Would B-29s have been better with R-2800s?

[stepwilk]

Ignoring for the moment the fact that it probably simply wasn't possible to build that many R-2800s for as many U. S. airplanes that needed them, and that it's never a good idea to rely on a sole-source engine supplier, would the trouble-plagued Boeing B-29 have been a more successful (i.e. more reliable) airplane if it had Pratt R-2800 engines?

Certainly the Wright R-3350 seems to have been plagued by near-intractable reduction-gear and engine-cooling problems right from the outset of its development, while the R-2800 ended up as reliable, and as pound-for-pound powerful, as a big-block Chevy Rat Motor...

Opinions?

[barit1]

There's a substantial bit of discussion on the big radials here (http://www.pprune.org/tech-log/409778-were-any-big-radials-reliable.htmlhttp://www.pprune.org/tech-log/409778-were-any-big-radials-reliable.html) and here (http://www.pprune.org/aviation-history-nostalgia/405281-largest-piston-airliner.html).

[Rory57]

I think the B29 would have needed at least six R2800 to replace four R3350. As soon as time allowed the R3350 was of course replaced by the P&W R4360.

Reduction gear design and manufacture, mixture distribution, cooling fin design, crankcase design, and, unbelievably, a cowling that could not be removed to allow engine maintenance without first removing the propeller.

There is some first class discussion of the many difficulties with this engine and some with the R2800 in the AEHS journal: see AEHS Home (http://www.enginehistory.org/)

(Look in particular for the articles by Kevin Cameron on the R3350s failings)

[stepwilk]

Thank you, Rory--excellent resource. Just joined the AEHS and I'm sure will make good use of it for an article on B-29 problems that I'm doing for Aviation History Magazine. Appreciate your help!

[stepwilk]

Let me ask, before printing something stupid: are we all convinced that Rory57 is correct in posting, above, that the B-29's cowl "could not be removed to allow engine maintenance without first removing the propeller"?

I'm looking at a B-29 photo, a general shot and not a nacelle closeup, that seems to me to show enough Dzus fasteners to remove the forwardmost cowling section, and certainly the panels aft of that.

But I could be totally wrong. Anybody know better?

[Brian Abraham]

stepwilk, don't be so quick to give the 3350 a bad wrap. Certainly had issues to start with, as did the 2800, but matured into a dependable form of motive power. The expediency of war time had it rushed into service on an aircraft that was itself a technological leap forward. Never put a new engine in a new airframe used to be the addage. That the engine lasted 100 hours in the demands made upon it getting a fully loaded 29 to altitude and doing it following a take off on a diet of coral dust in the tropical heat gets applause from this quarter.

The prop didn't have to be removed. Would have made it difficult to carry out the interminable maintenance that needed to be done - plug changes (following every flight at times), wrenching up dripping oil connections etc. A good film doco of Daryl Greenmeyers attempted 29 (Kee Bird) recovery from Greenland is available out there which gives a good insight into engine maintenance, cowling removal etc

[Jhieminga]

On the early B-29/R-3350 installations the prop did have to be removed to get the cowling off. This was later modified, which is why the fasteners are visible on most photos. I haven't got a reference handy for this right now but will try to look it up.

[Brian Abraham]

In August of 1939, Boeing had started work on the Model 341 project, which featured a new high-lift aerofoil for a high aspect-ratio wing of 124 feet 7 inches in span. The Model 341 offered a maximum speed of 405 mph at 25,000 feet. It was to have been powered by four 2000 hp Pratt & Whitney R-2800 radials. Weighing 85,672 pounds, the range was to have been 7000 miles with one ton of bombs. A maximum load of 10,000 pounds could be carried over shorter distances. The design was subsequently reworked to become the Boeing Model 345 aka B-29.

Boeing XB-29 Superfortress (http://www.joebaugher.com/usaf_bombers/b29_1.html)

[FunkyStick]

Having spent many hours pulling on and off the cowling of the B-29 Kee Bird, I can assure you that the propeller was indeed in place. Granted, this was a later model B-29 known as the F-13. By my experience, I can not even IMAGINE that there was ever a version that required pulling the prop first. Those engines needed way too much tender loving care. Needing to pull the prop just to change spark plugs is just totally illogical in my limited experience.

[stepwilk]

FunkyStick, you may be "new here," but you're sure a player. What an epic you were part of, sad as it turned out. You have my admiration.

