B-29 Modifications

[Joe Broady]

For these raids, I'm curious if they removed the gun-computer, or kept that. From what I remember, 300 B-29's were built without the gun-computer installed.

Computers, plural. There were five sighting stations (nose, tail, top, left, right) and each had its own computer. The last three had "double parallax" computers because they could control two turrets simultaneously. These computers weighed 130 lb each. The single parallax computers for the nose and tail stations were 25 lb lighter. Four of the five computers were in a well under the gunner compartment floor. The nose computer was above the navigator instrument panel.

"When you are operating in high temperatures give the computer a break. Pull up the floor boards over the computer well for better ventilation." Then as now, computers didn't like to get hot.

B-29 Gunner's Information File (no date)
T. O. No. 11-70A-1, The Central Fire Control System, Handbook of Operation and Service Instructions, 5 December 1944

I think I found that first manual online, maybe at the Army Air Forces Collection. The second one I paid for and it's a monster at 620 MB.

 

[GrauGeist]

Did they fly 5000' from Tinian to Tokyo and back? Or did they cruise higher?

They would typically climb to about 19 - 20 thousand feet and cruise, once they hit a waypoint offshore that turned them towards Tokyo, they'd start their descent and level off at 5 thousand before target.
This helped both with fuel consumption and maintaining speed for their egress and climb back to altitude for the trip home without burning excessive fuel.

 

[Zipper730]

Computers, plural.

Understood

The single parallax computers for the

When you say parallax, does that mean the fact that the computer would compute lead and drop, or lead, drop, and the position of the turret relative to the plane's centerline? I figure if the turrets were pointed sideways that would make the bullets seem to curve off behind the aircraft quite a bit. On a fighter it's not as complicated as you have the guns pointing wherever you're flying...

They would typically climb to about 19 - 20 thousand feet and cruise, once they hit a waypoint offshore that turned them towards Tokyo, they'd start their descent and level off at 5 thousand before target.

This helped both with fuel consumption and maintaining speed for their egress and climb back to altitude for the trip home without burning excessive fuel.

Did they climb back to the same altitude (19000-20000') or to a lower altitude? I remember some profiles for the P-47N when used as an attack-plane, which would see it cruise in at altitude and egress fairly close to sea-level.

 

[GrauGeist]

Did they climb back to the same altitude (19000-20000') or to a lower altitude? I remember some profiles for the P-47N when used as an attack-plane, which would see it cruise in at altitude and egress fairly close to sea-level.

Because of the distances involved, the B-29's pilot had to keep an eye on fuel management, so they would climb back to an altitude that provided a better cruise setting.
The air's density at sea level would mean a higher consumption of fuel, plus a disadvantage if intercepted by Japanese elements.

 

[Zipper730]

Because of the distances involved, the B-29's pilot had to keep an eye on fuel management, so they would climb back to an altitude that provided a better cruise setting. The air's density at sea level would mean a higher consumption of fuel, plus a disadvantage if intercepted by Japanese elements.

I wasn't trying to suggest they cruised out at sea-level -- I was just trying to determine if they climbed to the same altitude they entered at, or a lower altitude.

The P-47N's ground attack profile called for a cruise to target at 10000' followed by an egress at sea-level. So I was curious if they'd cruise home at around 10000' maybe 15000'.

 

[Joe Broady]

Understood
When you say parallax, does that mean the fact that the computer would compute lead and drop, or lead, drop, and the position of the turret relative to the plane's centerline?

To quote the book, "PARALLAX - To compensate for the distance along the longitudinal axis of the airplane between the turret and the sight which is controlling the turret." (Handbook of Operation and Service Instructions, The Central Fire Control System, December 1944)

A similar principle is at work when a fighter's wing guns are harmonized. They are at different distances from the gun sight, so no one adjustment will get the bullets to converge on the same point at every range. But the B-29 computer can continuously compensate for parallax due to range and direction of the attacker. For example, the two upper turrets, both under control of the upper sighting station, will "toe in" slightly so their fire converges on the attacker.

 

[Zipper730]

J Joe Broady
So, if you turn the turret sideways, the reticle will not be displaced further to the side to factor in that the bullets will appear to fly sideways as well as forward?

 

[Joe Broady]

J Joe Broady
So, if you turn the turret sideways, the reticle will not be displaced further to the side to factor in that the bullets will appear to fly sideways as well as forward?

Not sure what you're asking, but according to these magazine articles from early 1945 the relative wind at 30000 feet and 250 mph deflects a .50 caliber bullet 12 yards sideways at a range of 800 yards when blowing at right angles to the line of fire.

Popular Mechanics Feb 1945
Popular Science Feb 1945

The pipper of a B-29 gun sight has a fixed direction with respect to the sight. It is not displaced to compensate for ballistics and target motion. The situation is completely different when a manned turret has a computing sight. That sight has a fixed relationship with the guns, so the reticle must be displaced to effect the corrections. For instance, it lags behind the attacking fighter (compared to a ring and bead sight). To put the lagging reticle on the target therefore forces the gunner to aim ahead.

