Supercharger Development & Aircraft Design Policy (USAAC)

[drgondog]

According to documents that Joe Yancey has, the USA owned the engine designs that were paid for by US funds. So, the Allison V-1710 design was owned by Congress and any changes to the design were approved by Congressional vote. That's why when the Army ordered, say, a V-1710-31, the components were to be delivered as a V-1710-31, not some improved version. On several occasions, Allison asked the government if they wanted a 2-stage V-1710 and they were told each time, "No."

Allison, being a relatively small company, and even a small division after GM bought them, didn't have the resources to develop advanced versions on their own, and the government got exactly what they ordered. The research that WAS done was paid for piecemeal by the government. Allison finally DID develop an auxiliary stage on their own, but that was after asking several times for funding for a more advanced development and being denied. The aux stage was a stand-alone addition, not part of the engine. People may wonder why, and it is because they were not allowed to modify the design without approval.

The Packard Merlin engine team was never saddled with government oversight like the Allison was because the US government didn't own the Merlin design. They were free to add improvements as they came up. All in all, Allison didn't do badly at all considering the bureaucratic red tape that any improvement necessitated.

Most people have no idea of amount of paperwork involved. When Ben Kelsey wanted to just increase allowable boost, the paperwork required almost a six-month study and recommendations from several levels of authority. Allison was a good company and still is. I doubt they ever want to allow ownership of one of their designs by anyone ever again!

Hal Dupont led a GM post mortem dive into NAA electing to go with Merlin for the P-51B and the basic findings weren't pretty for Allison primarily because Hunt (Ch Allison) couldn't give GM a good reason for NOT asking for separate funding to develop the integral 2s/st supercharger - which would be expected to fit w/o massive airframe mods by Curtiss, Bell and NAA for the P-30, P-9 and P51. Series of letters and memos November 26 through December 10, 1942.

 

[P-39 Expert]

Hal Dupont led a GM post mortem dive into NAA electing to go with Merlin for the P-51B and the basic findings weren't pretty for Allison primarily because Hunt (Ch Allison) couldn't give GM a good reason for NOT asking for separate funding to develop the integral 2s/st supercharger - which would be expected to fit w/o massive airframe mods by Curtiss, Bell and NAA for the P-30, P-9 and P51. Series of letters and memos November 26 through December 10, 1942.

The two stage Allison with the separate auxiliary stage would have fit just fine with minor modifications in the P-39. And they actually put one in the P-40Q for trials. Don't see why it couldn't have been put in the P-51.

 

[Zipper730]

Actually, the main supercharger had a fixed gear ratio to the engine, while the auxiliary supercharger had high, low and neutral.

I'm not saying I disbelieve you -- I'm just surprised.

The Lycoming H-2470 was two O-1230s joined together.

I know that, I'm just listing various projects that I'm aware of at the time.

The H-2600 or X-1800 was intended for Army types. I'm not sure if an H-2240 existed, though that's what the Sabre would have been.

I thought both were funded by the Navy, though P&W submitted the ideas of their own initiative. I stand corrected, however.

The H-3120 and H-3730 were based on the X-1800 design.

Makes sense, sometimes two engines were produced of different sizes to cover different applications: In this case, it's large and gigantic.

Of those listed, only the Continental I-1430 had direct Army involvement in the design and development of the engine. The I-1430 was never a monoblock design.

No, I got that it wasn't a monoblock, I'm just surprised that they didn't change the specs to require it to be a monoblock: From what I recall...

1. Initially they wanted individual cylinders because it played a role in getting better engine performance
2. They made the individual cylinder
3. They made a 1008 in^3 cylinder inline prototype

I'm not sure exactly when the latter was produced, but I figure it wouldn't have been hard to have switched to mono-block somewhere between that point and the decision to go from flat-cylinder to inverted vee. It would have made the engine more compact, and lighter, both are important factors in aircraft design. I don't really think you need to know this, because I think you know it better than I do, but I'll throw it out there.

The Army had complete control of the specs of that engine, so if they wanted a monoblock they could have made it a monoblock.

So, they actually *did* have the latitude to change it at any point? I'm wondering why they didn't -- was it an example of mental inertia?

It was the Materiel Division.

Forgot about that detail...

 

[DarrenW]

Didn't the F6F have the engine mounted so that it canted nose-down a few degrees (?)

This would have made it fly more nose-up in S&L flight compared to just about all other aircraft where the engine was mounted at 90 degrees(?) to the vertical (bulkhead).

Yes that is true. Down-thrust helped mitigate the pitch excursions that occur with changes in engine power settings. This can be very helpful during carrier landings when there's a botched landing attempt and the pilot is forced to apply more throttle for another go around. It allowed the plane to not inadvertently nose up and stall.

 

[drgondog]

The two stage Allison with the separate auxiliary stage would have fit just fine with minor modifications in the P-39. And they actually put one in the P-40Q for trials. Don't see why it couldn't have been put in the P-51.

