nagano's flight, a 649km/h (403mph). Prior to striking the tar-
Power loading 5.30k9hp l1,7lb/hp rocket booster was fitted to the underside of get, he would put the Oka into a 50'dive that
each wing. At 3,505m (11,500ft) Nagano was would take the speed up to nearly 934km/h
released from the G4M2e bomber and (580mph). At the last moment, the pilot
Empty N/A
entered a good, stable glide. A few minutes would pull up the nose to strike the ship at the
Loaded 2,520k9 5,5551b
into the flight, Nagano activated the booster waterline.
Useful load N/A
rockets and almost immediately the Oka Oka pilot training was soon underway. Typ-
began to yaw. Nagano quickly jettisoned the ically, the pilotwould use a Reisen to practice
Mar speed 596km/h 370mph rockets and the problem disappeared. The the Oka attack routine flying the fighter with
at4,000m at l3,l25ft remainder of the flight went perfectly, the engine switched off. For many, they only
Cruise speed N/A Nagano bringing the Oka down without had the opportunity to become familiar with
Nlax range 200km l24miles mishap after releasing the water ballast. Sub- the Oka while it sat on the ground. A few were
Ceiling N/A sequent investigation showed that uneven fortunate to make an unpowered flight using
Fuel capacitv 400 litres 105.6 gallons
thrust from the rockets caused the yawing one of the MXYT trainer prototypes. As
0il capacity 16 litres 4,2 gallons
and Nagano is said to have stated that the Oka expected, accidents occurred and on 13
handled better than a Reisen. November 1944, the Oka claimed its first
As flight testing and production of the Oka casualty. Lieutenant Tsutomu Kariya exe-
800kg (1,763Ib) explosive warhead
got underway, 721st Kokutai was formed at cuted a perfect drop from 2,987m (9,800ft)

72 JlpnruEsn SecnEr PnoJscrs: ExpsRIN4ENraL AIRcnapr oF rHE IJA AND IJN 1939-1945

and was bringing the Oka down for a landing. Following the Oka's disastrous debut, Oka Model 22 in the colours of the example found
He inadvertently released the water ballast reviews of gun camera footage from the US at the close of the war and now on display at the
from the nose tank, leaving the rear tank full. Navy fighters and from pilot debriefings Smithsonian Air and Space Museum.
This immediately caused the nose to pitch revealed the existence of the newweapon for
up, putting the Oka into a stall that Kariya was the first time to the Allies. At first it was
unable to recover from, the plane crashing thought that the Oka was simply a large, anti-
into the ground. Kariya was pulled from the ship bomb. This would change when four to
wreckage but within a few hours had died six examples were captured near KadenaAir- Oka Model 43B - data (estimated)
from his injuries. field after the Allied victory at Okinawa. Only
By December 1944, Ktgisho had produced then was the aircraft's true nature made T1pe SpecialAttacker
151 Okas and the Dai-lchi Kaigun Kokusho known to Allied intelligence. The Oka was Crew One

production was also well under way. subsequently given the codename Baha by
Attempts were made to deploy the Oka lo the Allies, the word baka meaning 'fool' in Pow€rplant One Ne20 axial-flon turbojel developing 175kg

units outside of the Japanese home islands. Japanese. (1,047Ib) olthrust: one Tlpe I tr'lark I \'lodel 20 solid luel rocket.

Fifty were dispatched to the Philippines Production of the Oka Model l1 ceased in deleloping 256kg (565Ib) of thrust

aboard the carrier Shinono, but on 29 Novem- March 1945 with the Dai-lchi Kaigun Kdkl
ber 1944 the ship was sunk en route. Only a Gijutsu-sho having built 155 and the Dai-lchi
Span 8.99m 29.5f1
handful would reach other bases, notably in Kaigun Kokusho constructed a total of 600. Length 8.16m 26.8fi
Okinawa and Singapore, and none would see One Oka Model 11 was fitted with sheet steel Height 1,12m 3.7ft
combat. Even though the 721st had yet to see wings made by Nakajima but no other exam- Wing area 12.99m: 139.9ftr

combat, there were some who realised that ples were produced with this feature. To help Wing loading l74.79kgim, 35.8lbrfr

the G4M2e bomber would be easy targets for improve the training regimen, once the Dai- Pouerloading 5.18kgihp 12.1 lbihp

enemy fighters and the odds of actually Ichi Kaigun Koku Gijutsu-sho had completed
reaching the target were small. Conse- their run Oka production was switched to the Weights
Emptl i,l50kg 2.535tb
quently, morale dropped as the Oka was seen Oka MXYZ K-1 trainer. In all, 45 of the K-l
Loaded 2,270kg 5.0011b
as a waste of a pilot who could be used to bet- would be completed and placed into the pilot
Useful load I,l20kg 2,4691b
ter effect elsewhere. The vulnerability of the training program.
G4M2e was vividly displayed when the 721st Clearly, the G4M2e carrier aircraft was too Performance
went into battle for the first time on 2l March slow and easy prey for defending Allied llar speed 556kmih 315mph
1945. Attacking US Task Group 58.1, all 18 fighter protection. In addition, the short range at 4,000m at l3.l25it
bombers (of which 15 were Oka carriers) of the Oka Model 1l compounded the prob- Cruise speed NlA

were shot out of the sky by US Nar,y fighters lem. Consequently, Krlgisho decided to utilise i\'lax range 277km 172 miles

along with their fighter escort before they the superior Klgisho P1Y Ginga (Allied code- Ceiling

could getwithin attack range. Again, the story name Fronces) as the carrier aircraft and also Fuel capacitl 300 litres 79.2 gallons

0il capacitl 4.2 gallons

of this and subsequent Oka missions are to give the Oka a longer range. This adapta- 16 litrcs

beyond the scope of this book but the inter- tion was called the Oka Model 22.
ested reader can find many excellent sources The primary change in the Oka Model 22
800kg (1,763 lb) explosir,e narhead
of information on the topic. was the use of the Tsu-11 thermojet engine in

Inrpenral JepeNesE Navv l,)

Oka Model 53 - data (estimated) place of the rocket boosters. This consisted of planned on using the Nakajima G8M1 Renzan
a 100hp Hitachi Hatsukaze [Ha-I1-11] l1 4- (known as RrTa to the Allies) but with the fail-
Type Special Attacket cylinder, inverted inline engine driving a sin- ure of the Renzan to enter production, the
Crew One (or none) gle-stage compressor. Fuel was injected into Oka Model 33 was quickly shelved without
the compressed air that was then ignited, pro- any prototype being constructed. This was
ducing up to 200kg (440 lb) of thrust. To com- followed by the Oka Model 43A. Larger in
One Ne20 arial-florv turbojei developing 475kg (1,047Ib) ofthrust; one
pensate for the weight of the engine and fuel, dimensions in comparison to the Oka Model
Trpe 4 \'lark I solid iuel rocket, developing 267kg (;88lb) 0f thrust
the warhead had to be reduced to 600kg 22, rhe Oka Model 43A was designed to be
(l,323 lb). Finally, as the PlYwas smaller than launched from submarines such as the Sen
Span 6.43m 21,lft the G4M2e, it was necessary to reduce the Toku class. To facilitate storage on such boats
Length 7.77m 25.5fr wing span by lm (3.2ft), although the length the wings were foldable, but with the Allies in
Height 1.43m 4.7ft of the Oka Model 22 was increased by .8m complete control of the seas the oka Model
Wing area 8,99m' 96.8fr, (2.6ft). These changes improved its range of 43A was soon put aside and work begun on
Winq loadins N/A
up to 129km (80 miles), although 65km (40 the oka Model 43B instead.
Porver Ioading N/A
miles) or less was considered achievable This version was designed to operate from
under combat conditions. A rocket booster caves and launched by a catapult. It retained
could be fitted to the underside of the fuse- the folding wings to allow the production
Empty N/A
lage to increase speed during the terminal lines to be set up in cramped, underground
Loaded NlA

Useful load N/A dive. sites or caves as well. Unlike the previous
Once the design of the Oka Model 22 was models, the oka Model 43B was all metal,
Performance finalised, Ktgisho began a production run of used a central skid and in order to better facil-
N'lar speed NlA 50 aircraft even before flight testing was itate target penetration the pilot could iettison
Cruise speed N/A underway. Aichi Kokuki K.K. was contracted the wing tips. Like the Oka Model 22, a rocket
l,la\ range 277km 172 miles
to construct a further 200 Model 22 aircraft, booster could be carried under the fuselage.
Ceiling N1A
but due to US B-29 bomber raids Aichi's pro- A full scale wooden mock-up was completed
capacitt 400 litres 105.6 gallons
duction lines would never enter operation. in June 1945 and was promptly approved for
Oil capacitv l6 litres 4.2 gallons
Once the first handful of Oka Model 22 aircraft production. Aichi were tasked with construc-
had been made available their testing began. tion of the Oka Model 43B at their Gifu and
600ke (1.322lb) erplosire warhead Thanks to its short wings, a high siall speed of Oyaki factories but the war ended before the
334kmih (207mph) and high landing speed first prototype was completed. However, a
made a soft landing impossible. Test pilots catapult ramp was built at Takeyama, near
were instructed to abandon the Oka rather Yokohama, and pilots destined for the Oka
than make a landing. Lieutenant Kazutoshi Model 43B were being given instructions on
Nagano took the Oka Model 22upfor the first catapult launching as theywaited for their air-
time on 26 June 1945. The flight would also be craft to be delivered.
his last. After being released from a modified A hybrid Oka was considered which was
Kugisho PlYl at 3,658m (12,000ft), the Oka called the Oka Model 2 1 . The Tsu- 1 I engine
went out of control (another source states was to be removed from the Oka Model 22 to
that the wing rocket boosters fired acciden- be replaced by the standard rocket booster
tally, causing the Oka to crash into the Ginga, system as used on the Oka Model ll. This
Oka K-l - data damaging the Oka's controls). With no ability may have been contemplated as production
Type Trainer to regain level flight from the plummeting of the Tsu-l1 engine was slow and was not
Crew 0ne Oka, Nagano was able to extract himself from keeping pace with the Oka Model 22. The
the stricken aircraft but his parachute only proposal, however, never proceeded past a
Powerplant partially opened before he hit the ground and single prototype.
was killed. A second test model was ready in Whereas all previous Okas, with the excep-
Dimensions August 1945 but the war ended before it could tion of the Models 43A and 43B, required
Span 5.12m l6.8ft
fly. Although fifty Oka Model 22s were built, modified bombers to carry them aloft and
Length 6.06m 19.9ft
the carrier, the Kr-rgishd P1Y3 Model 33, would launch, the Ne 20 turbojet equipped Oka
Height l,l2m 3.7fr
never leave the drawing board. The com- Model 53 was designed to be towed into the
Wing area 6,00m' 64,6fr
pleted Oka Model 22 were retained in Japan air. As such any aircraft, with the addition of a
for use against the expected Allied invasion tow line and having enough power, could be
EmpI 730k9 1,6091b force. used to tow the Oka Model 53 into the air.
Loaded 2,l20kg 1,673tb K0gishO continued to investigate ways to Nothing came of this design due to the end of
Uselul load l50kg 330Ib improve the performance of the Oka and a the war. However, it is worth noting that some
series of models were planned around the contemporary illustrations show the Oka
Performance KugishO Ne 20 turbojet. The first was the Oka Model 53 without a cockpit, which would turn
['lar speed N/A
Model 33 which was simply the Oka Model 22 the type into a glider bomb. For guidance, it is
enlarged to accept the Ne20 (or as one
Cruise speed l47krilh 9imph
speculated that upon release from the tow
Landing speed 200k'ri,"h l24mph
source states using the Ne12B.iet engines aircraft, it was either radio controlled from a
that had been built prior to the shift to the parent plane or used infrared or acoustic
Ne20 development). For a carrier, Kugisho homing to guide itself to the target. This con-

T4 JepnNssE SscnEr PRo;Ecrs: ExpeRrn'rEnral ArncRnnr oF THE IJA AND IJN 1939-1945
cept has not been verified in rvartime Japan- G4M2e parent aircraft. In order to get some Oka +3 K-l Kai - data
ese sources and could be post-rvar conjec- use from the Okas, mcchanics planned to fit
ture. them lvith floats cannibalised from unser- Ttpe Trainel

A derivative of the MXYT K-1 rvas planned viceable or available floatplanes such as the Cren Ino
and this rvas known as the Oka Modcl 43 K-l Aichi E13A (hnown as Joke to the Allies). It is
Kai Wakazakura (meaning 'Young Cherry' in not knorvn exactly horv the lloats r,vere to be
0ne Tvpc 1 \lark 1 -solld hrel r0cket. de\elopinq 26llq (ritjlb) ollhm,'l
.lapanese). This was to be the definitive installed but crudc fittings could have been
Ir';rinel lol pilots destined for operalional Oka fabricated to attach a float under each ,"ving. Dirnensions
models. A second cockpit r'vas installed in the It is belicrcd that lhe float cquipped Okas Span i.l2rn 16.8fr
nose in place of the warhead, flaps were fit- were to be positioned along the Straits of L.enqth 6.0lin 19.ltil

ted to the rvings to help with landing and, like Johor that separate Johor from Singapore and Heiqht Ll2nr I ift
the K-l, the Nlodel 43 had a central landing be used in conjunction ',vith Shinyo special \\ inq iirea !A
skid with wing bumpers. It also included a attack boats. Another unknor,vn is holv they
single Type 4 Mark I Model 20 rockct in the lvould have performed given the short burn iVeights
lnrplr ll+kq
tail to allorv the student to get a taste of time of thc rocket boosters let alone handling t,.lt 1l lb

Loaded El{)kg t,i.q; lb

porvered flight. By the close of the r.var only qualities across watcr. It can be surmised that
I setul load llilkg i{ii llr
trvo of the Wakazakura trainers had been performance rvould have been very poor. By
completed. comparison, thc German Tornado attack Performance
Perhaps one of the more unusual uses for boat used two floats fi-om a Junkers \l:u -pce d !A
the Oka occurred in Singapore. The handful Ju 52l3mg5e and rvas pon'ered by an Argus Cruisc spccd l2llkm h E0mph
of oka Model I 1 aircraft that were received by 109-014 pulse-jet. Trials rvould prove a failure Landirr! spcd \,A
units in Singapore rvere, for the most part, as the boat could not operate on anything but
groi,rnded because they did not not have thcir calm seas rr illroul cap5isinq. fumament \one


Iuprrir,ql JaparEsn Nlvy 75


Oka Model 43 K-l Kai

shown in the
standard orange hue
for training and
experimental aircraft.

K-l Ktgisho MXYT Oka - data

Contemporaries Daimler-Benz Projekt E and Prolekt F (Germany), Kugisho oka N'lodel I 1 Kugisho oka Model I I

Messerschmitt N{e 328C (0ermanl) Originallv this okawas in the collection oithe \iictoryAjr l'luseum ]Ocated This, the onl,-v knom genuine oka in Japan (a replica is used at the

in l\'lundelern, Illinors. It closed ils doors in 1984 and the aircraft lvas sold Yasukuni Shrine in Chiroda. Tokroar nell a. lhe Okd Paik in Ka.hima

Deplolrnent off, It ruas lasl obtained by the Yanks Air I'luseum in Chino, Calilornia. Citr, Ibaraki Prefecture), is currently housed at ImmaAir Base, a Japan-

\raried. The oka I'lodel I I and NIX\'7 K"l san'operational seruice. The ese Air Self.Deience Force facilit.v in lruma, Saitama Prefecture. During
K[gish6 oka ]\'lodel 1 I
Oka \lodel 22 u as deplor ed bul did not .ee acriOn. Ihe oka luodel 2 I the \4ar it rvas at the IJA base IrumagarvaAirfield,
Shipped to India in September 1947 bt'the No,4 Squadron of the Indian
remained a single protot\De. The oka I'lodel 138 protot\pe uas incom-
Arr Force lollou'ing their duties in Japan as pa( of the British Common- Kugisho oka l\'lodel 22
plete at rvar's end. The oka N'lodel 33, 43A and 53 remained designs
nealth Occupation Forces, the Oka is cunentlv on display in the Indian Restored and on displa,r'at the Smithsonian Air and Space l,luseum.
0nh" The oka ['10del 43 K-l Kai $'as too late to be issued to training
Air Force Museum at Palam Air Force Station, Nen Delhi, India.
units. K[gisho I'IXY7 K-l
K[gisho oka \'lodel I I \Vith a serial number of 5100, the \lXY7 K-l is housed in the Nary
Another Oka in England, this time drsplaled at the FleetAirArm l'luseum at the Washington Naly Yard in Washington, D.C.
Museum in Yeorilton. Somersel, LK.
Kugisho MXYT oka (FE-N50)
Kugish6 MXYT K-l
The 'N' is assumed to signifr 'Nalr' as in the US Narr'. therebv denoting Kngisho oka Model I 1
This lrainer is cunenth on display at the National N'luseum of lhe United
the LrS Narryu'as evaluating this parliculal aircraft. This MXYT was listed This example olthe oka is in the collection ofthe tr'luseum ofScience Siates Air Force, located on the proped\' 0f the WrighlPattersonAir
on a l0 l'larch 10 1946 repofi allou'ing il to be released t0 a\iation indus- and industry in \'lanchester, UK, $'ith the registration number ol Force Base in Riverside, Ohio, which is just outside 0fDavton. Ohio. At
try. 0n 1 August 1 I 916, an inventorr reported it at \'lA['1A and bv I 8 Sep- 11996.53.10. Since 1961, this Oka has passed through a number olmuse one Iime itha. pdinted a' an operalional Oka but ha" rince been
tember i916 \\'as slated lor lhe museum and storage at Park Ridge. No ums beiore ii reached its currenl lOcatiOn. retumed t0 the orange colour scheme ofa trainer,
further trace ofthe [JXY7 is kno$n.
Kugisho oka tr'lodel I I
Kogisho oka l\,lodel {3 K-1 Kai
K0gisho oka \'lode1 I I (no tail number assigned) This oka is housed in the collection 0f the Deience Lrplosive Ordnance
At present. this trainer bearing the serial number 6l is in st0rage and
One of a number oiOka aircrait taken t0 lhe US, this one, serial number Disposal School. The school is cunentll located in Chattenden, Kent, but
unrestored at the Paul E. Garber facility.
1019, was obtained bI Ed$ ard \'lahonev. li $as restored and remains on is to be relocated t0 St. George's Barracks, Bicester in Oxfordshire, UK.

displav at his Planes ol Fame Museum in Valle, Arizona. (Noie: The Kugish0 oka \'lodel I I displa.ved at the Wings ofEagles
K[gish0 oka Model I I
museum in Horseheads, NY, is a replica,)
Kngisho oka Model I 1 The lourih oka in the British Isles is on display at the Rolal Air Force

Bearing the serial 1018, this oka is currenth'on displa| at the l'larine \luseum at Cosiord in Shropshire. Prior to this, it uas housed at the
Corp. Barp qron1 , o. near I rianqle. \'irqinia. Rocket Propulsion Establishment in\\restcott, Buckinghamshire.

76 Japallrse SEcnnr PRoJECTS: Explnlr'tEnraL AIRcRntr oF rHE IJA AND IJN 1939-1945
Kugisho MXYS and MXY9

The design and development ol the Mitsubishi the glider down, Inuzuka gave the MXY8 high Klgisho MXY8 - data
J8Ml Syusui presented a challenge. Despite the marks, having found the handling and flight
Contemporaries \lesserschmilt I{e 1635, Hcinkel He l62S (Cermanr')
information available to the Japanese on the characteristics to be very acceptable.
Messerschmitt Me 1638, upon which the J8M1 The IJA, who were also slated to fly the J8M1 Data is fot tlrc ,\lXY8. The specrficotions also applt lo the fahoi Ku'l 3.

was based, the concept of a tailless fighter, let as the Ki-200, was provided with the second trio specific datais at:ailable on the \IXYS Shuha,

alone a rocket powered one, was new and MXY8. Delivered to the Rikugun Kokuyijutsu
untested. What was required was a means to Kenkyujo, the pilot selected to test the MXY8 for T}?e Proof ol ConcepliTraining Glider

verify the design of the J8M1 and in doing so the IJAwas Colonel Aramaki, and like Inuzuka, Crew

provide a way to train pilots who would be fly- he felt that the MXY8 performed well. The only
ing a plane that was unlike any they had ever notable deficiency to be found by both Inuzuka
flown before. and Aramaki was the tendency for the MXY8 to
Therefore, Kugishd was given the task of cre- nose over into a dive. The third MXY8 to be built Span 9.50m 31.2ft

ating a glider that was to be a copy of the J8Ml. was delivered to the Naval Air Force. Length 5.82m lg.lft
The main purpose was to assess the flight char- The first MXYS did not match the combat Height 2.16rn Lift
acteristics of the tailless fighter given that the weisht of the J8M1. To this end the IJN wanted Wing area 17.6im 190ft

Japanese did not have extensive experience to modify the MXY8 so that it incorporated bal-
with such aircraft. Data collected from flying last tanks'nvhich could hold enough water to lVeights

the glider would in turn be reviewed and fully simulate the combat weight of the J8NII Emptr 90jkg 1,99;lb
applied to the J8M1 fighter prior to series pro- and be the definitive production model for use Loadcd l.037kg 2,286Ib
duction. In addition to seruing as a proof of con- in training pilots. With the completion of the ini-
cept vehicle, the glider would provide the tial three MXY8 gliders by Klgisho, a number of
\lax glide speed Lnknown
means to train new pilots in flying the aircraft manufacturers were organised to begin the \'lax tou' spced 295knrh l83mph
since it was like no other fighter then in service production of the revised 'hear'y' MXY8 glider to
in the IJN and IJA. By using the glider as a meet the training needs of the 312th Kokutai
trainer, pilots could better transition to the J8M 1 and other units that would be flying the J8M1
and therefore minimise operational mistakes. and Ki-200. Maeda KOk[ Kenkyujo was tasked Deplolment Kugisho built the thiee prolott pe MX\'8 gliden, \laeda
.ll to r4 \'lX\'8 trainers rvhile Yokoi produced
Krigisho assigned the glider construction, with producing the MXY8 for the IJN and the constructed 6 Ku-13 train'
ers. A number of l\'1XY8 gliders wcre operated bl the 312th Koklrtai.
called the MXY8 Akigusa (meaning 'Autumn MXYS was to be built for the IJA by Yokoi Kokti
\0 \'l\\9 $as constructed and the pr0ject remained a design onlt.
Grass'), to engineer Hidemasa Kimura. Kimura K.K. as the Ku-l3.
utilised wood with some cloth covered sur- Further flight testing by the 312th Kdklrtai
faces in the design of the MXY8 and ensured found that the MXY8 experienced aileron flutter Another version of the glider was investi-
that the glider was a near exact replica of the at speeds above 295km/h (183mph) (as a side gated by the IJN. Whereas the MXY8 was
JBM1 in order to provide the most accurate note, the same problem was encountered in unpowered, the new version would have some
flight data once testing got under way. By the the Messerschmitt Me 163A Vl first prototlpe means of propulsion. Designated the MXY9
close of 1944, the first MXYS was finished and during testing). This and other minor problems Shuka (meaning 'Autumn Fire'), the new
another two were nearing completion. were noted, analysed and corrected, and the design was to be an advanced trainer which,
In December, the first MXY8 was taken to the flutter issue was resolved by closing the gap because it had the means to propel itself when
airfield located in Hyakurigahara, which is between the wing and the aileron (the Me 163A it was airborne, would provide training with a
about 79km (49 miles) northeast of Toky6. It Vl was rebalanced). In the meantime, the modicum of power and offer longer flight
was here that the tJN's Hyakurigahara Air MXY8 was being flown by NavalAir Force pilots times. It was envisioned that once training in
Group was stationed, operating in the defence at the Kashiwa airfield in Chiba Prefecture. the MXY8 was completed, pilots would transi-
of Toky6. Also at the airfield was the 312th However, the pilots were less enthusiastic on tion to the MXY9 for advanced training before
Kokritai, a newly formed unit that was to be the design, especially after a crash involving moving to the JSMl or Ki-200. The propulsion
equipped with the J8Ml once it entered pro- one of the gliders that severely injured the pilot. method proposed was the Tsu-l1 thermojet.
duction. As such the 31 2th Kdkt-ttai was the per- Regardless, the Kaigun Koku Hombu assessed This was the same engine as used in the
fect group to begin testing the MXY8. The first all of the data from the test flights and formally Ktgisho Oka Model 22 (see Page 70). However,
tlight, scheduled for 8 December 1944, was approved the MXY8 on March 1945 and work the MXY9 was never
given to Commander One. Unfortunately, One proceeded with full production of the MXY8 realised.
was taken ill and was unable to fly so the mis- and Ku-13 that continued until the end of the
sion was assigned to Lieutenant-Commander
Toyohiko Inuzuka. On the day of the flight,
lnuzuka climbed into the cockpit of the MXYS
and once secure, a Kyrishir K1OWl of the 312th
Kdkr-rtai took the glider into the sky. At altitude
Inuzuka was released from the tow plane and z
began his descent. After successfully bringing

MXYS shown in the orange colouration as used on

trainers and experimental aircraft.

InpsRraL JapnrlrsE Navv 77

Kugisho R2Y Keiun


Given the expanse of the Japanese empire by order to proceed with the RlYl, K[gishO fuselage with the shaft running forward,
1942, the IJN found that they had a need for a decided to utilise two Mitsubishi MK10A through the middle of the cockpit, spinning a
long range reconnaissance aircraft that could radial engines. In so doing, the RlY1 took on four-bladed propeller in the nose. To cool the
operate from land bases and fly at high speed the appearance of the Kugishd P1Y1 Ginga paired powerplants, the He 119 used a wing
to render it immune to interception. In the and with the use of two radials and the result- surface evaporation system in which steam
same year, the IJN issued a 17-shi specifica- ine drag imposed by them the RlYl's calcu- from the engines was circulated through the
tion for just such a plane and Kugisho looked lated performance was projected to fall wings where it cooled and condensed back
to provide the response. below the 17-shi specification. Consequently, to liquid where it was pumped back to the
The 1942 17-shi specification called for the all work on the RlYl ended and the project engines. To cool the engines when on the
aircraft to have a maximum speed of was abandoned. ground or during take-off and landing (due to
667km/h (414mph) at 6,000m (19,685f1) along Even as Ktgisho was working on the RlYl, the lack of sufficient airflow across the wings
with a mission profile of long range recon- they were developing another design, the Y- at such times), a supplementary radiator was
naissance. The speed requirement stemmed 40, which was the result of an evaluation of installed under the forward fuselage. The
from the need to be able to avoid intercep- the Heinkel He 119, two examples being pur- He 119 Vl first prototype attained a top speed
tion; the intelligence it gathered would be chased from Germany in 1940. of 565km/h (351mph) at 4,500m (14,765ft).
useless if the aircraft was shot down before it The He 119 was an attempt to create a fast, Unfortunately, Heinkel was forced into
could return to base. The initial design, the unarmed reconnaissance aircraft whose high adding armament to the He 119 but this was
RlYl Gyoun (meaning'Dawn Cloud'; other speed would enable it to avoid interception done in a very minimal fashion. The V2, with
sources use Seiun, meaning 'Blue Cloud'), and elude pursuit. To accomplish this, the a full functional bomb bay, was able to reach
bore the designation Y-30 and was to be He1l9 used radical concepts to minimise speeds of up to 585km/h (363mph) at 4,500m
developed around a new Mitsubishi, 24-cylin- drag and thus enhance speed performance. A (14,765f1). The V4 was used as a record
der, liquid-cooled engine that was projected pair of Daimler-Benz DB 601 engines coupled breaker, briefly holding the record for speed
to produce 2,500hp. However, delivery of the together drove a single propeller shaft. The with a 1,000kg (2,205Ib) payload on a closed
engine was not expected to be rapid and in engines were placed in the rear of the He I 19 1,000km (621 mile) circuit with the average

78 hplruEse SrcRsr Pno:Ecrs: ExpEnn'rpNraL ArRcRnnr oF THE IJA arun IJN 1939-1945
speed of 504.97km/h (313.78mph). Later the aircraft. Following the defeat of the Japanese Contemporary sources show no less than
V4 was wrecked in a crash during an attempt in the Marianas Islands (following Operation four versions for how the R2Y2 may ultimately
to better that time. The record was set on 22 Forager), the fate of the Keiun was all but have appeared. The first had the Ne330
November I 937 and the successful aeroplane sealed. The IJN had no need for such a plane engines in underwing nacelles. The second
was listed as the 'He 606'. The V3 was built as as existing designs would be adequate for the version showed the two engines buried
a float-plane, intended to best the 1,000km dwindling Japanese holdings. In addition, the within the fuselage with wing root air intakes
(621 mile) seaplane speed record. Ultimately, need for fighters and bombers was rather and narrow jet nozzles. The third removed
the V5 through to the V8 would be the last more urgent than reconnaissance aircraft. the wing intakes and replaced them with a
He 119 aircraft built because the Luftwaffe But Otsuki and KDgishO did not let the nose intake, but it retained the narrow noz-
showed no further interest in the aircraft. Keiun fall by the wayside. In late 1944, zles. Finally, the fourthwas similar to the third
Following testing in the summer of 1938, a Klgishd approached the IJN and informed save that the engine nozzles were larger. The
delegation from the IJN was able to inspect them that the R2Y1's airframe was readily first design is considered by most to be the ini-
the He119 at Marienehe in Germany. The adaptable to other roles, including that of a tial R2Y2 concept while the other three are
Japanese were most impressed by the range fast attack bomber. To heighten the interest, subject to debate. In part, this is due to the
offered by the He 119 as well as its speed. Of itwas proposed that the IHa-70] 10 engine be fact that the Japanese had very little time to
interest were the coupled DB601 engines. replaced with two Mitsubishi Ne330 axial- explore various installations of turboiets in
After reporting their positive findings, nine flow turbojets, each of the engines being airframes. The easiest means to place turbo-
technicians from the Dai-lchi Kaigun Kokrl slung under the wings in nacelles. The fuse- jets on aircraft was by using nacelles and this
Gijutsu-sho flew from Japan to Germany to lage space vacated by the Aichi engine would was seen in the Nakajima Kitsuka, Nakaiima
study the He 119 further. Commander Hideo be replaced with fuel tanks. For weapons, the Ki-201 Karyu and proposed Klgishd Tenga
Tsukada arranged to obtain the manufactur- aircraft would carry one 1,800kg (7641b and Kawanishi K-200.
ing licence for the He 119 and also the pur- bomb) and have a cannon armament in the Even the Germans with their turbojet expe-
chase of the He 119 V7 and V8. Both aircraft nose. With the introduction of the Ne330 rience did not fully understand the effects of
were crated for shipment and sent to Japan engines, the maximum speed was expected a long nose intake feeding a high perfor-
arriving in May 1940. Reassembled at Kasum- to be 495mph, superior to the projected mance jet buried in a combat fighter's fuse-
igaura, Krlgishd began flight trials under the 720kmlh (447mph) top speed of the Aichi lage. Messerschmitt, when they began to
leadership of Major Shoichi Suzuki. During engine model. With these advantages in study how to start the P.1 101 second genera-
the brief trials, one He 1 l9 was lost to landing mind, the IJN approved that work should tion jet fighter, catalogued the obstacles that
gear failure (the He 1 19 used a special, begin on designing the R2Y2, the turbojet needed to be overcome. They included the
retracting telescopic oleo leg in order for the powered Keiun which was sometimes effects of engine operation on the fuselage
long landing gear to fit into the wings). In the referred to as the Keiun-Kai, as well as per- integrity, ensuring the nose intake was prop-
end plans to manufacture the He 1 19 in Japan mitting the R2Y1 to be completed as an air- erly positioned and shaped for maximum air-
did not come to fruition. frame demonstrator to test the handling flow, making sure the intake tube was made
Although the He 119 was rejected, the study characteristics. as smooth as possible to minimise air restric-
of this aircraft resulted in the development of In April 1945, the first prototlpe of the R2Y1 tions, how to protect the rear of the aircraft
the Y-40. Like the He 119, the Y-40 was to be a was completed and moved to Kisarazu in from the heat generated by the exhaust
fast, unarmed two-seat reconnaissance air- Chiba Prefecture to begin flight testing. Initial thrust, the effects of reduced airflow on thrust
craft using coupled engines placed within the taxi trials, conducted by KugishO test pilot due to flight angles and more. The Germans
fuselage behind the cockpit and driving a pro- Tereoka, showed that the nose wheel had a were at least able to devote some time to
peller via an extension shaft. The IJN's l8-shi bad shimmy when in motion and the Aichi investigating these problems and providing
specifications for a long range reconnais- engine was prone to overheating. The latter solutions to them. This was time however,
sance aircraft were based on the Y-40. was either due to a lack of airflow through the that the Japanese simply did not have. Up
The Y-40 project, by now called the R2Y1 radiators and inlets during taxi tests or until the construction of the P.1101 Vl and the
Keiun (meaning 'Beautiful Cloud'), was led through a poorly designed cooling system. planned Focke-Wulf Ta 183, all of the
by Commander Shiro Otsuki and his design Nevertheless, despite the problems testing of wartime jet designs flown by the Luftwaffe
team made good progress. The Keiun was to the Keiun continued. had nacelle mounted turbojets. The Japan-
be equipped with two Aichi Atsuta 30 engines On 8 May 1945, Lieutenalrt-Commander ese may not have been made fully prily to the
coupled together in a combination known as Kitajima, another Kr-rgisho test pilot, took the latest German jet engine technology as it per-
the Aichi IHa-701 10. The 24-cylinder, liquid- Keiun on its first flight. Kitajima noticed that tained to long intakes before the war ended.
cooled IHa-701 10 was rated at amaximum of the oil temperature was rising rapidly and he It is within reason to suggest that the R2Y2
3,400hp and drove, via the extension shaft, a cut short the flight, landing the Keiun before with the wing root intakes could have been
6-bladed propeller. The Keiun did not use the the engine suffered damage. Mechanics and under consideration since it would be a logi-
same method of cooling as the He1l9. engineers continued to try and solve the cool- cal development, especially since such
Instead, it relied on air intakes and a radiator ing problems, but a few days later the engine intake arrangements were not entirely new.
bath underneath the fuselage, and it also dif- caught fire during ground testing, completely The third and fourth designs may or may not
fered from the He 119 in that the Keiun used destroying the power unit. Then before the have been post-war conjecture.
a tricycle landing gear system and was not a Keiun could be returned to Krigish6 to receive Unfortunately for Klgisho and the IJN, the
tail sitter. a new engine, the aircraft was destroyed by a R2Y2 would never be brought to full produc-
By the fall of 1944, the war situation for US bombing raid. tion. With the end of the war, the second
Japan was deteriorating. With the loss of terri- Even before the destruction of the first R2Yl prototype remained incomplete and the
tory to the advancing Allies, the IJN no longer R2Y1, a second was being constructed and R2Y2 would forever remain a design board
saw a need for a long range reconnaissance design work for the R2Y2 was underway. aircraft.

Il,Ipenral JepnnEsp Navv 79

Klgisho R2Y Keiun - data
Contemporaries F
\'lesserschmitt Me 509 (GermanD, Tupolev Tu-91 (NAT0 codename

Boot) (Rusia), l'leserschmitt P, I 100 (Germany)

The specificot[ons in parcnthesis ate for the R2Y2 aith the underulng

Type Long range reconnaissance airclaft (altack aircraft)

Crew Tu'o

Powerplant OneAjchi[Ha-70] 10,24-c1'linder,liquid-cooled

engine developing 3,400hp at take-offand 3,l00hp at 3.000rn/9.845it
driving a 6-bladed metal propeller (trvo Ne330 arial-flow turbojeh

developing l,320kg/2,9 I 0 lb of thrust each)

Span 13.99m 45.9ft

(R2YD 13.99m 45.9ft

Length 13,01m 42.8ft

(R2Ya 13.04m 42.8ft

Height 4.23m 13.9ft

(R2YD 4.23m 13.9ft

Wing area 33.99m' 365.9ft'

(R2YD 33.99m' 365.9ft'

Wing loading 238.26k9/m' 48.81b/ftl

(R2Y2) 269.99k9tm' 55.3ib/fl'

Power loading 2,35k9/hp 5.21b/hp

(R2Y2) 3,22k9hp 7,1 lb/hp

Empty 6,0l5kg 13,261 lb

(R2\2) 5,700k9 12,5661b

Loaded 9,400kg 20,i231b

(R2YD 9,950kg 21,9351b

Fuel capacitv 1.515 litres ,li I gallons

(R2YD 3,2l8 lilres 850 gallons

NIa\ speed 20kn'h 447mph

at 10,000m at32,8l0ft
(R2Y2) 797kn/h 495mph

at mean sea level, estimated

Cruise speed 464knvh 288mph

at4,000m at l3,l25ft

Landing speed l66km/h lO3mph

(R2YD 158kn/h 98mph

Range 3,l39km 1,950 miles

IR2YD l.269km 788miles

Climb I0 min to 10,000m (32,81Oft)

(R2YD 7 min to 10,000m (32,810it)

Ceiling | 1,700m 38,385fI

(R2Y2) 10,700m 35.104f1

None (one 800k#1,764Ib bomb and a batten'of fomatd fiilng cannon)


Concerning the three other R2Y2 jet variants rvilh internal engines, little is

documented and much is open to conjecture. In some instances, the tdng

span. length, height and wing area are listed as being the same for the

R2Y1 but the speed is given as being a marimum oi800knr'/h (497mph).