But the B-29 sight is merely a means to inform the computer of the attacker's azimuth, elevation, azimuth rate, elevation rate, and range. The computer and servos in the turret do the aiming. An exception is an emergency mode you can select when the computer fails. Then the turret is slaved directly to the sight and the gunner must "aim off" with estimated corrections in the old fashioned way.

 

[BobB]

Going off on a bit of a tangent; a fellow who worked at a mod center near Denver said that he had a reputation for being good at fixing B-29 fire control computers. He said that as a last resort, he would take his shoe off and whack the computer with it. The computer was full of electro-mechanical relays and sometimes one was stuck but would get knocked loose by a whack.

 

[Zipper730]

Not sure what you're asking

Basically, when I said "displaced", I meant positioned to factor in lead: As described in the first link, at a distance of 800 yards, gravity will cause the bullet to drop 4.6 yards, so the reticle would be displaced so it matches where the impact point would be.

On a WWII fighter plane that had a lead-computing gunsight: The reticle would be set by selecting an aircraft's wingspan, and using a grip on the throttle to widen and narrow it, so as to encircle the target's wings. With the target's wingspan set, widening and narrowing the reticle infers range. With the aircraft in level flight, and 800 yards computed (something that would probably be useless on most aircraft, though the P-38, P-39, Mosquito NF/FB variants had centerline mounted guns), the reticle would be displaced such that it would produce an aiming point on the reflector site that would correspond to a location 4.6 yards below the centerline of the aircraft flying in a straight line.

With the aircraft pulling g-load (say a left hand turn), the increased vertical acceleration and directional changes would result in the reticle being repositioned down and to the right a greater degree so that the impact point is corrected for, when range is computed by widening and narrowing the reticle.

What I was curious was two things: Firstly, the definition of parallax, which seems to be factoring the difference between the sighting station and the position of the turret, and; whether the turret's gunsight would correct not just for g-load, but whether the reticle would be displaced as a result of the position of the turret relative to the centerline, and correct not just for drop/g-load, but also to the degree it was turned off the centerline. With the turret aimed left of the centerline, I was curious if the reticle would be displaced to the left as well as slightly down because that's the direction the bullets would appear to fly when you're pointing the gun sideways while flying at 250 mph.

The pipper of a B-29 gun sight has a fixed direction with respect to the sight. It is not displaced to compensate for ballistics and target motion.

So the reticle in the sighting station is pointed at the target, with the computer positioning the turret based on where the reticle is aimed?

 

[MIflyer]

When the B-29's were enroute to Japan in daylight they would sight vessels and provide reports on them. Trouble was, the USAAF crews proved to be very poor at identifying ships as to type and size. So some special "slide rules" were developed to enable them to use the gunsights to measure the size of the ships.

 

[pbehn]

Because of the distances involved, the B-29's pilot had to keep an eye on fuel management, so they would climb back to an altitude that provided a better cruise setting.
The air's density at sea level would mean a higher consumption of fuel, plus a disadvantage if intercepted by Japanese elements.

I didnt see this post when it was made maybe 'cos it was on my birthday, but as I understand it the pilot of a B-29 was aware of the issues of fuel management but it was the flight engineers job. The flight engineer on a B-29 took more training than the pilot from what I've read and most of the time he was telling the pilot what to do as far as airspeed and altitude goes.

 

[EAIAnalog]

This is for Zipper730. I would quote his post, but I can't figure out how to do that. Here is a visualization of the parallax problem in the gun turrets. Hold up both of your hands with the index fingers extended. The right hand represents the gun sight and the left hand represents the turret. The distance between the hands is the parallax distance. Now move your right hand to represent the gunsight tracking the target. Since the turret control system is designed to track the gunsight, the turret matches the gunsight exactly. However the turret is not pointing at the target. It needs a correction that depends on the distance between the gunsight and the turret. That correction rotates the turret so that its line-of-sight intersects the gunsight's line-of-sight (just rotate your left hand to see what I mean). The computer sits between the gunsight and the turret, and it makes that correction, and a host of others that have been mentioned above.

In the fighter example, the guns are fixed and the computer-assisted fixed gunsight moves the pip so the pilot can "move" his guns to shoot. In the B-29 the movable gunsight internals are fixed, and the computer-assisted turret moves to the target.

Weight reduction 1 turret installation: 2 amplidynes (30 lb each), 1 dynamotor (20 lb), 1 computer (130 lb), 1 gunsight (20 lb), 1 servoamp (10 lb), 1 turret installation (150 lb?), bullets (300 rounds per gun at .5 lb/round = 150 lb x 2). So ripping out 1 turret would knock off 690 lb. The turrets may have held 600 rounds/gun, which would make the weight 990 lb.

