B-29 Central Fire Control Computer

[Scott M. Kozel]

I have read about the Central Fire Control (CFC) system and computer
that was utilized on the B-29 Superfortress, in many different books and
publications over the years.

The 5 machine-gun turrets were directed by gunners from remote
mechanical aiming stations, and the gunners in the center of the
aircraft could be given secondary control of the front and tail turrets.

The computer took the guess-work out of "point and aim" because it
figured all of these out:
A. Range Estimation
B. Bullet Drop
C. Gunner's Deflection
D. Target Deflection
E. Windage Correction
F. Parallax compensation of guns to sights
G. Increased range far beyond the usual 600 yds by automatically
establishing the correct lead.

There are plenty of sources that explain conceptually how the system
worked. My question concerns the computer itself, and I haven't been
able to find any details about exactly how the computer itself operated,
and what kind of architecture it had.

I work in the information technology field, and I know that the state of
the art for electronic computers in 1944 and 1945, was primitive, and I
am wondering how they managed to develop a computer to do all that
real-time processing, and fit it into a box that I've read was about 6
feet by 3 feet by 3 feet.

That was a pretty amazing feat for that era.

[Martin Rapier]

"Scott M. Kozel" <***@comcast.net> wrote in message news:d1d8qi$ain$***@gnus01.u.washington.edu...
{snip}

Post by Scott M. Kozel
There are plenty of sources that explain conceptually how the system
worked. My question concerns the computer itself, and I haven't been
able to find any details about exactly how the computer itself operated,
and what kind of architecture it had.

It was an analog computer similar to those used for fire control
calculations in warships.

They were electro-mechanical but built for specific tasks and not
programmeable.

Cheers
Martin

--

[Carey Sublette]

Post by Scott M. Kozel
I have read about the Central Fire Control (CFC) system and computer
that was utilized on the B-29 Superfortress, in many different books and
publications over the years.

...

Post by Scott M. Kozel
I work in the information technology field, and I know that the state of
the art for electronic computers in 1944 and 1945, was primitive, and I
am wondering how they managed to develop a computer to do all that
real-time processing, and fit it into a box that I've read was about 6
feet by 3 feet by 3 feet.

Well, from the state of the art of the field it is possible to immediately
guess that the computer system was analog, not digital. All embedded
computing/control systems of the era were analog, nearly all mechanical or
electro-mechanical.

According to:
http://www.historylink.org/essays/printer_friendly/index.cfm?file_id=3828
in this case it was an electronic analog computer, and claims it to be "the
first mass production and use of electronic computers ", which seems about
right.

Analog computers are inherently very fast. Each mathematical equation being
processed is encoded by its own electronic circuit, all of which process
input signals in parallel. Even more important is the fact that computation
time of each circuit is simply its stabilization time to changing input
values. Analog computers inherently tend to be real-time devices.

Carey Sublette
--

[h***@bbs.cpcn.com]

Post by Scott M. Kozel
I work in the information technology field, and I know that the state of
the art for electronic computers in 1944 and 1945, was primitive, and I
am wondering how they managed to develop a computer to do all that
real-time processing, and fit it into a box that I've read was about 6
feet by 3 feet by 3 feet.

Just speculating here...

My guess is that it was a hard-wired analog (as opposed to
digital) processor. Analog computers use variable ratios
to come up with solutions and were more prevelant in those years.

I believe battleships used similar technology for fire control,
though may have used mechanical analog rather than electronic.

Based on my reading of wartime technology, my guess is that the
accuracy resolution of such devices was rather poor. That is,
the unit, like the Norden bombsight, was a help and better than
nothing, but far from perfect.

I also wonder how reliable and responsive such devices were.
Hopefully the electronic tubes were made to withstand the rigors
of jostling in an airplane and the crew could replace burned out
ones (the B-29 could fly for many hours). Also I wonder if the
mechanical gun positioning gear was reliable and fast enough to
respond to the aim of the human gunner

I suspect for the B-29 the intent was to have all guns firing
at the same target at the same time to increaes the chance of hitting
it.

One writer said of the war that only a tiny percentage of shells
and bullets fired actually hit their desired target. It was more
of putting up a wall of fire to keep people from moving forward.
--

[Bob Matthews]

Post by Scott M. Kozel
I work in the information technology field, and I know that the state of
the art for electronic computers in 1944 and 1945, was primitive, and I
am wondering how they managed to develop a computer to do all that
real-time processing, and fit it into a box that I've read was about 6
feet by 3 feet by 3 feet.
That was a pretty amazing feat for that era.

How effective was it? Did the USAAF ever do any analyses? Didn't they
eventually yank it out so they could haul more gas and ordnance? Was it
effective in Korea?