None. One prOt0tlpe 0[the R2Yl was built and flom while the second
R2Yl prototvDe was unfinished by the end of the nar, The R2Y2 slal'ed

on the design board,

80 .hp.qrrlEsE SEcREt PRorscrs: ExpsRIttpNral AIncRerr oF rHE IJA AND IJN 1939-1945
Kiigisho Tenga

Of the many Japanese experimental aircraft Despite the problems production got mov-
of World War 2, perhaps none is more of a ing and design changes saw the machine
mystery than the Kugish6 Tenga. The Tenga guns replaced with Type 99 Model 1 20mm
(which can mean the Milky Way as one trans- cannons and l3mm Type 2 machine guns.
lation of the kanji) was to be a first for Japan: Other changes included revised engine cowl-
a turbojet powered bomber. To realise this ings, replacing the retractable tail wheel with
ambition as quickly as possible, the Japanese a fixed wheel, moving from flush riveting to
intended to use one of their latest and best fl at-head riveting, incorporating a bullet-proof

bomber designs - the Klgisho PlY Ginga panel in the windshield and also replacing
('ginga' also means MilkyWay) - as the basis the Homare 11 engines with the Homare 12
for the Tenga. which could produce 1,825hp.
If one examines the aircraft available to the After these modifications, the IJN finally
Japanese during the war, the distinct lack of accepted the PlYl Ginga bomber into ser-
a medium bomber quickly becomes evident. vice. But the type was still nagged by prob-
Whereas most of the warring powers oper- lems, notably the Homare 12 engines which
ated medium bombers (for example, the rarely produced the horsepower they were
Martin B-26 Marauder, the Junkers Ju88 or rated for. Such issues delayed the Ginga
the Vickers Wellington), the Japanese were entering combat until the spring of 1945. Even
very late in bringing such aircraft to the front. though the Ginga would see battle for a mere
The IJA brought the Ki-67 Hiryu medium six months the design nevertheless proved to
bomber into service in 1944 and so the IJN be a capable medium bomber and one
looked to the Klgish6 PIY Ginga as their which the Allies respected when they
answer to the need for a medium bomber. encountered it. When the Allies first heard of
Development of the Ginga began in 1940 as the plane they thought it was a healy fighter
the Dai-lchi Kaigun Koku Gijutsu-sho's and assigned the codename Francr's to it
attempt to meet a 1S-Shi specification for a (after Francis 'Fran' Williams of the Material
medium bomber capable of high speeds, the Section of the Directorate of Intelligence,
ability to conduct low-level bombing and tor- Allied Air Forces, Southwest Pacific Area).
pedo missions, and the capability to perform However, when the Ginga was finally spotted
dive bombing. With Tadanao Mitsuzi and after 1943, itwas realised that itwas abomber
Masao Yamana at the helm the Y-20, as the and the name was changed to the feminine
Gingawas called at this stage, emerged as an version of Francis - Frances.
aerodynamically clean, mid-wing, twin- The Ginga was developed into several vari-
engine design. Despite its relatively small ants and there were plans to use the bomber
size, the Ginga had fourteen fuel tanks (of as a carrier for the Oka Model 21 and Oka
which only eight had some protection from Model 22 suicide aircraft. Kawanishi built a
battle damage), a modicum of armour for the night-fighter/intruder version as the PlY2-S,
pilot (which consisted of a 20mm thick plate which entered service with the IJN as the
behind his head), a light defensive armament Kyokko (meaning 'Aurora') in the summer of
of a 7.7mm machine gun in the nose and in 1944. The Kyokko was fitted with Mitsubishi
the rear of the cockpit, and the ability to carry Kasei 25a 1,850hp, 14-cylinder radials
a single 800kg (1,764 lb) torpedo or two 500kg because the Homare 12 could not be assem-
(1,102 lb) bombs. With the two Nakajima bled fast enough to meet demand. Weapons
Homare I 1 18-cylinder, air-cooled radial included two forward oblique mounted
engines developing 1,280hp each, speed was 20mm Type 99 Model 2 cannons firing
estimated at 556km/h (345mph). upwards and the nose cannon was removed.
The first prototype was completed in First flown in July 1944, it was found that the
August 1943 and flight testing began shortly Kyokko did not perform well at the high alti-
afterwards. Test pilots found that the Ginga tudes where the Boeing B-29s roamed. This
possessed excellent speed and also dis- revelation was so disappointing that the
played good handling qualities. Ground upward firing cannons were removed and
crews on the other hand had anything but the Kyokko returned to its bomber role as the
good things to say about the aircraft. The Ginga Model 16 (P1Y2). Nakajima also built a
Homare 11 engines and the hydraulic system similar night-fighter version as the P1Yl-S
used in the Ginga were a constant mainte- Byakko (which meant 'White Light'). The
nance hassle, requiring far more time and Byakko fared little better than the Kyokko and
effort to maintain than was considered rea- did not see service. Other modifications and
sonable. So bad were the problems, the IJN plans included upgrading the engines to the
postponed its acceptance of the aircraft. Homare 23, Kasei 25c orthe Mitsubishi MK9A,

In'tpERrnr- JnpaNEse Navv 8t

the idea of an attack model with ten to sixteen would have been the logical starting point. utilise as much of the Ginga as possible, offer-
forward firing 20mm cannons and using steel Quite possibly, RATO units may have been ing the benefit of an airframe already in pro-
and wood in the aircraft's construction. The needed to boost the Tenga off the ground. duction with proven aitworthY
most interesting was the P1Y3 Model 33. This Clearly the Ne 130 would have been a better characteristics. This would have reduced
version was to be built from the ground up to selection and with the Ne 330, the Tenga development time when the need for such a
carry the Oka and would have had a special would have enjoyed a noticeable speed bomber was most urgent. It would have also
bomb bay to accept the oka Model 21 or 22 improvement. Problems with the develop- served as a starting point for aerodynamic
with increased wing span and an enlarged ment of the Ne 30 engine are cited as a reason testing. Still, when one reviews the jet
fuselage. The PlY3 never left the drawing for the Tenga project being cancelled. bomber proposals of other nations, the num-
board. Indeed, the Ne 30, an off-shoot from the Ne 12 ber of piston engine to jet engine concepts
With the Ginga's success in terms of per- program, never advanced, being surpassed can be counted on a single hand. For the
formance, it's easy to see why there was by the Ne 20, Ne 1 30, Ne 230 and Ne 330 devel- maiority, the jet bomber was designed from
interest in converting it to turbojet power. The opments. the ground up instead of being adapted from
concept of the Tenga was certainly real. But But could the basic airframe of the Ginga an exisiting aircraft. The designers of the
outside of the name and the basic intent to be used with radial engines replacing with Tenga may have come to the same conclu-
replace the radial engines with turbojets, turbojets? It may have been attempted had sion had they had the opportunity to continue
nothing else is known. Therefore, one has to the Tenga advanced in design. Even changing their work. If so, the final Tenga design and
review other designs to make assumptions the radials for turbojets would have necessi- prototlpe may have borne little resemblance
on what kind of task the Japanese might have tated fairly significant adjustments in the to the Ginga with which it shares its name.
faced in making the Tenga a flying reality. wings to accommodate them but at least
The first point to consider would be that the redesigning a wing to accept turbojets is a
Kugishd Ne20 turboiets, then in production simpler task than redesigning the entire
for use in the Nakajima Kitsuka, would not aircraft.
have been sufficient to provide the Tenga However, if one examines the history of
with any meaningful speed if mounted one combat aircraft, you would be hard pressed
per wing. One Ne 20 produced 487kg to find a conventional combustion engined
(l,074Ib) of static thrustand two could propel bomber switching to jet power merely by Kugish6 Tenga - data
the Kitsuka to 623km/h (387mph), which was changing the engines and adiusting the
not particularly signifi cant over conventional wings. For instance, not even among the Contemporaries

high-performance propeller driven aircraft. dozens of jet bomber projects undertaken by Arado Ar234 Blitz (Germany), Heinkel He 343 (Cermany),
NorthAmerican B-45 Tornado (US), Illushin IL-22 (Russia)
The Kitsuka was a much lighter aircraft and a the Germans did a piston-engined bomber
twin turboiet Tenga using Ne20s would not switch its engines for turboiets without heavy Therc is no exact infomat:nn ouailable on the specilications for the
Kitgish1 Tenga. The dato prouided belou is based on the Tenga haD[ng
have been feasible. modifications, if at all. One such example
used the PIY Ginga abfiame ptetty much aefuotlm, apot't fr\m the
It would have needed some of the pro- was the Messerschmitt Me 264 which used change of engines (as the Tengo [s often depicted). Et)en then,
jected advances in the Ne20's development four Junkers Jumo 9-21 1 radial engines when information is fragmentary and subject to guessoorh,

to come closer to reality to provide the Tenga the first prototype was flight tested. However,
with a meaningful system of propulsion. The the proposed four turbojet engined version Type Medium Bomber
Crew Three
Ne 30 turbojetwas expected to generate up to bore little resemblance to the original design.
850k9 (1,873Ib) of thrust (better than the Ger- Perhaps the only notable propeller to tur-
Powerplant (planned)
man BI\,fW 003 turbojet rated at bojet design created byadaptating an existing Two Ne 30 turbojets producing 850kg (l,8i3 lb) of thrust each
800k9/1,763Ib) while the Ishikawaiima airframe was the Russian Tupolev Tu-12
Ne 130 was projected to produce 900kg whose heritage was owed to the Tu-2, one of Dimensions

(1,984Ib) of static thrust, comparable to the the premiere Soviet light bombers. Built from Span 19.99m 65,6ft

Length 14.99m 49.2fr

Junkers Jumo 004 engine. The Nakajima 1941 through 1948, the Tu-2 possessed fast
Height 4.29m l4.lft
Ne 230 and the Mitsubishi Ne 330 were esti- speed, excellent agility and had a substantial Wing area 55,00m' 592ft'
mated to be able to produce 885kg (1,951 lb) weapon fit and bomb carrying capacity. (likely to be different with turbojets)

and l,300kg (2,866|b) of thrust respectively When Tupolev answered the call to produce
with the Ne 230 sacrificing thrust for a lighter a jet bomber, he took the Tu-2 as the basis for Weights
Empty N/A
weight. his Tu-12. He used the fuselage, wings and Loaded N/A
It is said that the Ne 30 would have been the tailplane of the Tu-2 and adapted them to suit
initial choice to power the Tenga had it been the installation of two Rolls-Royce Nene-l tur- Performance
available. ln comparison, the German Arado bojets and the higher speeds that would Ma\ speed N/A

tu234B jet bomber used two Jumo 004 result. Although one can certainly see the lin- Range N/A

Climb N/A
engines. Itwas similarin size to the PlYl, the eage of the Tu-2 in the Tu-l2, the aircraft still
Ceiling N/A
notable differences being a smaller wing required a general redesign to cope with the
span, the loaded weight was nearly 680kg new engines and the associated handling Armament
(1,500Ib) lighter and the tu234B had far less characteristics and was not simply a case of Two 20mm cannons or l3mm machine guns, one in the nose, the other
rear firing; one 800kg (1,764Ib) torpedo or behveen l,000kg to l,560kg
wing area. Together, the two Jumo 004 swapping the radial engines for turbojets. The
(2,205Ib to 3,525Ib) ofbombs
engines could move the Ar2348 at speeds up design of the Tu-l2 began in 1946 and the first
to 742kmlh (46lmph). Certainly, two Ne30 flight took place in June 1947. Deplolment
engines would not have provided such a It is not unreasonable to conclude that the None. The Tenga existed onlyas a paper design.

speed when mounted to the PlYl but it initial Tenga designers may have tried to

82 JeperuEsn SEcRET PRoJECTS: ExpenrunntnL Atncnepr oF rHE IJA IJN 1939-1945

Iuppnnl .lnpANsss NA.w 83

Kyushu J?W Shinden

In an l8-shi specification
1943, the IJN issued (26,246f1) in nine minutes and the ability to cle landing gearwas employed, the front tyre
that included the requirement for a new inter- carry at least two 30mm cannons. To go with being 550xl50mm and the two main tyres
ceptor. Japanese aircraft tasked with inter- these two proiects, Tsuruno introduced the being 725x200mm in size.
ception roles had by this time begun to be Shinden to the IJN as a third competitor. The Shinden would carry four Type 5
eclipsed by the newest Allied fighters and the Despite some opposition to the design, the 30mm cannons. The Type 5, while heavier
IJN sought to ensure their edge was main- IJN was intrigued enough to accept the Shin- than the earlier Type 2 30mm gun, possessed
tained. Three contenders submitted their den proposal. However, the design had to a higher rate of fire at 500 rounds per minute
designs and it would be Kyrlshrl's that was the show promise and the canard configuration and had a higher muzzle velocity. Each can-
most radical of them all: the J7W Shinden. needed to be proven before the IJN would non was provided with 66 rounds. With less
The man behind the Shinden ('Magnificent authorise further development. Therefore, than eight seconds of 30mm rounds per gun,
Lightning') was Captain Masaoki Tsuruno, a Tsuruno designed a glider based on his Shin- one hit would be sufficient to cripple and
member of the Technical Staff of the IJN. Tsu- den concept as a means to test the canard shoot dow,n a fighter or bomber, therefore
runo conceived an interceptor that made use properties and handling. Kugisho was com- there was little ammunition to waste. There-
of a configuration rarely seen at the time of his missioned to build three gliders which were fore, there were two Type 1 7.92mm machine
design work, a design with canard fore- called the MXY6. Further details on MXY6 guns, one on either side of the gun camera, in
planes. Canards were not a new concept, development can be found on page 69. The front of the nose. The purpose of these guns
even in 1943. They were seen as far back as initial, positive results achieved with the was not offensive but to serve as a ranging
I 91 0 with a Gabriel and Charles Voisin design MXY6 convinced the IJN to move forward weapon for the cannons. Upon lining of his
and later a Bleriot tail-first aircraft had incor- with the Shinden project even before the target, the pilot would fire a short burst frorn
porated canards. (Both planes used the term completion of the glider testing by authorising the machine guns. If the rounds struck the tar-
'canard'which in French means 'duck'- the two prototypes of the J7W1. get, he would fire a burst from the cannons
1910 Canard Voisin and the 1911 Bl6riot The IJN gave the Shinden project to Kyushit and be reasonably assured of a hit, thereby
'Canard'). Canards would sporadically Hikoki K.K. even though Kyltshu had no expe- conserving the precious cannon ammuni-
appear in experimental aircraft right up to rience with high performance aircraft, let tion. Each Type I was provided with 75
World War 2, examples being the 1929 Raab- alone one like the Shinden. Unlike other rounds of ammunition in a saddle drum mag-
Katzenstein Rakate, the 1931 Focke-Wulf maior manufacturers however, Kyushrl's azine. It should be noted that sources often
Ente (the 'Duck') and the 1937 Beltrame Col- research facilities, personnel and production list the two Type I weapons for training pur-
ibri. Tsuruno felt canards could offer a num- capacity were not heavily taxed by the needs poses, i.e. for practice and gunnery training,
ber of advantages such as reducing the of the Japanese war machine. To assist not gun laying. While certainly possible, gun
chances of stalling, improved controllability Kyushl, the IJN placed a team from the Dai- laying would seem more plausible due to the
and manoeuvrability and easing some con- Ichi Kaigun K6kr1 Gijutsu-shd as well as Tsu- rapid ammunition consumption of the Type 5
struction concerns such as the engine instal- runo himself at the disposal of Kyushu cannons and in training there is no real rea-
lation and control linkage arrangements. engineers and managers to bolster their son why machine guns would be used as a
Besides the canards, Tsuruno introduced capability in handling the program. substitute for cannons. For payload, the Shin-
another feature in the Shinden that was cer- With everything in position work com- den had a modest bomb carrying capacity of
tainly new to the Japanese - the use of a tur- menced on the first prototlpe in June 1944. 120kg (264Ib).
bojet to power the aircraft. Of course, The heart of the JTW1 was the Mitsubishi BySeptember 1944, amodel of the Shinden
Tsuruno understood that a more conven- MK9D ([Ha-43]) 12 l8-cylinder radial engine was being tested in a wind tunnel to assess its
tional piston-engine would have to be used boosted by a supercharger. Although Tsuruno aerodynamic properties of the shape and
until such time as a suitable turboiet became wanted to use a turbojet he rejected the planforms. With the results deemed accept-
available, but a turbojet was incorporated Ne 12B (TR-12) as insufficient in terms of able, the first metal was cut on the prototype
into his original design to ensure that the tran- thrust production. And since further turbojet at the Kyushu Haruda factory located in
sition r,vould not present any difficulties. At developments were projected to show Fukuoka City. By May 1945, the Shinden was
this time, the Shinden was known as the X- I 8. improved performance, the Shinden would nearly complete although it lacked the
By the time Tsuruno's initial layout for the use this radial engine until such time as a suit- canopy, landing gear fairings, much of the
Shinden was complete the IJN had already able turbojet was available. The Mitsubishi main wings and other components. As the
issued its late 1943 18-shi specifications for engine and its superchargerwere mounted in MK9D ([Ha-43]) 12 was alreadyinstalled, test-
three classes of aircraft. The first of these cov- the rear of the fuselage. A six-bladed, metal ing of the powerplant commenced and trials
ered an air superiority fighter (18-shi Ko), the Sumitomo VDM propeller was mated to the showed that a cooling problem existed, prob-
second for an interceptor (18-shi Otsu) and engine through an extension shaft and placed ably in part because no airflow was reaching
the third for a night fighter (18-shi Hei). For in a pusher configuration. If required the pro- the engine during static testing.
the 18-shi Otsu competition, both Nakajima peller could be jettisoned to effect pilot bail- In June 1945, the first prototype was fin-
and Kawanishi had submitted designs: the out. On each side of the fuselage were air ished but the armament was not fitted.
single-engine J6K1 Jinpu ('Squall') and the intakes for cooling the engine. The main Instead, weights simulating the Type 5 can-
twin-engine J5Ni Tenrai (or'Heavenly Thun- wings were swept and on each was mounted nons were installed in the nose. Flight testing
der') respectively. These entries were based a vertical stabiliser situated in approximately was to commence immediately although the
on the rather sparse directives of the specifi- the middle of the wing. The pilot sat in a cock- problem with the engine cooling would delay
cation which called for a top speed of pit in the centre of the fuselage while the the first flight until 3 July. Tsuruno would be
665km/h (413mph), a climb to 8,000m canards were mounted on the nose. A tricy- the first to fly the Shinden. The aircraft was to

g4 JapanEsr SEcnEr PRo;Ecrs: ExpsnrMeNrar- AIRcnerr op rre IJA eNo IJN 1939-1945


IupEnraL Japai'ress Nnvv 85

.laplrinsE Srcner Pnotrcls: Erprnrnrsr"r'aL Arrtcnanr crr. rrrn I.JA axo I.lN 1939-1945

IuprnrnL JapnxEsE Nevv 87

take-off from the Mushiroda Airport in cannons remained. The flight would be made
Fukuoka City. The engine was started and by Kyushu's Yoshitaka Miyaishi. With the
Tsuruno climbed into the cockpit. On releas- torque now a known issue, Miyaishi was able
ing the brakes and commencing to taxi, the to compensate and the Shinden successfully
Shinden showed an unexpected heavy took to the air for the first time. The flight was
torque pulling to the right. Surprised, Tsurano for a very short duration and the aircraft was
was unable to stop the plane swerving off the not taken above 400m (1,312ft). On landing
runway where the propeller touched the Miyaishi reported that the take-off was rela-
ground bending several of the blades by as tively easy but rudder rise was experienced at
much as 28cm (1 1 inches). In addition, a por- 185km/h (1lSmph) with the plane lifting off
tion of the right wing's vertical stabiliser was the ground at l93km/h (120mph). As he
also damaged. The accident would delay climbed pushing the speed Io 222krn/h
flight testing for nearly a month while repairs (l38mph), ttre pull to the right caused by the
were made. To prevent the propeller from torque from the propeller was very notice-
striking the ground, tail wheels, as used on able. After levelling off at 400m (1,312ft) and
the KyushD Kl 1W Shiragiku trainer, were fit- at a maximum speed of 16lmph, the pullwas
ted to the bottom of the vertical stabilisers. still a problem. He also found the landing to
On 3 August 1945, the Shinden was pre- be atrickyprocess. The Shindenwas found to
pared for flight. Only 38 litres (10 gallons) of have a relatively fast landing speed at
fuel were loaded with 80 litres (21 gallons) of 240krn/h (149mph) and because of the
lubricating oil. The weights simulating the torque pull and the inclination of the nose,
bringing the Shinden down was not a simple
A second flight was made on 6 August 1945
KyushuJZW Shinden - data with Miyaishi at the controls. Manoeuvrability
was the mdn focus of the test, though the air-
Contemporaries Performance
craftwas flown up to 491m (1,61Oft). The pilot
Cuniss-Wright XP-55 Ascender (US), Henschel P,75 (Germany), Gotha Mar speed 750km,h 466mph
found that during the climb the nose wanted
Co P.50/l (Germany), Henschel P.87 (Germanv), Messerschmitt P.1 I l0 8,700m
at at 28,545fi
to dip. Again, the pull to the right was evident
Ente (Oermany), Miles M.35 and M,39B Libellula (UK), Mikoyan- 800krnh 497mph
(J7W2) atl0,000m and during landing if a slight rudder up posi-
Curevich MiG-S Uika (Russia), Arnbrosini SS,4 (ltaly) at32,8081t

The spec[f[cations in parenthesis ae for the JM2 Shinden-Kai in uhich Cruise speed 424knlh 263mph
tion was applied the nose would pop up just
the standad lM I aifrcme is used, at4,000m at 13,125f1 before touchdoum. It was also discovered
449km^ 279mph that the oil temperature rose as the flightwent
Type Interceptor (J7W2) at6,000m atl9,685ft on and a means to rectify the problem would
Crew One Landing speed 240kn/h l49mph have to be found.
Take-offlength 560m 1,837ft On 8 August 1945, the third and final flight
Powerplant One Mitsubishi MK9D (lHa-431) 12, l8-cylinder, Landing length 580m 1,902ft
of the Shinden took place with Miyaishi at the
air-cooled radial engine, developing 2,l30hp at take-off, 2,020hp at Range 85lkm 529 miles
controls. At lg3km/h (l20mph), the nose
I,l80n/3,870ft and l,160hp at 8,700n/28,545f1, driving a sk-bladed,
altemating stroke propeller (one Ne 130 arjal-flow turbojet developing

2.5 hours

10 min 40 sec to 8,000m (26,245ft)

wheel left the runway and at
(126mph) the plane lifted off. Miyaishi noted
900kg/1,984lb of thru$) Ceiling 12,000m 39,370[t
that the nose tended to drop regardless of
Dimensions Armament how fast or slow the engine revolutions were.
Span I L09m 36.4ft FourTlpe 5 30mm cannons with 66 rounds per cannon, two Tlpe I He also found that even with the nose down,
(JiWD ll,09m 36.4fr 7.92mm ranging machine guns wilh 75 rounds perweapon with the the Shinden still flew horizontally with a
Length 9,63m 3L6ft capacity to carry two 60kg (132 Ib) bombs or four30kg (66lb) bombs level track and slight application of the rud-
(r7wD 9.63m 3 l.6ft der. Landing speed was again 240km/h
Height 3.90m I 2,8fr Deployment (149mph).
(J7WD 3.90m 12.8ft None, Two JTWl Shinden aircraft lvere produced. Although the aircraft
In total, the first Shinden logged between
Wing area 20.49m' 220.65ft! was ordered into produclion, no further examples had been constructed
30 and 45 minutes in the air. In addition to the
(JiWD 20.49m' 220.65fr' bv the end ofthe war. The J7W2 remained a design only.

problems already noted, Miyaishi com-
Wing loading 241.19k9/m'
(Jiw2) 240.2lkgln' 49.2 tb/ft, SuMvors
plained of strong vibrations in the fuselage, in
Power loading 2,40kgArp 5,3lb,tp Kiush[ JilVl Shinden (FE-326) part due to the engine torque and from the
(J7w2) 5.44kg/ttp l2.0lb/hp This was the second Shinden prototpe and lvas captuled at Ky.rshu's extension shaft that mated the propeller to
main factory. It was jisted on the aviation industry Ielease report on the engine. With the flight results, Kyushu
Weights l0 March 1946 and to undergo restoralion at I\'IAMA by I August 1946. engineers set about the process of solving the
Empty 3,525kg i,771 lb FE.326 was moved to Park Ridqe in September 1946. 0f note is the torque and vibration problems as well as the
(J7W2) 3,465k9 7,6391b Shindenwas provided with a stipulation that it should be housed in such
cooling concerns.
way that it could be quickly removed from storage ifan engine and
Loaded 4,950k9 10,912tb a
However, even before the Shinden took
(J7W2) 4,930kg 10,868rb other parls were to be obtained to bring it to flight status. This never
flight the IJN was desperately in need of a
Ma,c Ioaded 5,272k9 ll,62tb happened but the Shinden was lortunate enough to escape the cutting

Garber facilitywhete it cunently

high performance interceptor. The Kawan-
Useful load l,425kg 3,141 lb torch and was moved to the Paul. E.

(J7W2) l,a6skg 3,229tb remains in pieces and unrestored (see page 86),
ishi J6Kl Jinpu failed to show any improve-
ment over the Kawanishi NlK2-J Shiden-Kai

88 JnpnNsss Sscnnr PRo.lrcrs: ExpERrn,teNrlL ArRcRnpr oF THE IJA AND IJN 1939-1945
(meaning 'Violet Lightning', known to the the BMW 003A, which was being developed peller clearance the Shinden could dispense
Allies as George) and the NakajimaJ5Nl Ten- by Ishikawajima-Shibaura. The Ne 130 was to with the tall landing gear, thus lowering the
rai was proving to be a disappointment by the have produced nearly double the thrust ofthe height of the aircraft. Aside from modifying
time flight trials commenced in July 1944. Ne20; however, the Nel30 would not be the landing gear, the vertical stabilisers, fuse-
With the failure of these two entries for the ready by the close of the war and as such the lage and canopy shape may also have
18-shi specification, the IJN ordered the JTWl J7W2 remained a design board aircraft. There required adjustment. It is likely that had
Shinden into production in May 1944 and in so is speculation concerning what the J7W2 development of the J7W2 commenced with
doing made the b,pe the only canard configu- would have looked like. One suggestion is the availability of the Ne 130, a J7W1 airframe
ration aircraft to achieve this status during that the J7W2 would have been the J7W1 would have been adapted to accept the
World War 2. By September 1944, the pro- with the Mitsubishi radial replaced with the engine and testing conducted on this, with
duction plans had been formulated with Ne 130 turbojet. A second suggestion con- other modifications coming into play as a pro-
Kyushu's Zasshonokuma factory expected to cerns the fact that without the need for pro- duction J7W2 was standardised.
turn out 30 Shindens per month while Naka-
jima's Handa plant would produce 120 Shin-
dens each month. In light of the war situation,
such production numbers would have been
very difficult to meet. As it was, the war ended
before production could get going. z
In addition to the Iirst prototype, the second
machine was also completed but it did not fly
before the end of hostilities. The war's end
meant that the modifications required to cor-
rect the problems found during ground and
flight testing were never made. As a side note,
nearly four decades later Colonel Bob
Thacker would construct a radio controlled
flying model of the Shinden. His initial proto-
type showed the same pull problem as the
original Shinden resulting in two crashes that
required the complete rebuilding of the
model each time. To solve this problem,
Thacker placed the front canards at 7.5" pos-
itive incidence, adiusted the centre of gravity
and pointed the extension shaft for the pro-
peller 3" to the right and 4.5" down from the
zero thrust line. The pull problem was suc-
cessfully eliminated by these adjustments
and the Kyushri engineers may have found
the same solution had they had the time to
implement it.
The Shinden was not an easy aircraft to fly.
Given the configuration of the aircraft, it
would have required a skilled pilot to use in
combat and Japan's forces were low on
experienced pilots by the time the Shinden
would have entered service. The same prob-
lem would plague the Germans when their
high performance turboiet fighters were
coming into operational use.
The end of the war also spelled the end of
the J7W2 Shinden-Kai. This was to be the tur-
boiet-equipped version of the Shinden. The
use of the radial engine had been a stop gap
until a suitable turbojet was available. The
Ne 128 was reiected as its power was consid-
ered too low to effectively propel the aircraft.
In any case, work was by this time under way
on the Kugisho Ne 20 turbojet that was based
on the German BMW 0034, engine, the only
turbojet built and flown in aJapanese aircraft:
the Nakajima Kitsuka (page 114). The J7W2
was to use the Ne 130 turbojet, also based on

IupEnreL Japer{Ess Nevy 89



t\ rF*
b \- &ir.
_..r ? 4 "- t
1* #:
_.. I r.:::

g0 Jlp,qr-nse Sucnrl Pno;rcls: Err,EnnrE:lrAr. Alttcnalr op lrrE IJA,qrn IJN 1939-1945

Mitsubishi GZM Taizan



r l!}* ?

f- F,
&... .;@
::'.t: *

Prior to the start of World War 2, there were 6,043km (3,749 miles) this r,vas insufficient to The illustration of lhe GTMI provided in this book
men rvho foresaw the need for long range attack targets in the United States or, if is based upon Kijiro Takahashi's design. It is
sholvn in the colours of the Yokosuka Kdkutai,
strategic bombers capable of striking across required, deep into Russia. Folloi'r'ing this
vast distances. Men like Generalleutnant revier'v, the Kaigun Koku Hombu put forth the
Walter Wever of Germany who pushed i.vith 16-shi specification for an attack bomber. whicli upon revierv rvas allor'ved to proceed.
urgency the need for such bombers despite Only trvo key specifications rvere stated. The Takahashi's version of the NI-60 \,vas to use
the veritable wall of opposition to such first'uvas that the marimurn speed had to be trvo 'Nu' engines. The Nu engine r,vas a 24-
endeavours. This r,vas also the case in Japan at least 580km/h (36lmph) and the second cylinder, horizontal-H, liquid-cooled engine.
rvhere it was clear that aggression against the r'vas a maximum range of at least 7,340km Simply put, a horizontal-H cngine is trvo flat
United States would require the capability of (4,598 miles). Nlitsubishi set about the task of engines placed one on top of the other and
hitting the US. Therefore in 1941, the Kaigun designing a bomber capable of meeting these geared together (a flat cngine is one in n'hich
Koku Hombu issued its 16-shi specification requirements. the pistons move horizontally). Despite hav-
for a long range bomber. Initially, Mitsubishi engineer Kiro Honjo ing a poor porver to rveight ratio, they ofTer the
A review of bombers in service',vith the IJN (who designed the G3UI and G4N{) proposed advantage of being more compact and,
by 1941 showed that none were capable of that the 16-shi bomber should be of a because of this, Takahashi elected to use
crossing the Pacific to attack distant targets. four engine design. Within Mitsubishi the them on his bomber. Each engine n'as rated
At the time, the Mitsubishi G3NI (codenamed bomber was knor,vn as the M-tj0. His pro- at 2,200hp at 5,000m (16,404ft).
lVe// by the Allies) was being phased out to be posal, however, was flatly rejected by the In appearance, Tahahershi's bomber bore a
replaced by the Mitsubishi C4M (known as Kaigun Koku Hombu. Instead, another strong resernblance to the Heinkel He 177
Betty Io the Allies and Hamaki or'Cigar'to the Nlitsubishi engineer, Kijiro Takahashi, put for- Greif (German fbr 'Giiffon') heavy bomber
Japanese). Although the G4N41 had a range of rvard his own design for the l6-shi bomber that first flerv on l9 November 1939. 'lhe nose

Irrpenral JapaNEsE Nariv 91

Mitsubishi G7M Taizan - data was rounded and fully glazed, a style unlike By 1942, Mitsubishi had completed the
any Japanese bomber then in service. The bomber's design and were ready to construct
Contemporaries wings were mounted mid-fuselage, each a full size wooden mock-up of the GTMI
Bri$ol Buckineham (UK), Lockheed P2V Neptune (US), Junken Ju 88H- wing sporting the Nu engine in awell-stream- Taizan, which was in due course completed.
1 and H-2 (Germanl, Junkers Ju 288A (Gemany) lined nacelle. On top of the fuselage, fore and Unfortunately, the Kaigun Koku Hombu had
The specificotions for the GTM\ Taizan are baed on the design aft of the wings, was a turret for a portion of now issued a 17-shi specification for a
dimensions Tnd estimated performance of the linal G7M1 proposal as the defensive armament. A fairly spacious tail bomber that Kawanishi was developing as
deriuedbl Mitsubishi. gunner position was fitted beneath the verti- the K-100 (which some sources designate as
cal stabiliser on the underside of the fuselage the G9K Gunzan, meaning'Mountain Group',
Type Long-Range Bombet
with a rear facing ventral gun station. A rela- but this has never been verified; other
Crew Seven
tive rarity in Japanese bomber design was the sources have the G9K as a 1944 19-shi
tricycle landing gear. Takahashi's perfor- bomber project). Mitsubishi was instructed to
Powerplant Trvo ['litsubishi MKI0A (Ha'42'l 1) lS'cylinder, ait-
halt all further work on the G7M1 until the
Ohp at 2,200m
mance estimates put the normal operational
cooled radial engines developing 2,000hp ior take'off, 1,8 I

range at 6,412km (3,984 miles) which, with a K-100 could be evaluated.

(7,21 7ft) and l,720hp at 5,400m (l 7,7l6fi); each engine drove a metal,
4.5m (l4.7it) diameter lighter payload, could meet the 16-shi specifi- Kawanishi completed the initial design of
four-bladed, alternating stroke propeller with a

cation. The maximum speed would have the K-100 bomber and the Kaigun Koku
Dimensions been 555km/h (345mph) with a relatively Hombu reviewed it along with the G7M1 in
Span 25.00m 82.Ofr light defensive weapon armament of two the summer of 1943. By this time, the GTMI
Length 20.00m 65.6fr
Type 99 20mm cannons and two TYPe 97 had suffered further range performance
Height 6.09m 20,0ft
7.7mm machine guns. reductions, dropping from a proposed nor-
Wing area N/A
Unfortunately for Takahashi, Operation mal range of 3,705km to 2,778km (2,302 miles
Barbarossa, the German invasion of the to 1,726 miles). This was caused in part by the
Wing loading N/A

Power loading 3.99kg/hp 8.8lb/hp

Soviet Union on 22 June 1 941 , would prevent heavier armament compared to the initial fit,
the required machine tools and equipment to removing the two Type 1 machine guns and
Empty 10,60Okg 23,368lb produce the Nu engine from being exported adding three more Type 2 machine guns to
Loaded 16,000kg 35,273Ib to Japan. Without the powerplant, the design make a total of six Type 2s. This was, to a
Lselul load 5,400kg l l,904lb was doomed. With Takahashi's proposal hav- degree, tempered by a higher speed of
Bomb load 800ks l,764lb marimum ing fallen by the wayside, Kiro Honlo resumed 544km/h (344mph) at 5,000m (16,404ft).
control of the M-60 proiect. This time, instead Based on the projected performances of
Performance of a four-engine bomber, Honio would utilise both aircraft, it was felt that neither design
Max speed 544knth 344mph
two engines and base his design heavily on would be suitable either for the 16-shi or the
17-shi specifications. The Kaigun Koku Hombu
at 5,000m at 26.246ft
the G4M.
Norma) range 2,799km 1,739 miles
The G7M Taizan (meaning'Great Mountain') was critical of the G7M1's design for concen-
\'lar range 7,400km 4,598 miles
as the design was later designated was to use trating much of the defensive weaponry in the
Climb l0 min to 10,000m (32,808f1)
N/A two 18-cylinder, air-cooled radial engines, the frontal arc of the bomber, thereby reducing the
Fuel capacily 4,497 littes l,188 gallons Mitsubishi MKl0A (Ha-42-11), developing aircraft's defences in the side and rear arcs. In
2,000hp each. The wings were mounted mid- addition, it was considered that the actual per-
Armament way on the fuselage and the aircraft was to be formance of the G7M1 would likely have been
Six 13mm Type 2 machine guns, two mounted in each oi hvo upper constructed of metal with fabric covering the little, if at all, better than the operational G4M 1 .
luselage tunets (one fOnvard. One alt of the wings) and two in aventral, ailerons and rudders. It was anticipated that Another nail in the coffin for the G7M1 was the
rear firing position: hvo 2Omm Tlpe 99 \'lodel 2 cannons, one mOunted
the Taizan would carry the same 800kg fact that the Kaigun Koku Hombu was looking
in the nose, the othel in the tail
(1,764Ib) bomb load as the G4M1 but unlike to four-engine bombers as the real means to
the Hamaki, the Taizan would have a far more achieve the necessary range (at least
poient defensive armament as the bomber 8,816km/5,478 miles, allowing for a one way
\one, Awooden mock-up was built befole the Taizan project was
would operate far from fighter protection. This trip from Tdkyd to Los Angeles). In fact as early
step also took into account the shortcomings in as 1938 the IJN had asked Nakajima to produce
the G4M1's protection. Of course, using less a four-engine bomber, the G5N Shinzan
powerful engines and a heavier weapon fit ('Mountain Recess') which was based on an
caused a revision in performance when com- imported Douglas DC-4E.
pared to Takahashi's design. A 31 October 1942 With the Kaigun Koku Hombu showing no
performance estimate gave ihe G7M1 a range interest in the G7M, Mitsubishi shelved all fur-
of 5,559km (3,454miles) ataspeedof 518km/h ther work on the bomber. Ironically, the G5N
(322mph) at 5,000m (16,404ft) with a weapon Shinzan would prove a failure and had a
fit of two Type 99 20mm cannons, two Type 2 worse range than the G4M. Onlywith the con-
13mm machine guns and two Type 1 7.9mm struction of the four-engine Nakajima G8N1
machine guns. However, as work on the G7M1 Renzan ('Mountain Range') which first flew in
proceeded and the design underwent further October 1944 would the original l6-shi range
testing, these estimates would continue to be specification be met. By then, the need for
revised. Unfortunately for Mitsubishi, the such bombers had passed as attention had
revised estimates did not see any expected turned to defending Japan and fighters/inter-
improvements to the performance but rather ceptors were required.
some deterioration.