[Brian Abraham]

A big :ok: Funky. The video is a treasured possession and the ending certainly came out of left field. Enough to make a grown man cry, particularly given the effort and deprivation suffered. And that's not to mention the death of a hard toiling colleague.

[Rory57]

Still looking for original evidence of my assertion about B29 propeller removal for maintenance. Bill Gunston refers to it briefly in Piston Aero Engines but I know I have read something at length about it, referring to the first deployment. (India / China?) Meanwhile, have a look at this superb set of B29 photos on Flickr.
B-29 Superfortress - a set on Flickr (http://www.flickr.com/photos/18532986@N07/sets/72157607351430364/)

[treadigraph]

FunkyStick, the DVD of the Kee Bird saga is one of the best and, sadly, one of the most poignant films I've ever seen. I really wish that Darryl had pulled it off... And as Brian says, would that Rick was still with us. He came across as a really nice bloke.

[Brian Abraham]

Rory, be very interested in a source for your prop removal assertion. Can find no mention anywhere. The 29 AFM includes an item about turning the prop by hand to check for hydraulic lock during the pre flight, and details the procedure for determining which is the offending cylinder, then removing the spark plugs on that cylinder, draining the oil and installing new plugs. It notes this check should be made early in the pre flight, and remedial action taken should a lock be found, so that the scheduled take off time can be met. To my way of thinking, if it was necessary to remove the prop there is no way that a lock could be fixed in short order. I suspect it may be a myth, perhaps generated by the fact that the props were the last thing to be fitted on the production line.

[Jhieminga]

Brian, I'll have a go tonight at finding a reference for that. I'm sure I've read it in more than one place but all my books are stored in boxes in the basement. That doesn't speed up the search unfortunately.

[stepwilk]

That's an excellent point: how can you routinely remove a plug to drain out the oil if you have to remove the entire prop first?

[FunkyStick]

<<To my way of thinking, if it was necessary to remove the prop there is no way that a lock could be fixed in short order. I suspect it may be a myth, perhaps generated by the fact that the props were the last thing to be fitted on the production line.>>

It was SOP to pull those props thru by hand when we were in Greenland. It was pretty routine to have a pull a spark plug for any engine start. And no props came off.

It amazes me to this day just how much interest this project created. For years I had non-aviation people tell me that 'I just wanted to cry at the end' of the program. The show did a good job of telling the story, but so much never made it to air for a variety of reasons.

By the way, I was a NOT a mechanic. I became an assistant to the assistant to the lead mechanic. I did the kind of jobs that didnt take brains or knowledge. It was an amazing experience that I"m glad I lived through but I'm not sure I would do it again. We were very lucky in many ways.

[stepwilk]

R-3350s were particularly susceptible to the lowest-cylinder oil leakage that created hydraulic lock. Certainly R-2800s weren't routinely pulled through on P-47s and all the rest of the airplanes that carried them.

And as you of all people doubtless know, an eight-foot-long prop blade can actually provide enough leverage that two guys pushing too hard on it can bend a rod. You don't even have to start the engine.

[FunkyStick]

Treadigraph
<<He came across as a really nice bloke.>>

I didnt "fit in" the group when I first arrived. I was a journalist, an environmentalist, not "a wrench" and not the TV crew. I was, to a lot of folks, deadweight. But when I returned to the Kee Bird from Thule, after passing up the opportunity to return home early, Rick made a point to tell me he was glad I was back. It was one of my favorite moments of the entire experience. Rick, to me, was an artist with a wrench. He amazed me time and again with his ability to solve problems and he never, ever, raised his voiced at anybody. Gaining his respect and appreciation was one of the best things that happened to me.

When the Kee Bird burned, I didnt shed a tear. When Rick was taken away on that stretcher, I put down my cameras and Cecilio (my "boss" and Rick's assistant) and I stood at the nose of the Caribou and we cried unashamedly.

Rick is, and always will be, the real tragedy of the Kee Bird.

[stepwilk]

I know you, I think.

[Brian Abraham]

Certainly R-2800s weren't routinely pulled through on P-47s and all the rest of the airplanes that carried them
Sloppy procedures if that is indeed the case. From 2800 notes - Up to 10 gallons of oil were trapped in the system following shutdown despite it being a dry sump. Some of this oil will inevitably find its way into the lower cylinders through leakage past the piston rings and oil trapped in the rocker boxes leaking through the valve stems. Being incompressible any excess oil in the lower cylinders has the potential to wreak havoc inside the engine. Bent connecting rods are not uncommon. As a guard against this kind of catastrophe the 2800 has a clutch on the starter, the aim of which is to prevent damage in the case of a hydraulic lock. The fierce acceleration induced into the engine upon engagement of the starter can cause heavy internal loading, and that is assuming nothing is amiss such as a hydraulic lock. Even with the protection of the clutch, serious damage can result if the engine is hydraulic locked.