==bob
--

[John Lansford]

Post by Bob Matthews
How effective was it? Did the USAAF ever do any analyses? Didn't they
eventually yank it out so they could haul more gas and ordnance? Was it
effective in Korea?

Much less so in Korea since the Chinese jets were faster than the
computer could keep up with. In WWII the computer, like everyone here
has said, was an analog device similar to the ballistic rangefinder
used on US battleships. Lots of gears that translated into the gunner
leading the target the proper amount. The Japanese quickly found that
their planes were not capable of easily bringing down the B-29, both
because it was flying so high and the heavy firepower it could bring
to bear on them. Later fighters began carrying more cannon and more
armor, but then the B-29's began flying at night on low level
firebombing raids. For those the guns were unneeded.

John Lansford
--
The unofficial I-26 Construction Webpage:
http://users.vnet.net/lansford/a10/
--

[Scott M. Kozel]

Post by Bob Matthews
How effective was it? Did the USAAF ever do any analyses?

That would be interested to see. One of the sources said that the tail
turrets accounted for about 75% of the fighters shot down, because of
the slower closing speeds from that angle, and because most of the B-29s
had a 20mm cannon in addition to the two 50-calibers in the tail turret.

Post by Bob Matthews
Didn't they
eventually yank it out so they could haul more gas and ordnance?

All the guns were removed except for the two 50-calibers in the tail
turret, but that was only for nighttime raids where the Japanese were
known to have virtually no night fighter capability. Without guns, the
CFC computers were unneeded.

Even then, all of the WWII B-29 models at least retained the tail guns.

Post by Bob Matthews
Was it effective in Korea?

One of the sources said that the system was too slow to target Mig-15
jet fighters when at full speed.

In both WWII and the Korean War, having good fighter cover is the most
important factor in protecting the bombers from enemy fighters.

The B-36 had a faster CFC system that could target jet fighters.

The chin turret on the B-17G was a predecessor of the B-29 CFC, and
utilized very similar technology. The bombardier had a control yoke to
guide and aim the guns.

[Merlin Dorfman]

Scott M. Kozel <***@comcast.net> wrote:

...

Post by Scott M. Kozel
There are plenty of sources that explain conceptually how the system
worked. My question concerns the computer itself, and I haven't been
able to find any details about exactly how the computer itself operated,
and what kind of architecture it had.

A quick web search doesn't turn up much. It was an electro-
mechanical computer, designed by General Electric, and apparently
produced in volume by Singer. This source seems to have more
information than the others I looked at:
<http://home.cfl.rr.com/featherweight/b29comp.htm>
--

[Scott M. Kozel]

Post by Merlin Dorfman
A quick web search doesn't turn up much. It was an electro-
mechanical computer, designed by General Electric, and apparently
produced in volume by Singer. This source seems to have more
<http://home.cfl.rr.com/featherweight/b29comp.htm>

I found that link several days ago; thanks!

I agree that a web search doesn't reveal much. I am looking for
technical details about how the computer operated; actually it appears
that there were 5 separate CFC computers, one in each turret.

This link has a photo of the computer, in the bottommost photo --

http://www.science.widener.edu/~schultz/b29/b29.html
.....

More details here --
http://www.ww2guide.com/b29ops.shtml

Remote-Controlled Gunnery System

Four companies competed for the contract including Bendix, General
Electric, Sperry, and Westinghouse. General Electric developed the
system used in the B-29 consisting of four turrets and a tail mount.

M5 Director General Electric fire control system

The bomber is provided with a system of electrically powered gun turrets
which are operated by a sighting system in order to concentrate a
broadside fire on enemy planes approaching from any angle. One of the
vital elements of this sighting system is the mechanical-electrical
computer whose function it is to make instant corrections for the speed
of the bomber, the speed and direction of the attacking plane, altitude,
temperature, windage, gravity, and the ballistics of the gun
projectile. In addition, it makes correction for parallax - the
distance between the sighting station and the remotely located guns.

Production of M5 Director by the Singer Company Elizabethport Works

Starting in September 1943, weekly production stepped up gradually,
reaching a rate of approximately 250 complete Computers per week at the
end of 1944, and over 500 per week in July 1945, at which time a total
of 20,794 complete Computers had been shipped out of the factory. About
one-third of these Computers have been Double Computers, incorporating
two computer mechanisms. In the Spring of 1945 the Double Computer was
discontinued but the cancellation of this quantity was partly
compensated for by some increase in the requirements of the Single
Computer.
....