92 JApANESE SEcRET Pno:Ects: ExpeRIntEIrtteL ArncRnnr OF THE IJA AND IJN 1939-1945
Mitsubishi J4M Senden


The concept of the J4M Senden ('Flashing ventional when compared to Kawanishi's of the turbochargerwas to boost the manifold
Lightning') was born of the need for a high design, the J3K1. The proposed plane, known pressure on the engine over and above oper-
performance interceptor that could operate within the company as the M-70, was a mono- ating pressures at sea level as a means to
at high altitude. The main catalyst for this was plane pusher design that featured twin maintain and improve performance at alti-
the American Boeing B-17 Flying Fortress. booms connected to vertical stabilisers by a tude. For landing gear, the Senden had a tri-
The bomber, in action in the Pacific Theatre low mounted horizontal stabiliser. The cycle arrangement with the nose gear
from 1941 to 1943, proved to be difficult to booms were slung under the low, fuselage retracting into the fuselage and the main
intercept since it normally flew at heights that mounted wings. The heart of the aircraft was wheels being housed in the booms. The pilot
operational Japanese fighters could not reach to be the Mitsubishi [Ha-431 12 MK9D tur- sat in the glazed nose of the aircraft in a cock-
or attain with difficulty. Even if an interception bocharged, radial engine. Rated at 1,650hp at pit that was blended into the fuselage. The
was achieved, the B-17 carried a formidable 8,000m (26,246fr), it was projected that this majority of the Senden was constructed of
defensive armament with which to protect engine would push the Senden to a top speed metal with fabric being used on the rudders
itself. To a lesser extent, the Consolidated of 704km/h (437mph) via its six bladed pro- and ailerons.
B-24 Liberator was also a factor when it peller. For weapons, there was a Type 5 The Senden came in two versions. The first
began to replace the B-17s still remaining in 30mm cannon and two Type 99 20mm can- was the project described above while the
the Pacific. ln 1942, two companies, Mit- nons. All three were arranged in the fuselage second variation replaced the blended cock-
subishi and Kawanishi, were given a 17-shi nose with the Type 5 being centrally mounted pit with a bubble canopy to improve the
Otsu specification by the Kaigun Koku and the two Type 99 cannons on either side of pilot's radius of vision. It also removed the
Hombu to develop an aircraft to meet the the fuselage. If required, the aircraft could protruding inlets and replaced them with two
need for a high altitude, high performance carry a small bomb load of up to 120kg bands of flush inlets thatwrapped around the
aircraft. (2641b). Mounted across the top of the fuse- fuselage, the first being directly behind the
Mitsubishi Jukogyo K.K.'s response to the lage behind the cockpitwere inlets to feed air cockpit and the second around the engine
17-shi Otsu directive was anything but con- to the turbocharger and engine. The purpose area just past the wings. Finally, the horizon-

Iupnnrlr. .llpnxEse Nlvv 93

Mitsubishi J4M Senden - data
Contemporaries o
Focke"Wulf mit BMW 803 (Germanl'), Vultee XP-54 'Su,oose Goose' z
(US), Belyayev EOI (Russia), Bell XP-59 (lS)

Specifications in parentheses rcfet to lhe J4Ml Project 2 Senden only

Tlpe Fighter/lnterceptol

Crew 0ne

Powerplant One l'litsubishi (Ha-43) N'IK9D iS-cylindet, air-

cooled radial deleloping 2, I 00hp at engine start, l,900hp at 2,000m

(6,561fi) and l,650hp at 8,000m (26,246t1), driving a alternating stroke,

six bladed propellerrvith a 3.2m (l0.5ft) diameter

Span 12,49m 4lft
Length 12.98 42.6fr tal stabiliser was moved to the top of the ver- the support of the IJN, Mitsubishi were told to
(J4M4) 12.49m 4lfr tical stabilisers. The remainder of the aircraft cease work on the Senden and instead further
He;ght 3,47m I l,4ll was basically the same between the two ver- develop the Mitsubishi AZM Reppr-r ('Hurri-
Wing area 22.00m' 236.8ft,
sions. The blended cockpitversion is credited cane') to meet the 17-shi Otsu standards. The
(J4M4) 24,69m' 265.8ft:
as the J4M1 Project 1 while the second, with result was the A7M3-J Model 34 Rifuku (Land
Wing loading l99.69kgh' 40.9lbifr
the bubble canopy and modified inlets, is Wind) that had not advanced beyond the
197.73kg/m' 40.5lb/ft
sometimes referred to as the J4M4 Project 2. design phase before the war ended.
After analysing the two designs, Mitsubishi Despite the fact that the J4M Senden did not
Empty (J41\'r4) 3,400kg 7,195.7 lb selected the original configuration, the J4M1, progress past a wind tunnel model, US air
Loaded 4,400kg 9,700.3Ib to develop further. To confirm their initial pro- intelligence was aware of the design mainly
(J4Nr4) 4,486kg 9,889,91b jections, a full scale model was constructed through captured documentation. In the Jan-
Mar loaded [J1\'r4) s,25sks 1 1,585.2 tb in 1943 and put to the test in a wind tunnel. uary 1945 issue of the US RecognitionJournal,
Unfortunately for Mitsubishi, the tests proved the J4M Senden was announced as a possible
to be a disappointment. Performance projec- adversary in the coming weeks of the war.
Nlax speed 756knlh 470mph
tions based on the testing were below the ini- No artist renderings of the Senden were
at 8,000m at 26, 246ft
tial calculations and problems with the MK9D included in the article. The J4M was given the
703krn'h 436mph
(J4M4) at 8.000m at 26, 2461t
in terms of not reaching its horsepower rating codename Luhe in anticipation of Allied
Cruise speed 462knv'h 287mph only added to the concerns. pilots encountering the aircraft in combat,
(J4N{4) 199knttir 3l0mph However, the Kaigun Koku Hombu and the something which was never to occur.
Landing speed (J4M4) l47kr["h 91mph IJN ensured that Mitsubishi would not have to As a note, although there are artist impres-
Endurance 2.2 horus concern themselves further with the Senden. sions of a iet-powered Senden (as shown
Climb l5 min to 8,000m (26, 2461t) ln 1943 as Mitsubishi was working on the here) there is no evidence to support the
(J4l\'14) l0 min to 8,000m (26, 24 6ir)
Senden, the Kaigun Koku Hombu issued an notion the J4M was ever revived or consid-
Ceiling 12,000m 39,370it
18-shi Otsu specification. From it, the Kyushl ered for turbojet power as there were other
designs being considered (for example, the
11,000m 36,089ft
JZW Shinden resulted (page 84 for details).
With the 18-shi Otsu requirements being simi- J7W2 and the Ki-201) which offered better
lar to the 17-shi Otsu specifications and with prospects and capability.
One Type 5 30mm cannon with 100 rounds of ammunition and t$'0

Tlpe 99 20mm cannons rdth 200 rounds ofammunition each: ptovision the J7W showing far more promise and having
for hvo 30kg (66 lb) bombs or hto 60kg (132 lb) bombs

The Senden did not advance past a wind tunnel model.



94 Japnrursl SpcnEr PRo;Ecrs: ExprnlutEnteL AtRcRepr oF rHE IJA AND IJN 1939-1945

InrpERrer- Jnpar.rEsE Navv 95

Mitsubishi J8M Syusui


In 1943, the Japanese were only too well deployment and training as well as the coor- and that developing such a plane and its
aware of a threat looming on the horizon. dination of the various contractors and test engine would consume much needed
That threat was the Boeing B-29 Super- centres involved in development and pro- resources. Despite these objections, the sup-
fortress. Witii ttre development of the B-29 duction of the Me 163. During the tour EKdo porters for the Me 163 won out.
starting in 1939, the Japanese were in no 16 personnel explained to the Japanese the The Japanese swiftly and successfully
doubt that once ihe bomber entered produc- temperamental nature of the Walther HWK negotiated the licences to manufacture both
tion it rvould eventually appear over Japan, 5094 rocket motor and the dangerous and the Me 1638 as well as its HWK 509A rocket
The problem for the Japanese was that they explosive properties of the two fuels the motor. The motor licence alone cost the
did not have an effective countermeasure motor used. This did nothing to dissuade the Japanese 20 million Reichsmarks. In addition
against the B-29 and feared theywould not be Japanese who saw the answer to their needs to the two licences, Germany was to provide
able to have one ready in time for its antici- right before their eyes. To them, the benefits complete blueprints for the Me l638 and the
pated arrival. Fortunately, the answer was of an interceptor able to climb rapidly and HWK 509A, manufacturing data for the air-
found in one of the most radical fighters ever possessing a very high speed overrode any craft and engine, one complete Me 1638,
to achieve operational status. concerns about the fuels or the engine. The three HWK 509A motors, and two sets of sub-
Towards the middle of 1943, representa- Japanese wasted no time in entering negoti- assemblies and components by no later than
tives of the Japanese military in Berlin were ations to obtain the N{e 1638. 1 March 1944. Also, Japanese military
notified of the development and progress of However, not everyone was in agreement attach6s in Berlin were to be notified of any
the Messerschmitt Me 163, a point defence about the value of the Me 163. Detailed reports improvements to the Me 163 design so
interceptor powered by a rocket engine. had been sent to Japan from Germany regard- changes could be incorporated into the
Interest was expressed immediately. In short ing the findings of the attaches which overall Japanese version. The Japanese also
order Japanese attach6s from the IJN and the were positive; nevertheless, some argued that requested to overseethe manufacturing
IJA visited Bad Zwischenahn in Germany it would not be possible to produce the fuels processes for the Me 1638 and the rocket
where Erprobungskommando 16 was sta- the aircraft required in sufficient quantity to motor as well as being allowed to study and
tioned. This unit had been created earlier in support operational requirements. Others review Luftwaffe operational procedures for
1943 to develop Me 163 combat tactics, criticised the unorthodox nature of the Me 163 the fighter. Three submarines were tasked

96 JapaNesE SecRsr PRoJECIs: ExpsRIltlNlal Alncnnrr oF rHE lJA,qro IJN 1939-1945

with shipping the materials to Japan - the nearby Philippine island and reported the frame. In addition to the Mitsubishi men, rep-
RO-500, RO-501 and I-29. loss. resentatives of the Yokosuka KOkittai were
RO-500 was still named U-511 when it Technical Commander Eiichi lwaya, upon involved, namely Captain Kumamoto and
departed from Lorient in France on 10 May leaving the I-29, did not take all of the docu- Commander One, who was tasked with test
1943 bound for Penang, Malaysia. Aboard mentation he had for the Me 1638 (or the flying the J8M1 upon completion. Technical
were four Japanese including Vice Admiral Me 262) and the loss of the I-29, along with Commander Eiichi Iwaya was also a part of
Naokuni Nomura and Major Tam Otsu Sugita that of the RO-501, delivered a major blow to the overall development team given his famil-
of the IJA medical service. AIso aboard was the development program. However, the iarity with the Me 1638 acquired during his
the data for the Me 1638. During the transit, information Iwaya had preserved, combined time in Germany. One last meeting was held
U-511 was named Satsuki 1 ('satsuki' mean- with what was received from the RO-500, was on 7 August 1944 to finalise the development
ing the month of May). On 16 July, U-511 enough to keep the project alive and in July of the Syusui and then work began.
reached Penang where Nomura, Sugita and 1 944 the IJN issued a I 9-shi specification for a The first stage was the wooden mock-ups.
the otherJapanese passengers disembarked rocket powered interceptor. This decision On 8 September 1944, the full scale mock-up
and returned to Japan by air. U-51 1 departed was based on the analysis of the documenta- of the cockpit was completed and on 26 Sep-
Penang for Kure, Japan, on 24 July 1943 and tion on hand for the Me l638 and the current tember 1944, the mock-up of the Syusui was
arrived in Kure on 7 August 1943 where the construction capacity and capability of the air completed. Both the IJN and the IJA
submarine was presented to the IJN as the industry, and also down to the drive of Vice inspected them and suggestions were made
RO-500. Admiral Misao Wada who supported the for possible alterations to the design. These
RO-501, a Type IXC/40 submarine, was for- development of the rocket aircraft. changes were incorporated and Takahashi's
mally U-1224. On 15 February 1944, U-1224 Upon issuing their 19-shi specification, the team laboured day and night to produce the
rvas handed over to the IJN who gave it the Kaigun Koku Hombu assigned the project to detailed blueprints for the J8Ml. Three proto-
name Satsuki 2, and on February 28, it was Mitsubishi. Mitsubishi were initially reluctant types were to be built; the first would be for
commissioned into the Imperial Nar,y as to accept the design, but further considera- load testing while the remaining two would
RO-501 with Lieutenant Commander Norita tion and the need to adapt the Me 1638 design be used for the flight test program. As the
as captain. On 30 March 1944, RO-501 to Japanese production capability saw the rocket motor was not yet available, two of the
departed from Kiel, Germany, with the man- manufacturer agree. Even though the IJN was prototlpes would be weighted to simulate
ufacturing data and blueprints for the behind the aircraft, the IJA would also be the motor and fuel. To hasten construction,
\'le 1638 among other cargo. At 7.00pm on 13 involved in the development of both the air- when one portion of the aircraft was drafted
\'lay 1944, north west of the Cape Verde craft and rocket motor. The Japanese rocket and finalised, a copy was sent to the assem-
Islands, the USS Francis M. Robinson, a Buck- interceptor was to be called the J8M1 Syusui bly shop assigned to construct the compo-
ley class destroyer escort, reported a sonar (which means 'Autumn Water') and in IJA nent so work could begin without delay.
contact 755m (825 yards) from the ship. The service the Syusui was to be designated Externally, the J8M1 was unmistakable in
Francis M. Robinson immediately initiated an Ki-200. its lineage but Takahashi and his group had to
attack, launching 24 Mark 10 Hedgehog On 27 July 1944, all personnel involved met make modifications as they adapted the
bombs and five salvos of Mark 8 depth to discuss the Syusui and it was agreed to fol- Me 1638 design. For example, the Me 1638
charges. Sonar reported four explosions sig- low the design plan of the Me 1638 as much used two MK 108 30mm cannons which were
nifi'ing the death of the RO-501. as possible. The key reason was that the heavier and shorter than the 30mm cannons
l-29 of the Imperial Japanese Narry design was proven and worked and thus crit- the Japanese were to use. Fuel capacity was
departed from Lorient, France, on 16 April ical time could be saved. The same applied to similar to the German aircraft and so were the
1944. She caried on board a HWK 5094 the rocket motor. A second reason for adher- dimensions, althorrgh the J8M1 was slightly
rocket motor, the fuselage of a Fieseler Fi 103 ing to the Me 1638 design was that Japanese longer due its more pointed nose and had a
and a Junkers Jumo 0044 turbojet, again with fabricators had almost no experience with wider span and smaller wing area. (The
other cargo. Technical Commander Eiichi the tlpe of aircraft that the Me l638 was. But Syusui unlike the Me 1638 did not use a nose-
Irvaya, a passenger, carried with him the not everyone was in full agreement. installed generator, the space being used for
plans for the Me163B and Me262 while The IJA saw flaws in the Me 1638 and felt radio equipment.) The wing thickness was
anotherpassenger, Captain Matsui, had plans that Japanese industry could not fully pro- also increased. The main difference, how-
for accelerators used for rocket launching. duce the Syusui to the specifications of the ever, was the weight: the Syusui was 363 to
Between the two of them, they also had plans German aircraft. Modifications to meet the 408kS (800-900Ib) lighter than the Me 1638.
for a glider bomb and radar equipment. On 14 current capabilities of the Japanese aviation This was not due to any effort to purposely
July 1944, the I-29 arrived safely in Singapore. industrywould be required to both the rocket lighten the Syusui as it lacked armour protec-
Here, Iwaya and Matsui disembarked, along motor and the aircraft which, as a conse- tion for the pilot and carried less ammunition
rvith a portion of their documents, and con- quence, would force changes to the design. for its cannons than the German interceptor.
tinued on to T6kyo by air. On 15 July, Allied As such the IJA argued that in the end a new For weapons, the J8M1 was to be equipped
code breakers intercepted a message from design would be required anF/vay. The IJN, with two Type 5 30mm cannons in the wings
Berlin to T6kyo regarding the cargo that the however, would hear none of it and was while the IJA's Ki-200 would use two Ho-155
I-29 carried and on 26 July 1944 aI 5:00pm adamant that the Me 1638 design would be 30mm cannons or two Ho-S 20mm cannons.
near the western entrance of the Balintang followed. Because the Japanese lacked the experi-
Channel, Luzon Strait, the USS Sarofish spot- Mitsubishi forged ahead with assembling a ence in flying tailless aircraft, Krlgisho was
ted the I-29 on the surface. She fired four tor- team to develop the J8M1. The project was tasked with creating a glider version of the
pedoes and three struck the Japanese led by Mijiro Takahashi at Mitsubishi's Nagoya Syusui. In part, the glider would provide per-
submarine. I-29 sank almost immediately and plant. Under Takahashi was Tetsuo Hikita formance data, findings from which could be
only one sailor survived who swam to a who would be the lead designer for the air- incorporated into the Syusui, but would also

lrrpERrnr- JepeNess Navv 97

serve as a trainer for rocket aircraft pilots. T-Stoff formed from eighty per cent hydrogen mance calculations for the Syrisui based on
Therefore, the MXY8 Akigusa and MXY9 peroxide with the remainder Oxyquinoline the thrust rating of the KRl0, they found that
Shuka were developed, as described else- and pyrophosphates to act as stabilisers. Ko the lighter weight did not totally offset the
where in this book on page 77. rvas the oxidising fuel. The second, Otsu, was lower thrust. Regardless, the estimated speed
While work was undetway on the first the Japanese equivalent of C-Stoff. Otsu was and climb rate was considered exceptional.
three prototypes, a production plan for the the reductant fuel and was composed of On 8 January 1945, a Nakajima B6N1
fighter was put together and was completed thirty per cent hydrazine hydrate with the (known as Jrl/ to the Atlies) towed the Sylsui
by October 1944. By March 1945, 155 Syusui remainder being methanol, water and potas- into the air from the Hyakurigahara airfield and
were to be produced with another 1,145 built sium-copper cyanides. Together, Ko and after a successful flight the design was vali-
by September 1945. Ultimately, by March Otsu were a hypergolic fuel combination, dated. Work quickly proceeded on further
1946 at least 3,600 Sylsui were anticipated to which meant that when the two fuels were production of the Sy[sui, this time with the
be in service. combined they spontaneously ignited. The KR10. However, the motor program was at
In addition to developing the Sy[sui, Mit- problem with Ko and Otsu was that they were least three months behind schedule and itwas
subishi was also assigned the task of creating colourless and, of course, when they came not until 11 April 1945 that the KR10 was suffi-
the Japanese version of the Walther HWK together, the result was explosive. This ciently developed to enable it to function with
509A rocket motor and both the IJN and the required strict handling procedures and con- some measure of reliability. With the possibil-
IJA were involved in the motor program. To tainment methods. Both fuels were stored in ity of powered flight, Captain Shibata, com-
assist the engineers in Mitsubishi's engine special ceramic pots. To produce both fuels, mander of the 31 2 Kokutai due to be equipped
department, personnel from the IJA's First three chemical companies were contracted. with the J8M1, sought to speed up the process
Army Air Arsenal engine section were They were the IJN's First Fuel Arsenal, Mit- for testing. In discussions with the S1'usui

assigned to the firm. The resulting motor was subishi Kasei and Edogawa Kagaku. In the development team it was decided that if the
called the KRl0 but was also known as the in wing and
Syusui, the fuels were stored KR10 could produce thrust for at least two
Toku-Ro.2. Components for the KR10 were fuselage mounted tanks. The pilot sat minutes without mishap, the motor should be
constructed by four companies: Hitachi, between two 91 litres (24 gallons) tanks of Ko fitted to the Syusui so that powered flight test-
Ishikawajima, Mitsubishi and Washimo. while behind him in the fuselage was a 961 ing could commence. 22 April1945 was set as
Washimo, for example, was responsible for litres (254 gallons) tank and a 8 litre (2 gallon) the deadline for the first powered flight.
the fuel flow control mechanisms and the tank of Ko. Each wing housed two tanks of Meanwhile, Germany made another
relief valve for the Ko fuel tank. Otsu, the capacity of each tank in each wing attempt to send more material to Japan
Mitsubishi faced several problems in build- being 64 and 197 litres (17 and 52 gallons) in including documents and parts for the
ing the KR10, the main issue being that the the two tanks respectively. Me163. These items and other cargo were
HWK 509A used a nickel-chromium alloy in By December 1944, the second and third loaded onboard U-864 that departed from the
the fuel injector atomiser, regulating valves J8M1s had been completed but as no engines Bruno U-boat pen located in Bergen, Norway,
and reliefvalves. Since the Japanese did not were ready for installation, ballast was used on 5 February 1945. However, having past
have access to this alloy they had to use plain to simulate the weight of the KRl0 with full Fedje the submarine developed a misfire in
chromium steel. It was expected that the fuel tanks. Earlier, the first J8Ml had been one of her two MAN diesel engines and itwas
KR10 would be ready for testing by October completed and load tested on 1 December necessary to return to Bergen to effect
1944, but the lirst prototlpe exploded imme- 1944. However, the 7.9 magnitude Tonankai repairs. The British submarine HMSVenturer,
diately when it was started for the first time, earthquake that struck the Tokai region of dispatched to deal with U-864, spotted the
partly believed to have been caused by the Japan at 1.30pm on 7 December 1944 German submarine's periscope on 9 Febru-
metal used. A deviation was made from the destroyed the aircraft and the testing facility ary 1945. Korvettenkapitiin Ralf-Reimar Wol-
original HWK 509A plan in that the KR10 that housed it. The remaining J8Ml aircraft fram realised he was being followed and
motor used wider supports and included a were transferred to the IJN's First Naval Air began to take evasive action, moving in a zig-
bearing in the middle for ihe Ko fuel com- Technical Arsenal. From there, the aircraft zag fashion. James S. Launders, Venturer's
pressor. This revision in the KRl0 resulted in were shipped to Hyakurigahara, located captain, decided to press home the attack
the KR12 but the addition of a second version about 79km (49 miles) northeast of Toky6. and fired all four of his loaded torpedoes in a
of the motor risked compounding any pro- December would also see delays due to the spread pattern. U-864 crash dived, dodged
duction problems. Indeed, testing of the KR12 increasing B-29 bomber raids. Attacks against three of the torpedoes but turned into the
also resulted in an explosion. Mitsubishi engi- Mitsubishi's Nagoya facility resulted in the fourth which struck the submarine. The
neers discovered that a bearing seal had KR10 program being moved to the Dai-Juichi resulting explosion split U-864 into two.
failed that allowed the Ko fuel to leak into the Kaigun Kokusho complex at the Hiro Naval Unfortunately for the Sylsui, the deadline
motor and then come into contact with the fusenal in Kure, Hiroshima. Here, work con- for the KRl0 would not be met. In exhaustive
bearing lubricant with catastrophic results. tinued on the motor supervised by Professor testing, another motor detonated after having
Given that it offered no real advantage, the Kasai of the Kyushl University (although achieved two minutes of burn time. ln addi-
KRl2 was shelved and work focused solely another source states the entire engine devel- tion, fears of B-29 raids saw the KR10 team
on the KRl0. These accidents, their subse- opment group was moved to an underground being moved to the Yamakita factory com-
quent investigations and the resulting revi- facility in Natsushima in Yokosuka prefec- plex in Hakome prefecture while the Mit-
sions put the KR10's development further and ture, overseen by the Dai-Juichi Kaigun subishi Syusui development group was
further behind. Kokusho). relocated to the IJA research and develop-
For fuel, the Syusui used two ingredients During testing, the KR10 delivered less ment centre in Matsumoto in Nagano Prefec-
which, when combined, provided the com- thrust than the HWK 509A. Although the ture. These moves consumed precious
bustion and resultant thrust. The first, Ko, was Syusui was lighter than the Me1638, when research time throughoutApril and May 1945.
the Japanese version of the German fuel Takahashi and Hikita completed perfor- Both groups were eventually able to continue

98 JapaNEse SEcREt Pno;pcls: ExpERIurnral AtRcRapr oF THE IJA axp IJN 1939-1945
work on the KRl0 in an attempt to enhance two minute requirement, plans were swiftly in the second week of June 1945 at Mit-
its reliability and, in June, success was prepared to install the Yamakita KRl0 into a subishi's Number One Plant in Nagoya. The
achieved. A KRl0 from the Yamakita group J8Ml while the Matsumoto motor was to be Syr.rsui lacked much of its operational equip-
functioned for four minutes while the Mit- placed into another airframe that would be ment including weapons and was trans-
subishi group in Matsumoto managed three completed as a Ki-200. ported to Yokoku airfield. This site was
minutes. With these motors now meeting the The J8M I 's installation was completed first favoured because it was situated along a
shoreline, which meant that if the pilot had to
c ditch the aircraft he could do so into the
= ocean, offering a better chance for survival as
- well as possibly lessening the damage to the
Syrlsui. The Syusui arrived at Yokoku at the
beginning of July and ground testing began
immediately. Secured to the tarmac, the tail
of the Sylsui was removed exposing the KRI 0
and motor running tests commenced. It was
found that the motor did not burn fuel evenly,
generating plumes of light red smoke from
the combustion chamber as it ran. By 5 July
1945, technicians and engineers had cor-
rected the burn problem to the point that the
KRl0was deemed readyand the Sylsui's first
powered flightwas scheduled for 7 July 1945.
In front of a crowd of onlookers, the Syusui
was moved to the start of the 1,200m (3,93ift)
nrnway, the longer of the two at Yokoku. It
was then fuelled with 568 litres (150 gallons)
of Ko into the fuselage tanks and 159 litres (42
gallons) of Otsu into the wing tanks as the
mixture ratio was approximately 10 to 3.6. At
4:55pm, the pilot, Lieutenant-Commander
Toyohiko Inuzuka, fired the engine and
within I I seconds and after only 320m
(1,049ft) of runway, the Syusui lifted off the
ground and into the air, Inuzuka releasing the
dolly and raising the nose to provide a 45'
angle climb. Then, at 350m (1,148ft), a puff of
black smoke issued from the motor, sput-
tered and went out. The speed that had been
built up carried the Syusui up to 500m
(1,640ft) where Inuzuka levelled off and
banked to the right ready to return to the run-
way and land. As Inuzuka continued his right
hand bank, the Syrlsui began to drift and air-
speed rapidly dropped off. As he approached
the runway, Inuzuka raised the nose of the
Syrlsui to try and avoid colliding with a build-
ing but it was too late. A wing clipped the side
of the building, putting the Syhsui into a crash
so forceful that it broke apart, scattering
pieces across the south-west edge of the air-
field. Both wings were ripped away and the
front of the aircraft was completely
destroyed. Inuzuka survived the impact and
was extracted from the wreckage. However,
the extent of his injuries was so severe that he
died the following day.
No time was wasted in trying to find the
cause of the motor failure. Mechanical issues
were ruled out and it was surmised that the
puff of smoke and the subsequent loss of
power from the KR10 was due to fuel being
cut off from the motor. Miraculously, the fuel

Irr.rpsRler, JepeNesE Nnvv 99

tanks did not explode on impact and it was Mitsubishi J8M Syusui - data
found that at least half of the fuel loaded prior
to take-off remained. It was determined that Contemporaries
l'lesserschmitt Me l638 Komet (Germany) None. A total of seven J8Ml aircraft were completed with one to be
the culprit was the fuel line from the Ko tank.
finished as a Ki-200. The 3 1 2 Kokutai were t0 receive the J8['ll had it
Due to poor design, when the Syusui went Specificalions in parenthesis pettain to the l8M2 only and are based on
entered production, No J8M2 was ever built nor were any Ki-200 airctaft
into its climb the fuel in the tank shifted away M ibubishi's est[noted data,

from the line which starved the motor of the Survivors

needed oxidiser and thus the KRl0 cut out. Type Interceptor/Fighter
Mitsubishi J8Ml Sl'usui (FE-300)
Crew 0ne
While the investigationwas being carried out, One of three brought fiom Yokosuka on 3 November 1945, this Spsui is

bench tests of two additional KR10 motors aircralt N0.403 and is thought t0 have been captured at l'litsubishi's No.l
(one each at Matsumoto and Yamakita) plant in Nagoya. Appearing on the l0 March 1946 report for aircraft
One Toku-R0.2 (KRl 0) bi-fuel rOcket motor developing i,500kg
resulted in both exploding. This left a single (3,30iIb) ofthrusi
releasable to the aviation industry, the Sy.rsui rvould be made available

for display purposes on I August 1946 appeadng to the public in

KRl0, the one slated for the Ki-200.
Holllwood, California. The aircralt was later obtained and restored by
Flight testing of the Syusui was suspended Dimensions
Edward Maloney for displav at the PJanes oiFame l'luseum in Chino,
until the problem with the fuel system could Span 9.47m 31,lfr
California, where it remains to this day.
be resolved. A further four Syusui aircraft had Length 6.03m I 9.8fr

been completed by Mitsubishi by the time a Height 2,68m 8.8fr Mitsubishi l8M1 Slrsui (tail number 24)
After being received at NAS Patuxent River, the aircralt lvas moved to
solution was found. These changes were Wing area 17.72m' 190.8ft,

21922k9'n' 44.9tbift, NAS Glenvierv in Glenview, Illinois (a suburb ofChicago, Illinois), rvhere
incorporated into the KR10 engines then \Ving loading
(J8MA 2l9.70kgh' 45 lb/fr, it rvas on display by 3 October 1946. This Slusui eventually reached a
under development and flight testing was derelict state and was scrapped.
scheduled to resume in late August 1945.
However, on 15 August 1945, Japan surren- Mihubishi JSMl Syusui (tail number A"25)
Emptl' 1,445k9 3,1 85 Ib
Nothing is knorvn about this particular Slusui other ihan it likely ended
dered. All further work on the Syusui ceased (J8MD l,5lokg 3,328tb
up as scrap.
and no further flights were made. At the end Loaded 3,000kg 6,613 tb

of the war the Ki-200 remained engineless, its (J8MD 3,650k9 8,046tb Mitsubishi J8MI Syusui

KRl0 never having been installed. Aside from l\'larimum 3,870k9 8,53t lb Mitsubishi has recently restored a J8Ml and it is currentlv on display at

the seven J8Ml aircraft built - including the (JrNzl 3,900k9 8,598Ib the company's Komaki Plant Museum. A potion of the restoration

l,545kg 3,406Ib contains components irom a badlv damaged J8M1 fuselage found in a
one to be finished as the Ki-200 - another six Useful load
(J8[,r2) 2.l40ks 4,71 7 lb cave but it still required significant custom fabdcation 0f new parls to
were in various stages of completion. A fur-
iinish the project. Prior to Mitsubishi obtaining the fuselage, the lemains
ther four KRl0 motors had been completed
Performance had been on display on the grounds of the Japanese Air Self-Defence
with another two nearly finished. Enough
Mar speed 900krdh 559mph Force's Gifu Air Base.
components had been constructed to assem- at 10,000m at 32,810ft
ble a further twenty motors. Cruise speed 699krr/h 434mph
Another variant of the J8M had been Landing speed 150kn/h 93mph
planned which was called the J8M2 Syusui- Range 3 min 6 sec oipowered flight
Ki-200 - data (estimated)
Kai. The J8M2 lost one of the Type 5 30mm at 599km,ih 372mph
Type Interceptor/Fighter
cannons/ammunition to be replaced by addi- Mar range 5 min 30 sec of powered flight
Crew 0ne
tional fuel tankage. It was hoped that this Climb 40 sec to 2,000m (6,561ft)

2 min 8 sec to 4,000m (l3,l23ft)

would increase the endurance of the aircraft. Powerplant One Toku-R0.2 (KRl0) bi-fuel rocket motor
3 min 8 sec to 8,000m (26,246ft)
The end of hostilities would see the J8M2 developing l,500kg (3,307Ib) of thrust
3 min 50 sec ro 10,000m (32,808ft)
remain only a preliminary design though pro- Ceiling 12,000m 39,370tt
duction of the J8M2 was a certainty had it Fuel capacity I,l8l litres (312 gallons) of Ko and

been completed. As mentioned earlier, the 522 litres (138 gallons) of Otsu
Span 9.47m 31.1fr

IJA was not pleased with the Ki-200 and it Length 5.88m l9.3fl

would undertake development of its ownver- Annament

Height 2.68m 8.8fr

Wing area 17.69m! 190,5ft,

sion of the J8M, the Ki-202 Sy'r-rsuiKai, to right Two Tlpe 5 30mm cannons with 53 rounds ofammunition per gun

the wrongs it felt were evident in the Syusui. (one Tlpe 5 cannon wilh 53 rounds ofammunition)
For more details, please see the chapter on Empty l,505kg 3,317 lb
the Ki-202 (page 40). Marimum 3,870ks 8,531 lb
A note regarding the use of Syusui as the
name for the J8M. The kanji for the aircraft Performance
(Syu and Sui) translate as 'Autumn Water'. Ma\ speed 800-900knr/tr 49l559mph

However, Shlsui has been used in many at 10,000m at 32,808f1

sources with translations ranging from Cruise speed 35lkn/h 2l8mph

min 30 sec otpowered flight
'Sword Stroke' or'Swinging Sword' to'Rigor- Range 2

illax range 7 min ofpowered flight

ous Sword', but the name Shrlsui is not cor-
Climb 3 min 40 sec to 10,000m (32,808it)
rect. The use of Shlsui evolved from the Fuel capacity l, l8l litres (312 gallons) of Ko
metaphor that Sylsui represents - the wavy and 522 litres (138 gallon9 otOtsu
pattern on the metal blade of a highly sharp-
ened sword as well as the brightness of the Annament
polished metal which reminds :tne of the Tlvo Ho-155 30mm cannons (or two Ho-5 2Omm cannons)

waves on a body of clear water.