Modifications: I'm pretty sure that the aircraft came from the factory without the radar installation. Radar installation was done at modification centers. Because of the secrecy associated with radar, this seems reasonable. It's difficult to verify because most factory photos are from above, and the wings and hull screen the radar location. Another modification would be the installation of the electronic countermeasures equipment and the associated antennas.

 

[mjfur]

https://digitalcollections.museumofflight.org/items/show/46311

 

[MIflyer]

By the way, while the B-29's did use the low altitude night attacks on Japan they also continued high altitude daylight attacks. So presumably the entire force did not have the gun turrets removed.

I recall reading where B-29 pilot said, "I flew that great big beautiful bomber over that burning city, looked down into the flames and said, 'Who did you think you were messing with?'"

 

[EwenS]

LeMay's orders were to remove the guns, ammunition and gunners for the low level night fire raids on Japan. Note, no mention of removing the turrets, which would have been a much more onerous task. BUT it is not clear just how many Bomb Groups followed those orders in full. Many seem to have left the guns in situ. Most seem to have left the ammunition behind but 3 groups flew with ammunition in the tail turret except for their homing aircraft.

Note also from that Gunners Manual pages 19-21 just which sighting positions had primary and secondary control of the various turrets. Contrary to popular opinion, not every gunner could control each and every one of the turrets. So for example the side blister gunners had no control over the upper turrets.

The 311 lightened B-29B version was interspersed on the Bell Atlanta production line (small batches and sometimes single aircraft) with normal B-29. The normal AN/APQ-13 B-29 radar was replaced on these aircraft with the aerofoil shaped AN/APG-7 Eagle radar and the tail turret was fitted 3x0.5" guns with an AN/APG-15B radar control system. Most went to the 315th Bomb Wing which began to arrive in the Marianas at the end of May 1945, flying their first mission a month later. The Eagle radar needed a 70 mile bomb run but was very accurate. So the group flew at night or in bad weather to avoid interception. Their principal targets were oil related, refineries and storage facilities on Honshu, earning them the nickname "The Gasoline Alley Boys".

 

[SaparotRob]

I didnt see this post when it was made maybe 'cos it was on my birthday, but as I understand it the pilot of a B-29 was aware of the issues of fuel management but it was the flight engineers job. The flight engineer on a B-29 took more training than the pilot from what I've read and most of the time he was telling the pilot what to do as far as airspeed and altitude goes.

Happy birthday!

 

[manta22]

In the '70s a forward section of a B29 fuselage arrived in a local Tucson salvage yard. Nose art identified it as "The Dutchess" and the salvage yard owner maintained that it had been one of the planes involved in the A-bomb program. If this were true it should have been placed in the Pima Air Museum or some other place to preserve it. I undertook to try to trace The Dutchess' history and talked to Paul Tibbets on the phone for 45 minutes about it. Tibbets was friendly and quite helpful and gave me some things to check to see if it had been one of the 509th Composite Group's planes. The most significant feature was the single bomb-bay rather than two shorter bomb bays on a standard production B29. He also related that many other changes were also made to the Wichita production run of these special versions. He gave me Chuck Sweeney's phone number and I talked to him as well. He did not know of a B29 named The Dutchess but said that it might have been one of the Los Alamos research planes.

I never did find out the history of The Dutchess B29 and it disappeared from the salvage yard thereafter.

Update: In going through some old paperwork, I found a reply from the Air Force Museum to my inquiry about a B-29 named "The Dutchess". The penciled notes are what I wrote while I was on the phone with Paul Tibbets. I thought I would pass this along as someone may be interested.

 

[Zipper730]

First,
I'm not sure what post this was on, but I remember there being issues with the cowl-flaps causing very large amounts of drag when extended (as well as control problems). While I've heard of other aircraft having control problems with the cowl-flaps out, the B-17 seemed have less issues at high altitude as the B-29 when it came to fuel efficiency (despite a similar climb-rate).

What in particular was wrong with the cowl-flaps?

Second,

The 311 lightened B-29B version was interspersed on the Bell Atlanta production line (small batches and sometimes single aircraft) with normal B-29. The normal AN/APQ-13 B-29 radar was replaced on these aircraft with the aerofoil shaped AN/APG-7 Eagle radar and the tail turret was fitted 3x0.5" guns with an AN/APG-15B radar control system. Most went to the 315th Bomb Wing which began to arrive in the Marianas at the end of May 1945, flying their first mission a month later. The Eagle radar needed a 70 mile bomb run but was very accurate. So the group flew at night or in bad weather to avoid interception. Their principal targets were oil related, refineries and storage facilities on Honshu, earning them the nickname "The Gasoline Alley Boys".

Wait, at 300 mph that would take 14 minutes for the whole bomb run... what kind of CEP did that have compared to the traditional Norden bombsight, and the LORAN aid (I think they were used in the pacific theater).