[swordfish41]

I was the director and producer of the documentary about the Kee Bird and Darryl. It exists in two versions one for NOVA in the US and one for Channel 4 in the UK which was broadcast with the above title. I think I know Funky Stick also. Brian Abraham is right about the need to turn the engines, and in the second trip when Darryl wanted to use the frozen lake as a runway the props were being turned every half hour it seemed to me, and i can definately confirm that it was possible to remove the plugs to drain the lower cylinders without removing the props. When the Kee Birds Squadron first started operating from Alaska there were several crashes after takeoff. It was standard procedure in the Pacific theatre to open the cowl flaps fully on take off to prevent the engines from overheating. This had the reverse effect in Alaska so after the first crashes they took off with cowl flaps closed. There are of course four of these engines still up at the crash site in Greenland with very low hours on them. I know at least one person who has been up there recently to take a look. The saga continues.
sf41

[Brian Abraham]

It was standard procedure in the Pacific theatre to open the cowl flaps fully on take off to prevent the engines from overheating.
Someone was smoking something when they told you that swordfish :E As you say overheating during prolonged ground operation was a problem. Take off was not permitted once the cylinder head temp had exceeded 220°C. The procedure was to set the cowl flaps at 15° upon entering the runway and reducing to 7.5° or less by the time the aircraft leaves the ground. This permits rapid increase of airspeed and should keep the cylinders below 260° (max for take off 260, climb 248, cruise 232). If 260 is exceeded cowl flaps may be opened to a maximum of 10°, larger openings provide little if any additional cooling. Power may be reduced on the offending engine but not until the aircraft has reached 170mph.

To illustrate what might happen with a take off with the cowl flaps open, a Boeing Stratocruiser (basically a passenger version of the B-29) ditched in Puget Sound following take off from Seattle because of such an error. The open cowls create a huge airbrake and the turbulence produced severe vibrations, loss of lift and lateral instability, to the point the Captain thought he might lose control and this prompted his ditching. These symptoms however did not show themselves until the wing flaps were retracted from the take off setting of 25° and passing through the 10° position. The crew thought they may have had a asymmetric flap condition. Subsequent tests showed that the effect of the open cowl flaps on aircraft performance was the same as if an engine had failed and the cowl flaps were in their normal closed position. The root cause of the accident was thought perhaps to be the engineer who was mainly experienced on the Constellation, in which the switches operated in the reverse order to the Stratocruiser.

http://i101.photobucket.com/albums/m56/babraham227/b0052.jpg

[Feather #3]

In case it makes a difference, didn't the Strat have R-4360's?

G'day ;)

[Brian Abraham]

Affirm, as did the re-engined B-29 which became the B-50. The B-29 contains a cautionary note regarding severe vibration with the cowl flaps open in flight. As an aside every degree of cowl flap above 10° opening increased fuel consumption by at least 15 gallons per hour. With four engines and flying 16 hour missions to Japan it took precise flying to get the most out of the aircraft. Reducing drag (intercooler doors and cowl flaps) is a prime concern through out the manuals because of its tremendous impact on performance. Cowl flaps open 6° reduced speed by 10mph and so expected cowl flap cooling requirement had a considerable effect on flight planning.

[swordfish41]

Brian,
thanks for the clarification. I'm not an engineer obviously, so take your point about the cowl flap settings in the Pacific. However the information about the early crashes in Alaska, and the reason for them was given in a book about the history of the squadron. What I understood, (and what do I know) was that the temperatures were so low that the normal setting of the cowl flaps prevented the engines from delivering the necessary power. I think I still have the book, some where in storage. This may of course have been nonsense, there was a great deal about the official history of the Kee Bird that I found to be erroneous.
sf41

[Brian Abraham]

swordfish, I wonder if the following contains a germ around which your story may have been based? Is it possible they got airborne with the flaps closed and cooked things?