100 J,qpnrussr Slcnpr PnoLECrs: ExpsRnrsNrnr- ArRcRLpr oF THE IJA AND IJN 1 939-1 945
Mitsubishi-Payen Pa.400 and Suzukaze 2O

When Allied intelligence discovered an illus- delta planforms as well as canards and ogival repairs and was consequently abandoned by
tration of the Suzukaze 20 in a Japanese mag- (bullet shaped) flight surfaces. Throughout his the Germans.
azine, it was unlike anything so far seen in life, Payen would design a large number of air- Prior to the outbreak of war, the Japanese
Japanese aviation design. Despite the radical craft in a wide array of configurations but, had civilian and military personnel in France
appearance, it was felt the Suzukaze 20 was a despite the prolilic nature of his studies, only who studied and reviewed French aviation
bona fide aircraft and might be encountered two were built before the end of World War 2. progress for possible use by Japan. This prac-
in action. As it was, the plane was a work of Both used Payen's Flechair (an English con- tice went as far back as 1919 when Japan
fiction and so the Suzukaze 20 was later traction of auion fleche or'arrow aircraft') invited French military aviation instructors to
stricken from the publications on Japanese configuration that consisted of a trapezoidal teach the fledgling Japanese Army air force.
aircraft identification and coding. However, fore-wing that housed the ailerons and a rear The French also brought with them some of
Allied intelligence may or may not have been delta wing which contained the horizontal the latest aircraft that their country had avail-
a.,vare of the very real inspiration for the artist control surfaces. Payen had to rely on his able. This training would forge a link between
of the Suzukaze 20. salesmanship to gain access to material, wind Japan and France that would last for many
At the time, because of the relative difficulty tunnel time and other resources to build his years, and it was by these means that the
in obtaining information on Japanese military aircraft as he had little money of his own to Japanese would learn of Payen's work.
matters, intelligence services relied on vari- fund projects. Of course, the nature of his In 1938, Payen received a letter from Mit-
ous publications such as newspapers and designs often made it a hard sell to the more subishi expressing an interest in his designs,
magazines as a means to glean data on the conservative aviation industry. The Payen notably the deltawing so often used in his air-
Japanese military machine. In April 1941, the Pa.100 Fleche Volante ('Flying turow') was craft concepts. A meeting was held between
Japanese magazine Sora (translated as 'Sky') his first aircraft to be built and was intended to Payen, Commander Koshino and the captain
published a number of illustrations of various be a racer to compete in the Coupe Deutsch of the UN coruette Sumikawa to talk about
aircraft in a section entitled 'Dreams of Future de la Meurthe. Payen was able to borrow a the Pa.112. During the discussions the IJN
Designers'. Included in the selection of art- 180hp Regnier R6 but the engine was later inquired as to whether the Flechair design
n'orkwas the rendition of the Suzukaze ('Cool returned. He was then able to acquire a larger could be adapted to that of a two-seat, carrier
Breeze') 20. The 25 December 1941 issue of engine, a 380hp, 7-cylinder radial Gnome- borne, light bomber. The specifications
the US magazine Fllghf would also feature the Rh6ne 7KD Titan Maior, but was too large for required the aircraft to have the ability to take-
Suzukaze 20, along with three other aircraft: the Pa.100. Payen had to seek donations offfrom and land on a deck space 80m (262ft)
the Nakaiima AT27 (codenamed Gus), the (which he received) and rebuilt the Pa.l00 long, to have a range of at least 800km (497
\litsubishi T.K.4 Type 0 (codenamed Franh around the 7KD to create the Pa. I 01 . Unfortu- miles), be capable of carrying a 800k9
IhenHarry) and the T.K.19 (codenamedJoe)' nately for Payen, the Pa.101 failed to meet (1,763Ib) torpedo or bomb and be fitted with
The Suzukaze 20 would receive the code- expectations. It finally took to the.air on 17 up to 180kg (396Ib) of armament.
name Omar. April 1935, but on 2TApril a hard landing broke Payen took the specifications and worked
The illustration of the Suzukaze 20 depicted the port landing gear and a fire broke out in up a study to meet the IJN requirements. The
a single-seat fighterwith the striking feature of the resulting crash gutting the Pa.101. The design was called the Payen-Mitsubishi
having a cockpit blended into the vertical sta- accident saw Payen's flight insurance Pa.400. This would have used two 670hp
biliser that was itself in the form of a half-delta. revoked and so he went to work at the Bloch radial engines mounted one behind the other
Another notable feature was the use of two factory constructing a mock-up of the Pa.112 driving two, two-bladed, contra-rotating pro-
radial engines, one mounted behind the fighter that used two 150hp Salmson engines pellers. For weapons, in addition to carrying
other, driving two, contra-rotating propellers. in tandem buried in the fuselage. The out- the required torpedo or bomb, a nose
fumament appeared to be healy with four break of World War 2 saw the French military mounted cannon (firing through the propeller
\\ieapons fitted in each wing. Its speed was show no interest in this design. hub), two machine guns per rear wing and a
given as 769km/h (478mph), loaded weight The second aircraft was the Pa.22 which tail mounted machine gun were proposed.
2,858kg (6,3001b), wing area 13.37m' (144ft) was the test bed for Payen's proposed Pa.1 12 Endurance was to be 1l-12 hours with a max-
and wing loading 21 4.82k9/m') (441b/ff). fighter. Originally to be powered by a ramjet, imum speed of 580km/h (360mph). Unlike his
As the war dragged on it became evident no such engine was available and a 180hp other Flechair designs, Pa.400 used staggered
that the Suzukaze 20, along with the other Regnier R6 engine was used instead. Payen wings (his earlier offerings had the two wings
three aircraft illustrated with it, were works of constructed the Pa.22 in 1939 and the Ger- level with each other). The study was
fantasy and thus all four were removed from mans would later capture it after the invasion reviewed and Payen was asked by Japanese
Japanese aircraft intelligence bulletins, the of France on 12 June 1940. The Germans, representatives to obtain from the French gov-
last of them disappearing by June 1943. showing some interest in the design, test flew ernment the authorisation to export the tech-
Despite the Suzukaze 20 being a fictional air- iton 18 October 1941 and found modifications nical information for the Pa.400 study. This
craft, there was a kernel of truth behind it that were needed to correct poor longitudinal sta- would have allowed the Japanese to further
perhaps germinated in the mind of the artist bility. The aircraft was moved to Rechlin in develop the Pa.400 in Japan. The authorisa-
that drew the Suzukaze 20. The kernel could Germany and after adiustments had been tion was granted on 28 September 1938,

have been the works of the French aircraft made to the cockpit position and the vertical signed by the head of the cabinet of the Min-
designer, Nicholas Roland Payen. stabiliser had been rebuilt the Pa.22 flew in istry of Air. However, with the cloud of war on
Payen was born in France in 1914 and the summer of 1942. After a number of short the horizon, Payen decided not to send the
became interested in aviation early in life. By flights, the aircraft was wrecked in a crash requested documents to Japan and it would
the 1 930s, he had begun to focus on the use of landing. ThePa.22 was returned to France for appear the Japanese did not follow this up. To

IN4penInl JapaNssE NnvY t0r



This illustration
depicts the
Suzukaze 20 Omar.

102 Jnpnrussl SscRsr PRolEcrs: ExpenrMeNrnr- ArncRLnr oF rHE IJA AND IJN 1939-1945
all intents and purposes, the Japanese ical design of the Pa.400 that saw more con- Mitsubishi-Payen Pa.400 - data
seemed to have lost interest in the Pa.400. servative IJN officials directing Mitsubishi to
Why would the Japanese show an interest focus their efforts on more conventional air- Contemporaries
Horton HoX (Germany), Messerschmitt P.l 106 (Germany), Lippisch
in the Pa.112 and the Pa.400 only to abandon craft projects.
Pl3a (Germany), Bl\{W 803 engine (Germany), Wight R-2160 Tornado
it on the brink of receiving the technical infor- Enter the Suzukaze 20. A photograph of
engine (US), Pratt &Whitney R-4360 Wasp Maior engine (US), Butler-
mation? There were several factors which the what was likely the Pa.101 appeared in Japan
Edwards'Steam Dart' (UK),Scroggs'Dart' (L'S)
Japanese may have become aware of upon in a printed document in the late 1930s. The
further review of Payen's initial Pa.400 study. caption for the photograph read 'French for the Milubishi-Payen Pa.400, based on the des[gn study conducted

by Poyen.
The first was that Japanese radial engines of Brand New Model Pey-yan 266th. Airplane No-
the time did not have sufficient horsepower mu 400 Horse Power'. The 'Pey-yan'was the
Type Light Carier Bomber
and, more importantly, were not of the correct phonetic spelling in Japanese of Payen while
Crew Two
size to fit into the Pa.400's fuselage. Thus, the 'No-mu'was the phonetic spelling for Gnome.
Japanese would have had to either construct How it got to Japan is open to speculation but Powerplanl
a new radial engine or adapt the Pa.400 to use the two prevailing theories are that Payen, in Two Cnome-Rh6ne l4 M4/5 radial engines, each developing 670-680hp
a Japanese engine, radial or not. A more per- trying to drum up funds for his work, made it marimum, driving two c0ntra-rotating, two-bladed propellers
tinent problem was the use of tandem radial available to a French diplomat to take to
engines. To make such an arrangementwork- Japan to shop around to Japanese industry. Dimensions

able required a considerable feat of engineer- Alternatively, the photograph was given to the Span 6.58m 2 1,611

Length 10.00m p.8ft

ing and such designs making it to prototype IJN by Payen during the discussions over the
Height 2.68m 8.8fr
form were exceedingly rare. Another factor Pa.400.
Wing area 25.00m' 269ft?
concerned the poor visibility afforded the pilot The photograph - and perhaps other
given that the cockpit was situated far back in sources because Payen's aircraft were sholvn Weights
the fuselage which made landing a challenge in pubfications such as Brll Barnes: Air Aduen- Empty 2,1 igkg 4,784Ib
at the best of times, let alone landing on a turer (from April 1935) - likely played a part in Loaded 5,860k9 12.9r9rb
moving and rolling aircraft carrier. The rear the rendering of the Suzukaze 20. The similar-
wings and the long nose blocked side and ities to Payen's designs cannot be ignored. For Speed

downward vision, a serious liability in aerial one, the Suzukaze 20 utilised two radial Max speed 570kn/h 360mph

at 4,950m
combat, and the relatively short wingspan of engines in tandem driving two, contra-rotat- at 16,240fi

the Pa.400 would not have offered much ing propellers. Also, the Suzukaze 20's cockpit Nlar speed (one engine) 447kn/h 278mph

at 4,950m at 16,240
agility, a trait favoured by Japanese pilots and was blended into the large, half delta-shaped
Landing speed 89km/h 55mph
designers. In addition, the Japanese may have vertical stabiliser, another Payen trait seen in
Endurance I 2-14 hours
learned that the Pa.101 was a flawed design the Pa.l00/Pa.101 and Pa.22. The artist likely Ceiling 4,950m 16,240tt
and since the French military paid Payen no removed the rear delta wing and slid the for-
attention may have concluded there was ward wings back and enlarged them since Armament
nothing worth pursuing where the Pa.400 was their shape is reminiscent of the Pa.400. With Five machine guns, two in each wing and one in the tail; one cannon

concerned. Finally, it may have been the rad- the exception ofthe nose, the fuselage shape firine throush the propeller hub; one 800kg (1,764 lb) torpedo or bomb

of the Suzukaze 20 was similar to that of the

Pa.400. Even the horizontal stabilisers of the Deploynent
The Pa.400 depicted here is in the colours and None. The Suzukaze 20 was a fictional aircratt while the Pa.400
Suzukaze 20 had a delta shape, perhaps a nod
markings as used on a Nakajima B5N2 torpedo remained a design only.
bomber (known as Kdt€ to the Allies) flown by Petty
to the Pa.101. However, whether or not the
Officer First Class Toshio Takahashi from the artist based the Suzukaze 20 on Payen's
carrier Hiry[ during the attack on Pearl Harbor. designs may never be known for certain.


lupEnrnl JapnNesr Navv 103

Mizuno Shinryu II


Fsffi-T -,€: t

In June 1944, the first Boeing B-29 Super- aircraft being investigated would be a glider
fortresses appeared over Japan. It was the was. The Kaigun Koku Hombu envisioned
start of a bombing campaign that would see that gliders would be launched with rocket
key Japanese cities, infrastructure and indus- boosters from caves or shore positions and
tries reduced to ashes through conventional pilots would guide the aircraft and the 100kg
and firebombing raids. With the aircraft (220\b) explosive payload inside it into Allied
industry being a priority target, the Kaigun ships or tanks should the Japanese home
Koku Hombu looked to ways to combat the islands be invaded.
B-29 menace. One concept was a point The Kaigun Koku Hombu assigned the Dai-
defence interceptor that could quickly rise to Ichi Kaigun Kdku Gijutsu-shd at Yokosuka the
meet the bombers and so the Mizuno Shinryr-r task of turning the glider into reality. The pro-
II was born. However, the development of the ject was led by Shigeki Sakakibara who
Shinryu II (Shinryu meaning 'Divine Dragon') staffed a number of teams that would each be
began with designs for a far more conven- responsible for one part of the glider. The dif-
tional craft. ferent sections were the wings, the fuselage,
In November 1944, the Kaigun Koku control surfaces, aerodynamic testing and
Hombu looked into the possibilities of an air- test flights once the prototype was complete.
craft to undertake shimpu missions. While The Kaigun Koku Hombu gave instructions
the mission was not unique, the fact that the that the glider must be built from as much

104 Japnrursn Stcner PRorEcrs: ExpERruEnral Arncnarr op rHs IJA aruo IJN 1939-1945
wood as possible. This restriction was was able to cut the rope and proceed with the ceptor rather than a glider; Suganuma was
imposed for two reasons. The first was that in test flight. When Narabayashi put the Jinryu especially interested in this idea since he had
using wood and keeping the use of metal to into a dive and had reached 300km/h access to rocket engines that promised 32 sec-
an absolute minimum, the glider could be (186mph), the glider began to vibrate to such ond burn times. Two people were retained
manufactured in any small shop using only a degree that he was unable to read the from the Jinryu project: Sakakibara, the lead
wood working tools, and secondly, as a con- gauges. Pulling the nose up to bleed off designer, and Yoshio Tonsho who would over-
sequence, what metals were available would speed, Narabayashi discovered that the vibra- see the construction of the prototype. Ytjird
be conserved for other military uses. tions ceased. During his descent Narabayashi Murakami was tasked with the aerodynamic
Much of the glider's design was conceived examined the vibrations and after landing the testing of the Shinryu II. All of those assigned to
by Yoshio Akita. A number of concepts were issue was reviewed. The conclusion was that develop the Shinryu II were ordered by Sug-
discussed and sketched and after much the tailwas not sufficiently reinforced and the anuma to maintain the utmost secrecy.
deliberation among Akita and his teams the vertical stabiliser was too small. The Jinryu Unlike the Jinryr1, the Shinryu II was to be
design was complete by May 1945, and was modified by adding some strengthening built from the outset as an interceptor. Sakak-
Mizuno, a small aircraft manufacturer, had in the tail and the fitting a second stabiliser. ibara would use a canard design that made
almost finished the prototype. The glider was The changes were later validated both in the this the second Japanese aircraft to be devel-
very simple and used a high-wing monoplane air and in the wind tunnel testing of the mod- oped during the war with such a feature (the
form. The straight and flat wings were wide ified Jinryu model. Interestingly, before flying first was the Kyushrl J7W Shinden on page
but had a short span and were designed to the Jinrlu, Narabayashi had suspected that 84). In addition, the main wings had a plan-
ensure that the glider was easy to handle the aircraft would have stability problems form similar to a cropped delta. These design
given that inexperienced pilots would be at which, as was seen, proved to be the case. features were included as a means of ensur-
the controls. Also, the planform would be With the handling and flight characteristics ing stability in flight as well as good handling
able to accommodate the rocket engines that of the Jinryu proven, the testing moved to the characteristics. Since the average Japanese
were to be used to boost the glider into the next phase - that of powered flight. The glider pilot had little experience with canard
air. The pilot sat in an open cockpit. was relocated to an airfield in Kasumigaura, equipped aircraft, the Shinryu ll had spoilers
The design was sent to the Kaigun Koku about 19km (12 miles) north of Ishioka. Here, fitted into the top of each main wing. Each
Hombu for review. Sakakibara studied the the Jinryu was modified to accept a group of spoiler was able to rotate between 60' and 90'
plans and projections and after his analysis it three Toku-Ro 1 Type I rocket engines that and if the mechanism for controlling the
was felt the glider was flawed and changes together would produce 661 lb of thrust dur- spoilers was damaged, they would automati-
were necessary. After these had been made ing a l0 second burn. Testing of the rocket cally return to the closed position. The pilot
the design was approved. Work began on the array showed two serious flaws. The first was was provided with a canopy covered cockpit.
revised Jinrlu as the glider was now called by the quality of the rockets that resulted in a For power, the Shinryu II was to use four
the middle of June 1945. To hasten the con- number of failures. The second was the Toku-Ro 1 Type 2 rocket engines located in
struction, the finalised blueprints and work inconsistency of the burn times. Narabayashi the rear ofthe fuselage. Each engine provided
plans for the Jinryl were drawn up even as noted his concerns and forwarded them to a 30 second burn time and all together up to
the components for the first prototype were Major Suganuma who had been placed in 600kg (1,322 lb) of thrust could be delivered.
being built. Construction of the Jinr5,u was charge of the Jinryl project. ln addition to Two rockets would be used to get the Shinryu
again given to Mizuno. Working around the expressing his doubts about the rocket ll airborne while the other two engines would
clock, the company completed two proto- engines, he also stated that the Jinryu would be used when making the attack. There was
types with such speed that wind tunnel test- be unsuited for shimpu missions because, a concern regarding the operating tempera-
ing of the design was still underway. In fact, despite the changes made to the glider to tures of the Toku Ro rockets and two meth-
the first flight of the Jinry[ occurred even improve the flight characteristics, it was a ods of cooling the engines were considered.
before the results of the testing had been pro- challenging aircraft to fly. Narabayashi sug- The first would have utilised an air-cooled
vided to Tonsho and Sakakibara. gested that instead of being used for shimpir combustion chamber that would have
Tashiichi Narabayashi was the pilot who operations the glider should be modified to required an air inlet using a bayonet mecha-
flew the maiden flight in mid-July 1945 at the take six rocket engines each with a 30 second nism in order to maintain air flow across the
airfield in lshioka, a city located in Ibaraki pre- burn time. He estimated that at maximum chamber. It also would have required specific
fecture, about 90km (56 miles) northeast of burn the Jinryr1 could attain a speed of positioning of the fuel iniectors so as not to
Tokyo. The Jinryu was towed into the air by a 750km/h (466mph), and forweapons he envi- have the air flow disrupt the injection
Tachikawa Ki-9 (known to the Allies as sioned that it could carry ten explosive process. The second method would use
Spruce) piloted by Saburd Fujikura, a man charges adapted from artillery shells used by injectors which sprayed a water and alcohol
known for his skill in flying gliders prior to the the IJA in their 100mm guns (likely the Type mixture onto the rocket nozzle, cooling it. In
beginning of the war. 92). Not only did Narabayashi agree that the reviewing the two solutions for cooling, it was
For the first test, Narabayashi assessed the Jinrlu could be used against tanks and ships determined that the water/alcohol system
Jinryu's handling. On landing, his opinion was but added that it could also be used to attack would be the simplest to implement.
that the gliderwas stable and possessed good US B-29 bombers. Despite the issues with the No provision was made for a wheeled land-
handling characteristics. For the second flight rockets work continued on preparing the Jin- ing gear system and design skids were used.
Narabayashi would investigate the Jinryu's ryu for powered flight. A nose skid was provided with a basic spring
diving capability and after a few bounces on Major Suganuma, however, would become suspension to absorb the landing forces.
the ground the Ki-9 and the Jinryu took off. At the catalyst for the Shinry'r-r ll's continued Under each wing was a non-sprung skid
a height of 2,300m (7,545f0, Narabayashi development. Taking Narabayashi's concerns arrangement supported by two struts. For
went to cast off from the Ki-9 but found that onboard, Suganuma formed a team to revise take-off the Shinryu II was to use a two-
the tow rope release had stuck; however, he the Jinryu and produce a design for an inter- wheeled dolly similar to the one used by the

IvpeRrnr, JaparuesE Nnvv t05

MitsubishiJ8M Sylsui. Once airborne the pilot Japan). Like the J8M and Ba349 and due to Mizuno ShinrSr II - data
could jettison the dolly. In addition to con- the limited range afforded by the rocket
ventional runway take-off procedures, other engines, the Shinryu II would have to be posi- Contemporaries
Blohm und Vos BV40 (Germany), Zeppelin Fliegende Panzeriaust
methods for launching the Shinryu II were tioned close to targets that were likely to be
(Germany), Heinkel P. I 077 Julia (Germanl'), Junkers EF 27 Walli
bombed. And like the J8M, the Shinryu
considered but what exactly these were is not II
[Germany), Messerschmitt lie l638 (Germant'), BereznyakJsat'ev BI
known. lt can be speculated that towing the would have used a jettisonable wheeled dolly (Rusia)
Shinryu II aloft was one consideration. to take-off while firing a pair of its rocket
for the Shinn'i ll and are estimates onll
Another may have been air dropping the Shin- engines. Unlike the J8M which burned up all Specilications in parenthesis arc

ryu Il in the same manner as the Kugishd of its fuel at once, the Shinryu ll had a second
Type Special Atlack Glider (lntercept0I)
MXYT Oka. In both cases this may have pre- set of rocket engines which could be used to
Crew 0ne
serued two of the rocket engines which sustain flight endurance or to increase speed
would have been used up had the Shinryu II during the attack. In the same way as the Powerplant Three Toku-Ro Ttpe 1 rockets with a total combined
taken off from the ground. Ba349, the Shinryl II would be armed with thrust of 400kg (881 lb) rvith a burn time of l0 seconds (Four Toku-Ro
In order to combat the B-29, which could rocket projectiles, likely fired as a group to Tlrpe 2 rockets rvith a total combined thrusl 0f600kg/1,322Ib lvith a burn

operate at altitudes up to 10,241m (33,600f1), affect a spread pattern, to bring down the time of 30 seconds)

the Shinryrl II was to be equipped with a pres- bomber target. Finally, akin to the J8M, once
surised cockpit or, if such a cockpit proved the fuel and ammunitionwere expended, the Dimensions

problematic, the pilot would wear a pressure Shinryr II would glide back to its base to be Span 7,00m 22,91r

Length 7,60m 21,9ft

suit. For weapons, the Shinryfi II was to be recovered, refuelled and rearmed.
armed with eight rockets. Attached to the The Shinryl II would never be built Height 1.80m


Wing area I 18.4f1

inside of the rear landing skid arrangement because the end of the hostilities in August
Wing loading VA tN/A)
were four tubes, one on top of the other and 1945 terminated any further work on the Pou'er loading )i/A (N/A)
angled downwards, which contained the design. Likewise, the Jinryu glider would
rockets. never fly under power. After the failure of the Weights
There has been some conjecture as to the rocket motors during ground tests, the war Emptv NlA (\/A)

mission objective of the Shinryu II. Some came to a close before more suitable and reli- Loaded N/A (),1/A)

sources make the case that the Shinry,'u II was able motors could be acquired and tested. \'laximrrm N/A (N/A)

to be used like the Oka while others come to Mizuno completed a total of five Jinry0 glid-
the conclusion that the Shinryu Il was to attack ers. As a note, Jinryu is the known name for
the f irst Mizuno glider. For the purposes of this
['lar 300knh l86mph
armoured ground targets such as tanks. In ll0kn'h
Cruise 68mph
both cases these sources state that the nose of text, the author used Shinryu II to differentiate
Range 4km 2 miies
the Shinryir II contained an impact fused the interceptor from the glider. The kanii is Endurance (Shinnuil)1.3minutes
explosive warhead and once the rocket arma- the same for both spellings but is pronounced Ceiling 400m l,312lt
ment was expended, the pilot would crash differently. Both translations of Jinryu and
the aircraft into his final target using the war- Shinryl mean 'Divine Dragon'. Shinryl II is l00kg (220Ib) ofexplosive (8 unguided rockets)

head to deliver the coup de grdce. However, also used in contemporary texts and as such,
analysis of the Shinryu II shows that neither is used here for recognition purposes. Deployment None. A total of five protoptlpe Jinryu gliders were

mission was likely. The aircraft would have Whether the interceptor would have carried built br' \'lizuno. The Shinryu II remained a design boatd aircrafi.

been far more complex to build than the the same name as the glider is unknown.
Showa Toka or Oka and the Shinryl II was
constructed for manoeuvrability, high altitude
operation and the means to land. In addition,
using the Shinryu II for shimpu missions
against tanks makes little sense when there
were other simpler and more effective means
(both already in service and under develop-
ment) to eliminate armour. Perhaps this is a
case of the Jinryu glider's role being applied to
the Shinryr-r II, or an assumption based on the
fact that, like the IJN's other special attack air-
craft such as the Nakajima Kitsuka, Kawanishi
Baika and Showa Toka, the Shinryu II pos-
sessed no letter/numerical designation. So, by
extension, the ShinryTr II must also have been
a special attack weapon. This, of course, is not
to say that the pilot could not choose to use
the Shinryr-r II as a shimprl aircraft.
As an interceptor, the Shinryu II had a simi-
lar role to the Mitsubishi JSM Syusui and the
Gennan Bachem Ba349 Natter for which the
Japanese were aware of and obtained data
on (although the plans never made it to

106 JnpaxEsr SEcnnr PRo;ecrs: ExpsnrueNrnr- AIncRapr oF rHE IJA AND IJN 1939-1945
IlrpEnrll JapnNsse Nevv 107
Nakajima Fugaku

Japan shared a fundamental flaw with Ger- When the tide of war turned against the selice as a combat aircraft, let alone reach
many in regards to not developing a bomber Japanese, it was soon realised that some America and return.
capable of long range missions. The lack of means to attack the US mainland had to be It was to be Nakajima who would attempt to
this capability is considered by some to be a acquired, not only to destroy the war industry provide a strategic long range bomber capable
pivotal nail in the coffins of each country dur- of America but to ravage the civilian popula- of bringing the war to America. The man
ing the war. In both cases, efforts to develop tion centres to reduce morale and bring the behind the project was Chikuhei Nakaiima,
such a bomber came too late to affect the out- war to the US doorstep. In consequence, the chairman and engineer of Nakajima Hikdki
come of the conflict. Although the Japanese US would have to allocate or divert resources K.K. Motivated by his fears over the inability of
had considered the need for such a bomber to increase the defence of the homeland the Japanese to reach and destroy US indus-
at the outset of hostilities - as had the Ger- which would affect the war on other fronts. trial capacity, Chikuhei tried to convince the
mans with the disastrous Heinkel He177 As historywas to show, the Japanese did suc- IJN and the IJA of the need for a strategic
Greif -verylittle happened until the need was ceed in launching attacks against America, bomber. However, officials from both services
dire, and by then the noose was tight, chok- but only in the form of the Fu-Go balloon refused to consider his ideas. Thus, without
ing any hope for putting a long range bomber bombs and isolated attacks on the west coast official sanction or request, Chikuhei invested
into service. from submarinelaunched float planes. None a portion of Nakajima's resources to draft
The main cause of this apathywas the early had much of an effect. designs for a bomber that could take-off from
success in the Pacific theatre where the short There were some early attempts to pro- Japanese bases, cross the Pacific, attack tar-
and medium range bombers then in use by duce a long range bomber - for example, the gets on the West Coast of America and return
the Japanese were adequate to fulfil the Mitsubishi G7M1 Taizan (a 16-shi proiect) - to either their original bases or elsewhere in
needs of the IJA and IJN. With the entry of the plus designs that were actually built such as Japanese orAxis held territory. Nakaiima gave
United States into the war a formidable prob- the Nakajima G5N Shinzan and Nakajima the design work the name 'Project Z'.
lem arose. Geography put the military G8N Renzan. The Shinzan was not a success On 29 January 1943, Nakajima began the
machine of the US far out of reach of Japan. and the Renzan failed to reach operational ta-sk of assembling drafts and studies for the

108 .lApANEsE SEcRET PRolEcrs: ExpERItteNrnl AIRcnapr op rHE IJA arln IJN 1939-1945
design of the bomber, along with reports was in part due to his belief that Japanese air Nakajima Fugaku - data
which studied the feasibility and problems of forces were not strong enough to repel a
production. On the completion of this stage in bombing raid. Another facet of the thesis was Contemporaries
Conrair B-36 Peacemaker IUS), Convair XC-99 (l,S), I{essenchmitt
April1943, he again pitched the concept to the use of the bomber to attack the US war
['1e264 and N'le264B (Oermany), Junkers ju390 (Germany), Junkers EFl00
both the IJA and the IJN. This time, neither industry. Without materials and oil, the US
(Germany), Nlesserschmitt ['le P 08,01 (Germany), Tupolev Tu-4 (NATO
selice turned Nakajima away. However, could not produce aircraft, tanks and other
codenameBui/) (Russia), Tupolev Tu-85 (NATO codename Barge)
despite the information Nakajima had assem- weapons. More importantly, he added, the (Russia), Tupolev Tu-70 (Russia), Vickers.Armstrong Victory Bomber (UK)
bled for the proposed bomber, and despite Japanese should use the bomber to destroy
Specifications in parenthesesrefer to the Fugahu specifical\'.
both services now accepting the need for the Soviet military industry as a means to sup- All other

specificalionsrefer lo the primon Nahajina Project Z design.

such an aircraft, the IJA and the IJN also pro- port Germany. This implied that Nakajima
duced their own ideas. Not surprisingly, the could provide Germanywith such long range
Tlpe Long range slrategic bomber
hvo services had differing opinions on the bombers. Crew Six to ten Geven to eight)
requirements for the bomber. The IJA desired The Project Z bomber would employ an all
a tlpe that could operate at 9,998m (32,800ft) metal structure with the wings mounted in Powerplant Six Nakajima Ha-54 36-cvlinder, air-cooled radials
and carry a heavy defensive armament. By the mid-fuselage position. The plane was developing 5.000hp at engine start, each driving two alternaling stroke,
contrast, the IJN wanted a bomber capable of envisioned to be powered by the Nakajima contra-rolating three bladed propellers of 4.5m (i4.7ft) diameter/Six

fl-v'.ing at a height of 14,996m (49,200f1), an alti- Ha-54, 36-cylinder radial engine, also known Nakajima NKI lA {Ha-531 18-cllinder, air-cooled radials developins

tude where interception would be minimal as the D.BH. The Ha-54 was, in fact, two Ha-44 2,500hp at engine start and each driving a altemating stroke, four-bladed

and thereby allowed a lighter defensive 1 8-cylinder radials paired together. It was pro-
propeller ol 4,8m (l 5.7ft) diameter

\veapon load to be carried. Furthermore, the jected that the Ha-54 engine could produce
IJN wanted the bomber to take-off from up to 5,000hp and that six of these would be
Span 64.98m 213.2f1
Japan, bomb any target within the US, then sufficient to propel the bomber to a generous (Fugaku) 62.97m 206.6ft
utilise bases in Germany or German held ter- marimum speed of 679km/h (422mph). Each Length 44.98m 147,6f1
ritory to land, as opposed to making a round engine would drive two contra-rotating, (Fugaku) 39.98m 13l,2ft
trip. three-bladed propellers with a 4.5m (14.7f1) Height 8.77m 28.8ft

Though there were a number of variations diameter. The Ha-54 engine, however, would (Fugaku) 8.77m 28.8ft

of the aircraft during the Project Z develop- not be ready for some time (and as events Wing area 352.01m: 3,766.8ft'

ment, three basic designs of what became turned out, by war's end it was still only a pro- (Fugaku) 330.00m' 3,552ft'

the Fugaku emerged. The project presented totype engine and problems with cooling the Wing loading, 348,60k9m' 71,4lb/ft'

normal (Fugaku) 126,91kg/n: 261bilt'

b1,' the IJA used a 'tail sitter' undercarriage, power unit through the use of a ducted cowl-
Wing loading, 456.99k9/m' 93.61b/fl:
featured dual vertical stabilisers and bore ing were never solved). Therefore, Nakajima
loaded (Fugaku)211.89k9m' 43.4lb/fl'
some resemblance to German designs. It also had to settle for the experimental NKllA Power loading, 44kgihP lbih'
5 12
had a rounded off nose similar to the Boeing (Ha-53) which, while also in development, normal (Fugaku) 3,76kghp 8.3lbihp
B-29 Superfortress and Messerschmitt Me 264 was expected to be ready for trials. The drar.r. Porver loading, 7.2lks/hp l5.9lb1hp
'Amerika' bomber. The IJN's proposed back was that the NKI 1A was expected to loaded (Fusaku) 6.30ks/hp l3.glblhp
design used a tricycle landing gear arrange- muster only 2,500hp and this would certainly Aspect Ratio tz.t (NlA)

ment and rounded nose but utilised a single have lowered the performance estimates.
vertical stabiliser. Nakajima's proposal kept The introduction of the NKl1A meant that a Weights

the single vertical stabiliser but had a stepped revision of the Project Z airframe became Emptv 65.000kg 143,3001b

nose much like that used on the G5N Shinzan necessary.

(Fugaku) 33,800kg 74.5161b

n'hich the company had previously worked

Loaded 122,000k9 268.9631b
The bomber's ceiling was estimated to be
(Fugaku) 42,000k9 92,5941b
on. 1 5,000m (49,212tt) and it was believed that a
N'lar loaded 160,000kg 352,7391b
By June 1943, Nakajima had received plans healy defensive armament was not neces- (Fugaku) 70,000k9 154,3231b
from the IJA and IJN, reviewed them and sary as the high altitude would offer protec-
begun work on drafting a final design. To con- tion from fighter opposition. To a lesser extent Performance
tinue the research and further development the projected speed would also reduce vul- Mex speed 679krn'h 422mph
and study the Project Z aircraft, the Army and nerability. Consequently, the bomber would 10,000m at 32,808f1

\avy Aviation Technical Committee was carry at least four Type 99 20mm cannons, but i79knt' 184mph
(Fugaku) at 10,000m a132,808ft
formed on 9 August 1943. The IJA delegation contemporary illustrations of the bomber
Take-olf run (Fugaku) 1,020m 3,347ft
n'as headed by Captain And6. Later in August, often show a much heavier armament. This
Range 16,499km 10,252 miles
Chikuhei Nakajima prepared a thesis entitled may be a result of having to settle for the less
(Fugaku)16,499km 10,252miles
'Strategy for Ultimate Victory'. Chikuhei used powerful NK11A and any speed/altitude il'lar range 17,999km ll,l84miles
his personal clout to make sure his document advantage would have been lost, so an (Fugaku)19,400km 12,054miles
reached not only IJA and IJN officials, but also increased weapon load would have been Ceiling 15,000m 49,21211

politicians and even Prime Minister Hideki necessary to protect the aircraft. Typically, (Fueaku) i5,000m 49p12fi
T0j0. His thesis was laid out in six chapters illustrations show two cannon mounted in
and contained Chikuhei's plan for defeating the tail, two in the nose, two twin-cannon tur- Armament

the US as well as defending Japan. The key rets placed in the front and rear of the fuse- Four Tlpe 99 2Omm cannons (up to l2 cannons depending on the

component was the Project Z bomber which lage top and at least one belly turret. source) and a marimum bomb pa,vload 0f20,000k9/44,092Ib

he proposed could be used to destroy US air- Variations included waist gunner stations. For
fields as a means to deny the US the ability to a normal bomb load, the bomber was
None. The 'Project Z' and Fugaku aircralt existed onlv as paper designs.
launch raids against Japan. This suggestion expected to carryup to 20,000kg (44,0921b) of

IrrpEnru Japanrse Navv 109

bombs, but in the case of anti-shipping mis- for the Fugaku was to be destroyed to prevent of the cannon equipped Fugaku could bring
sions, torpedoes could be carried (see the information being handed over to the down 100 bombers, the area covered by the
below). For attacks on the United States, the Allies. Papers on the Fugaku that survive to cannons from one plane being 2.5m (8.2ft)
bomber would carry only up to 5,000kg this day, including a number of drafts for var- and 3km (1.86 miles) long. A system of
(1 1,023Ib) of bombs. ious Project Z/Fugaku proposals, were mis- ground radar stations would give advance
work continued on the Project Z, plans
As laid or kept for safe-keeping by individuals. warning of the incoming enemy bomber
were made to assemble and house the Since the war it has been claimed that the force, allowing time for the Fugaku to inter-
bomber's production line. By the fall of 1943, Misawa Air Base would have been used by cept and destroy the bombers before they
these plans had been completed and con- Fugaku bombers to launch raids against the reached Japan. This was all very impressive
struction of the new facility had begun. By US. While Misawa was used by the IJA and on paper but had it been put into practice the
January 1944, the Project Z moniker was operational IJA bombers flew from this facil- results were likely to have been less than stel-
dropped and changed to the Fugaku which ity, there has been no definitive evidence to lar, especially when considering the failure of
means'Mount Fuji'. support or refute Misawa being considered as the Mitsubishi G6M1 heary escort fighter (a
As it was, more pressing demands on a Fugaku base. G4M converted into a gunship to provide
Nakajima resulted in less and less work being Bombing was not the only mission that was cover for bomber formations). Finally, the
done on the Fugaku. To compound the prob- envisioned for the Fugaku and during the Pro- Fugaku was considered as a transport which
lem, by the time the design was nearing com- ject Z brainstorming three other concepts would have provided a significant healy lift
pletion, Japan was on the defensive and arose and later formed part of Chikuhei's the- capability. It was estimated that one Fugaku
chances of producing the Fugaku, let alone sis. The first was an attack design that had 400 transport would be able to carry 300 soldiers
using it to attackAmerica, were about nil. The Type 97 7.7mm machine guns crammed into with full equipment, about equal to one
IJA believed that there was no probability of the aircraft. The front and the back of the infantry rifle company with a heavy weapon
the Fugaku being built and therefore aban- bomber would accommodate 40 machine platoon. Chikuhei envisioned a grand
doned the project, leaving the IJN as the sole guns arranged in ten rows. The intention was scheme of a raid against America where four
remaining party involved. Even the Gunjushd to rain thousands of rounds of bullets down hundred transports would deposit 120,000
(the Ministry of Munitions) felt the Fugaku on to enemy ships with the theory that a men (equivalent to a Japanese Army, which
was impossible to realise and ordered swath of destruction 45m (148ft) wide and equates to a US and British Corps) on US soil
Kawanishi to design a new long range 1Okm (6.2 miles) long could be achieved by to take over the Seattle-Tacoma airport
bomber. Unfortunately, the Gunjusho failed 15 Fugaku aircraft. Once the decks of these located in Washington. After landing the
to inform the IJA, IJN and Nakajima about the ships were swept of personnel, nine Fugaku troops would move overland to attack and
Kawanishi bomber, which was known as the bombers, each with twenty 907kg (2,000Ib) destroy Boeing's B-29 producing Renton Fac-
TB. When the new bomber project was dis- bombs or torpedoes, would deliver the coup tory in Renton before returning to Japan.
covered, a hail of protests and arguments de grAce, covering a path 200m (656ft) wide There is no evidence to suggest any of
erupted ihat hampered not only the develop- and lkm (0.62 miles) in length with high these concepts made it to the final Fugaku
ment of the Fugaku but all long range bomber explosive. designs. However, if any of the three ideas
projects including the TB r,vhich was soon Another version had the Fugaku loaded were supported, a transport may have topped
cancelled. with 96 Twe 99 20mm cannons. The front the list for possible consideration given the
However, it was the fall of Saipan in 1944 and the back of the aircraft would contain 12 late war need for aircraft capable of bringing
that sealed the Fugaku's fate. The Japanese cannons arranged in eight rows while raw materials into Japan to feed the war
air forces no longer had need of a super long another 36 cannons lvere fitted on each side industry that was slowly being starved. As a
range bomber and demanded more perti- of the aircraft. This particular variant was to note, although the designations G10N and
nent aircraft to protect the mainland. As such, target enemybombers flying missions against G 10N I have been used in print for this aircraft
all work on the Fugaku was stopped and the Japan and would use hidden bases for many years, there has been no confirma-
plans, calculations and drafts were shelved. untouched by the Japanese airfield bombing tion in historical sources that confirms this
Work on the production facility was halted campaign. By flying over the enemy bomber was the case.
prior to completion and left unfinished. With formation and unleashing a withering fusil-
the Japanese surrender, all documentation lade of cannon fire, it was speculated that ten

110 JApAr'JESE Srcnpr PRo;Ecrs: ExpERmenr,ql ArRcRapr or rue IJA nNo IJN 1939-1945

lrrpnnrx- .lnplruEsE Navv l1l

Nakajima JIN Gekkd

In the spring of 1938, the IJN issued a 13-shi upward and forward at a 30' angle and two unknown. What is known is that it did not
specification for a long range fighter. This additional cannons firing downward and for- progress past the drafting board.
came about as a result of combat experience ward at a similar angle. Called the J 1 N 1 -C Kai,
in China that showed that Japanese fighters the field modification proved a success and
did not have the range to escort bombers on with that success, the IJN became interested.
missions deep into Chinese territory, the In short order, Nakajima was instructed by
result of which were high losses. What Naka- the IJN to produce a dedicated night fighter
jima produced to answer the specification version of the J1N. ByAugust 1943, the assem-
would have been a complete failure had it not bly of the first model, the JlN1-S Gekko
been for one redeeming feature. (meaning 'Moonlight'), had begun at Naka-
The l3-shi specification was a very strict jima's Koizumi plant. The main changes saw
one with a number of very demanding crite- the glazing over the crew compartment
ria. Nakajima's Katsuji Nakamura would pro- reduced and the step removed. The ring
duce the initial prototype. It was an exhaust collector was also removed and the
aerodynamically clean 3-seat monoplane engines used individual exhaust stacks. In
with low mounted wings. Each wing housed addition, it was realised that the downward
a Nakajima Sakae radial, developing 1 ,130hp. firing cannons were ineffective and in the
However, one wing was fitted with a Sakae 21 J I N 1-Sa were removed. Both the J 1 N 1 -S and

engine, the otherwith the Sakae 22, and their J1N1-Sa were sometimes fitted with a centi- Nakajima JIN Gekko - data
propellers rotated in opposite directions to metric radar in the front of the aircraft, the
reduce torque. To complement the nose external antenna for the set being situated on
Curtiss XP-87 Blackhank (US), l,lesserschmitt P.1099 (Germany)
armament, twin remote controlled barbettes, the tip of the nose. Others had a searchlight in
each with two 7.7mm Type 97 machine guns, the nose while a number lacked the radar No exact specifications ate hnoun fot the turbojet pouered I I N .