Because the wing would have to move two feet forward, cockpit and firewall with it... to name a few design/tooling issues. NAA studied it in late 194q/early 42.

 

[wuzak]

Actually, the main supercharger had a fixed gear ratio to the engine, while the auxiliary supercharger had high, low and neutral.

I'm not saying I disbelieve you -- I'm just surprised.

It would be pointless to have 2 sets of gear change systems for the 2 stages of supercharging, as it would weigh more. The two options, then, are too have both first and second stages on the same driven shaft and rotating at the same speed at all times (the Rolls-Royce Merlin/Griffon system), or have a fixed main supercharger stage and multi-speed auxiliary stage (teh system favoured by Pratt and Whitney and Allison).

 

[wuzak]

No, I got that it wasn't a monoblock, I'm just surprised that they didn't change the specs to require it to be a monoblock: From what I recall...

1. Initially they wanted individual cylinders because it played a role in getting better engine performance
2. They made the individual cylinder
3. They made a 1008 in^3 cylinder inline prototype

I'm not sure exactly when the latter was produced, but I figure it wouldn't have been hard to have switched to mono-block somewhere between that point and the decision to go from flat-cylinder to inverted vee. It would have made the engine more compact, and lighter, both are important factors in aircraft design. I don't really think you need to know this, because I think you know it better than I do, but I'll throw it out there.

1 is not correct. The Hyper cylinder development started with a modified Liberty L-12 cylinder. The L-12 was a WWI design, complete with separate cylinders.

Single cylinder engine development is commonplace because it is cheaper o build the prototype and quicker to make changes. Only after the engineers are satisfied with the performance of the single cylinder will they move to a multi-cylinder configuration.

(Note: Rolls-Royce did not do any single cylinder development on the Crecy, though it was based on the work of Harry Ricardo with his single cylinder mule. Rolls-Royce had come up with a different way of driving the sleeves, so they built V-2 test engines so that the mechanism could be evaluated as they tested their engine design.)

According to Wiki, the first multi-cylinder test engine in the O-1430/I-1430 program was an opposed twin. This would have been used to test the connecting rods, and maybe some other details.

Then they made the full engine.

I'm not sure that the engine ran in its 1,008cid horizontally opposed configuration, or the O-1430 configuration.

The idea behind individual cylinders was not an aid to performance, but for ease of manufacture and transitioning from single cylinder mule to the full engine. It was on less design problem to solve.

 

[Shortround6]

The two stage Allison with the separate auxiliary stage would have fit just fine with minor modifications in the P-39. And they actually put one in the P-40Q for trials. Don't see why it couldn't have been put in the P-51.

They put several in XP-40Q(s) for trials. there were 3 XP-401 aircraft.
Minor modification was a 2ft difference in length (to the inch) from the P-40K and Ns they were modified from.

They did put one in a P-51, The XP-51J, engine reliability was such that the aircraft was loaned to Allison for engine development.
Since the light weight P-51 series did differ from the standard P-51s and since there seems to be little recorded about the XP-51J we don't know a whole lot about the difference between the XP-51J and normal Mustangs, however the photos show a rather different aircraft than the older Allison Mustangs.

 

[BiffF15]

They put several in XP-40Q(s) for trials. there were 3 XP-401 aircraft.
Minor modification was a 2ft difference in length (to the inch) from the P-40K and Ns they were modified from.

They did put one in a P-51, The XP-51J, engine reliability was such that the aircraft was loaned to Allison for engine development.
Since the light weight P-51 series did differ from the standard P-51s and since there seems to be little recorded about the XP-51J we don't know a whole lot about the difference between the XP-51J and normal Mustangs, however the photos show a rather different aircraft than the older Allison Mustangs.

View attachment 567811

Loolks very similar to a P51H. Taller tail, smaller gear, more streamlined appearing nose...

 

[GregP]

Wing root dogtooth also missing like P-51H. Since this was a North American project, I'm sure they shared aero data among the lightweight P-51s.

Joe's documents show Allison DID ask for 2-stage S/C funding and was denied twice, perhaps more, perhaps not.

It could be done today, too, if only the funding were there. But the market today is WAY less than it was when funding was originally denied.

I'm not crying about it. History is what it was, and what "might have been" fuels a LOT of current discussions. Still, it would have been good to see that development ... at that time, not today when it doesn't really matter. And, without any bias either way, the Merlin was and IS a superb engine. I just think the Allison was and IS, too. It didn't always have a great time, but the issues were cleared up well before the war was over for the "E" and "F" engines. Not too sure the "G" series were ever going to be bulletproof at high MAP, but ALL very high-output piston engines have those issues. Late-model 2,200+ HP Merlins were also prone to breakage.