[h***@bbs.cpcn.com]

Post by Scott M. Kozel
I agree that a web search doesn't reveal much. I am looking for
technical details about how the computer operated; actually it

appears

Post by Scott M. Kozel
that there were 5 separate CFC computers, one in each turret.

Some libraries may have old army Technical Manuals in their
collection and there may be some on this topic. It was probably
classified in the distant past, but not now.

You could check military libraries. If you can get access to
the inter-library loan central service (IIRC it's called OCLC)
they could do a better search for you.

Indeed, if anyone knows a good library source of WW II era
Technical Manuals, would you share it with us? I've seen
different isolated manuals here and there at various libraries,
but no coherent collection.

Post by Scott M. Kozel
Four companies competed for the contract including Bendix, General
Electric, Sperry, and Westinghouse. General Electric developed the
system used in the B-29 consisting of four turrets and a tail mount.

Some corporations maintain archives and may have information.
Sperry and Westinghouse no longer exist, though the Sperry archives
may be at the Charles Babbage museum. IIRC, Westinghouse archives
are stored someplace but I don't recall where.

Corporate archives are a fascinating source of historical
information, including the WW II era. Many companies published
books detailed their wartime contributions and make for interesting
reading about the home front. Unfortunately, due to the expense
and potential lawsuit liability*, corporations aren't as friendly
about public access to their archives as they once were. But a
call to GE wouldn't hurt.


(*People have gone through archives and sued corporations
over something that happened 50-60 years ago. The lawsuits lose,
but generate unfavorable publicity and needless expense. This
has ruined it for historical researchers.)
--

[Scott M. Kozel]

Post by h***@bbs.cpcn.com

Post by Scott M. Kozel
Four companies competed for the contract including Bendix, General
Electric, Sperry, and Westinghouse. General Electric developed the
system used in the B-29 consisting of four turrets and a tail mount.

Some corporations maintain archives and may have information.
Sperry and Westinghouse no longer exist, though the Sperry archives
may be at the Charles Babbage museum. IIRC, Westinghouse archives
are stored someplace but I don't recall where.

Thanks for the ideas... I see that the General Electric website has a "A
History of GE" section where they direct you to the "GE Hall of History"
--

GE maintains an archive of information and photography through the GE
Hall of History.
If interested, contact:
Hall of Electrical History
15 Nott Terrace Heights
Schenectady, New York 12308
518-382-7890

http://www.ge.com/en/company/companyinfo/at_a_glance/history_story.htm
....

I'm going to give them a call and ask about this. From the rather light
amount of material available about the B-29 CFC, I surmise that this was
an electronic analog computer that had the programming hard-wired into
the circuitry. Given that General Electric was the prime contractor,
that would seem to indicate that this computer was primarily electrical/
electronic in its architecture, perhaps with vacuum tubes for its logic
centers. The transistor was invented at Bell Laboratories in December
1947, so obviously a WWII computer did not utilize transistors or
integrated circuits.

One of the Internet links that I found was for a museum of analog
computers, but the B-29 CFC was not mentioned. The site said that
analog computers are somewhat short-shrifted in the historical record,
since they were almost completely replaced by the advent of digital
computers, that digital computers is a much more popular topic, and that
the effort by historians to detail the history of analog computers is so
far rather inadequate.

It may take some digging, but I am interested in finding more details
about the inner workings of this computer.

[Carey Sublette]

...

Post by Scott M. Kozel
One of the Internet links that I found was for a museum of analog
computers, but the B-29 CFC was not mentioned. The site said that
analog computers are somewhat short-shrifted in the historical record,
since they were almost completely replaced by the advent of digital
computers, that digital computers is a much more popular topic, and that
the effort by historians to detail the history of analog computers is so
far rather inadequate.

The short shrift of analog computers is quite true. For example, while it is
true that "they were almost completely replaced by the advent of digital
computers", for control applications - like the B-29 CFC - this didn't
happen until the 1970s, 30 years after WWII. And, when I say that digital
computers replaced analog computers for control in the 1970s, I don't mean
"that's when the last analog computers were retired", I mean that's when
they pretty much stopped fielding NEW analog computers for new applications!
Aircraft being built in the 1970s still used *mechanical* analog computers
for engine and flight surface control.

Analog computers are still in use. I think one reason why they have gotten
little attention of late is that with the rise of digital computing as the
commonplace standard model of computing, people stopped recognizing analog
systems as even being "computers."

Carey Sublette


--

[h***@bbs.cpcn.com]

Post by Carey Sublette
Analog computers are still in use. I think one reason why they have gotten
little attention of late is that with the rise of digital computing as the
commonplace standard model of computing, people stopped recognizing analog
systems as even being "computers."

True and good points.