The specifications aiuen belou are for the JlNl-5.

were placed behind the pilot's cockpit. Des- and searchlight to be replaced by a single for-
ignated the JlNl, the first two protob,pes ward-firing 20mm Type 99 cannon.
T1pe Night Fighier
were delivered and put under test. A myriad In combat, the Gekkd proved satisfactory
Crerv Trvo
of problems were noted - the plane was over- against the Consolidated B-24 Liberator, but
weight, the novel propeller arrangement against the faster Boeing B-29 Superfortress Powerplant
caused numerous difficulties, the hydraulic was hard pressed to make a single attack. Two Nakajinia ),JKlF Sakae 21 l4-cylinder, air-cooled radial, each rated
system was too complex, the barbettes were Therefore, it should not be surprising that the at l,J30hp for take-off, l,l00hp at 2,850m (9,350it) and 980hp at 6,000m

too difficult to aim and the entire arrange- JIN was considered as a candidate to be (19,685ft), driving a three-bladed, c0nstant speed metal propeller

mentwas too hear,y. To top it all off, manoeu- equipped with two turbojets. Successful as a
vrabilitywas poor and all the JlNl had going reconnaissance platform as well as a night Dimensions

for it was its range. fighter, replacing the radial engines with tur- Span 16.97m 55.7fi

Rejected as a hear,y fighter, the JIN was bojets would have provided the JIN with a
Leogth 12.74m 4l.8ft

given new lease of life as a reconnaissance

Height 4,54m l4,9ft
superior speed that would have served it well
Wing area 40.00m' 430.;fl:
aircraft. Nakajima stripped the aeroplane of in either role. A contemporary illustration of Wing loading 175,27k{n' 35.9lb/ft'
its weapons and the barbettes, cut the fuel the turbojet equipped JI N shows the fuselage Poner loading 3.62k9/trp 8lb/hp
capacity (which, in part, was made up by of aJl N 1-Sa. The illustration lacks the upward
using drop tanks) and replaced the Sakae 22 firing cannon and radar which suggests it was Weights
engine with another Sakae 21 powerplant. equipped with the nose mounted cannon as Empt"v 4,840k9 t0,670Ib

The armament was a single, rear firing its armament. Perhaps it may have worked in Loaded 7,01Okg i5,454Ib

l2.7mmType 2 machine gun to be was used conjunction with radar and/or searchlight \'larimum 8,l84kg I8,043Ib

by the radio operator. Now called the J 1 N 1 -C, equipped JlNl-S and J1N1-Sa during combat
following successful trials the aircraft missions, or it may have been considered as
Speed 507krrt,'h 3l5mph
received approval and entered service with a fast special attacker fitted with two 551 lb
at 5,840m at 19,160it
reconnaissance units from August 1942. On bombs (as surviving J1N aircraft were at the Cruise speed 3nkn/h 207mph
encountering it in combat, the Allies code- end of the war). The wings seem to have at 4,000m at 13. I 25ft
named the plane lrulng, thinking it was a been unaltered outside of the required mod- Climb 9 min 35 sec to 5,000m (16,405ft)

fighter. The IJN renamed the plane JlNl-R, ifications to fit the nacelles for the turbojets. Range 2,544km 1,581 miles

equipping some examples with a turret As to what iet engines were to be introduced ['1ar range 3,779km 2,348miles

mounting a 20mm Type 99 Model 1 cannon. is not known. The Klgish6 Ne 20 was cer- Ceiling 9.330m 30,610ft

In early 1943, CommanderYasuna Kozono, tainly a candidate as was the Nakajima

who led the 251st Kdkutai operating from Ne 230. The adaptation of the Jl N to jet power
Tlvo upnard firing 20mm Trpe 99 cannons and hvo dolr,nlard firing
Vunakanau Airfield in Rabaul, Papua New lvould likely have been an easier task than
2Omm Type 99 cannons installed in the luselage
Guinea, believed that the J 1 N 1 -C would make that faced by Kugisho in adapting the PlYl
an excellent night fighter. Mechanics Ginga into the turbojet bomber Tenga. Deployment
replaced the observer's station and installed Exactly when the proposal to modify the None. The turbojet porvered J1 N remained a paper project
two 20mm Type 99 cannons that fired JIN to a turbojet aircraft was made is

112 Jnperuese SEcnEr PRoJecrs: ExpnRrrvrsNrnL ArncRerr oF rHE IJA AND IJN 1939-1945






lrrpEnral JnpnrriEsE Navv 113

Nakajima Kitsuka


Even though the genesis of the turbojet began compressor and, despite relatively little inter- 1938 saw German firm BMW begin their
long before World War 2, it would take the war est being shor,rn in it, he patented his concept. research into turboiets and the Heinkel He 178
to accelerate the development of this new In 1933, German Hans von Ohain designed a Vl prototype was built to test the HeS 3 turbo-
powerplant to the point that by the close of hos- turbojet similar to Whittle's bui it would not be jet that was being developed from the earlier
tilities jet aircraft had been blooded in battle. until 1936 that Ernst Heinkel took an interest in HeS 1. In late 1938, Messerschmitt started work
Germany can, by some, be considered the the engine and hired von Ohain to continue his on what would become the world's first iet
leader in turbojet technology during the war, work. By March 1937, this resulted in the fighter to enter squadron seruice, the Me 262. In
but the US and Britain were not far behind. Heinkel HeS 1, the first German jet engine Japan, and despite the lack of interest being
Japan, too, was not idle in producing its own although in fact a hydrogen demonstrator. The shown in jet propulsion, Captain Tokiyasu
turbojet but it would take German knowledge following month Whittle tested his first jet Tanegashima was appointed as the head of the
to give their industries a boost. One such engine, the WU orWhittle Unit. Engine Test and Field Support Shop of Kt1gish6.
results was was an historic aircraft in the Around this time, Rear Admiral Koichi Hana- He was issued with a meager sum to fund jet
annals of Japanese aviation history: the Naka- jima became aware of Whittle's work as well as engine research although, with the assistance
jima Kitsuka. that of Secondo Campini, an Italian who began of Professor Fukusaburo Numachi, he would
Because the Kitsuka (which in Japanese work on a thermojet and an aircraft to use it: the initially focus his efforts on turboprops. Both
means 'Wild Orange Blossom') was probably Campini Caproni N.1 in 1934. This rekindled his men were able to source the Ishikawaiima-
the most important Japanese aircraft to use a interest in jet propulsion and using his position Shibaura Turbine Company and Ebara Seizo
jet engine as its powerplant, it seems apt to as head of the engine division of the Dailchi Kai- K.K. to help build a number of test engines that
provide a general overview of Japanese turbo- gun Koku Gijutsu-sho, saw to it ihat studies were used compressors and gas turbines, but these
jet development in this section. The first axial- conducted in such engines. Hanajima reached labours did not bear fruit.
flow turbojet was patented in 1921 by out to the Tdky6 Imperial University and Mit- By 1939, BMW had tested its first axial-flow
Frenchman Ma.xime Guillaume. However, the subishi Juk6gyd K.K. and together all manner of turboiet design and on 27 August of that year
technology of his day was not enough io realise rocket and iet engines were investigated such as the He 178 Vl made its first flight, the first tur-
a working model. In 1930, Englishman Frank ramjets. To Hanajima's disappointment, little bojet powered aircraft to fly. In February 1940,
Whittle designed a turbojet using a centrifugal officiai interest was generated from the results. the British Air Ministry ordered two examples

114 JapaxesE SEcRsr PRo;Ec'r's: ExpsRruENter- AtRcRarr oF rHE IJA ar"rl IJN 1939-1945
ofthe E.28l39 research aircraft from the Gloster Wada was the man who oversaw the develop- from Germany. In May 1944, the Japanese
Aircraft Company to serve as testbed aircraft ment of a turbojet and the first result was the negotiated for the manufacturing rights to the
for Whittle's engines. 1940 also saw the ltalian TR-l0. This had a single stage, centrifugal com- Me 262 and the Germans initially agreed to the
N.l fly forthe first time and Heinkel began glid- pressor with a single stage turbine and was, in release. However, the deal was not concluded
ing tests of the He 280 jet fighter prototype as it essence, built by adapting a turbosuper- due to the large number of modifications that
rvaited for its two HeS 8 turbojets now under charger. The engine was constructed by Ebara the design was found to require after its flight
development (the He 280 did not enter pro- Seizo K.K. When the TR-I0 was first tested in testing. It was not until July 1944 that orders
duction). November would see Junkers test the summer of 1943 its performance did not were given to provide the Japanese with blue-
the Jumo 004 turbojet and Gloster's jet fighter meet expectations. The TR-10 was renamed prints of the Me 262 fighter and the Junkers
proposal, the Meteor, was ordered in February the Ne 10 and the engine was further devel- Jumo 004 and BMW 003 turboiets.
1941. Also in November, Lockheed com- oped by adding four axial stages in front of the On 22 July 1944, Reichsmarschall Hermann
menced work on the L-1000 axial-flow turbo- engine inlet. This reduced the load on the cen- Goring authorised the licensing of the Me 262
iet, the first American jet. Finally, in December, trifugal compressor, lowered the engine RPM to Japan and the delivery of one sample air-
[/hittle's W.lX turbojet, a flight ready engine, and produced more thrust. The revised jet craft. However, the Japanese submarine I-29
n'as tested for the first time. engine was designated the Ne 12. The problem had left Lorient, France, on 16 April with a sam-
Japan though, was not idle in 1940. Early in with the Ne 12, however, was its great weight ple Junkers Jumo 004 turbojet and plans for the
the year, Tanegashima, with the help of the Mit- and so steps were taken to lighten the engine, Me 262 and BMW 003 turbojet among its cargo.
sui Seiki Kogyo K.K., created a free piston com- which resulted in the Ne 128. Also aboard the submarine was Technical
pressor for a gas turbine based on a Junkers 1944 was an ominous year for Japan. When Commander Eiichi Iwaya who carried on his
design, but it was not a success as a means for the Mariana Islands of Saipan and Tinian were person a portion of the documentation on the
aircraft propulsion. Another attempt was tried wrestled from the Japanese by US forces in July German fighter and turbojets. By 14 July, the
b1'a different department. Under the leader- and August, Japan found herself well within I-29 had arived in Singapore. Iwaya, seeking to
ship of Lieutenant Commander Osamu striking distance of the Boeing B-29 Super- reach Japan as soon as possible, disembarked
\agano, head of the Kugish6 aircraft engine fortresses. Prior to this, B-29 raids had to fly from I-29 and took only aportion of the German
div'ision, and Masanori Miyata, who led the from remote bases in China and India and so documentation. From Singapore, Iwaya fl ew to
Kugishd electric parts section, built a tiny free the bombing of Japanese targets was relatively Tokyo. On 26 July, Allied code intercepts pin-
piston compressor gas turbine, generating one rare. Staging from Saipan and Tinian, B-29s pointed the location of I-29 and the USS Solo-
tenth of a horsepower at 1 2,000rpm that drove were far closer, could be more active and the /lsh sent her to the bottom near the Balintang
a magneto that lit a lamp. Despite this measure Japanese were only too aware of this. In addi- Channel in the Luzon Strait, taking the precious
of success, apathy on the part of the IJN con- tion, it was surmised that it would only be a cargo with her.
tinued to stymie progress. Tanegashima soon matter of time before the main Japanese When Iwaya arrived in Japan, all he pos-
realised that the Japanese industry was not islands were targeted for invasion. In August with regards to the
sessed of the German files
capable of constructing a free piston engine 1944, the Kaigun Koku Hombu called for a Me 262 and turbojets was a single copy of a
and switched his studies to axial flow jets. meeting to discuss changes in air strategy to cross-section of the BMW 003A turbojet. The
On 1 5 May 1941 , the Gloster 8.28/39 flew for combat the air and land threat as well as to subsequent news of the loss of I-29 was a
the first time, but previously in April, Heinkel's consider the aircraft that would be used. The crushing blow, but not a fatal one by any
He 280 Vl had flown under jet power on its Kaigun Koku Hombu invited aircraft designers means. In studlng the BMW 0034 document,
maiden flight, the first jet fighter to fly. 1942 saw from both Nakajima and Kawanishi to attend the Japanese found it to be of a similar design
the Junkers Jumo 004 under test while BMW and the outcome of this meeting was the pro- to the Ne 12 but instead of the centrifugal com-
focused efforts on the BMW003 Sturm. Heinkel posal for three classes of aircraft termed pressor the German engine used an eight stage
n'as instructed to concentrate on developing Kokoku Heiki (one literal translation being axial-flow compressor. lt was adjudged that
the HeS 011, a turbojet that was to power the 'Empire Weapon'). The first class, or Kdkoku this method was superior to the Ne l2 and as
second generation of German jets. On l8 July, Heiki No.1, was the adaptation of cunent air- such, efforts should be concentrated on build-
the Messerschmitt Me 262 flew under turbojet craft to accept a 800kg (1,7601b) bomb with ing the Japanese equivalent to the BMW 003A.
porver, becoming the second jet fighter to fly, which their pilots would undertake shimpl Despite the decision against it, work on the
and on 2 October, the American Bell XP-59 missions and target enemy invasion ships. If Ne 128 continued. Four companies were
.dracomet jet fighter made its maiden flight. By the bomb overloaded the carrying capacity of involved in the development of the new turbo-
this time, Japanese engineers and scientists the aircraft, then RATO (Rocket Assisted Take- jet. Each was to be provided with a copy of the
had learned of the flight of the He 1 78 as proof Off) units would be used to get them airborne. BMW 003A cross-section and other available
that an aircraft powered by a jet engine was Kokoku Heiki No.3 was to be a conventional, data and to build their own versions. Ishikawa-
feasible. This was just the boost the flagging radial engine aircraft designed by Kawanishi as jima-Shibaura Turbine Company was to
.lapanese jet engine research desperately the Tokkoki, which would be used for shimpl develop the Ne 130, Nakajima Hikoki K.K. the
needed. missions, but this proiect was soon abandoned Ne 230, Mitsubishi JnkOgyO K.K. the Ne 330, and
As a result two different paths were taken (perhaps because the IJN was to build the sim- Klgishd would move forwards with the Ne 20.
u'ith renewed vigour. The first employed the ilar Nakaiima Ki-115 as the Showa Toka). Following the August conference with the
principle of the thermojet (as used by Secondo It would be Kokoku Heiki No.2 which provided Kaigun Koku Hombu, Ken'ichi Matsumura,
Campini) and was called the Tsu-l 1 . While this the seed for the Nakajima Kitsuka. This chief designer for Nakajima and with the assis-
engine was to be selected for use in the 'Empire Weapon' was to be an aircraft that tance of Kazuo Ono, produced a number of
Kugisho Oka Model 22, iI was found to be used the Tsu- 1 I and, when available, the Ne I 2 concept drawings for the Kokoku Heiki No.2.
unsuited as a powerplant for a iet aircraft. The turbojet. Within Nakajima, the new aircraft was given
second route, that of a pure jet engine, was However, three months prior to the meeting, the codename Maru-Ten. On 14 September
pursued further. Kugish6's Vice Admiral Misao efforts were underway to obtain the Me 262 1944, IJN representatives met with Nakajima at

IrrpEnrnl JnpeNsse Nevy 115

their Koizumi plant to discuss the concepts duction schedule was revised. Nakaiima were On 28 January 1945, the wooden mock-up of
which had been put forward. The design that requested to produce the first prototype by the Kitsuka was finally ready for inspection at
stood out was based on a description of the February 1945 for use in static iesting. It was Nakajima's Koizumi plant. Vice Admiral Misao
Me 262 as provided by Technical Commander also during this meeting that the aircraft's spec- Wada and his staff visited the plant and
Eiichi Iwaya who, while in Germany, was able ifications underwent a revision. Instead of a inspected the mock-up with both Matsumura
to view and study the German jet. Thus, Mat- fixed bomb, the bomb could now be released and Ono in attendance. It rn'as made clear to
sumura's drawing bore an outward resem- by the pilot. The role of the aircraft was also the Krlgisho inspectors that the Kitsuka was a
blance to the Me262. After reviewing the changed. No longer was it to be used for a very simple aircraft that could be constructed
concept, the design was approved as the shimprl mission but instead for close air sup- in 7,500 man-hours. By comparison, it took
Kokoku Heiki No.2. In keeping with the shimpu port, the aircraft acting as a fast attack bomber. 15,000 man-hours to build a Mitsubishi A6M
mission of the aircraft, the initial design had no As a consequence ofthese changes, the design Reisen. Following the inspection, Nakaiima
landing gear and was to be launched from cat- had to incorporate a landing gear. The IJN was told to make two slight adiustments to the
apult ramps, boosted with MTO units. The cal- issued its specifications for the new iet, which Kitsuka. The first involved the windscreen.
culated range was a mere 204km (127 miles) was now called the Kitsuka, and the docu- Originally, the front windscreen was rounded
due to the designated engine, the Ne 12, which ments requested: but now it was desired that it should be flat
burned fuel at a rapid rate. At sea level the esti- Span: no more than 5.3m (17.3ft) with the wings panelled. This change may have been sug-
mated speed was 639km/h (397mph). A single folded gested to allow for the future installation of a
bomb fixed to the aircraft was the only arma- Length: no more than 9.5m (31.1ft) reflector gun sight because such a sight
ment. Another feature was the inclusion of Height: no more than 3.lm (10.1ft) requires flat panels to avoid sighting problems
folding wings to allow the aircraft to be hidden Powerplant: Two Ne 12 jet engines due to canopydistortion. The second alteration
in caves and tunnels and protected from Maximum Speed: 513km/h (3l9mph) with was to make the canopy slide to the rear
bombing attacks. 500kg (1,102Ib) bomb instead of opening to the side. At the conclu-
On 8 October, Kugisho ordered Kazuo Range: 204km (127 miles) with a 500kg sion of the meeting, Nakaiima was iold to
Yoshida, plant director for Nakaiima, to have a (l,l02lb) bomb or 278km (173 miles) with a cease all work on the Nakaiima J5N1 Tenrai
wooden mock-up of the aircraft completed 250kg (551 lb) bomb and the company was also informed that they
and ready for inspection by the end of the Landing Speed: 148km/h (92mph) could expect the Nakajima G8Nl Renzan to be
month. In addition Nakaiima was told to have Take-offRun: 350m (i,148ft) using two 450kg ierminated as well. These changes in produc-
the initial structural plans finished by the same (992Ib) RATO bottles tion and development were done to speed the
date. This was ordered so that production of Manoeuwability: The aircraft had to be highly coming production of the Kitsuka. The close of
the aircraft could begin without delay. Unforiu- manoeuwable, have a short turn radius and January also saw the final design draft of the
nately, delays would be a major problem. The be stable at speed to facilitate target tracking Ne 20 completed and almost immediately
IJN promised that the Ne 12 would be ready for Protectioni Shatter proof glass for the canopy. work began to build the first engine. Klgisho's
testing by November 1944 and in short order Front windscreen to have 70mm of bullet Aero Engine Division provided 400 machine
thereafter, production engines would be avail- proof glass. 12mm of steel armour plate tools and engineers and labourers began to toil
able. Based on this assumption, Nakajima was below and behind the pilot. Fuel tanks to be day and night to realise the Ne 20.
to construct thirty aeroplanes by the end of 22mm sandwich types February 1945 opened with the Japanese
December 1944. Because of the rush to pro- Basic lnstrumentation: Tachometer, altimeter, naval docks in Singapore targeted and
duce the aircraft, a myriad of problems arose artificial horizon, airspeed indicator, Model O destroyed by B-29 bombers along with contin-
with the design which necessitated changes. A Tlpe 1 flur gate compass, fuel pressure ued fighting in the Philippines. A second
major issue was the lack of critical war materi- gauge, oil pressure gauge, oil temperature inspection of the Kitsuka was called for on 10
als which required the use of substitutes and gauge, tail pipe temperature gauge and a pitot February. Present at the inspection, among the
brought additional delays. To compound the tube electric heater other engineers and Kugishd personnel, were
problem, Nakajima was concerned that the Basic Equipment: Type O parachute, automatic Technical Commander Iwaya and the man
Ne 128 would not be ready despite the IJN's fire extinguisher, Tipe 3 dry battery, Tlpe 3 who was destined to fly the Kitsuka, Lieutenant
promises. radio receiver, Tlpe I life raft and a reserve Commander Susumu Takaoka. The Kitsuka
Meanwhile, Kugisho proceeded with the weight of 30kg (66.1 lb) was given final approval and production was to
Ne 20. The engineers were forced to use alloys commence at once, even before the Kitsuka
which were not to the standards of the German 1945 would open with more misfortune for the had been flight tested. The first five Kitsuka air-
engine and would be a source of problems dur- Japanese war machine. Japanese troops were craft, No.1 through No.5, were to serve as pro-
ing testing. The design of the Ne 20 was smaller pushed out of Burma from 5 January and B-29s totypes and none would be fitted rvith armour
than the BMW 003A but it retained the com- would bomb Tokyo the next day. Two days ear- plating or self-sealing fuel tanks. In addition the
bustion chamber shape of the German engine. lier, Matsumura and Ono, along with others first two aircraft would not to be equipped with
While it used the same size of burner as the involved in the Kitsuka proiect, discussed the the bomb carrying apparatus. February would
BMW 003A, it only used twelve instead of six- possibility of using the Ne 20 turbojet in place of also see the Ne l28 tesied for the first time.
teen due to the smaller size. Kugish6 would the Ne 12. ln the debate, some suggested that Unfortunately for the Kitsuka, US bombing
draft and refine the design of the Ne 20 through the Ne 20 was not as far in development than ensured that production did not go smoothly.
December. the Ne 12 and would delay progress if used. On Due to the ever increasing number of strikes
On 9 December 1944, the IJN called a meet- the other hand, some argued that the Ne12 against the industrial centres of Japan, it was
ing to discuss the progress and outlook of the was not achieving significant results. In the felt that it was only a matter of time before the
Kokoku Heiki No.2. Based on the problems end, the consensus was that the Ne 12 should Nakajima Koizumi plant would attract the
Nakajima were having with the aircraft, not to remain as the powelplant only because it was attention of US bombers. Therefore, on 17 Feb-
mention the doubts about the Ne 12, the pro- projected to be ready before the Ne 20. ruary, engineering staff for the Kitsuka was

116 JnpeNsse SEcnEr PRorpcrs: ExpEnrunIlral AIRcRRpT or rHE IJA nlo IJN 1939-1945
tain Tokiyasu Tanegashima, laboured to refine to the Kitsuka design. The final blow was that
moved to Sano in Tochigi Prefecture. Despite
the Ne 20. The worsening bombing situation saw high grade aviation fuels would only be avail-
the move, a sizable portion of the Kitsuka com-
at the Ne 20 team moving to Hadano in Kanagawa able for the Honrare series of radial engines' All
ponent construction remained Koizumi
Prefecture, a three hour drive from Yokosuka. other engines, including the Ne 20, would have
n,hile the wings, tail assembly and the centre
and aft portion of the fuselage were con- Set up in warehouses belonging to a to make do with poorer quality fuel. This, cou-
tobacco factory, the Ne 20 group comprised pled with defeat after defeat for the Japanese
structed by Kugisho in Yokosuka' In the face of
10 officers and 200 men. Here, two bench military, cast a very serious cloud over the Kit-
further bombing attacks, production was dis-
testing stations were created and Ne 20 devel- suka project and some no longer saw value in
persed among silkworm factories and build-
opment and testing continued. The process continuing with the aircraft. Others however'
ings in Gunma Prefecture (northwest of
revealed numerous flaws. At one stage the had a strong desire to see the Kitsuka taken to
I oKyoj.
pressure of the axial-flow compressor was completion because it would put Japan into
\'larch arrived in a blaze of smoke and fire as
found to be too low. Nagano came to the con- the jet age.
the US ramped up their incendiary bomb cam-
clusion that the camber of the stators was not On 25 June 1945, the first Kitsuka was com-
paign against Japan's cities. T0ky6 and Nagoya
correct and so he took them out, bent them pleted but without its engines. Although exter-
u.ere particular targets, the burning cities light-
on an anvil and then reinstalled them. These nally the Kitsuka bore a resemblance to the
ing the night sky. On March 26, the first Ne 20
were tested in the second Ne 20 to be built. Me262,that was as far as it went. The wings of
engine was successfully test run from a cave
Yet another difficulty arose with the thrust the Kitsuka had a total of 13'sweepback, the
set into a cliff in Yokosuka. With the success of
the Ne 20, the Kitsuka engineering team began bearings on the compressor which was burn- centreline of the wings being at 9"' Wing tip
ing out very quickly. Nagano solved the prob- slots eliminated the tip stall discovered during
to seriously consider replacing the Ne 12B with
the Ne20. It was clear that the Ne20 outper- lem by revising the bearings and bearing wind tunnel testing and split flaps and droop
formed the Ne 12B and, based on the higher rings. One problem that reappeared was ailerons were fitted to compensate for the
blade cracking. The blades were made from heavy wing loading. Nakaiima K series airfoils
thrust potential, it was decided that the Kitsuka
should use the Ne 20 even if it meant a longer manganese-chromium-vanadium steel and were used - a K 125 airfoil at the wing root and
not the more suitable nickel alloy. These a K 309 airfoil at the wing tip' The wings were
delay while the engine became available.
Although the current Kitsuka production did blades were then welded to the disk and, as
not yet involve the engine mountings, a revi- such, the blades did not have the strength to
Interceptor initial concepts - data
sion of the aircraft design plans was required to withstand the operating stresses of the motor'
accommodate the Ne20. By l\4arch 31, these After one to two hours of operation, cracks The specifications in parenthesis rcfer t0 the modified uing uariant
revisions were complete and the Kitsuka pro- would appeare on the blade roots at the point
gram entered a stage of finalitY. where they connected to the disk. The solu- T]?e Interceptor

With the revised Kitsuka, some of the speci- tion was to thicken the blades but this low- Crew 0ne

fications were adjusted as follows: ered the efficiency of the engine. However,
Maximum Speed: 620km/h (385mph) with a the Ne 20 was abte to run for four or five hours Powerplant TrvoKrigishoNe20axialflo$turboiets,
each developing 490kg (1,080Ib) oithrust
500kg (1,102 lb) bomb at sea level before cracks appeared and while the engine
Range: 351km (218 miles), at sea level, at full could have run longer, there was no guaran-
power tee when blade failure would occur. With 10.00m 32.8fi
Take-off Runr 500m (1,640ft) with two 450kg these improved results, work began to pro- 9,23m 30.3ft
(992lb) RATO bottles duce a small number of engines. Height 3.01m lOfr

Landing Speed: 92km/h (57mPh) 25 April 1945 would see the first Kitsuka fuse- Wing area 13.19m' l42fr,

Bomb Load: 500kg (1,102Ib) as normalwith the lage completed. This was then subiected to (modified) l4.5lm' l56,2ft'

ability to carry a 800kg (l,763Ib) bomb; a Type stress and load testing which began on 20 May,
3 rack would be used for the larger bomb but with the stipulation that the fuselage was Weights

Protection: Reduce the bullet proof glass not to be damaged during tests. Nakajima was Empty 3,920k9 8,612.1 lb

scheduled to produce 24 Kitsuka aircraft by (modified)2,980kg 6.569 7lb

thickness to 50mm and add 12mm of armour
June 1945 and with the availability of six Ne 20 Loaded 4,l52ks 9,153.51b
to the front of the cockpit, while the fuel tanks (modified) 945k9 2.083.31b
Useful load
would incorporate an automatic fire engines. On the surface, the Kitsuka proiect
Fue) capacit.-v 725 litres 191,5 gallons

suppression sYstem looked to be moving along. The reality was a 1,450 litres 383 gals
n'ith drop tanks
far different story.
Engineers working on the Ne 20 found that, On 13 June, Vice Admiral Wada held a meei- Performance (estimated)
although the initial test of the engine was a suc- ing to discuss the Kitsuka. Wada addressed a \'loispeed 698knXh 434mPh

number of issues that were becoming prob- al 6,000m

cess, there were many issues to solve. At first, at 19,6851t

the blades were prone to cracking but this was lematic. Nakajima's G8N1 Renzen program 684knYh 425mPh

had to be stopped in order to free up produc- (modified) at 6.000m at 19,6851t

soon overcome. An electric starter was Iitted
tion capacity for the Kitsuka as both a special Range 608km 378 miles
into the compressor spinner that could spin the
at 6,000m at 19,685f1
engine at 2,250rpm; the engine would reach attack aircraft and an interceptor' More trou-
594km 369 miles
ma,ximum RPM within 10-15 seconds of engine bling was that unless the stock of aluminium (modilied) al 6,000m at 19,685ft
start. Gasoline was used to start the engine and was conserved the supplywould be exhausted l2,l00m
Senice ceiling 39,698ft

once running the fuel was switched to a pine by September 1945. At best, even with conser- (modified) 12,300m 40.3ilft
root distillate using 20-30 per cent gasoline. vation, by the close of 1945 there would be no
What was becoming a problem was how to more aluminium available. As a result, only Armament

position the tail cone. Lieutenant Commander steel and wood would be left and to use such One Tlpe 5 30mm cannon uith 50 rounds ol ammunition

Osamu Nagano and his team, along with Cap- materials would, again, have caused a revision

InrpsRIal JepaNPsE NnvY 117



1 18 JapaNrse Secner PRo:Ecrs: ExpoRnrrrurar- Arncnapr op rne IJA aruo IJN 1 939-1 945
of double spar construction with nine main blades in one of the compressors. The damage Kitsuka Special Attacker - data
support ribs, all covered with steel and duralu- to the engine was so extensive that repairs
min skinning. Mitsubishi A6M Reisen flap were simply not possible and replacement the Spetifintions in parctilhesis refer to the rail launched rcrsion.

hinges were used on the trailing edge flaps and only option. This delayed the flight for many
Tlpe Sperial Attacker
the wing tips were fabricated from wood and days. As the Kitsuka was being repaired, the
Crew 0ne
steel sheeting. The outer wing folded upwards. personnel for Ihe724 Kokr-rtai, which had been
The Kitsuka had a slight gull wing form thanks designated a special attack unit and which Powerplant Tno K[gisho Ne 20 arialflou'turbojets,
to 5" dihedral ofthe centre span and 2" dihedral would fly the Kitsuka in selice, had been each developing {90kg [1,080Ib) ol thrusl
of the outerwing. All control surfaces were fab- assembled at Yokosuka after its formation on I
ric covered. The fuselage had a slight triangu- July 1945. On 15 July, the new unit moved to Dimensions
lar shape, being composed of three sections Misawa Air Base where it began training using Span 10.00m 32.811

(nose, centre section and aft). The centre sec- Aichi D3A1 and D3A2 carrier bombers (known Length 9.23m 30,31t

tion had the centre wing span built into it and as Va1 to the Atlies), which had been relegated Heighl 3.0{m l0ft

much of this and the other two sections were to the training role. \\ring area l3.l9m 112ft'

(Rail) 270.00k9hr'
constructed from sheet steel due to the On 27 July, Lieutenant Wada conducted \Ving loading 5;.3lbril

unavailability of duralumin in quantity. Twenty- some successful tari tests with the Kitsuka.
four bulkheads were contained within the High speed tari tests, however, were prepared
Emptl 2,300kg 5.0i0.61b
complete fuselage with two bulkheads coming by appointed Kitsuka test pilot Lieutenant Loaded 3,550k9 7.826..1 lb
together where each section met, which were Commander Susumu Takaoka. Two days after (Rail) 1.080kg 8,99.1.8 tb

then bolted together to complete the fuselage. the initial taxi tests, Takaoka ran the Kitsuka up
Trvo fuel tanks were fitted, one in front of and to 129km/h (80mph) and then applied the Performanre (estirnated)

the other behind the cockpit. The tail of the Kit- brakes to test their effectiveness. He found that l\{ax speed 6i6knrih 120mph

suka was fairly conventional and the aft fuse- their stopping power was not adequate, at 6,000m at 19,685ft

888kn,h 5i2mph
lage line was kept high so that the stabiliser though he felt the problem was not so severe
n.ould not be effected by the jet efflux. For the
(Rail) at 10.000m atil2.808ft
that flight testing had to be stopped. Ground
Landing speed (Rail) lTlkmih l06mph
tricycle landing gear, the main gear (to include testing was finally completed on 6 August, the
Range 583km 362 miles
the brake system) from a Mitsubishi A6M same date that Hiroshima was devastated by at 6,000m at 19,685ft
Reisen was modified to suit the Kitsuka and the the'Little Boy'atomic bomb dropped from the (Rail) 8llkm 506 rnilcs. mar
600mm x l72mm-sized wheels retracted into B-29 'Enola Gay'. News of this strike soon Clinb (Rail) I I min 50 sec to 6.000m (19,685it)
the rving. The 400mm x l40mm-sized nose reached the Kitsuka crews, technicians and Ceiling 10.700m 35, r 0Jfl

n'heel was taken from the tail wheel of a engineers. l2.l00m 39,698ft

KugishO PIY Ginga and it retracted into the rear 7 August 1945 would see excellent flying
of the nose. conditions and the Kitsuka was made ready for tumament Onc 250kg [55 I lb) bomb, one 500k9 (t.I 03 lb)

After being assembled the Kitsuka was then flight. Weather reports stated a 24krnlh Oor5. or 1tu6 Type 99 2Omm cannons

broken down, loaded into trucks, and moved (15mph) southwest wind and a crosswind
to Nakaiima's Koizumi plant where two Ne 20 blowing from the right across the 1,692m
engines awaited it. By 27 June, the Kitsuka had (5,550f1) length of Runway 20 that pointed Trainer Kitsuka - data
been put back together and the engines towards Tdkyd Bay. The Kitsuka was only given
T)pe Trainer
installed, and two days later weight and bal- a partial fuel load to keep the weight to 3,l50kg
Crew Two (Student and lnslructor)
ance checks had been completed. The Kitsuka (6,945Ib); this allowed for approximately 16
u'as then declared ready for flight testing. On minutes of flight time. No RATO bottles were Pon'erplant Two Kugisho Ne 20 arial-flow lurbojeh,
30 June 30 I 945, both Ne 20 engines on the Kit- fitted so that the take-off characteristics of the each dereloping 490k9 (1,080Ib) oithru$
suka were started and run for a short time. aircraft could be assessed. Takaoka climbed
Flight testing could not be conducted at the air- into the cockpit and made ready to take-off. On Dimensions

field at Koizumi because the runway was too his signal the Ne 20 turbojets were started and Span 10.00m 32.8ft

short and had many approach restrictions. Mis- he was soon ta-riing out to the start of the run- Length 9.23m 30.3ft

Height 3,04m
an'a Air Base (Misawa HikdjO), in Aomori Pre- way. Once there, he extended the flaps to 20" 1{)ft