 

[gjs238]

It would be pointless to have 2 sets of gear change systems for the 2 stages of supercharging, as it would weigh more. The two options, then, are too have both first and second stages on the same driven shaft and rotating at the same speed at all times (the Rolls-Royce Merlin/Griffon system), or have a fixed main supercharger stage and multi-speed auxiliary stage (teh system favoured by Pratt and Whitney and Allison).

Or use a torque converter system.

 

[Deleted member 68059]

It would be pointless to have 2 sets of gear change systems for the 2 stages of supercharging, as it would weigh more.

I understand why you have made that statement, but suprisingly enough there are advantages to doing just that - and Jumo engineers in 1944 were designing just such a system for the 2-stage Jumo213. The first stage was just geared to the crank, and the second stage was variable by hydraulic coupling.

Actually much more compact and lighter than having a separate external "auxilliary" stage, because you share a lot of the housing structure and so on.

To do that really means you need a swirl-throttle to do it properly, something the Allies all failed conspicously to develop with (ironically) the sole exception of the French.

(AM35 /38 swirl throttle pic below - the inlet eye is 165mm diameter)

Ugh.... how do you make photos on here NOT fill up the entire page.....

 

[Zipper730]

To do that really means you need a swirl-throttle to do it properly, something the Allies all failed conspicously to develop with (ironically) the sole exception of the French.

Honestly, I thought just the Russians used the swirl-throttle. It's nice to know that somebody else stumbled upon the idea

 

[drgondog]

Wing root dogtooth also missing like P-51H. Since this was a North American project, I'm sure they shared aero data among the lightweight P-51s.

Joe's documents show Allison DID ask for 2-stage S/C funding and was denied twice, perhaps more, perhaps not.

It could be done today, too, if only the funding were there. But the market today is WAY less than it was when funding was originally denied.

I'm not crying about it. History is what it was, and what "might have been" fuels a LOT of current discussions. Still, it would have been good to see that development ... at that time, not today when it doesn't really matter. And, without any bias either way, the Merlin was and IS a superb engine. I just think the Allison was and IS, too. It didn't always have a great time, but the issues were cleared up well before the war was over for the "E" and "F" engines. Not too sure the "G" series were ever going to be bulletproof at high MAP, but ALL very high-output piston engines have those issues. Late-model 2,200+ HP Merlins were also prone to breakage.

I missed Joe's comments. The question might be 'when'. Te GM docs copied to NAA. GM and Allison were late 1942.

 

[gjs238]

I understand why you have made that statement, but suprisingly enough there are advantages to doing just that - and Jumo engineers in 1944 were designing just such a system for the 2-stage Jumo213. The first stage was just geared to the crank, and the second stage was variable by hydraulic coupling.

Actually much more compact and lighter than having a separate external "auxilliary" stage, because you share a lot of the housing structure and so on.

To do that really means you need a swirl-throttle to do it properly, something the Allies all failed conspicously to develop with (ironically) the sole exception of the French.

(AM35 /38 swirl throttle pic below - the inlet eye is 165mm diameter)

View attachment 567927

Ugh.... how do you make photos on here NOT fill up the entire page.....

In my ABB/Alstom gas turbine world, these are called VIGV's, Variable Inlet Guide Vanes.

 

[tommayer]

I know from some of your previous posts that you are an experienced bike racer. I also know that you know the difference between single stage and two stage superchargers.
I have no first hand knowledge of how a two stage system with a mechanical stage suffers from "lag" in it's turbo auxiliary stage. I would suspect that the throttle response is better than a turbo alone. I have read that the P-38 may have suffered and that there were two methods to get around it. The discredited high rpm, low boost cruise which used more fuel, wore out the engines quicker, over cooled the intake mixture and kept the large GE turbos at low rpm (and a GE turbo is absolutely huge compared to a modern car or bike turbo). or the preferred low rpm, high boost method of cruise, the turbo is spinning thousands of RPM higher, the intake mixture is not puddling in the manifolds and there is a fair amount of boost available to increase power (torque) as the the throttle is opened (or prop control moved?) to accelerate the engine/props.

You also have how the propeller acts as it goes from cruise to high speed. I believe (but could very well be wrong) that the prop is in coarse pitch (or close to it) to give the most speed for a low rpm. when the throttle is opened up and rpm increased the prop should go to a finer pitch as the engine gains rpm and then as the power increases and the speed increase the prop should go back to coarse pitch as high speed is approached?

Most of the people trying to compare modern cars/bikes to WW II engines seem to forget the aircraft engines were operating in rather thin air insead of the "soup" that most cars and bikes operate in. The AIrcraft engine at 20,000ft is only getting about 53% as much "air" (by weight) per cubic ft as the engine at sea level.

With C/S prop, assuming you are in level flight, opening the throttle will produce the following: the RPMs will remain constant, manifold pressure will increase, the engine will make more power, the prop blades will coarsen in order to keep the RPMs constant, the plane will accelerate.