Cold Weather Operations
Below –23°C the cowl flaps should be closed soon after the engine fires regularly, as they will assist in raising the cylinder head and oil temperatures. When either the oil temperature reaches 70°C or the cylinder head temperature reaches 170°C the cowl flaps should then be opened.

Regardless of the degree of cold weather encountered, the cowl flaps should be opened one third (10°) for take off.

there was a great deal about the official history of the Kee Bird that I found to be erroneous
And that's why historians will never be out of a job. Even the participants in an event come away with differing perceptions as to what went on quite often.

[swordfish41]

Brian.
No. What I think was that it was so cold that once the aircrfat had started its takeoff run the engine temperatures dropped. Thad Dulin, one of the flight engineers on Fifi was in the engineers seat when the Kee Bird started moving, be interesting to talk to him. But if you are prepared to wait a couple of days I will try to hunt out the book I am refering to. Seems we are the only people interested in this thread so I might pm you.
sf41:ok:

[henry crun]

swordfish41: You are not the only ones, there are other like myself who find this thread interesting but do not know enough to comment intelligently, so please continue to post in public.

[FunkyStick]

<<Regardless of the degree of cold weather encountered, the cowl flaps should be opened one third (10°) for take off.>>

......this is a minor point in the grand scheme of things, but in hours of conversation with the co-pilot of a B-29/F-13 that crashed in Fairbanks due to the intense cold during takeoff, I was told that after his crash cowl flaps were to be kept closed during intense cold operations. The temp where the cowls were to be completely closed escapes my memory at the moment... but I know at -30 in Fairbanks strange things start happening to cars and trucks. At -20, my cameras stopped working regardless of how warm the batteries were.

On this particular crash during the night of December 11, 1946 it was so cold (-57F) that the fuel didn't vaporize and so there was no initial explosion. The copilot literally unbuckled and walked away because so much of the cockpit around him was gone. As the temperatures increased during the day, the fuel began vaporizing and a significant fire erupted eventually reducing the wreck to a smoking hulk.

It needs to be pointed out that at the time, B-29/F-13 operations in that kind of cold had not been experienced. The crash I just described was the first for the squadron. Despite its recon mission, the squadron was something of an Arctic guinea pig and the "book" had to be modified to reflect local operating conditions.

In an interesting bit of trivia, the Kee Bird successfully took off moments behind the crashed aircraft and unknowingly overflew the crash because of ice fog in the area. The pilot of the crashed aircraft would eventually be part of the crew of the Kee Bird when it went down in Greenland.

[swordfish41]

I'm about to go and search for the book about the Kee Birds Squadron, but Funky Sticks post chimes very well with what I remember. I have a feeling that in fact there were three crashes all told before it was realised that the cowl flap settings might be a contributory factor.
sf41

[swordfish41]

Well that was a wasted three hours. I am afraid after searching through boxes of production files I couldn't find the book I was looking for. Funky sticks account of what happened in Fairbanks is going to have to be the best source for the effect of cold on the B29. Not sure the original question of this thread has been answered.
sf41

[stepwilk]

Actually, I got a very good answer to the original thread question, which I posed, from an acquaintance who saw it and answered me privately by e-mail.

In short, the answer is that the R-3350 was an incredibly fuel-economical engine--probably one reason it had such cooling problems, since it was burning no extra gas for combustion-chamber cooling--and my friend had done a number of SFC calculations for various R-3350-versus-R-2800 situations that showed that only at very low power settings did the R-2800 fuel economy match and slightly exceed the R-3350's. In order to come anywhere near matching the R-3350's typical mission output, the R-2800 would have had to burn substantially more fuel and the B-29 would no longer have been a very-long-range bomber.

[swordfish41]

I had no idea that the B29 was considered to be unreliable, or indeed what that means. I have just written a book about a Lancaster bomber crew, who in their first thirty missions experienced four seperate engine failures, (RR Merlins) involving five individual engines. The flight engineer didn't seem to think this was out of the ordinary, and I suppose it wasn't.
sf41

[stepwilk]

The standard joke was that Curtiss-Wright killed more B-29 crewmen than did the Japanese. Unfortunately, it wasn't a joke: 147 B-29s were lost to enemy fire, and 267 were lost to "operational causes"--typically engine fires and failures that led to either the destruction of the airplane in the air or crashes on overloaded takeoffs. Do the math and that shows that nearly two B-29s were lost because the airplane was basically unreliable to every one lost to Japanese fighters or antiaircraft.