Winq area l3,l9m l12ft,

fecture 684km (425 miles) north of TdkyO, was and kept the brakes set. So as not to cause a
also considered since it had open approaches compressor stall, Takaoka slowly eased the
and was rarelyvisited byAllied long range fight- engine throttles forward and when both had Loaded 4,009k9 8,838,31b
ers. However, because of the great distance it reached 11,000rpm, he released the brakes
n as ruled out. Finally, it was settled that the air- and the Kitsuka began to roll. Twenty-five sec- Performance (estimated)
field at Kisarazu Air Base (Kisarazu Hikoio) onds later and after a run of 725m (2,378f1), the Mar speed 72lkn,h 148mph

ivould be the location for the first flight Kitsuka was airborne and went into the history at 6,000m at 19,685f1

because it was far closer to Yokosuka than Mis- books as the first Japanese jet to fly. Landing speed l66kmih l03mph

arva. Once more the Kitsuka was disassem- At 610m (2,000ft), Takaoka levelled off. He Range 66km at 4ll miles

6.000m at
bled, loaded into trucks, and moved to the was instructed to not retract the landing gear 19,685f1

Ceiling 12.000m 39,37011

Kisarazu airfield, adjacent to Tokyo Bay. nor exceed 314km/h (195mph). As a test pilot
Sen'ice ceiling 10.700m 35,101ft
On arrival, the Kitsukawas reassembled and he was used to hearing the roar of a conven-
made ready for its first flight. Unfortunately, on tional aircraft engine and used such noise as a Armament
l4 July, during engine testing, a loose nut was means to detect problems. However, Takaoka Likeh't0 carry lhe pa_-vload as per the Kitsuka
ingested which completely shattered the was not prepared for the whine of the turbojets

IupenrlL .JapanEsE Navv t19

Reconnaissance Kitsuka - data that told him almost nothing outside of what At four seconds into the take-off roll, Takaoka
his insiruments reported. He circled Kisarazu activated the RATO units. Immediately, the
Type Reconnaissance airfield, keeping it in sight in case of a failure acceleration caused the nose of the Kitsuka to
Crew Tlvo (Pilot and 0bseruer)
and because the airspeed kept rising, Takaoka pop up, the tail slamming onto the runway.
had to constantly throttle back to keep from Takaoka fought to get the nose down by jam-
Powerplant Trvo K0gish0 Ne20 arial-florv turbojets,
exceeding the gear down speed limit. A brief ming the stick forwards but he received no
each developine 490kg (1,080Ib) ofthrust
test of the control sensitivity showed that the response from the aircraft's elevators. The two
rudder was stiff, the ailerons were heary but MTO units burned for a total of nine seconds
Span 10,00m 32.8ft
were working and the elevators were overly and during eight of those seconds Takaoka
Length 9,23m 30.3ft responsive. When his flight time was up, was helpless and unable to correct the nose up
Height 3,04m 1oft Takaoka was wary of how he would land. He condition. One second prior to the units burn-
Wing area 13.19m' l42ft, did not want to drop the turbojets to below ing out, the elevators finally took effect and the
6,000rpm since that risked a flameout from nose came down so hard Takaoka was sure
Weights which he would likely not recover in time. the front tyre had blown when it contacted the
4,24lkg 9,349,81b
Therefore, he chose a long, shallow drop, low- runway. Takaoka felt a sense of deceleration
ered his flaps 40'and brought the turbojets as the Kitsuka reached the halfway point on
Performance (estimated)
down to 7,000rpm. On touchdown, he only the runway - his speed at that point was
Max speed Zlktr/h 448mph
needed moderate braking to bring the Kitsuka 1 66km/h/1 03mph. A second later, with the feel-
at at 19,685ft

Landing speed 169kn/h lO5mph to a stop using only a little under 610m (2,000ft) ing of deceleration still present, Takaoka
Range 666km 414 miles of runway. Takaoka brought the Kitsuka back decided to abort the take-off and he cut the
at6,000m at 19,68511 to the ramp amid throngs of cheering men. The power to the engines. Unfortunately, the brake
Ceiling 12,000m 39,370ft total flight time was 1l minutes. In his immedi- issue Takaoka had discovered during high-
Service ceilnq 10.700m 35,104ft ate report on the flight, Takaoka stated he had speed taxi tests now came back to haunt him.
experienced no problems with the engines Despite maximum application of the brakes,
and had no recommendations for improving the Kitsuka showed no signs of slowing and
None (posibl.v trvo Type 5 30mm cannons)
the aircraft. During his debriefing, technicians Takaoka was rapidly running out of runway. As
had removed the cowlings to the Ne 20 turbo- he neared one of the taxiways, Takaoka held
jets and examined each engine. They found no the left brake in an attempi to make the Kitsuka
faults and so gave the Kitsuka clearance for bring its left wing down into the ground to
another flight, scheduled for 1 0 August 1 945. bleed off speed (known as a ground loop). The
For the second flight, more fuel was to be Kitsuka's nose turned slightly but this then put
Interceptor definitive version - data
stored and RATO bottles used; this would the aircraft on a crash course with a group of
Type lnterceptor allow for a longer flight and test the RATO units hangars and buildings. Takaoka reversed the
Crew 0ne as boosters. Takaoka would again pilot the Kit- braking, holding the right brake. The Kitsuka
suka. Prior to the flight Takaoka examined the came back around onto the runway and
Powerplant Trvo lshikawajima Ne 130 or two Nlitsubishi Ne 330 MTO bottles which were fitted to the under- despite Takaoka working the brakes, it was to
arial-liow turbojets, each developing 900 or J,300kg (1,984Ib or 2,866Ib) side of the fuselage and found fault with the no avail. The aircraft ran out of tarmac and
of thrust respeclively
angle at which they were set. However, to crossed the 100m (328ft) of grass overrun
adjust them would have taken too much time before the landing gear caught in a drainage
and so instead of 800kg (1,763 lb) of thrust, the ditch and collapsed. The Kitsuka slid along its
Span 10.00m 32.8ft

Length 9,23m 30.3ft

bottles were reduced to 400kg (881 lb) each. belly until finally coming to a halt by the edge
Height 3.04m 10ft On the day of the second flight, Allied air of the water of Tokyd Bay. The damage to the
Wing area l3.l9m' 142ft' power was highly active and any flighi attempt Kitsuka was extensive. In addition to the man-
\[ring loading 302,71k9h' 62lb/ft' was bound to be spotted putting the Kitsuka at gled landing gear, the two Ne 20 engines were
risk. Consequently, it was decided to waii until badly damaged, having been iarred from their
Weights the following day on 10 August. However, it mounts but still remaining attached to the
Empty 3,060k9 6,746.1 tb
would be remembered for the drafting of the wings. Initial assessments suggested that the
Loaded 4,232k9 9,329.9tb
Imperial Rescript on the Termination of the damage was so severe the Kitsuka could not
Useiul load 940k9 2,072.3tb
War by the Japanese cabinet at the behest of be repaired. On the positive side, the aircraft
Emperor Hirohito, though the populace had no did not catch fire and causes of the accident
Perf ormance (estimated)
knowledge of this. were swiftly looked into. IJN Captain It6, who
NIar speed 713km/h
(Ne 130) ar 6,000m ar 19,685ft 1 1 August 1945 shared a similar weather pat- was present for the flight, was thankful that the
Landing speed I 54kr[/h 96mph tern to the day the Kitsuka had first flown. The Kitsuka did not become airborne with the nose
Range 594km 369 miles difference was that several IJN and IJA officials high attitude during the RATO burn. Had that
792km 492 miles at cruise of high rank had arrived to witness the second happened and once the RATO bottles cut out,
speed flight. Once more Takaoka climbed into the the Kitsuka would have most likely crashed
Climb I I min 18 sec ro 6,000m (19,685fr)
cockpit, signalled for engine start and taxied into the ground. A motion picture camera cap-
Ceiling 12,300m 40,354ft
out to the runway. As before he extended the tured the flight and the film developed to see if
flaps 20", and after receiving the signal to take- it could shed any light on the crash.
off, he slowly opened the throttles until the On 15 August, the film of the ill-fated flight
Trvo Tlpe 5 30mm cannons or two Ho-155 30mm cannons (lJA); one

500kq (1,102 lb) or 800kg (1.763 lb) bomb (if used as a fighter-bomber)
engines had reached 1 l,000rpm before releas- was studied but proved inconclusive as to
ing the brakes and the Kitsuka rolled fotwards. whether or not the Kitsuka was airborne once

120 JapaNEsE SEcnEr Pno:Ecrs: ExpeRtuENral ArRcnnnr oF THE IJA AND IJN 1939-1945
the RATO bottles were exhauasted, as was
suspected. This would have explained the o
heary impact of the front landing gear on the o
runway and the sense of deceleration experi- J
enced by Takaoka. In any case, the Kitsuka
rvould never fly again for at l2.00pm the Impe-
rial Rescript on the Termination of the War was
broadcast on the radio bringing World War 2 to
a conclusion.
The end of the war would see none of the
Kitsuka production plans realised. Naka.jima,
bv the close of December 1945, was to have
produced 200 Kitsukas. In reality, Nakajima
completed only one with a further 22 under
construction. The KyushuHikoki K.K. was, also
br,the end of the year, to have turned out 135
Kitsuka aircraft but was only able to begin con-
struction of two aircraft, started in July 1945,
rvhich remained unfinished by the close of
hostilities. A third producer, the Sasebo Naval
.\senal (Sasebo Kaigun Kdshd), was sched-
uled to have begun production of the Kitsuka in
September 1945 with 115 completed by the
close of December. The fourth production line trl!
u'as to be at the Kasumigaura NavalAirArsenal

u'ith the commencement of Kitsuka construc- :


tion scheduled for October 1945; 80 aircraft

u ere to have been completed by the end of
A number of variants of the Kitsuka were
planned, none of which would see completion
come the capitulation. One of these was a two-
seat trainer. Given the nature of the Kitsuka, it appreciated that a trainer would be
required to help the conversion of pilots used
to conventional piston engined aircraft to the i::ti:rt1
peculiarities of a turbojet powered aircraft.
Five of the Kitsuka airframes under production
bi' Nakajima were modified by including a sec-
ond cockpit for the instructor. Outside of the
inclusion of the additional cockpit, it is
unknown exactly whai other changes were
made in the Kitsuka to accommodate it. If
there were a parallel to the German
\le 2628-1a two-seat trainer, the rear fuel tank
n-ould have been removed to make room for
the instructor's cockpit. The German solution
to the loss of fuel was to utilise the two front
bomb racks for drop tanks. Whether Nakajima
considered the use of drop tanks (as the Kit-
suka could use them) or simply accepted the
reduced endurance for the sake of expediency
is not known. The two-seat trainer would be
the only variant of the Kitsuka to reach the pro-
duction phase.
It was planned that some of the two-seaters
were to be modified for reconnaissance roles.
The instructor's cockpit was to be removed
and replaced with a crew position for an
obseler. He was to have a Type 96 Modet 3
radio set at his disposal for use in relaying tar-
get information to other aircraft. It is unknon'n

InpeRral JapanssE Nnvv t21

Nakajima Kitsuka - data nons would be used for self-defence and for fir-
ing at the target before ramming the aircraft
Contemporari€s Messerschrnitt I'le 252A-1:ftJ3 and I'le 262A-5a Deplolment into the victim using any remaining fuel and
(Germanl), \'lessenchmitt 2tj2A-2aitl2 (Germanl). \'lesserschmitt None. One finished aircralt lras built bv Nakajima (hno if one counts the
ammunition as the secondary explosive ele-
Me 2628-la (Germanr'), l\'lesserschmilt Mc 262C-1 a. ['1e 262C-2b, airirame completed lor load testing) rvith a iudher 2l in various slages
ment. A variant of this Kitsuka was to utilise a
and l\'le 262C-3 (German.-v) ol conslnrction bv \\'als end.
200m (656ft) launch rail that Klgisho had been
designing and which they expected to have
T}?e Attack Bomber Sunivors
Crerv 0ne Nakajima Kitsuka (tail numbcr A.l 03)
ready for testing by September 1945. Using a
With the loss 0f lhe onh,completed Kitsuka follo[ing its crash on I I rocket booster, the Kitsuka would leave the
Powerplant TuoK[gish0Ne20arial-flosturbojets, August 1915, examples captured bl the tls lolloning the uativete from launch rail at 220kmih (137mph) at an acceler-
each deleloping.l90kq (1,080Ib) olthrun lhe stocks of incomplete Kitsuka aircrafl found in Nakajima and ation of between three to four'g'.
KvushI s plants. Th;s particular Kitsuka anived at ]iAS Patuxent Rii'er In regards to lhe 724 K6kutai, wiih the end of
Dimensions and on l8 0ctober 1946 was shipped to San Dicgo. California. The the war they would never see their Kitsuka air-
Span 10.00m 32.8ft aircraft erentuall.-v lound rts uat back to the Paul. E. Carber facilitl
craft. Ii was planned that by November 1945 the
5,21m I 7.2ft (rvings folded) nhere it is believed to be the second Kitsuka held in storage there,
unit would have been based near Yokosuka at
Length 9.23m 30.3ft Some sources hai'e the serial for this Kitsuka as 7337.
a site along the Miura Peninsula, west of Tokyo
Heighl 3.04m l0ft
Nakaiima Kitsuka (tail number A-l 04) Bay. It was expected that by then the unit would
\Ving area l3.l9m ll2ft-
Also at NAS Patuxent River u'ilh A-103, it u'as lalel shipped to NAS have received sixteen Kitsukas. In addition, the
\\iing loading 268.i3kgim' S5lbrlt'
\\iillou, Grove in \Villolv Crove, Pennsit'ania on 23 October 1946. unit was to use one of the handfuls of Kawan-
Power Ioading J.6kgrkg 3.7 lbllbst
Folloning its arriral, no iurther trace of the Kitsuka is knonn.
ishi E15K1 Shiun (meaning'Violet Cloud'; code-
Weights \akaiima Kitsuka (no tail numbcr assi{ned) named lVorm by the Allies) reconnaissance
Emptl 2,300k9 5,0i0.61b This Kitsuka was received in the I S and appearcd on a storage manifesl floatplanes, which were removed from active
Loaded 3,5i0kg 7,826.4Ib in 1950 bcing housed al NAS Norlolk in Noriolk, Virginia. 1n 1 960. it u,as service follorn'ing their disastrous combat debut
Mar loaded 4,3l2kg 9.506.3Ib shipped to the Paul E. Garber facilit\'. Fron 1972 until the lacility closed
in 1944. The Shiun, operating from a nearbyhar-
llseful load 1.249k9 2,i53.5 lb to the public. it rvas on displav hanlinl from the ceiling Gee page I 18).
bour, would locate the shipping targets, mark
them and then loiter in the area to broadcast
radio signals. The Kitsukas would then be
['1ar speed 62lkmih 386mph at sea level

6i9kmih 122mph
rapidly launched and, by means of the radio sig-
at 6,000m at 19,6851t
if any cameras were to be fitted but itwould not nals received through the Kuruku system,
695knrh 432mph be unreasonable to conclude ihat the obselver attack the ships at low level with bombs and
at 10.000m at i'l2,80llit would at least have had a hand-held camera. ramming tactics. Had the reconnaissance ver-
Mar speed r09kmih 3l6rnph An interceptor version of the Kitsuka was sion of the Kitsuka gone into production, the
at sea level, nith 500kg (l,l02lb) bomb discussed, as previously mentioned, and a 724 Kdkltai was to receive it as a replacement
speed i48kn'h 92mph
number of general arrangements for it were for the far more vulnerable Shiun.
at autight of3.95Okg (8,708.2Ib)
considered. One of these was the inclusion of Finally, with the close of the war, none of the
Landinq speed lsskmih 98mph
a single Type 5 30mm cannon with 50 rounds projected turbojet successors to the Ne 20
at a rveight ol2,57()kg (5.665.8Ib)
of ammunition installed in the nose. A second would enter production. One prototlpe of the
Take.olflength 501m 1.653.5ft

u,ith RATO u'ith zero nind at

design was to feature enlarged and extended Ishikawajima Ne 130 had been completed by
a n'eight ol 1,200m (9,259,1 lb) wings incorporating flaps and double-edged June 1945 but testing was unfinished by the
1,363m 1,{71,ift leading slots. A more definitive interceptor was time the war ended. Nakajima started devel-
lrithout RATO rvith zero rvind at to replace the Ne 20 engines with either Ne 130 opment of the Ne230 in May 1945 and had
a ueight oi 3,950kg (8,i08.2 lb) or Ne 330 turbojets. A second cannon was to be three under construction byAugust 1945. How-
Range 583km 362 miles added in the nose. Interestingly, it appears that ever, none of the engines were completed or
at 6,000m at 19,685tt if the IJA had used the Kitsuka, the Tlpe 5 can- tested. Mitsubishi was unable to construct a
888km 552 miles
nons would be replaced with two Ho-155 Ne 330 and so it remained on the design board.
at 10,000m atlJ2,808it
30mm cannons. This may have been a stopgap A note about the use of the name Kitsuka as
203km 126milcs
or fallback if the IJA's own Ki-201 Karyu failed opposed to the more commonly used Kikka.
\\'ith ma\imum bomb load
555km 345miles
to materialise. With the heavier weight the Kitsuka is the proper translation of the kanji
nith 551 1b bomb load structure of the Kitsuka, including the landing characters. However, it is pronounced'kikka'.
918km 589 miles at cruise speed gear, would have been strengthened. A fighter- Kikka was used in post-war reports as phonet-
Climb l2 min l6 sec to 6,000m (19,685ft) bomber model was envisioned for the defini- ically it approximated to Kitsuka and thus has
32 min 12 sec to 10,000m (32,808ir) tive interceptor by including a fitting for a single become the accepted name of the aircraft.
Ceiling 12,000m 39,370ft 500ke (1 ,102 lb) or 800kg (1 ,763 lb) bomb. Neither name is incorrect. Also, some sources
Senice ceiling 10,700m 35,l0llt As originally planned, a model of the Kitsuka use the J9Y1 (or sometimes J9N1) designation
Fuel capacit.-v i25 litres 191.5 gallons
lvas proposed for shimpu missions. Similar to for the Kitsuka. While logical for the intercep-
1,1i0 litres 383 gals u,ith drop lanks
the Kitsuka as constructed, this version was to tor version of the Kitsuka, there is no evidence
carry either a 500k9 (1,102 lb) bomb, a 250kg in wartime Japanese sources to support the
(551 lb) bomb or two Type 99 20mm cannons. designation. One may also find the designation
0ne 500k9 (1,102 lb) or 800 (i,i63 lb) bomb
With the latter, it could be assumed the can- J8N1 used but this is not supported.

122 JapeNssr SEcRor PRoLecls: ExpsRrMeNrnL ArncRnpr oF rHE IJA AND IJN 1939-1945
Other Aircraft

I tr

The Rammer Whereas the Japanese would simply use The aircraft depicted here sports the colours and
available aircraft to conduct ramming or markings of the 53rd Sentai operating in defence
of the Japanese homeland.
The practice of tai-atari, which literally means tai-atari attacks, the Germans took it a siep
'body crashing', was not unique to Japan. The further by producing and designing aircraft
deliberate ramming of one aircraft by another specific to the task. One operational example appeared. One of them was the Zeppelin
aircraft has happened as far back as World was the Focke-Wulf Fn' l90A-8/R7. To enable Rammer.
War 1 when Imperial Russian Air Force pilot the fighter to penetrate bomber formations The Rammer was a small, single seat air-
Pyotr Nesterov used his 1912 Morane- the Rammjiiger (or Sturmbocke) was fitted craft that was towed into the air. As it was
Saulnier monoplane to ram an Austrian Alba- with armour plating to enable it to weather designed for ramming, the constant cord
tross B.ll on 26 August 1914. In World War 2, defensive fire as well as possibly surviving main wings were strengthened by the use of
the first ramming attack went to Lt. Col. a successful ram. Sturmstaffel 1 was the three tubular spars along the forward edge of
Leopold Pamula who used his stricken PZL first Luftwaffe unit to operate the fighter. the wings. The remainder of the aircraft was
P.lic to down a Messerschmitt Bf 109 on 1 As the war went on and US bombers filled bolstered to allow it to withstand impact
September 1939. It would be the Russians, the skies, and the Luftwaffe was more forces. A single Schmidding 109-533 solid fuel
Germans and Japanese who would make and more hard pressed to stem the tide, a rocket developing 1,000kg (2,205Ib) of thrust
ramming a part of their war doctrine. number of dedicated rammer aircraft designs was fitted into the tail and the motorprovided

O lupR ArncRapr 123

a total oftwelve seconds of thrust. In the nose MXYT Oka which on such a small aircraft a target. Thiswould provide enough power to
were fourteen R4M rockets or a single SG 1 18 pushed the maximum speed to an estimated grant a second pass with sufficient speed to
battery. The cockpit was within an armoured 1 .1 25km/h (699mph or just over Mach 0.9 1) - allow for significant damage to be inflicted on
tub providing 28mm of armour plate in the i.e. a transonic speed. With a speed in the the bomber when it struck.
front and sides with 20mm in the rear. transonic range, this aircraft would have pre- However, the Japanese rammer would
The glazing was 80mm of bulletproof glass in sented a formidable challenge to the Japan- remain a paperproject only. It is unclear if the
the front and 40mm thick on the sides. The ese given that even the Germans had only iust design was to be the definitive rammer
method of attack was to tow the Rammer started investigating the problems of high- model or simplv a proposed concept.
within .48km (0.3 of a mile) from the target speed flight when their jet and rocket aircraft
and release it. The pilot would ignite the began to push into such speeds with the
rocket motor to boost speed and fire off the resultant issues of compressibility. lt is
nose weapon at the first target before making unknown if the Japanese rammer had swept
a ram attack on a second (or the same) tar- wings because the designers understood the
get. After engaging in the ram attack, the pilot principles in relation to overcoming com-
would glide back down to earth. The Ram- pressibility problems at transonic speeds, or if
mer was test flown, without power, in Janu- the shape was chosen as a means to provide
ary 1945 and a pre-production batch of an angled cutting surface to facilitate ram-
sixteen aircraft was ordered. However, Zep- ming attacks, or as a drag reducing planform.
pelin's production facilities were destroyed The wings were strengthened to withstand
by US bombing before their construction the high impact forces experienced when
could commence. striking the enemy bomber. Even though the
The Japanese would use aircraft already in rammer could rely on speed as a defence
operational seruice for ramming attacks such when under power, it still had to contend
as that Kawasaki Ki-45 and even stripped with the defensive armament of the B-29 and
down Kawasaki Ki-61 Hein fighters. It was thought the pilot had some measure of
long thought that Japan never developed a armour plating and bulletproof glass to pro-
dedicated rammer aircraft of its own but this tect him. The aircraft was certainly capable of
is no longer the case. Recently discovered in gliding back to base to be refuelled and
the archives of the Japanese National Insti- relaunched once it had conducted its attacks. The Rammer - data
tute for Defence Studies is just such a project. Given the small size of the plane, no landing
The aircraft was a joint venture between gear was fitted. As such, it is likely the under- As lhe design uls neaer buill, the specifications arc estimales based on
the IJA and the IJN, something that occurred side of the fuselage was reinforced or had a the oiginol design sketch and dalI.
with more regularity towards the close of the skid installed. How it was to be launched is
war. The design was based on the Syusui- unknown - it could have been towed aloft, Tlpe Rammer

shiki Kayaku Rocketto (meaning 'Autumn catapult launched or perhaps even vertically Crew One

Water'-type ram attack rocket), a project launched.

Powerplant Four Trpe 4 N'lark I l\'lodel 20 rockets rvith a
started in March 1945 for a unmanned, In a ram attack, typically the tail would be
combined l, l02kg (2,232 lb) of thrust
remote controlled anti-bomber missile. The targeted because the loss of the tail assembly
plan was to ground launch the missile, guide would send the bomber out of control. Strik- Dimensions
it remotely towards the target, engage the tar- ing the wings and engines was another focus Span 4.41m l4.5ft
get via ramming, and then recover the missile of ramming attacks. Finally, the aircraft fuse- Length 2.89m 9.5ft

(if it suruived the collision) for reuse. Design lage was the other key area to strike. The Height N/A

work was carried out by the Kokukyoko (the probable mission profile of the rammer flying \Ving area N/A

Aeronautical Bureau) and, although a mock- from a ground base would include being Wine loading Ni/A

up was completed, the war ended before positioned within very close proximity of Polver loading

finalised production plans could be com- likely bombing targets. With the short burn
pleted, let alone the missile ever being tested. time of the rockets (8-10 seconds) the air-
Emptv N/A
The missile's design borrowed heavily from craft's operational radius would have been Loaded N/A
the Mitsubishi J8M Sylsui in terms of its very limited. After launching, as bombers \'larimum N1A

shape. Interestingly, the Messerschmitt Enz- came into range the pilot would attempt to
ian anti-aircraft missile had a shape influ- ram into either the tail or wing of the target Performance
enced by the company's Mel63 rocket with the ob.iective of severing it from the fuse- Speed l,l25km/h 699mph

interceptor, of which the J8M was the Japan- lage. If enough speed momentum remained Climb 32 seconds to 10,000m (32,808ft)

ese version (see page 96). after the initial hit, another ram attack would Range N1A

The piloted version used much the same be made. Should the aircraft remain in flyable Endurance N/A

design as the missile and was a small, tailless condition and if the pilot did not elect to ram Ceiling

aircraft featuring low mounted 45' swept his entire plane into a target, he would return
wings. The fuselage was bullet shaped with a to base where the rockets would be replaced.
large vertical stabiliser into which the cockpit If the bombers were still close by, he could fly
was blended. Located in the back of the fuse- another sortie. If the rammer was towed into Deployment
lage were four Type 4 Mark 1 Model 20 rock- the air, the rockets would most likely have None. The rammer remained a design dralt only

ets, the same as those used on the Kugisho been fired on approach and again after hitting

124 JapanEss SEcnsr PRo:scrs: ExpnRrupnraL AIncRapr oF THE IJA,qto IJN 1939-1945
The Kamikaze Airplane

Kamikaze, which in Japanese means 'divine their descriptions that the illustration of the VMF(N) 531 while the second was a photo
wind', was not a term used by the Japanese aircraft was created. The result was certainly reconnaissance unit. With the Tigercat being
to reference their special attack units but unlike any plane then in the theatre but there new to the Pacific Theatre, it can be surmised
came into common use in the West. Instead, is some precedent in terms of genuine air- that some sailors and crew mistook the twin-
shimpl and shimbu were more often used by craft being misidentified. One example was engine fighter for a Japanese plane and sub-
both the IJN and the IJA respectively. May the'Kawasaki Type 97 Medium Bomber' that sequently described something other than
1944 would signal the beginning of dedicated was given the codename Julia. Because of what was actually seen. However, the
suicide attack missions against the US and very inaccurate illustrations of the plane that Kamikaze Airplane was written off as a spec-
her allies. The effect of being on the receiving were derived from awritten description,Julta tre of the imaginations of sailors who had
end of such attacks was horrific and would was in fact the Kawasaki Ki-48 (frly). It may borne the brunt of kamikaze attacks.
take its toll on US sailors. very well be that the Kamikaze Airplane was
A poslwar report by the US Strategic actually a US Grumman F7F-2N Tigercat of
The Kamikaze Airplane illustrated is derived from
Bombing Survey bluntly stated that kamikaze which two US Marine Corps squadrons the magazine artwork and is shown in the colours
attacks were effective and given the situation equipped with the plane began operating of the l8th Sentai, operating from Kashiwa Airfield'
for the Japanese, very practical. Statistics of from Okinawa in September 1945. One was Japan, spring 1945.
sunk and damaged US ships during the
Philippines campaign (October 1944 to Janu-
ary 1945) showed that kamikaze attacks were
farmore effectual. Of course, being subjected
to such attacks wreaked havoc on morale as
rvell as a surge in mental illness. The US Naly,
the focus of the majority of kamikaze attacks,
saw such illness rise by 50 per cent in 1944
rvhen compared to 1941. So bad was the
problem that during the Okinawa campaign,
the US Naly stopped warning crews that
kamikaze attacks were to be expected as
they only added to the sailor's stress levels.
General George C. Marshall wrote in a 1945
report to the US Secretary of War that, 'The
American soldier has a very active imagina-
tion... and is inclined to endow the death-
dealing weapons of the enemy with
extraordinary qualities...' And thus the
'Kamikaze Airplane'appears on the scene.
Following the close of World War 2, an illus-
tration of an aircraft appeared in either Popu-
Iar Mechanics or Popular Science magazine.
The aircraft, labelled a kamikaze plane, was a
curious mixture of what appeared to be the
tail of a Mitsubishi A6M Reisen, a fuselage not
too unlike the Nakajima Kitsuka and a canopy
similar to aircraft such as the Kawanishi N1Kl
Kyofu (meaning 'Mighty Wind' but known as
Rex to the Allies) or the Nakajima Ki-84 Hay-
ate (meaning 'Gale'; Franh to the Allies). Even
more curious was the fact that the aircraft
used air-cooled radial engines, one in each
rving, but in a pusher configuration.
The Kamikaze Airplane was said to have
been sighted by some US Naly crew mem-
bers as it flew over their ships. It was from

The Kamikaze Airplane - data

.\io specilications uere proDided fot the atct1ft.

\one. The Kamikaze Aircrafl $'as strictlv fictional,

OrHER Arncnnpr 125

Mitsubishi T.K.4 Tlpe 0



The illustration provided here is an interpretation At the outset of hostilities in the Pacific, Amer- Area, would assign the T.K.4 the codename
of the Tlpe 0 as described in the O.N.I. 249 manual ican intelligence had very little information on Franh, Iaking his ornrn first name. It was
and is shown in the colours and markings of the
3rd Chutai, 26th Sentai. .just exactly what aircraft the Japanese were McCoy who arrived at the method of assign-
fielding. In part this lack of knowledge sprang ing names to Japanese aircraft in order to
from poor intelligence management and cen- simplify identification.
sorship of periodicals and other publications At some point, the T.K.4 Type 0 fighter took
by the Japanese authorities. In scrambling to on a completely different appearance. When
document Japanese aviation, invariably US the Japanese Aircraft Manual, O.N.l. 249
intelligence officers turned to Japanese mag- (Office of Naval Intelligence), was first pub-
azines as a means of gaining information. lished in December 1942, the Mitsubishi Type
However, there were pitfalls to using such 0 was no longer called the T.K.4. Although no
sources and the Mitsubishi T.K.4 Type 0 was illustration was provided in the manual, the
just one example. Type 0 was described as an army fighter
The T.K.4 appeared in a section of the based on the Dutch Fokker D.)filll. The
Japanese aviation magazine Sora entitled D.XXIII was a twin-engine fighter that
'Dreams of Future Designers'. The issue was mounted the engines in the fuselage in a
published in April 1941. The T.K.4 was push-pull configuration. It was also a twin-
depicted as a twin-engine fighter whose boom design that was under development
design was rather similar to the German and in flight-testing until the German inva-
Messerschmitt Bf 1 10. The aircraft, although a sion. The manual stated that the Type 0 used
fighter, was shown with a glazed nose along two German BMW engines, each developing
with the expected glazing over the pilot and 750hp, but that a redesign of the aircraft
crew positions. Each of the low mounted would see it using two 1,000hp Mitsubishi
wings sported an inline engine in a very Kinsei air-cooled radial engines. No further
streamlined cowling, each motor driving a information was made available.
three-bladed propeller. What weapons the When the original T.K.4 Frank failed to
T.K.4 carried was unknown nor was the crew materialise in combat, McCoy removed his
compliment listed, although two or three name from the T.K.4 and reassigned it to the
could be estimated. Also lacking was any Nakajima Ki-84 Hayate (meaning 'Gale')
data on the performance of the T.K.4. which was a fighter very much in use, being
lnformation on the T.K.4 would also appear first encountered in combat in early 1944. The
in a US magazine. The 25 December 1941 Type 0, as described in the O.N.l. 249 manual,
Mitsubishi T.K.4 Type issue of F/rghf mentioned the aircraft as a was given the codename Harry after Colonel
0 - data
twin-engine, twin-tail monoplane fighter. Harry Cunningham, a friend of McCoy's, who
No informalion on the specifications, if ony et:er existed, is auailable for From these sources, the US intelligence was the intelligence officer for General Ennis
the T.K.4or the TWe 0. determined that the T.K.4 was a bona fide Whitehead. At the time it was believed that
fighter that would be encountered in combat. the Type 0 would eventually be seen in action
None. The T.K.4 existed only as an illustration in a magazine while the
Maior Frank T. McCoy, Jr., the head of the but, just like the T.K.4, it never would and the
Type 0 was a descdption in a manual.
Material Section of the Directorate of Intelli- Type 0 was dropped from the intelligence
gence, Allied Air Forces, Southwest Pacific publications.

126 SEcnEr Pnotects: ExpnnrMenral ArRcRnpr oF rHE IJA AND IJN 1939-1945
Nakajima AT27

The T.K.4 was not the only Japanese aircraft was afforded some protection by the engines NakajimaAT2T - data
uncovered in the pages of Sora. The same in front and behind him but the AT27 could
Contemporaries Republic (l.S), Fisher XP-i5 Eagle (US)
'Dreams of Future Designers' article in which also carry additional armour not only for the XP-72

that image was unearthed also included the pilot but for the engines as well. What tlpe of
Tlpe Fight€r
Nakajima AT27. Several months later, the 25 weapons the AT27 was to carry were not
Crew 0ne
December 1941 issue of Flight, a US maga- known.
zine, would also feature these planes, along Based on its appearance in the magazines. Porverplant Tnol2-c1'linder,water"cooledV-engineseach
rvith several others, lending credence to the the AT27 was believed to be a bona fide deleloping l,2i0hp and each driving a three-bladed propelleL
idea that they were genuine aircraft in use by fighter that could be encountered and thus
the Japanese. Allied intelligence gave the AT27 the code- Dimensions

The AT27 was novel in a number of name Gus. However, the AT27 would never Span

rvays. On the outside, the fuselage was be seen or met in battle since, as was later Length \,A
sleek and well streamlined. The wings were discovered, the aircraft was flctitious. Gus Height \1A

low-mounted with a conventional tailplane. was soon dropped from the Japanese aircraft Wing area 22.0lmr 2371t

Wing loading 239.23k$m' .l9lb/ft

Inside, however, the AT27 featured two intelligence rolls.
Porver loading NlA
i 2-cylinder inline engines each rated at Interestingly, the AT27 was very similar to
1,250hp and was reported to obtain a mari- the very real Kawasaki Ki-64 whose develop- Weights
mum speed of 660km/h (4i0mph). One ment began in October 1940. Both used 12- Emptv 3,629kg 8,000 tb
engine was in the nose while the other cylinder inline engines, one in the nose and Loaded 5,262k9 I l,600lb
rvas situated behind the cockpit. Contra- the other behind the cockpit, driving contra- Usable load l,633kg 3,600 1b

rotating propellers were used, the rear engine rotating propellers. In addition, both used a
driving its propeller via an extension shaft. surface evaporation system. Perhaps by Performance

To maintain the excellent aerodynamic prop- sheer coincidence, one ofthe Japanese illus- ['lili speed 660kn'h 1l0mph

erties, the engines were reported to have trations of the AT27 that was published in Cruise speed N/A

Range 2,0l2km 1,250 rniles

been provided with a'steam cooling'system. Soro and later in Flrghl showed it sporting the
Endurance N/A
This may have been a surface evaporation number 64 on the fuselage.
Climb NlA
system. Such a system took the steam Ceiling NlA
created after the water had passed through
the engine and ran it through piping in the The 3-view illustration of the AT27 is in the
markings and colouration of an aircraft of the
rvings where the cooler airflorv would con-
244th Sentai operating in the defence of Tarkyq
dense the steam back into water that was 1944-1945. The side view below depicts the AT27 Deplolment \one. The AT27 was purell a fictional aircralt

cycled back through the engine. The pilot as it appeared in Sora magazine."