I thought of example of an analog computer: the mechanical meter
on the front of gasoline pumps that calculates the total money due--
based on an adjustable price per gallon and the total gallons
purchased.
The higher the per gallon price the faster the total price dial moves
per gallon pumped.


They're pretty much obsolete now, replaced by electronic displays,
but IIRC they were out around the war years. Before them the attendant
filled a big glass jar with fuel--the jar had graduations on it
to indicate price. (The consumer also could see the gasoline was
clear, FWIW). Obviously the calculating pump was much easier and
faster. I'm sure plenty of older glass jar pumps were still in use.

(I know of one station still using the older pumps (from the 1970s),
apparently they're able to handle $2+ per gallon).

The pressure of war production and advanced weaponry created a number
of inventions during the war that had widespread consumer and
industrial
use afterwards.
--

[mike]

Post by h***@bbs.cpcn.com
I thought of example of an analog computer: the mechanical meter
on the front of gasoline pumps that calculates the total money due--
based on an adjustable price per gallon and the total gallons
purchased.

Another is that the older Automatic Transmission in cars and trucks,
before electronics overtook that job

The Valve Body takes a limited set of inputs, like the Engine Vacuum,
gearshift and throttle position, and RPMs from the Governor, has fluid
run thru its pre-programmed logic set by springs,check balls,pistons
and valves, and that activates hydraulic servos to put pressure on
different bands on the planetarys to change the gears at the right
time.

This adjusts the up and downshifts as engine RPMs and Load changes
to get a smooth shift all the time.

Hmm. WWII Content: General Motors(Oldsmobile,IIRC) built HydraMatic
automatic transmissions for use in Stuart and Chaffee Tanks.

**
mike
**
--

[h***@bbs.cpcn.com]

Post by Scott M. Kozel
GE maintains an archive of information and photography through the GE
Hall of History.
Hall of Electrical History
15 Nott Terrace Heights
Schenectady, New York 12308

Sounds like an interesting place to visit. Thanks for the reference.

Post by Scott M. Kozel
I'm going to give them a call and ask about this. From the rather light
amount of material available about the B-29 CFC, I surmise that this was
an electronic analog computer that had the programming hard-wired into
the circuitry. Given that General Electric was the prime contractor,
that would seem to indicate that this computer was primarily

electrical/

Post by Scott M. Kozel
electronic in its architecture, perhaps with vacuum tubes for its logic
centers.

I assumed it was electronic but analog computeres were built with
precision gears, too. GE built plenty of mechanical things, too.
(One common 1960s GE product was a strictly mechanical setable timer.
Photo labs used them. Our school had one in case the clock system
failed and we had manually sound the bell between class periods.
This timer had to be handwound. How GE ended up making something
like that I have no idea.)

Post by Scott M. Kozel
The transistor was invented at Bell Laboratories in December
1947, so obviously a WWII computer did not utilize transistors or
integrated circuits.

Transistors were difficult to manufacture in quantity and were
unavailable outside of a lab until the mid 1950s. However, some
semi-conductor research (things like diodes main from semi-
conductor material instead of a vaccum tube) were available earlier;
though I don't know if available during the war itself.

Post by Scott M. Kozel
One of the Internet links that I found was for a museum of analog
computers, but the B-29 CFC was not mentioned. The site said that
analog computers are somewhat short-shrifted in the historical

record,

Post by Scott M. Kozel
since they were almost completely replaced by the advent of digital
computers, that digital computers is a much more popular topic, and that
the effort by historians to detail the history of analog computers is so
far rather inadequate.

That's pretty much true. Analog computers were an important
research tool before digital computers came along. I believe they
were invented by Vaneuver (sp?) Bush, who became FDR's science
advisor during WW II. But compared to digital computers they were
rather limited.

It should be noted that the earliest digital computers were
extremely expensive, difficult to use, and very rare. After
the war there was a tremendous demand for computations as
the economy exploded and the Defense Dept sought more sophisticated
weapons like jet planes. To meet these needs, around 1949 IBM
developed a "poor man's computer", the CPC (Card Programmed
Processor). It was inspired by a customer and consisted of
IBM's 604 electronic calculator hooked up with some memory and
control circuits to allow it to do multiple and sophisticated
calculations. (The 604 by itself was also widely used and
welcomed). These machines were relatively cheap and fit in
with an organization's existing tab machine layout.

Also during the war years themselves, some IBM customers
modified their tabulating machines to create calculators--
this was done at Los Alamos for the Man. Project.

Post by Scott M. Kozel
It may take some digging, but I am interested in finding more details
about the inner workings of this computer.

They played a role in WW II. Please share your findings with
the newsgroup if you get anything.
--