[swordfish41]

Point taken

[Jhieminga]

Let me ask, before printing something stupid: are we all convinced that Rory57 is correct in posting, above, that the B-29's cowl "could not be removed to allow engine maintenance without first removing the propeller"?

I'm looking at a B-29 photo, a general shot and not a nacelle closeup, that seems to me to show enough Dzus fasteners to remove the forwardmost cowling section, and certainly the panels aft of that.
I finally got around to finding the relevant book "B-29 Superfortress in detail and scale, part 1", by Alwyn T. Lloyd (which is the only B-29 book I've got at the moment). From page 17:
The cowlings were comprised of a fixed cowl ring, a set of side panels, and ten moveable cowl flaps. On early engines, the propellor, cowl ring, and side panels had to be removed in order to change the 36 spark plugs. Later cowls were manufactured with removeable side panels - a feature which was also retrofitted.And on page 47:
http://www.vc10.net/div/Engine.jpg

There are some photos in the book of early B-29s and they do not appear to have the dzus fasteners on the cowling side panels. The photos are not clear enough to be certain but I'm inclined to believe it. Here is no.2 engine from a YB-29:
http://www.vc10.net/div/Engine2.jpg

[Brian Abraham]

Thanks for that Jhieminga, I must confess I find it hard to believe that the Americans, with all their experience to draw on, and engineering talent, would put out such a bastard child maintenance wise.

[Rhino1]

You might check and post the question on the Warbirds Information Exchange (WIX to many). There are several guys there who work on FIFI.

Ryan

[Brian Abraham]

Unlike a turbo charger a power recovery turbine has a coupling to the crankshaft. That is, instead of the turbine driving a compressor to pump air to the cylinders it is connected to the crankshaft.

[stepwilk]

Brian, is there some kind of hydraulic clutch/transfer case between the impeller and the gear drive to the crank? It's not pure direct-drive, is it?

[barit1]

I believe the turbocompound engine had direct coupling of the 3 turbines through quill shafts to reduction gearing to the aft end of the crankshaft.

No hydraulic coupling.

See p.18 of this brochure (http://www.enginehistory.org/Wright/TC%20Facts.pdf)

[stepwilk]

Wonderful booklet! Thank you for posting it.

[Rory57]

Thank you Jhieminga, I was beginning to believe I had imagined all of this. It does seem hard to believe that such a mistake could have been made but it is typical of what can happen in the first iteration of engineering design projects: that's why the development stage is as important as the design stage.
We have to remember the context of the design of the B29: 38 years after the Wright Flyer, war everywhere, every design and engineering resource stretched to a level we can nowadays only imagine, the most complex aircraft ever, and by a huge margin too. An astonishing achievement.

Barit1, Stepwilk, there IS a hydraulic coupling between the power recovery turbines and the crankshaft, see fig11/page 18 and text on page 19 of the brochure. The coupling is the item with radial fins on the left of fig 11. It couples the large bevel gear to the pinion that then drives the large diameter gear (splined to the crankshaft?)

[stepwilk]

Oh,good...I couldn't imagine how those things could transmit direct-drive and stay in one piece.

When I sell my racecar, I'm thinking my next project might be rebuilding an R-3350 to run on a motor-mount trailer. Must be a number of runnable by unairworthy examples still around...

[Brian Abraham]

Reproduced in the event anyone missed it.

Fluid Couplings
Intermittent exhaust pulses impinging on the turbine buckets tend to set up torsional (twisting) vibration in the turbine drive system. Fluid couplings are used to absorb this vibration and prevent it being transmitted to the crankshaft. At the same time the couplings prevent crankshaft vibration from being transmitted to the turbines, and help absorb the inertia loads during speed changes. The turbo compound fluid coupling utilises a vortex principle which clears the couplings of sludge by swirl action.

The design size of the coupling limits normal slippage to between one and two percent.

Each of the three couplings is supplied with oil from the main engine oil supply through an oil pressure regulating valve located in the supercharger front housing. The pressure is set at 50 psi at the factory or overhaul facility. No fluid coupling pressure gage is used in the aircraft.

When the engine is shut down, oil gradually drains from the couplings so they are momentarily disconnected on a cold engine start until oil pressure is obtained. Because of this, engine starting takes no more power than with a non compound engine.

[411A]

Having flown DC-7 and 1649 Constellation types, I found that the turbocompound engines were a pain at times.
Yes, the engine was very fuel efficient, however, PRT life was many times limited to about 200-300 hours.