OrHeR AIRcRapr t27


I li-



F _@

128 JapaNgse SEcRrr PRo.lscrs: ExpeRruENlal Arncnarr op rHE IJA aruo IJN 1939-1945
S-31 Kurowashi

As we have already seen, the April 1 941 issue With a wingspan of just over 33m (107ft),
of Sora misled Allied intelligence over the the Kurowashi was by no means a small air-
nature and extent ofJapanese aircraft design. craft. The plane was just under 21m (70ft) in
As the war continued, none of the four aircraft length and a height iust shy of 6m (20ft).
that feature in the April issue were encoun- These dimensions were very similar to the
tered and subsequently dropped from Allied Boeing B-17 bomber. The Kurowashi sported
intelligence publications. However, Soro con- horizontal stabilisers which ended in ovoid
tinued and so did the section responsible. shaped vertical stabilisers.
How often the 'Dreams of Future Designers' The Kurowashi was certainly not lacking
portion of the magazine appeared is not forweaponry. A total of eight 7.7mm machine
known but one issue from either I 944 or 1945 guns and four 23mm cannons were carried
contained a design that was nothing short of by the bomber. 7.7mm was a calibre used by
fantastic. This was the S-31 Kurowashi, or both the IJA and the IJN, but on the other
Black Eagle. hand, the Japanese did not field a 23mm can-
The Kurowashi was a four-engine hear,y non in any form, either in aircraft or on the
bomber concept. What was unique about the ground. The IJN did use a 25mm anti-aircraft
aircraft was that all of the engines were cannon (Type 96) but did not apply the
housed within the fuselage and the weapon to aircraft. The IJA also experi-
Kurowashi used a push-pull configuration. mented with a 25mm aircraft cannon but
Both in the front and the rear of the fuselage abandoned it in favour of the 30mm cartridge.
were two 2,500hp, 24-cylinder, liquid cooled, Why the creator of the Kurowashi decided to
inline X-engines, driving a pair of contra-rotat- use 23mm as the calibre for the cannons
ing, three-bladed propellers via a gearbox. remains unknown.
With this powerplant the Kurowashi was to What is known is the novel arrangement of
boast a top speed of 689km/h (428mph), but the defensive armament. Fitted directly into
such a powerful engine of this type would not the leading edge of each wing were two ball
see service with the Japanese air forces. trlrrets. The outer turret contained one 23mm
However, this was not for a lack of trying: it cannon while the inner turret sported two
may or may not be that the originator of the 7.7mm machine guns. Directly opposite
Kurowashi was aware of the Yokosuka YE3 these front-facing turrets was another set of
series of engines. ball turrets. As the trailing edge of the wing
In 1940, the IJN initiated development of was too thin to allow the turrets to be inter-
the a 24-cylinder, liquid-cooled,
YE3A, nally mounted, each turret was fitted into the
X-engine that was to produce 2,500hp. An end ofa nacelle that extended from the back
X-engine is produced by having paired V- of the wing. Therefore, each wing was fitted
block engines horizontally opposed to each with four turrets for a total of two 23mm can-
other with the cylinders in four banks driving nons and four 7.7mm machine guns. To con-
a common crankshaft and thus, when trol these turrets the Kurowashi relied on two
viewed head-on, appearing as a 'X'. The gunners, the bombardier and co-pilot. Both
major benefit to using such a configuration is gunners had positions facing to the rear of the
that the engine is more compact than a com- aircraft behind the bomb bay. The first gunner
parable radial engine or standard V-engine. station was in the upper portion of the fuse-
However, X-engines are far more complex lage while the second was in a ventral station.
to construct and service and are heavier. It Weapon sights were provided along with the
was not be until October 1943 that the first controls to manipulate and fire the turrets.
YE3B [Ha-74 Model 01] (also known as the The bombardier and the co-pilot stations
Ken No.l) was completed and tested. The were also provided with a sight and turret
YE3B was designed to be housed inside the controls so that if they were not occupied
wing. A second model, the YE3E [Ha-74 with other duties they could man the
Model l1) (Ken
No.2), was rated at 3,200hp weapons. It is likely that the bombardier and
and was slated to be completed in March co-pilot had control of the forward facing tur-
1944. Unlike the YE3B, the YE3E was rets while the two gunners maintained con-
designed to be housed within the fuselage. As trol over the rear facing weapons.
it was neither engine would see service by the For its war load the Kurowashi could carry
time the war ended. Interestingly, a surviving just over 7,257kg (8 tons) of bombs, about
YE3B engine was fitted with a two-stage 1,814kh (2 tons) less than the Boeing B-29.
reduction gear and a extension shaft that The bomb baywas divided into two and each
would have been used to drive two, contra- section could hold six bombs to give a total of
rotating propellers. twelve. Beneath the main bomb racks were

OlleR AtRcnapr 129

hinged panels, one per side. Each panel held S-31 Kurowashi - data
four bombs for total of eight. When the bomb
bay doors opened, the bombs suspended Contemporaries
Daimler-Benz Schnellbomber mit DB P83 0ruppenmotor (Cerman,v)
from the panels would be released and the
panels swing aside so the remainder of the
Type Healy Bomber
bombs could drop. This arrangement was
Crew Five
created to maximise the payload space avail-
able. Situated directly above the bomb bay Powerplant Four24-c,vlinder.liquid-cooledX-engines,each
were fuel tanks and it was likely the wings deleloping 2,500hp, each pair driung [vo, metal 3-b]aded contra-
also housed fuel. rotating propellers

The Kurowashi used a tricycle landing gear

system with the nose wheel retracting into Dimensions

the fuselage while the main landing gear Span 32.82m I 07,7fr

Length 21.09m 69.2fr

went up into the wings. However, because
Heighl 5.88m I 9.3ft
of the healy tail and to prevent damage to
Wing area 1.33.00m' 1.431.6ft,
the rear propellers while on the ground or
during take-off and landing, a large,
retractable tail wheelwas fitted to the back of Loaded i 7,850k9 39,352 tb
the fuselage.
For its crew the bomber had five men: pilot, Performance
co-pilot, bombardier and two gunners. One ['lai speed 690knyh 429mph

of the gunners served as the radio operator as Cruise speed ;89knth 366mph

the radio station was situated in front of the Landing speed 145knh 9Omph

5,900km 3,666 miles

upper gunner's position. Range

Ceiling 15,100m 19,540ft

It may very well be that Allied intelligence
FuelWeight 8,000kg 17,636Ib
was aware of this design and it was also likely
that by 1944-1945, intelligence officers were Armament
no longer taking aircraft illustrated in the Eight 7.7mm machine guns and four 23mm cannons (see text for
'Dreams of Future Designers' section in Sora anangemenl); up to 8,000kg (l 7,636 lb) of bombs
The Kurowashi depicted here is in the markings of
magazine at face value. The Kurowashiwas a the 2nd Chutai, 62nd Sentai. Historically, this unit
creation that would have been very difficult to Deployment None, The S-31 Kurottashi was purell a paper, ii operated Mitsubishi Ki-21 (SaIlyJ, Nakajima Ki-49
execute in reality and may not even have not impractical, design in a magazine, Donryt (Helen) and Mitsubishi K-67 Hiryn(Pegry)
been feasible.


130 JnpaNEss Secnor PRorecrs: ExponnrnNtaL AIRcnarr oF rHE IJA AND IJN 1939-1945



\q .l

It was difficult for the US and her Allies to applications for their charges. Each wing of The T,K,l9 depicted here is shown in the colours of
acquire intelligence about the Japanese war the T.K.l9 showed what appeared to be three the 77th Hiko Sentai during operations in Burma,
industry as far as HUMINT (HUMan INTelli- weapon ports, totalling six machine guns or t94t-1942.
gence) was concerned. This was due to the cannons. There were also ducts in each of the
relative difficulty associated with either turn- wing roots, ostensibly to cool the engine. A in many aircraft), moving the wings higher up
ing a Japanese source or inserting a foreign radiator bath may have been located in the the fuselage, adding pronounced wing root
spy into Japan who was capable of avoiding nose of the aircraft. Given the flush canopy, fairings that extended from the nose of the
detection. Once it was broken, the PURPLE the T.K.19 may have used a system similar to aircraft to rear of the cockpit (the latter being
code (as used by Japanese foreign offices) the Stal'-6 in which for take-off and landing set behind the wings) and having a main
and the laterJN-25 code (as itwas labelled by the canopy hood was hinged upwards and landing gear reminiscent of the Brewster F2A
the US) used by the IJN provided a wealth of the pilot would raise his seat. Whether the but with landing gear doors. No weapons
information and intelligence, but human T.K.19 used a periscope vision system for were shown but air intakes were illustrated in
intelligence and cipher cracking were only a the pilot once in flight as was proposed in the the wing roots.
part of the overall processes. One avenue Soviet Lavochkin LL fighter was not known. From reviewing the information, US intelli-
used prior to the war was the gathering of Like many of the other aircraft in this sec- gence made the assumption that the T.K.l9
publications such as books and magazines. tion, the T.K.19 would later appear in the was a bona fide Japanese fighter that was in
Besides being relatively innocuous to pur- American magazine Flight in the 25 Decem- service or was soon to be in service. Thus, it
chase in Japan, such sources could be ber 1941 issue. The description made no was codenamed Joe after Corporal Joe Grat-
obtained outside of Japan and were therefore mention of the more striking features of the tan, one of the team members responsible for
easier to gather. It was one such publication plane as described and shown in the Japan- assigning codenames to Japanese aircraft.
issued just prior to the start of hostilities that ese magazine. Instead, the article, which con- The T.K.19 failed to appeare in Japanese
revealed the T.K.19 to intelligence officers. tained no illustration, reported the T.K.19 was sources despite remaining in US intelligence
The illustration of the T.K.l9 appeared in of orthodox appearance save that the aircraft bulletins. It eventually became clear that the
the April 1941 issue of the Japanese aviation had a twin row radial engine in the rear of the T.K.l9 was nothing more than a fictional air-
magazine Sora in a section entitled 'Dreams fuselage and was cooled via ducts. From this, craft and Joe was removed from future intel-
of Future Designers'. The T.K.19 showed a a drawing evolved that took the basic shape ligence publications.
fuselage that was elongated and ovoid in of the Japanese T.K. I 9 and made it more con-
shape. More interestingly, it showed a canopy ventional, to the point that it bore a slight
that could be lowered to fit flush with the top resemblance to the Curtiss P-40 Warhawk, a T.Kl9 - data
of the fuselage thereby eliminating the drag of plane that first flew in 1938 and one that
a standard canopy. This same concept was Japanese pilots first encountered in combat No infonnation, if ony, on the specif[cations arc auailable for the T,K,]9
seen in the Russian Bisnovat SK and Bartini in late 1941. The changes from the Japanese
Stal'-6, both of which were high-speed air- T.K.l9 included doing away with the flush
None. The T.K,19 existed only as an illustration in a magazine.
craft whose designers were seeking military canopy (providing a more standard style seen

OrHrR Arncnanr 131

Weapon Systems
and the JB series of missiles. Other Allies, (2nd Bureau of Gunpowder) and the Kure
Japanese Missile and
such as the British and the Russians, would Kosho (Kure Arsenal). Ultimately, it would be
Guided Munitions Projects not spend nearly as much resources on the Klgisho that was given the Funryu project in
The aim of tactical missiles, specifically subiect as did the Germans and Americans. early 1944. To accomplish the task, Ktlgisho
guided munitions, is to increase accuracy. It The British would squander the potential of formed the Funshin Kenkyu-Bu (Rocket
takes a considerable amount of conventional the Brakemine surface-to-air missile and stall Research Bureau) and was staffed with up to
bombs or torpedoes to strike a ship and inflict the Fairey Stooge while the Russians would 200 technicians led by a research team made
enough damage to cripple or sink the vessel. only test and reject the promising Korolev up of40 officers (all engineers and/or techni-
Likewise, anti-aircraft cannons have to put a Type 212A (built in 1937), waiting until the cians) from the IJN. In all, Krlgishd would
significant amount of shells into the air to close of World War 2 to revive its missile investigate and put forvrard four Funrlm
bring down a single plane. Another benefit of development work. In some cases the Soviets designs.
using missiles is the measure of protection used the fruits of German labour as their The first was the Funryu 1 and the design
afforded to the user by way of range. A fighter basis, for example, developing the R-1/SS-I was an airto-surface missile (ASM) whose
combating bombers has to attack at such a Scunner from the V-2 missile and the Type specific role was anti-shipping. Funryu 1 was
range that his weapons are effective and 1OCh from the V-1 flying bomb. much like a miniature airplane. The warhead
therefore within range of the defensive arma- An example of the greater accuracy of mis- contained 882Ib of explosive and guidance
ment of the target. The fighter also has to con- siles and guided munitions can be seen in the was via radio control. Testing of the Funryu I
tend with escorting fighters before he even 27 December 27 1944 mission flown by the US was conducted with the missile being
has a chance to press home an attack on the to attack the Pyinmana rail bridge in Burma. dropped from a modified Mitsubishi G4M
bomber. The same is true of attacking ships. Nine VB-1 AZON guided bombs were enough bomber. However, it was seen that the
To improve accuracy, a torpedo or dive- to destroy a bridge that for two years previ- means to effectively control the missile in
bomber has to be close enough to the ship to ously had failed to be hit by thousands of con- flight would require a significant amount of
ensure a hit. Of course, this also puts the air- ventional bombs. Likewise, the Germans time to perfect and with the increase in US
craft in the uncomfortable position of being were able to successfully attack shipping tar- bombing raids against Japan, it was decided
within range of the many anti-aircraft can- gets using the Henschel Hs 2934 and Fritz-X that efforts should be directed towards sur-
nons and machine guns carried by the ship, using less aircraft and with a higher hit and kill faceto-air missiles (SAMs). Thus, the Funryu
as well as attack by fighters providing cover ratio than if the same attacks had been made I was shelved and was to be the only ASM of
for the vessel. Guided munitions eliminate using conventional bombs and torpedoes. the Funryu family.
some or all of these problems. With these benefits in mind, it is not sur- The Funryr-r 2 was to be a SAM built around
Without doubt, the undisputed leader in prising that Japan also devoted considerable a solid fuel rocket using a radio guidance
World War 2 missile development was Ger- effort to producing such weapons themselves system. Despite three rocket motors being
many. Missiles such as the Fieseler Fi 103 (the (while Japan did receive some German mis- available already (the Ro-Tsu, Ro-Sa and
V-1), EMW A4 (better known as the V-2), sile technology, it is unknown how much of it Ro-Ta), it was decided that a new motor was
Ruhrstahl-Kramer X-l Fritz X and the Hen- found its way into the IJA and IJN missile pro- required. This motor was capable of produc-
schel Hs 293A were used operationally with a grams). Both the IJA and IJN funded the ing 2,400kg (5,291 lb) of thrust during its 3.5
measure of success. This was iust the tip of development of missiles as a means to both second burn time. The shape of the missile
the iceberg. Many more designs came close combat the bombers that tormented the was relatively simple. Four wooden wings
to seeing seruice or were in the latter stages homeland and to attack Allied shipping. were fitted to the body of the missile and each
of testing at the war's end. Such weapons was equipped with an elevon (elevons con-
included the EMW C2 Wasserfall, Rhein- trol both pitch and roll). Four fins, making up
metall-Borsig Rheintochter, Henschel Hs I 17 The Funryu (IJN) the tail, were fitted to the outside of the noz-
Schmetterling, Ruhrsahl-Kramer X-4 and The Funryu ('Raging Dragon') was the name zle for the motor. This final shape and config-
many more. The US was not lacking in missile given to the IJN's missile program that com- uration was a result of numerous tests of
and guided munition technology of its own. menced in 1943. The initial study for the Fun- various missile bodies and wing/fin arrange-
Operational weapons included the ASM-N-2 ryu was conducted by the Kaigun Gijyutsu ments in a wind tunnel. The radio guidance
Bat, GB-l/GB-4 and the VB-1 AZON (Mimuth Kenkyujyo (Nary Technology Laboratory) but system was initially to consist of a single
ONly). Proiects included 'Little Joe' (intended three other groups would review the study transmitter but a second was fitted to ensure
as a ship-borne missile to combat soon afterwards and thev were the Dai-lchi a measure of accuracy. The first transmitter
kamikazes), the McDonnell LBD-1 Gargoyle Kaigun KOku Giiutsu-shd, Dai-Ni Kayaku-Sho was used for target detection while the sec-

132 hpeNEsE SEcnEr Pnorscrs: ExprRruElrrnl Alncnepr or rHs IJA aNo IJN 1939-1945
ond would control and steer the missile to the a facility in the lzu Peninsula, west of Toky6, The I-Go (IJA)
target. To maintain attitude, the Funryu 2 con- The Funryu 4 was to be built around the Toko While the IJN put its focus on SAMs, the IJA's
tained two gyrocompasses and 50kg ( 1 1 0 lb) Ro.2 (KR10) rocket engine, the very same resources went into developing Air-to-Sur-
of explosive was housed in the nose. The engine used in the Mitsubishi J8M Slusui face missiles (ASMs). The culmination of
basic operation consisted of the Funryu 2 rocket fighter. As testing of the Toko Ro.2 was these developments, begun in 1942 by the
being launched from a rail set at an angle of already underway and would soon be put Koku Hombu, was the I-Go series of missiles.
80". Once launched, radio receivers fitted in into production, Mitsubishi could devote less The majority of the research on the I-Go was
the wings would receive signals from the time to engine concerns. The Funr1,,u 4 was to carried out by Rikugun Kokugijutsu Kenkyujo
ground transmitter, steering the missile onto use a mixture of the Ko fuels (concentrated located in Tachikawa. Once the preliminary
the target. hydrogen peroxide) and Otsu (hydrazine work for the missiles was completed, the
The war situation in 1944 was starving hydrate solution in methyl alcohol) as used by Koku Hombu reached out to Mitsubishi,
Japan of critical war materials and Krlgisho the J8M. The engine would provide up to Kawasaki and the Aeronautical Research
found they were unable to obtain the neces- 1 ,500kg (3,307 lb) of thrust and move the Fun- Institute of Tdky6 University to commence
sary quantity of duralumin to build the Funrlu ryu 4 to a maximum speed of 1,099km/h final development of the I-Go as they saw fit,
2 prototypes. It took theft from a warehouse (683mph). using the initial data assembled by Rikugun.
to obtain the required metals. Using the The guidance system selected for the Fun- Sumitomo Communication Industry Co. Ltd.
absconded material, a number of Funryu 2 rlr.r 4 was far more sophisticated than that was the provider of the autopilot and the
missiles were constructed with one being used in the Funryu 2. Whereas the latter transmitter/receiver system for the first two I-
used for continued wind tunnel testing and relied on radio, the Funryu 4 would use radar. Go missiles with T. Hayashi designing the for-
the remainder used for actual field tests, the Two stations would be used to deliver the mer and K. Nagamori the latter.
latter being conducted near Mount Asama Funryu 4 to the target. One station would The I-Go-1-A (Ki-l47) was the Mitsubishi
(located near Ueda). Testing commenced in track the target while the second would track version of the I-Go. The final design of the
the spring of 1945. The first launches of the and control the missile. The intention was Ki-147 was completed by the end of 1943.
Funryu 2 were unguided, conducted solely to that the two radar signals would coincide on Work began on the missile using a basic air-
evaluate the rocket motor performance and the target, thus bringing the missile to impact. plane configuration and its construction was
the general flight characteristics of the mis- To control the missile, a radio signal of made of wood and metal. It was propelled by
sile. In July 1945, the first test of the Funryu 2 1,000MH2 was to be used with five frequen- a rocket engine built by Nissan Jid6sha KK
n'as undertaken using the radio guidance sys- cies. Each frequency corresponded to con- which produced 240kg (529 lb) of thrust with
tem. With IJN personnel in attendance, the trolling the pitch and the roll with the fifth a burn time of 75 seconds, providing a top
Funryu 2 successfully lifted off from the being the detonation command. A variation speed of 550kg (342mph). The warhead was
launch rail and was directed towards a of this system is used today known at retrans- substantial using 800kg (1,764 lb) ofexplosive
ground target. Using the radio signals, the mission homing or Track-via-Missile (WM). triggered by an impact fuse. Guidance was by
Funryu 2 was guided to within 20m (65ft) of Funryu 4, like the Funryu 2, used two gJ,ro- radio from the carrying aircraft. The first
the target when impact was made. Although compasses and carried wing radio receivers Ki-I47 missiles rnere completed in 1944 and
a direct hit was not achieved, the test was for the commands sent to it from the ground. by mid-year unguided test drops had com-
considered a success. It was to be, however, It carried a far heavier warhead of 200kg menced at Ajigaura, Atami and Shiruishi. The
the last flight of the Funryu 2 because the war (4401b) in comparison to the 50ke (1101b) carrier aircraft lvas a modified Mitsubishi
ended before any further launches could be warhead of the Funryu 2. The shape of the Ki-67-l HiryU bomber. By October 1944,
made. Funryu 4 was also more streamlined and it guided test drops of the Ki-l47 had begun.
The Funryu 2 was 2.2m (7.2ft) long, .28m only had two of the elevon-equipped wings Despite the testing, the Ki-147 did not enter
(0.9ft) in diameter and had a span of .88m and two tail fins. Launch would occur from a production and only fifteen were built. The
(2.9ft). Total launch weight of the missile was rail set at a 45'angle. Ki-l47 had a length of 5.8m (18.9ft), a span of
3i0kg (816Ib) and its maximum ceiling was The Funryu 4 was 4.0m (13.1ft) long, .6m 3.6m (1 1.8ft) and a launch weight of I,400kg
5.000m (16,404ft). The maximum speed of (1.9ft) in diameter and its span was approxi- (3,086 rb).
the missile at full burn was 845km/h mately.8m (2.5ft). Fullyloaded itsweightwas The I-Go-1-B (Ki-l48) was the Kawasaki I-
(525mph). 1,900kg (4,1891b), range 30km (18.6 miles) Go. Smallerthan the Ki-l47, the Ki-l48 used a
Even as the Funryu 2 was being investi- and ceiling of 15,000m/49,215f1. HTP rocket motor that developed 150kg
gated, a variant of the missile, the Funryu 3, Nagasaki Arsenal was tasked with building (331 lb) of thrustr,vith an 80 second burn time.
rvas proposed using a liquid fuel rocket in the Funryu 4 and this did not begin until the The wings were constructed of wood while
place of the solid fuel motor. However, initial late summer of 1945. The first ground test of the body and fins were made from tin. As a
discussions on the new rocket engine led to the missile and its motor commenced on 16 consequence of the smaller size, the war-
the conclusion that there was no time or August, but the close of the war prevented the head comprised only 300kg (661 lb) of explo-
resources available to study, design, con- Funryu 4 from being launched or its guidance sive and it used a direct-action fuse. For
struct and test such a propulsion method. As system fully tested. guidance, the Ki-148 used the same radio sys-
such, the Funryu 3 was shelved. To prevent the Allies from learning of the tem as the Ki- 1 47. Following wind tunnel test-
With the commencement of flight testing Funryu developments, the IJN forbade any of ing with full- and half-size models, Kawasaki
of the Funryu 2, work got underway on the personnel involved with the Funry'u from produced a number of missiles at their Gifu
another SAM that was to be far more discussing the project with anyone. In addi- factory for testing to begin in late 1944. Ki-148
advanced. This missile was designated the tion, documents, test data, constructed mis- test launches were made from four modified
Funrlu 4. Design work was carried out by siles, the launching apparatus and the Kawasaki Ki-48-ll Otsu bombers at Ajigaura in
engineers from Mitsubishi and from the Air- facilities in which the Funrlu was developed lbaraki Prefecture. By December 1944, up to
craft Equipment Factory of Tdky6, all based in were all burned and destroyed. 20 Ki-l48 missiles were being launched per

WEapoN Svsrrn'ts 133

week from the bombers. Despite the rela- sile and in most cases would be 4km (2.5 above, resulting in the Kurai No.6 which
tively successful testing, the Ki-148 was never miles) from the target when the missile hit. improved the gliding ability but not to a satis-
put into production and total deliveries of While the handling characteristics of the factory level. Underwater testing was con-
the pre-production/test Ki- 148 missiles weapons were found to be good, analysis ducted by releasing the torpedo from a boat
amounted to 180. Had the Ki-148 gone into showed that the missiles tended to fall either and the results showed a 17' dive angle at a
service, the Kawasaki Ki-102 Otsu was to be 300m (984ft) short of the target or 100m speed of 5-6 knots. Colour dye released from
the designated carrier aircraft. (328ft) past the target. The reason forwas that the nose of the torpedo assisted the evalua-
Ki-I48 had a length of4.1m (l3.4ft), a span the operator had to rely on his own vision and tors in determining how the weapon worked
of 2.6m (8.5ft) and a launch weight of 680kg clear conditions in order to guide the missile. underwater. The maximum depth the tor-
(r,499Ib). He was not provided with any form of special pedo could reach before the pressure over-
The I-Go-l-C would be the final I-Go pro- optics nor did the missile carry a means to came it was 100m (327f1). From the tests, it
ject. The Aeronautical Research Institute of mark itself in flight such as using burning was determined that a form of gyrostabilisa-
Toky6 University decided to take a com- flares or smoke which the operator could use tion would be required.
pletely different approach to guidance. to maintain sight of the weapon. The only The designated carrier plane for the Kurai
Deciding that anti-shipping would be the measure of this kind ever employed was a tail was the Nakajima B6Nl and B6N2 Tenzan
main use of the I-Go-1-C, the missile dis- light which was used at night so the operator (known as JrI1 to the Allies). No modification
pensed with the radio guidance method and could track the missile. Had the Japanese of the plane's torpedo rack was required. In
instead employed a novel system that used given further consideration to the operator's operation, the pilot had to visually sight the
the shockwaves produced by naval cannons needs, accuracy may have been improved. A target and release the torpedo from a height
as the means to direct the missile. In essence, factor against the use of the Ki-147 and Ki-148 of 100m (327fr). An air spun vane would arm
the missile would guide itself to the target by was that the launch aircraft had to be within the weapon after it was released and the tor-
sensing the shockwaves developed in the air llkm (7 miles) of the target and had to pedo would go into a 20o downward glide.
by large naval cannons during firing. Since remain in the area to proceed with the attack. Once in the water, it would enter its circular
shockwaves travel outwards from the can- With the heavyAllied air presence, getting to dive with a diameter of 79m (260ft), making
non, the missile could determine direction the launch range would have been a formi- one revolution through a depth of 79m (260ft)
and adjust its flight path accordingly to bring dable task and this may have been a factor in after which it would continue to circle and
it onto the target. The main benefit of the sys- the Ki-147 and Ki-I48 failing to enter service. dive until it was crushed under pressure or
tem was that the missile was a fire-and-forget had struck the submarine before reaching
weapon. As long as naval ships engaged in 100m (327ft). A magnetic proximity fuse was
bombardment, the I-Go-l-C would be able to to be used on the torpedo.
track and attack them on its or,r.'n. Testing of Kurai Aerial Torpedo No.6/No.7 In all, 100 of the Kurai No.6 were built with
the system got under way in 1945 and the ini- In April 1944, the IJN initiated a design for an many of them expended in testing. Kugishd
tial results showed promise. However, the air dropped, anti-submarine torpedo. While constructed the rudders and the metal com-
missile body was never built as the war not a true guided weapon, what made the ponents (nose and wing braces) while the
ended before testing of the guidance hard- design unique was that the torpedo was Marunimoko Company (located in Fut-
ware was complete. The proposed I-Go-1-C winged and once in the water, entered a cir- sukaishi, Hiroshima Prefecture) produced
was to be 3.5m (11.4ft) long with a diameter cular pattern as it dived down. Work on the the torpedo body and wings.
of l.6ft. Other specifications for the missile, torpedo was carried out by the First Technical The poor results of the Kurai No.6 resulted
such as its warhead size, rocket motor, per- Arsenal Branch at Kanazawa under Com- in slight modifications to the design. The wing
formance and weight are still unknown. The mander Fukuba. The torpedo body was span was increased and the rudders made
I-Go-l-C is sometimes called the Ki-149 but made of wood with the exception of the taller and set at 6'rather than 8', and the nose
there is no evidence to support the use of this metal nose. The torpedo had no means of was thickened to enhance the ability of the
name. propulsion, either for flight or in the water. torpedo to penetrate a submarine hull. This
Since the Ki-147 and the Ki-148 achieved The wooden wings, each 1.5m (4.9ft) in new model was designated the Kurai No.7
flight testing and both used the same radio length, were glued to the main torpedo body and in January 1945, eleven test drops were
guidance system, the procedures to launch at a 20' upward angle. The wing span was made, but the weapon showed little improve-
and control the missiles were basically the approximately .9m (2.6ft) and the nose car- ment over the Kurai No.6. The end of the war
same. The Ki-67 and Ki-48 bombers used in ried a 100kg (2201b) warhead and the total brought the entire proiect to a close, the engi-
the testing were modified to accommodate weightwas 271kg (5971b). The wood rudders neers having run out of time to solve the poor
the missile operator as well as the equipment were fixed at an 8o angle which, once in the gliding performance and finalise the special
needed to guide the weapon. Operationally, water, imparted the counter-clockwise circu- fuse.
the missiles would be dropped at an altitude larpath. The rudderwas coveredwith awood Two Kr-rrai No.6 mock-ups were captured
of 1,500m (4,922f1),1lkm (6.84 miles) from fairing to stop it affecting the freefall glide of and shipped to the US for delivery to TAIC.
the intended target. By the time the missile the torpedo. Once the torpedo entered the What their fate was is unknown.
was 5km (3.11 miles) from the target, the alti- water, the aluminium pin holding the fairing
tude varied between 30m to 150m (98ft to on sheared off and the fairing came free. The Ke-Go
492ft) depending on the preset of the altime- The first tests were carried out to evaluate As the war progressed the Japanese came to
ter. The operator would guide the missile via the gliding properties of the torpedo. In all, see the invasion of the home islands as avery
a joystick and just before it passed over the forty drops were made of which fifteen were real threat and various means of attacking the
target, the missile would be put into a dive, complete failures, the torpedo tumbling or invasion fleet were investigated. One possi-
bringing it down onto its target. The launching spinning out of control. The wings were mod- bilitywas abomb capable of homing in on the
aircraft had to remain within sight of the mis- ified according to the specifications given heat emissions of naval vessels. The plan

lB4 JnpnrvEst SEcner Pno.rncrs: ExpsRr[4eNral ArncRnrt on rnE IJA eNo IJN 1939-1945
called for the bombs to be used in night release. The strikers seled as the impact because the smaller span was considered to
attacks when heat signatures of the ships fuse, but a delayed fuse was also installed be one of the main causes of the poor perfor-
would not be masked by the sun and other should the bomb strike water, in which case mance of the Ke-Go 106 and 107. Itwas antic-
ambient heat sources. The project was it would explode beneath the waterline of a ipated that the 108 and 109 bombs would be
undertaken by the IJA and given the name Ke- ship. In short order, the Ke-Go 107 joined the ready for test drops by September 1945. Of
Go and work commenced in March 1944. The program with a total of 30 bombs. The 107 course, the end of the war ensured this would
First Military Arsenal's Omiya Department in had similar dimensions to the 106 version and never happen and neither bomb was built.
Tdkyo undertook development of the weighed 726kg (1,600 lb). Had the Ke-Go made it into service, the car-
bolometer that would detect heat in the The Ke-Go was guided in flight by the rier was to be the Ki-67 Hiryu (codenamed
infrared wavelength and bench testing bolometer. The bolometer was part of a unit Peggy by the Allies). The bomb required a
showed that the mechanism was workable. that consisted of the bolometer itself, a mir- special brace that was fitted into the bomb
In fact, the sensor could detect the heat from ror, motor, distributor, amplifier, relay box bay, the apex of the brace extending a few
a man's face at 100m (328ft)! For the bomb to and battery. Heat sources passing through the inches past the exterior of the doors. The
carry the bolometer, three different tl,pes bolometer would strike the mirror that, bomb was secured to this and the lower wing
were investigated: the B-1,8-2 and B-3. Of the depending on where it struck, would trigger could fold to give ground clearance. On take-
three, only the first showed potential and the ihe amplifier which operated the ailerons, off, a crewman had to lower the wing and did
remainder rn'ere abandoned. A myriad of steering the bomb towards the heat source. so by means of a crank that activated a servo
men and departments were assigned to the The use of dive brakes slowed the plummet in the bomb, which then moved the wing.
task of realising the B-1. Under the Military of the bomb, allowing time for the bomb to The bombardier would use bombing tables
Ordnance Administration Board was Maior make adjustments in flight before impact. appropriate for the weight of the weapon and
Fujita who oversaw the gyro and airframe, The ailerons could move up or down 20'. For on release the bomb's arming wires were
\'lajor Hizuta who also worked on the air- awarhead, the Ke-Go caruied between 200 to pulled out along with a wire that released the
frame research, and Major Sonobe who 300kg (440 to 661 lb) ofshaped charge explo- dive brakes. Typically, the bomb would be
addressed the amplifier. The Seventh Military sive, depending on the model. Perhaps in dropped from 2,000m (6,562f1). After the
Laboratory encompassed Professor Konishi error, a 1946 US Naly report lists the warhead bomb was away, the Ki-67 saw a 64km/h
of the Shikan Gakkd in Osaka who handled as ranging from 20 to 30kg (44 to 66lb) of (40mph) increase in speed.
the mathematicalwork, Professor Sano of the explosive which is very light for the size of the All of the bombs were built from wood. The
Osaka Imperial University who conducted Ke-Go. only metal components consisted of the nose
the electrical design and Dr. ltakawa of the With enough bombs ready and available, cap containing the bolometer and the dive
Aeronautics Research Laboratory who field testing began without the bombs con- brakes. In all cases, the bomb bodies were
undertook the aerodl,namic design studies. taining any payload. The test used a floating built by the Atsu Department of the Nagoya
All wind tunnel testing was held at Rikugun raft which measured 10m (32.8ft) by 20m Arsenal, the gyroscope by Hitachi Co., the
Kokugijutsu Kenky'ujo. (65.6ft) and was anchored in Lake spring/gear parts for the timing mechanism
The first two bomb versions, the Ke-Go 101 Hamanako (located in Shizuoka Prefecture). by the Hattori Jewellery Co. and the electrical
and 102, were built in small numbers: ten and On the raft wood and coal in a 4m (13ft) by contacts for the timing mechanism by Sumit-
five respectively. The hydraulic and mechan- 4m (13ft) pile was set alight. Both Ke-Go 106 omo Communications Branch.
ical linkages for operating the bomb ailerons, and 107 bombs were released from altitudes Interestingly, the US would develop a simi-
coupled with the use of an electric gyro that varying between 1,524m (5,000ft) and 3,048m lar weapon, the VB-6 Felix. This was a 454kg
interfered with the bolometer (A bolometer is (10,000ft) but the results proved very disap- (1,000|b) bomb fitted with an infrared seeker
a device used to measure the energy of inci- pointing. In all, around 60 drops were made in the nose that controlled the fins and guided
dent electromagnetic radiation and was but only 5-6 of the bombs displayed the zig- the bomb. It was meant for night warfare,
invented by Samuel Pierpont Langley in zag flight path that indicated the bolometer notably against ships at sea. The VB-6 was put
1878.), saw both of these versions aban- was sensing the heat from the fire and guid- into production in 1945 but would not see ser-
doned. Continued study resulted in the Ke- ing the bomb. The remainder simply veered vice by the close of hostilities.
Go 1 03, I 04 and 1 05. However, none of these away from the raft. The dismal results were
designs left the drawing board although the blamed on faulty equipment although the
103 was to use an air gyro instead of an elec- data collected was not sufficient to properly
trical version. The gyro only existed to ensure identify the cause of the failures. It was found
the bomb did not spin faster than 360'in 50 that the terminal velocity of the Ke-Go107
seconds. The Ke-Go 106 would be the first to was 539km/h (335mph). In July 1945, further
be built in numbers, fifty examples in all. testing ceased, though work continued on
Overall length of the 106 was 4.7m (15.5ft) improving the bomb.
long and the wing span was 2m (6.5ft). Four The Ke-Go 108 was developed but it was
main fins were fitted 1.7m (5.7ft) down the with the Ke-Go 109, the last to be designed, in
length of the bomb body with a smaller set of which the Japanese hoped the problems
four fins 1.2m (3.9ft) behind the main fins, would be resolved and the bomb would
and the tail of the Ke-Go contained dive finally prove successful. The Ke-Go 109 bomb
brakes. In the nose was the bolometer while was 5.5m (18ft long), 5m (1.6ft) in diameter,
the shape charged warhead was contained had a wingspan of 2.9m (9.4ft) and weighed
behind it. Two strikers that protruded out 800kg (1,7641b). It was estimated that the ter-
from the nose were tipped with small pro- minal velocity for the Ke-Go 109 was 579km/h
pellers which spun and armed the bomb after (360mph). lt featured a larger wing span

Wnnporu SvsrsN,ts 135


z From left to right, the Funrt'u 2, FunrSr 4' Ki-147'
o Ki-148, and the Ke-Go.

136 JapexrsE SrcRer Pnorncrs: Expsnrl,tnnrel Atncnanr oF rHE IJA AND IJN 1939-1945
Specialised Bombs MK.I9 (IJN)
This was an air-to-air rocket-propelled bomb
Both the IJA and the IJN developed spe- design intended for use against bomber for-
cialised bomb munitions during the war, mations. It weighed 7.5kg (16.5 lb). Although it
some of which remained projects while oth- was never built, the work eventually resulted
ers saw service. Several were dedicated to the in the Mk. 28 rocket bomb.
practice of air-to-air bombing in which bombs
were used to attack enemy bomber forma- 'Silver Paper Scattering Bomb' (IJN)
tions. Although not a complete list, the follow- This was the Japanese version of the German
ing are some of the more interesting Diippel, better known as 'Window'. A 2kg
examples. (4.4Ib) sheet steel cylinder was packed with
200 pieces of metal coated paper measuring
Ko-Dan .3mx 8.4m (lftx 27.5ft). The bomb used a pull
The Japanese expended some effort in devel- fuse that the crewman yanked to arm it and
oping the Ko bomb. This was developed in was then thrown out of the aircraft, the sub-
response to obselations made by one sequent charge expelling the papers into the
Kobayashi of high-speed photographs of det- air to act as false radar targets. This anti-radar
onations of various experimental shaped weapon was used heavily from 1944-1945.
charges. The basis of his theory was that the
energy produced by an explosion is projected To-3 (IJA)
along horizontal lines with the impact sur- Developed and produced from l94l to 1942,
face. By extension, the amount of energy cre- the To-3 was a cluster munition and initially
ated is proportional to the quantity of the designed to combat aerial targets. However,
explosive used. To achieve a wider impact the To-3 turned out to be more effective
area against a target, the Ko bomb had a 2mm against ground targets. Thirty to forty To-3
thick rubber nose that was mated to a 2mm bomblets could be carried in a dispensing
thick steel case. On impact the nose of the container or modified underwing bomb racks
bomb would flatten and expand, thereby could carry several of the To-3 bomblets.
enlarging the surface area and increasing the Once released, either from the container or
effects of the explosive pellets contained the rack, the To-3's vane would spin, arming
within. The Ko would be used against hard the munition. Use of the To-3 was performed
targets such as concrete emplacements. In on a limited basis and production ended in
tests, a Ko bomb with 50kg (1101b) of TNT 1942.
and ultropine in an even split mixture could
penetrate a lm (3.2ft) slab of concrete and Ta (IJA)
utterly destroy it. lt would take a regular 25Okg Evolving from the To-3 was the Ta. Each Ta
(551 lb) bomb to accomplish the same effect. bomb was a 40mm hollow charge weighing
A part of this concept is known as the Misz- .33kg (.741b). It used fixed tail fins. When
nay-Schardin Effect and is in practice today attacking airborne targets, canisters contain-
with modern HESH (High Explosive Squash ing between 30 and 76 of the Ta bombs would
Head) munitions and EFP (Explosively be released. A burst fuse would open the can-
Formed Penetrator) weapons. isters, usually just after release, to scatter the
Ta bombs. The loaded canisters weighed
Mk. 7 (rJN) 30kg (66 lb) and 50kg (1 I0 lb) respectively and
Conceived around 1936 by Dai-lchi Kaigun the Ta was first used operationally in 1943
KOk[ Gijutsu-sho, the Mk. 7 chemical bomb with Japanese pilots claiming a number of
was not given priority since at the time there kills using the weapon. Typically, the Ta was
was little need for such a weapon and it was far more effective when used against more
thought the work needed to bring it to fruition densely packed bomber formations.
would be too great. The idea surfaced near
the end of thewar, the Mk.7 then being denied Ta-105 (IJA)
a second time in favour of guided missiles. The Ta-105 came from the need to attack
The design was based on the l kg (2.2 lb) prac- Allied shipping during the expected invasion
tice bomb. It had a rubber nose and within a of Japan. The Ta-105 was the Ta bomblet
special tail was a glass bottle of Bacillus bac- enlarged to carry a more substantial payload.
teria, most likely B. anthracis, which causes The bomblet had a 100mm hollow charge that
anthrax. It had no fuse or explosive charge, was capable of penetrating up to 140mm of
the impact being enough to break the bottle armour plate. Unlike the Ta, the Ta-l05 used
and release the bacteria. The planned colour folding tail fins that deployed upon release.
for the bomb was a purple nose with a purple Each canister contained 21 Ta-105 bomblets.

WEnporu Svsrrl,rs 137

Type 2 No.6 21-Go Model I (IJN) Type 5 No.25 Mk.29 (IJN) together, each cluster weighed 50kg (1 10 lb)
This munition was the IJN's version of the T Designed for use by special a aircraft, and the average single engine fighter could
lt was a heavier bomblet and weighed lk the Mk.29 contained 1,100 packed in carry up to four clusters. The usual tactic was
would carry up to 40 of the
(2.21b). A canister (110lb) of whi After fly- to drop the To-2 bombs into the path of
Model I bomblets. Like the Ta, a fuse would rng formation, the pilot would oncoming bombers. To increase the depth of
open the canister deploying the bomblets use a pull cord to detonate the bomb while it the bomb spread, some To-2 bombs had a
onto their target. was still attached to the aircraft. The bomb smaller parachute that would result in a faster
could also be released where the tail fuse descent. The bomb was armed with an
Type 2 No.6 2l-Go Model2 (lJN) would trigger the bomb explosion. Unlike impact fuse that, regardless of where the
This was the same bomblet but instead of the other pellet dispersing munitions, the Mk.29 bomb struck, would trigger the detonation.
hollow-charge in the Model l, the Model 2 flung them at right angles to the bomb casing Usually the cable would be hit by the bomber
used a solid black powder charge. This instead of in a downward cone. Initial testing with the bomb swinging up and against the
change was made because the IJN felt that an was begun in 1944 but only one prototype was plane, exploding the bomb. The To-2, how-
armour piercing charge was not necessary air dropped and the type did not enter service. ever, was not developed further because the
against aircraft that were typically unar- The Mk.29 was 1 .6m (5.1ft) long and weighed cable was typically cut if hit by a wing, it relied
moured. This did, however, raise the weight 2s0kg (551 lb). on a high altitude to be effective and the Ta
of the Model 2 and a canister could only hold munition showed far greater promise.
36 of the bomblets, the total weight of the Type 99 No.3 Mk.3 Sango (lJN)
loaded canister being 49kg (1 08 lb). Designed in 1938 and entering service in 1 939,
this anti-bomber bomb contained 144 white
Type 3 No.6 Mk.3 Model I (lJN) phosphorous-filled steel pellets. Its tail fins
This was basically a simplified Type 99 No.3 were offset to impart spin as it fell and this
Mk.3 bomb as it used a wooden nose and did armed the tail fuse. The fuse would then trig-
away with the impact fuse. The bomb was 1 m ger a burster tube down the middle of the
(3.3ft) long and weighed 56kg (1241b). Its bomb, scattering the pellets. A charge in the
design commenced in 1943 and entered ser- tail ensured the pellets were directed down-
vice in 1944. Colouration was the same as the wards. Should the tail fuse fail, an impact fuse
Type 99. was proved in the nose of the bomb. A later
modification of the bomb added fins to the
Type 3 No.6 Mk.27 Model I (IJN) body to impart a faster spin. For identification
Development of this anti-bomber rocket purposes the nose of the bomb was painted
bomb began in January 1944. The cone- silver and the fins red. The bomb was .7m
shaped nose contained 140 iron pellets (2.3ft) long and weighed 34kg (74lb).
embedded in 4kg (8.8lb) of white phospho-
rous. A 1Okg (22 lb) propellant charge moved Ro-Ta (IJA)
the bomb to a velocity of 270mlsec One of the problems the Japanese had with
(885.8fVsec). On detonation the pellets were using cluster munitions was that they relied
scattered in a 60" radius and the bomb on the attacker having a higher altitude than
was provided with the fittings for rail launch- the target in order to rain the bomblets down
ing by fighters. The IJN accepted it for upon them. However, manyJapanese fighters
selice in February 1945 and final testing was struggled at the altitudes at which the B-29s
completed in April 1945. Testing was con- usually operated. The IJA sought to solve this
ducted by DaiJchi Kaigun Koku Gijutsu-sh6, dilemma with the Ro-Ta. The same Ta
the Second Powder Factory Arsenal, and bomblets were used but the canister was
Kashima Bombing essentially a rocket that was launched
) long and towards the target. A timed fuse would trigger
an explosive charge that scattered the Ta
bomblets. In effect, the Ro-Ta was like a buck-
Type 3 No.6 Mk.28 Model I (IJN) shot shotgun shell. Because the Ro-Ta could
Using the Mk.19 as a basis, the Mk.28 was a fired like a rocket, there was no longer a
, anti-bomber rocket bomb with a . requirement to be above the intended target
warhead A pro- in order to use the weapon. Luckily for the
pellant charge mb to a maxi- Allies, the Ro-Ta was still in development
mum velocity of 1,3l2fVsec. Dai-lchi Kaigun when the war ended.
Kdk[ Gijutsu-shd built the bomb and its
clearence testing was undertaken by Dai-lchi To-2 (lJA)
Kaigun Kokrl Gijutsu-sho, the Second Powder Even before the To-3 and Ta series of cluster
Factory Arsenal, and Kashima Bombing bombs, there was the To-2 parachute bomb.
Experimental Field. Testing was completed Developed in 1935, the To-2 was a 1 .8m (4lb)
late in 1944 but the rocket bomb was not bomb that was suspended from a length of
accepted for service. Its length was .7m steel cable that was itself connected to a silk
(2.3ft) and weight 7.3kg (16Ib). parachute. Ten To-2 bombs were clustered

138 JepafrEss SEcRrr PRorecrs: ExpeRrupntel AtRcRrnr or rHE IJA alll IJN 1939-1945
would now take 50-70 hours for it to traverse to prevent gas leakage as the untreated Washi
Specialised Bombs: Fu-Go it
the Pacific Ocean and arrive over the US was porous.
Japan was able to achieve where Germany mainland. General Sueyoshi Kusaba was put After being brushed with the glue, each
failed and that was to bomb the US mainland. in command of the project to tackle the dis- gore was inspected for flaws. This was done
On 9 September 1942, the Japanese subma- tance issue and put the Fu-Go into operational by laying the gores over a panel of frosted
rine I-25 surfaced west of Cape Blanco, Ore- use. glass beneath which was a light. The glue
gon, and a single Kugisho E14Y1 floatplane To overcome the problem of maintaining would appear blue and any uneven applica-
(known as Glen Io the Allies) was assembled. altitude as the balloon travelled by day and tions of the glue showed up as a lighter area.
Pilot Warrant Officer Nobuo Fujita and night, a ballast mechanism was designed. All flaws were marked and patched. Once the
observer Petty Officer Okuda Shoji climbed This consisted of a cast aluminium ring gores had passed inspection, they were taken
into the plane that was loaded with two 77kg around which 322.51o3.2kg (5.5 to 7lb) sand- to the final assembly area. A large room was
(l 70lb) incendiary bombs and took off for the bags were secured. Fuses were fitted to pairs required with sharp objects padded so not to
US coast, heading towards the Oregon side of of sandbags, the fuses powered by a small rip the gas envelope. High school girls were
the Siskiyou National Forest. Once over the batteryand connected to aneroid barometers. often employed for assembly, using the glue
forest, the bombs were released in the hope a When the balloon sank to an altitude of to affix the gores together to create the com-
forest fire would start. However, recent rains around 9,144m (30,000f1), the aneroid barom- pleted gas envelope. Each girl had to ensure
coupled with the bombs having been eter would trigger a switch. In turn, this trig- her nails were trimmed, that she wore gloves
dropped too low resulted in a few small fires gered a fuse which in turn would fire two and socks, and that she did not wear hair pins
and these quickly put out. The attack would small charges that would each release a sand- as these could damage the gas envelope.
be repeated again on 29 September 1942 but bag, one across from the other to maintain Once the envelope was finished, it was taken
once more the results were disappointing. balance. The balloon would then rise to an to another building, often a sumo hall or the-
The first attack marked the only time in history altitude of around 11,582m (38,000ft) where a atre (buildings specific for the task were later
an enemy aircraft bombed the US but another gas release valve set into the bottom of the built), and inflated to check for leakage. After
plan was in the works, one far more ambitious envelope discharged some of the gas to keep successfully passing the final inspection, the
and ingenious, but ultimately fruitless. the balloon from rising higher. Eventually it Fu-Go was completed.
The plan was called the Fu-Go, 'Fu' being would drop again triggering another release of The finished Fu-Go had a scalloped waist-
the first kanji of the word 'fusen', meaning bal- sandbags followed by a rise, vent and the band around the gas envelope to which the 19
loon. The 'Go' simply meant tlpe. Originally cycle would repeat. By the time the last pair of shroud lines, each 14m (45ft) long, were
conceived in 1933 by Lieutenant General Rei- sandbags was dropped, it was estimated that secured. The lines were brought together and
kichi Toda of the Japanese Military Scientific the balloon should be over the US mainland tied into two knots from which the bombs and
Laboratory, the bomb dropping balloon Fu-Go where its destructive payload would then be the ballast system were hung.
rvas part of a series of studies into possible released. With the problem of keeping a relatively sta-
new weapons. Erperiments with balloons The balloon could lift a maximum of 136kg tic altitude solved, the next hurdle was to
capable of maintaining a stable altitude were (300Ib) at 9,144m (30,000ft). The typical determine where and when to launch the bal-
initially allowed to proceed but by 1935 the Fu- munition payload was one Type 92 15kg loons. Towards the close of 1943 and into the
Go projectwas cancelled. However, following (33 lb) high explosive bomb, one Type 100 Skg early part of February 1944, the Japanese
the famous Doolittle raid on 8 April 1942, the (11 lb) incendiarybomb and one Type 97 12kg launched balloons equipped with radios
Fu-Go was revived as a means of exacting ret- (26.4Ib) incendiary bomb. A 29m (64ft) fuse which were tracked so their courses could be
ribution for the attack. was connected to a charge on the gas enve- monitored. Two stations set up in Hokkaido
The Fu-Go was to have been launched from lope and once the bombs were dropped, and in Chiba Prefecture could track the bal-
submarines situated 998km (620 miles) off the fuse was lit which, in turn, destroyed the loons only through the first portion of their
the US coast. In March 1943, a 6.1m (20ft) balloon. flight, but once over the open ocean all con-
diameter balloon was successfully launched The spherical gas envelope could store tact was lost. The Japanese were aware that
and remained aloft for at least ten hours, 538m'(19,000ft') of hydrogen gas. The diame- the west-to-east wind speeds were at their
enough to make the submarine launch plan ter at full inflation was 10m (33ft). Early gas peak from November through to March, top-
rvork. The main problem was the effect of envelopes were constructed from rubberised ping out at 298km/h (185mph). In addition, a
temperature on the hydrogen gas used in the silk but this was too costly to construct and the shortage of meteorological data on weather
balloon. At night and in cool temperatures, the production Fu-Go used Washi paper made patterns over the ocean and at high altitudes
gas pressure was reduced and the balloon lost from the kozo bush. While Washi was inex- limited the ability to plan trajectories for the
buoyancy while by day, in higher tempera- pensive and already produced by hand in balloons. While the winds were higher, it was
tures, there was the risk of the gas expanding paper mills across Japan, there was no means also winter throughout most of the launch
and rupturing the gas envelope. Because of to ensure a constant level of quality. There- window. In addition, the balloons had to be
this launches of the balloons had to take place fore, the Fu-Go program had to develop released in clear, cloudless weather with little
either by day or by night and not straddle the mechanical methods to make Washi as well surface wind. If balloons were sent up in over-
hvo times of day. However, the IJN's need for as laminate it. It took four to five layers of cast skies with precipitation laden clouds,
submarines to support operations in the Washi to make a gore (a segment) and 38 to moisture would collect on the balloons which
Pacific left no room for launching the Fu-Go 64 gores glued together to make the sphere. would freeze at higher altitudes, adding
and the projectwas again cancelled inAugust The adhesive used, called konnyaku-nori, weight resulting in the balloons being unable
1943. was made from konnyaku, a potato. As the to reach the US. Three major launch sites
Remaining undeterred, the Fu-Go engi- glue was relatively clear, colouring was added were selected: Nakoso (Fukushima Prefec-
neers looked into a solution where the bal- so workers could check for evenness in the ture), Otsu (lbaraki Prefecture) and lchin
loon could be launched from Japan, although application. The glue also served as a sealant Omiya (Chiba Prefecture).

Wsnporu SvsrEus 139

On 3 November 1944, the Fu-Go balloon The Fourth Army Air Force was tasked rvith ger forest fires rvere launched from two sites
bombing campaign was officially opened. In the detection and interception of the balloons. in England on 20 March 1942. The balloons
all, between 9,000-10,000 balloons were avail- In addition, Project Firefly was initiated to were of simple construction, reached a height
able and by 20 November, the first en masse position aircraft and troops to respond to for- of 4,877m (16,000ft) and inexpensive at 35
launchings had taken place. Prior to launch, est fires. Project Lightening was set up shillings each. More of a nuisance to the Ger-
the sandbag release mechanism was set through the Department of Agriculture to be mans, the greatest triumph of the operation
based on the estimated wind speeds to on alert for biological attacks against crops was the explosion of the Bohlen power station
ensure the balloon was over the US before and livestock. Successful interception of the caused by the failure of an overload switch
releasing its payload. The gas envelope was balloons proved difficult owing to the highalti- when a balloon struck a main power line near
only partially filled to allow for expansion of tudes at which they travelled, poor ground Leipzig. The last launch was made on 4 Sep-
the hydrogen at an altitude of 4,877m reporting and inadequate weather reporting tember 1944.
(16,000ft). On a good day crews could launch (this would also hamper the ability of the US Post-war, the US would use Fu-Go technol-
up to 200 balloons. March 1945 would see the to accurately determine launch points from ogy in the E77 balloon bomb project.
highest number of balloons deployed, 3,000 in Japan). In fact, only two Fu-Gos were shot
all, and the final launch was made on 20 April. down over the US mainland. Only US Army
Typically included in batches of balloon and US Navy assets in the Aleutian Islands had
launches would be a radio equipped balloon a higher tally as the balloons often passed over
to allow for tracking. the territory. With the problems in locating the
The first balloon was found on 4 November balloons, a final plan, Project Sunset, was ini-
1944by a US Nary patrol boat. It had crashed tiated to create a web of radar sites across the
into the sea 106km (66 miles) southwest of coastline of Washington State. When balloons
San Pedro, California. Nothing much was were detected, interceptors would be scram-
made of it until another turned up in the sea bled to engage them. As it turned out, the plan
two weeks later. Also, balloons were found up was put into action in April 1945, the same
in Montana and Wyoming and the US military month the Japanese ceased launching the Fu-
realised the purpose of the balloons. Federal Go. In any case, radars had a difficult time in
and state agencies were put on alert (espe- detecting the balloons due to their low signal
cially forest services as the threat of forest fires return.
from the incendiaries was very real) and steps The cost to produce one Fu-Go was approx-
were taken to prevent news of the balloon imately 10,000 yen. If the added expense of
bombs reaching the general public. This was the design and production of the mechanical
done to prevent panic in the populace since equipment to produce Washi and the erec-
no one could be sure when or where a bal- tion of buildings to inspect the balloons, the
loon would release its bombs. In addition, by price of the Fu-Go project was high in com-
keeping the story from the press, the Japanese parison to the results it brought. Still, Japanese
would be denied any information on the propaganda broadcasts played up the Fu-Go
effectiveness of iheir attacks. The censorship prior to the project being cancelled in April
was very effective and only one story con- 1945.
cerning a balloon bomb was published Ironically, on 10 March 1945, a balloon from
appearing in Thermopolis in Wyoming. This one of the last launchings struck a power line,
was also reported in a Chinese newspaper. resulting in the loss of power to the nuclear
From this, the Japanese judged the Fu-Go plant in Hansford, Washington. This plant cre-
campaign was a success and the balloon ated ihe atomic material used in the Manhat-
campaign continued. tan Proiect's atomic bombs, which would
The near total shutdown of public informa- ultimately be used against Japan. The loss
tion on the balloon bombs had one severe was barely noticed as back-up systems came
drawback for the Americans and that was that online to maintain the reactor. Another irony
the public had no knowledge of their exis- occurred on 13 March 13 when two Fu-Gos
tence and consequently were not be warned returned to Japan, although they touched
of the dangers. The censorship would be down without causing any damage.
reversed after an inevitable tragedy occurred. In all, 296 balloon sightings or incidents
On 5 May 1945, near Bly, Oregon, Reverend were reported across 17 US states, 5 Canadian
fuchie Mitchell, his wife and five children provinces and Mexico out of the 9,000-10,000
from his church group were enjoying a picnic launched. Hundreds remain unaccounted for
in a wooded area. One of the children, Joan and even today, some may still lurk in unpop-
Patzke, found a balloon stuck in a tree and ulated areas or in dense forest presenting a
tried to pull it down. The subsequent explo- danger to anyone encountering one.
sion of the bombs it carried killed all but Rev- Interestingly, the British would use a similar
erend Mitchell. The deaths rvould be the only weapon against Germany. Called Operation
The illustrations depict the following weapons,
US mainland casualties from enemy action. Outward, hydrogen filled balloons equipped from left to right: canister for the Ta bomb with a
Following the incident, the censorship was with a trailing steel chain to short out power Ta bomb beside it, Ko-Dan rubber bomb, Type 99
lifted to ensure public safety. lines and three 6lb incendiary devices to trig- No.3 Mk.3 Sango and the Fu-Go.

140 JnparussE Socnlr Pno:Ecrs: ExpERruENrar- ArRcRnnr op rHE lJA,qNl IJN 1939-1945

WeapoN SvsrEus 141


German Technical Exchange

\Mith Japan: A Brief Overview
It should not be surprising given Japan was operationally. This clause denied Japan
allied to Germany as part of the Axis powers access to the advanced research being con-
that there were numerous requests for and ducted by Germany's war industry. In addi-
the exchange of war materials between the tion, there were some in the German
two countries. What was perhaps surprising industrial sector and government who were
was that the bulk of the exchange would go reluctant to share the fruits of their labours
one way with very little going in the opposite with Japan. Nevertheless, Japan was able to
direction. obtain a considerable amount of war mater-
Japan's relationship with German aircraft ial for her army and navy forces. A third agree-
and manufacturers went as far back as ment, the Patent Rights Agreement drafted in
1915 when the Isobe Kaizo Rumpler Taube, December 1944, was meant to protect tech-
a copy of the Taube aircraft, was built and nological advancements and prevent confis-
flown by the Imperial Flying Association. cation ofpatents. The Japanese dragged their
In the 1920s and 1930s, Heinkel and Junkers feet on the agreement and it was never
were the dominant German firms, both of signed.
whose designs were much in demand by It would take a decree by Reichschancellor
the Japanese. Other German firms such as Adolf Hitler in January 1945 to remove the
Dornier, Rohrbach and Hansa-Brandenburg restrictions of the 1943 and 1944 agreements,
also sought to make sales prior to the following which Japan had full access to the
start of World War 2. Before the war the German military industry including experi-
Japanese military also entertained contacts mental projects. However, by this time it was
with British, French and American aviation too little, too late, because both Germanyand
firms. Japan lacked the capability to ship material to
However, with the advent of hostilities and Japan by sea or by air.
Japan having sided with the Axis powers, the Perhaps surprising given the verylong list of
country no longer had access to this broad technical exchanges that left Germany for
spectrum of aviation companies and aircraft Japan is that there was very little that went the
designs. Of course, through their acquisitions other way. Germany was content to receive
of aircraft prior to the war and their subse- currency in exchange for the designs and
quent study of aircraft design, the Japanese data - Germany needed raw materials for her
were capable of producing their own indige- war industry. One of the very few examples of
nous aircraftwith a good measure of success. Japanese technology that was acquired by
Prior to the war beginning, and continuing Germany was a single Nakajima E8N float
through until 1943, the Japanese obtained a plane (codenamed Daue by the Allies) that,
number of German aeroplanes. Some would oddly, ended up disguised in British markings
see series production such as the Biicker and was used by the German merchant
Bii 131 Jungmann basic training aircraft, raider Orion. The only other occasion when
while others were obtained for evaluation or Germany attempted to acquire a Japanese
as comparison aircraft to be pitted against aircraft, the Mitsubishi Ki-46 (codenamed
Japanese planes. Dinah), the Japanese ensured that the nego-
Formal agreements between Japan and tiations with the Luftwaffe for a manufactur-
Germany did exist during this time, but it was ing licence went nowhere.
the signing of the Economic Agreement of Japan would receive all manner of war
January 1943 and, later, the Manufacturing goods and data from Germany during the war
Rights Agreement of March 1944, which and it would make for a long list were every-
paved the way for increased German techni- thing to be included. As such, the list pre-
cal exchange. These agreements, however, sented here is limited to aircraft and
restricted Japan to only acquiring technology selections pertinent to the experimental
that Germany's war machine was fielding nature of the subjects in this book.

142 hpaNEse SscRtr PRoJECTS: ExpERruEllral AIRcRnnr ol rHE IJA,cNl IJN 1939-1945
Pre-World War 2 German Focke-Wulf Fw2005-l 25 as the basis for their entry into a 'made in
Announced with a major publicity stunt, the Japan' passenger/transport contest operated
Aircraft Exchange (l 914-38) Fw2005-1 'Brandenburg' took off for Japan by the Aviation Bureau of the Department of
Dornier Do C transport from Berlin, Germany, on 28 November 1938 Communications. The AB-1 could be con-
ln 1927, Kawasaki imported seven aircraft and arrived in TOkyd in a little under 48 hours. verted to a seaplane if required and proved its
from Dornier including the Do C Komet. The Japanese were impressed by the Fw200 worth going on to win the contest and seeing
Although rejected by the Japanese military, and early in 1939 the Dai Nippon Kabushiki several years' service in private hands.
Kawasaki used the Komet to fill an order for Kaisha (Japan-Manchuria Aviation Company)
three passenger transports from the newspa- contracted Focke-Wulf for five Fw2008 Con- Heinkel HD 26 single-seat float plane
per service Asahi Shimbun Sha. The three air- dors. Attached to this was an order for a sin- A single example was built by Aichi (also as
craft, known as the Kawasaki-Dornier Komet, gle Fw200B for the IJN as a maritime the Aichi Nar,y Type 2) while the Navy also
were built from imported components. reconnaissance aircraft. However, with the acquired a Heinkel constructed HD 26. Both
Kawasaki later imported a Dornier Merkur, an outbreak of the war in Europe, none of these were tested in 1926, but like the two-seater,
updated Komet, leasing the aircraft as the orders were fulfilled. The Allies felt, with good the model was made obsolete by catapults.
Kawasaki-Dornier Merkur to Asahi Shimbun. reason, that Japan would soon be using the Both aircraft were turned over for civilian use.
However, this Merkur would instead see ser- Fw200 and gave it the codename Trudy.
vice with the fumy as the Aikoku No. 2 after Heinkel HD 28 three-seat float plane
its conversion to an ambulance. Used during Hansa-Brandenburg oat plane
fl The Navy purchased one HD 28 from Heinkel
the Manchurian Incident, the Aikoku No. 2 Four were built between 1 925 and I 926 as the in 1926. Tested by the Naly, problems with
was very active in casualty evacuation and Aichi Type 15-ko Mi-go. Based on the German the Lorraine-Dietrich engine saw the HD 28
upon retirement became a monument. plane, Aichi refined the design but failed to failing to meet expectations and in 1928 the
When the Russians invaded Manchuria in win the Navy's competition for a reconnais- Naly withdrew their interest in the design.
1945, the plane was burned by the retreating sance seaplane and so work ceased.
Japanese troops. Heinkel HD 56 seaplane
Hansa-Brandenburg W 33 float plane To meet a 1929 Navy need for a catapult
Dornier Do J Wal flying boat As part of reparations from World War 1 , Ger- launched seaplane, Aichi once more turned
In 1 929, Kawasaki received an order for a pas- many sent Japan a single W 33 float plane. to Heinkel and imported the HD 56. Meeting
senger flying boat. As they already had an Highly impressed with the W 33, the IJN the needs of the Navy after modification, it
imported Dornier Wal flying boat from which requested a copy which resulted in the Type was accepted into selice in I 931 as the Aichi
to source a design, Kawasaki essentially pro- Hansa. 310 were built and served with the IJN E3A1 beginning with the first deliveries out of
duced their own version using imported com- from 1926 to 1928 before being retired; many an eventual total of 12 aircraft. The E3Al saw
ponents from Dornier. A total of three were were converted to passenger float planes for combat during the Sino-Japanese conflict,
built by 1930, being flown as liners for regular civilian use. operating from Jintsu-class cruisers. It did
passenger services. not, however, remain in service long, being
Heinkel HD 23 carrier fighter replaced with superior aircraft. Some E3A1
Dornier Do N heavy bomber In 1926, Aichi contracted Heinkel to design a seaplanes were retained by the Naly as train-
ln 1924, the Army asked Kawasaki to produce carrier fighter for entry into a Navy competi- ers with the remainder released to the civilian
a bomber to replace types already in selice. tion io replace the Mitsubishi Type 10 fighter market.
To this end, Kawasaki enlisted Dornier and then in service. Called the HD 23 by Heinkel,
BMW as coilaborators on what would become it was known as the Aichi Type-H Carrier Heinkel HD 62 three-seat float plane
the Type 87 Heavy Bomber. Dornier designed Fighter in Japan. Two were built by 1927 and Aichi turned to Heinkel to produce a long-
the Do N and the plane was buitt by Kawasaki, much emphasis was placed on the capability range reconnaissance float plane to meet a
the first completed in 1927. After a year of test of the plane to ditch at sea, including jettison- 1931 request by the Nary for just such a craft.
ing, the bomber, now called the Tlpe 87, was able landing gear and the ability for the Called the Aichi AB-5, testing showed the
accepted into army service, the last of the 28 engine to stop the propeller in the horizontal design to be sound but with room for further
bombers being delivered in 1932. Kawasaki position. However, performance-wise, the refinement. This would result in the 48-6 of
produced the BMW VI for the Type 87 under Aichi Type-H Carrier Fighter showed up 1932 which, despite being successful, was
licence. Equipping a handful of bomber lost th-E ceff56'Tition to Naka not accepted into selice having been beaten
chutai, the Type 87 would see brief action a Kawasaki design.
ing the Manchurian Incident of 1931. Heinkel HD 25 two-seat float plane
Licence-built in Japan as the Aichi Navy Type kel He 50/He 66
Focke-Wulf Fw42 bomber this aircraft saw service on a number of 933, the Navy issued requests for a carrier
The Fw42 was a design for a canard- dive bomber. Aichi elected to utilise Heinkel's
equipped twin-engine bomber. Various revi- their service life was short due to the advent He 50 which was then entering service with
sions of the design were made with wind of catapult launched seaplanes. The few clandestine military air force training units.
tunnel models undergoing testing from 1932 Aichi Naly Type 2 aircraft were then sold to Called the He66 for export reasons, Heinkel
to 1933. Once the final revision was settled, a the civilian market. Later, in 1930, three Type delivered a single He 66 to Aichi who, after
full scale mock-up was built. The Japanese 2 aircraft were converted to transports for some modifications, submitted it to the Nar,y.
were invited to review the mock-up, Focke- civilian use. Winning the Naly's competition, the modified
Wulf hoping to entice sales of the bomber or He 66 was then licence built in Japan, entering
licence rights. The Japanese showed no inter- Heinkel HD 25 transport selice with the Naly in 1935 as the Aichi
est and the Fw42 was soon cancelled. Built as the AB-1, Aichi used the Heinkel HD D1A1. The DlAl would see action in China in

GERnreru TEcHrurcn,I- ExcHnncn wrrH JApAN - A BnrEp OveRvrEvl 143

1937. The Allies, thinking the DlAl would be cations, blueprints, manufacturing data and L.V.G. D.IX
met in battle, assigned it the codename Susre. licence to build the aircraft as a bomber, The D.lX provided the inspiration for the
As it was, the DlAl remained in Japanese ser- known as the K 51, the export version of the Seishiki-l which used an imported Mercedes
vice only as a trainer, having long since been G 38. Mitsubishi sent designers to Germany in Daimler 100hp engine, later licence-built in
withdrawn from frontline seruice. 1928 to study the G 38 and production tech- Japan. The Seishiki-1 was completed in 1916
niques, and by 1930, the necessary tools, jigs but the biplane's poor performance resulted
Heinkel He 70 and material were imported and in place. in further development being cancelled.
light reconnaissance bomber Junkers sent a team of engineers to Japan to
In the late 1930s, a He70 was imported to assist with the production. The first bomber, Rohrbach R flying boat
Japan. Few details are available but the wing the Ki-20, was completed in 1 93 1 . A total of six The Nalywas very interested in the metal air-
form of the He 70 would later provide the were built from 1931 to 1935. Kept in secret, craft construction techniques used by the
influence for the wing design of the Aichi the general public was not made aware of the German company Rohrbach. Mitsubishi was
D3A1 carrier bomber, known to the Allies as Ki-20 until 1940 when three Ki-20s partici- asked to study the techniques and the two
Val. pated in a parade fly-over. The Ki-20 did not companies would form Mitsubishi-Rohrbach
see action. GmbH in Berlin in June 1925. A total of three
Heinkel He I l2B-0 fighter Rohrbach flying boats were to be imported,
In 1937, the IJN was seeking new fighters to Junkers K 37 bomber the R- I , R-2 and R-3, known collectively as the
combat the increasingly modern aircraft A single K 37 bomber was imported through Mitsubishi Experimental Type-R flying boats.
being encountered in China. Heinkel was at Sweden and was donated to the Army as Although these aircraft would prove to have
the time looking to export the Hel12, a Aikoku-l. Used during the Manchurian Inci- poor take-off and alighting that denied them
design which was ultimately rejected by the dent, the Army was impressed with the K 37 military service, they did provide invaluable
Luftwaffe. The IJN placed an order for 30 HE and in 1932 asked Mitsubishi to make a simi- experience to Mitsubishi when it came to
112B-0 fighters, known in Japan as the ATHel lar bomber. Using the K 37 as a basis and cap- metal stressed skin construction.
Type He Air Defence Fighter. I 2 of the 30 were italising on experience from the Ki-20, the
delivered though Japanese pilots disliked the prototype Ki-1 heavybomberwas completed Rumpler Taube
performance of the plane and mechanics had in March 1933. Despite problems with the The Imperial Flying Association purchased
difficulties maintainingthe liquid-cooled engines, the Ki-1 was adopted to replace the two Rumpler Taube aircraft prior to the out-
engines. As Japanese fighters of improved old Type 87. The Ki-l-ll soon appeared in an break of World War 1. As Japan was part of
capability were entering service, the attempt to fix issues with the Ki-l but it was the Entente Powers, and thus against Ger-
He 1 12B-0 never saw combat and the Japan- not liked by the crews that flew it. Another air- many, the Japanese Army bought the two
ese cancelled the remaining l8 aircraft. Ser- craft, the Ki-2 light bomber, was also built Taubes from the Association for use in action
viceable He 112B-0 were used as using the K 37 as a basis. The Ki-2 and later in the Tsingtao campaign. With no aircraft,
instructional aircraft and as means to study Ki-2-ll (both being built from 1933 to 1938) the Imperial FlyingAssociation built their own
German manufacturing techniques. Figuring proved very successful, seeing action in version of the Taube calling it the Isobe Kaizo
the He l12B-0 was in active selice, the Allies China and later as trainers into the late 1930s. Rumpler Taube. The solitary aircraft first flew
codenamed the aircraft Jeny although it was The Allies thought the Ki-2ll was still in ser- in 1915 before being wrecked in a crash later
never encountered. vice when the war began and assigned it the in the year. The remains were cannibalised
codename lourse. and used in the Ozaki Soga-go of 1917.
Heinkel He I 18 dive bomber
With the Luftwaffe uninterested in the Junkers Ju 90 transport
He 1 18, Heinkel found the Japanese receptive On 25 July 1938, Mitsubishi entered into nego-
to ihe plane. The IJN placed an order for a sin- tiations with Junkers on behalf of the IJA to
gle He 118 to be delivered by February 1937 work with the German company to produce
along with the licence to manufacture the a bomber version of the Ju90 transport. Ten
He 1 l8 in Japan. The IJA also purchased a sin- were to be completed and flown to Japan.
gle He118 for delivery by October 1937. The IJA even allocated the designation Ki-90
Known in Japan as the DXHel, Hitachi for the bomber. However, Junkers eventually
Seisakusho was to be the company that
would produce the licensed aircraft. The
He 118 V4 was shipped to Japan and
bled at Yokosuka for the IJN, but during a
flight it broke up in the air and the IJN Junkers Ju 160 transport
doned the plane. The IJA received the He I The Ju160 was an improved Junke 60,
V5 months later but it too lost interest and latter having lost to 70 in
Heinkel received no further orders. the market. Lufthansa pur-
chased 1 1 Ju 160A-0 and 10 Ju 160D-0 6-pas-
Junkers G 38/K5l healybomber senger aircraft, putting them into service in
The Army sought to have their own version of 1935-36. Two would end up in Manchuria,
the Junkers massive G 38 airliner and in registered as civilian aircraft. The IJN pressed
September 1928, Mitsubishi entered into a them into service as the LXJ.
contract with Junkers for the design specifi-

144 JapnNrse SEcRsr PRo;lcrs: ExptRrugNral ArRcnenl op rHe IJA AND IJN 1939-1945
World War 2 German Aircraft

Arado Ar 196 float plane z

It was reported byAllied int e selvlces o
the Japanese tu196 float
gest this occurred. The Germans operated a z
submarine facility at Penang, Malaysia, and
the unit used the Ar 196 in Japanese colours
which may have led to the confusion in the
intelligence report.

Arado Ar234 Blitz

jet reconnaissance bomber
The Ar234 was a twin-turbojet, single-seat
reconnaissance bomber that entered sewice
with the Luftwaffe in September 1944. Allied
intelligence intercepted a communication
between Germany and Japan in March 1944
that confirmed that the Japanese had data on
the Ar234. It was assumed that the data
related to the Ar234A, which was not as
advanced as the subsequentAr2348 models.
Another report went so far as to say produc-
tion plans were in place to build the latest
Ar 234 aircraft but this was based solely on the
fact another report stated that the FuG 136
Nachtfee visual command indicator equip-
ment (used in the Ar234P series night fight-
ers) had been delivered to Japan in January
1945. It would become clear that the Ar234
was never produced in Japan and it is
unknown exactly what data Japan did
receive on this aircraft.

Bachem Ba349 Natter rocket interceptor

The Natter (meaning 'Viper' in German) was
a rocket powered point defence interceptor -
in essence, a manned rocket launched verti-
cally towards enemy bombers where it
would use its high speed to avoid enemy
fighters and launch a salvo of either 73mm
Hs217 Fohn or 55mm R4M rockets at the
attacking bombers. The pilot would then
eject from the 8a349 and return to earth via
parachute along with the engine portion of
the aircraft. The Ba 349 required little in terms
of critical war materials and could be con-
structed by semi-skilled workers. Several
unmanned test flights were flown but the only
recorded manned flight ended in the death of
the pilot. Despite a handful being deployed,
none saw action. Allied intelligence surmised
that the Japanese were provided with infor-
mation on the Ba349 and they were correct.