An Allison V-1710 with a two-speed supercharger in 1942 (13 Viewers)

[tomo pauk]

No, it just speaks to the apparent relative ease of fitting an existing (if by now, 2 year old) engine with a two speed supercharger into one of the existing US fighters.

I don't think that anyone is questioning the ease of fitting of an existing engine with a 2-speed S/C on the existing US fighters; at least I am not. What I'm questioning is the Allison's ability to come out with an engine that is similar to the Merlin XX (including the power above 12000 ft) in 1942, if there is no reduction in work load at Allison's.

I'm not sure I buy that, but maybe at this point it might be worth taking a closer look at everything Allison R&D department was spending their time on in 1941 and 42. It seems to me that there could have been some overlap in design goals. For example, while it may have taken a lot of effort to create the 9.60:1 ratio supercharger, but that could have been the second gear for a theoretical two speed supercharger.

It could've been okay as a 2nd gear for the S/C drive. Upping the ante and developing also a 2-speed drive for the V-1710 is increase of work load of the design, testing and production teams of the company, not a reduction.

Then the priority becomes the impeller and whatever changes need to be made to carb etc., and a housing. It again seems to be like many of these components could be copied from the Merlin XX or one of the P&W superchargers.

Copying the supercharger section from the Merlin XX would've been good.
IMO, going on with a big supercharger - ~12 in impeller, as used on the aux S/C of the 2-stage V-1710s - with the hydraulic drive would've been even better.
Going to the 2-stage S/Cs ASAP would've been great, allowing the non-turbo V-1710 to finally compete with the best engines of the world.

Maybe to get this done we take some Allison R&D resources off of the project to make the quasi two stage supercharger for the V-1710-93 used by the P-63.

Quasi two stage S/C?

 

[tomo pauk]

It can be downloaded from NASA website:
Effect of Three Modifications on Performance of Auxiliary-Stage Supercharger for V-1710-93 Engine - NASA Technical Reports Server (NTRS)

Thank you.
Unfortunately, that seems about the V-1710, not about the 'post-war Mustang manual'.

 

[don4331]

Going back on the numbers:

Allison signed a contract with USAAF (I don't have number, so I'll use $20M for the contract; if someone has actual number, I'll update)
$20M/837 engines ≈ @$23,900 per engine; we'll assume 15% profit: so, the engine cost ≈ $20,800 per engine to manufacture.
So, when USAAF claws back their $1M to allow for foreign sales; and demands engines for 66 more P-38s, the contract price per engine is now ~$19,600/engine...i.e. Allison has to find ~$1,200 in efficiency improvements or they are losing money on each engine.
They're probably making a little on each engine as part of the French contracts, but they are constructing building, buying machinery, training people, etc.
​

So, all the majority efforts of the company are directed toward reducing cost/improving efficiency...otherwise Allison is going bankrupt.

USAAF isn't paying more on the contract to pay French gov't royalties*, and they aren't paying Allison for engineering/analyst/new machinery, new employees, etc to incorporate 2 speed drive. They have spent $M on turbochargers. Last time I checked, American companies were in business to make a profit.

If, in '39, you were saying France would fall in '40 in 6 weeks, they would have had a padded cell for you. Then when France falls in '40, all the unpaid invoices are null and void. Companies often pay 60-90 days after receipt, so Allison was on hook for $M until UK/USAAF took over the contracts.

So, it is mid '40, at best before Allison could start designing the Farman 2 spd into their V-12 without having to pay French gov't. And the Farman 2 spd doesn't really improve power because the compressor is so inefficient. So, you need to find an American Stanley Hooker and get them to work with your existing engineers to redesign the rear accessory section and compressor. The Merlin XX doesn't come out until mid '40 either, and it going to be a few months (mid-'41?) before its compressor's advantages are known/UK is going to release information about it, even to allies.

*The patent royalties aren't cheap: Patents are a depreciating asset - people find way around them; they only last 19 years. So, if you have a patent a better mousetrap e.g. Farman 2spd drive, you make the most of it. So, the French gov't is charging thousand(s) of $/engine (The French people want money back for the nationalizing of Farman brothers company).

 

[tomo pauk]

Going back on the numbers:
Allison signed a contract with USAAF (I don't have number, so I'll use $20M for the contract; if someone has actual number, I'll update)
$20M/837 engines ≈ @$23,900 per engine; we'll assume 15% profit: so, the engine cost ≈ $20,800 per engine to manufacture.
So, when USAAF claws back their $1M to allow for foreign sales; and demands engines for 66 more P-38s, the contract price per engine is now ~$19,600/engine...i.e. Allison has to find ~$1,200 in efficiency improvements or they are losing money on each engine.
They're probably making a little on each engine as part of the French contracts, but they are constructing building, buying machinery, training people, etc.
So, all the majority efforts of the company are directed toward reducing cost/improving efficiency...otherwise Allison is going bankrupt.

USAAF isn't paying more on the contract to pay French gov't royalties*, and they aren't paying Allison for engineering/analyst/new machinery, new employees, etc to incorporate 2 speed drive. They have spent $M on turbochargers. Last time I checked, American companies were in business to make a profit.

If, in '39, you were saying France would fall in '40 in 6 weeks, they would have had a padded cell for you. Then when France falls in '40, all the unpaid invoices are null and void. Companies often pay 60-90 days after receipt, so Allison was on hook for $M until UK/USAAF took over the contracts.

Good points.

So, it is mid '40, at best before Allison could start designing the Farman 2 spd into their V-12 without having to pay French gov't. And the Farman 2 spd doesn't really improve power because the compressor is so inefficient.

(my emphasis)
I'd say that V-1710's compressor was efficient - it have had the curved vanes, it was semi-shrouded, it have had a proper diffusor, and it was with an axial intake that was not cluttered. We can recall that one of the main things the S/C section of Merlin XX/45 had over the Merlin III was the un-cluttered intake elbow - a thing that Hooker figured out - while the impeller on the Mk.XX also turned faster than that of the Mk.III (driven at 9.49:1 at 8.588:1; Mk.XII and 45 turned at 9.089:1).

What the compressor on the V-1710 lacked were two things - size (at 9.5 in, the impeller was smaller than the 10.25in that Merlin or DB 601 had), and it turned too slow until late 1942. Introducing the 'faster' gearing almost haves the difference between the legacy V-1710s and the Merlin XX or 45. Stick the supercharger from the 1-stage V-3420, with it's 10 in impeller, and turn it at 9.60:1 = the V-1710 equals the Merlin XX and the 45.

 

[tomo pauk]

Note the intake elbow on the Merlin I/II/III (bottom right of the picture), that forces the air do funny curves after it passed the throttle plate and before it is taken over be the impeller:



Also here:



Similar elbow was also on the Mk.XII and Mk.X.

See here the revised elbow from the XX (actually, the post-war Mk.500, that was a member of the Mk.XX family):

 

[Steamed_Banana]

Going back on the numbers:

Allison signed a contract with USAAF (I don't have number, so I'll use $20M for the contract; if someone has actual number, I'll update)
$20M/837 engines ≈ @$23,900 per engine; we'll assume 15% profit: so, the engine cost ≈ $20,800 per engine to manufacture.
So, when USAAF claws back their $1M to allow for foreign sales; and demands engines for 66 more P-38s, the contract price per engine is now ~$19,600/engine...i.e. Allison has to find ~$1,200 in efficiency improvements or they are losing money on each engine.
They're probably making a little on each engine as part of the French contracts, but they are constructing building, buying machinery, training people, etc.
​

So, all the majority efforts of the company are directed toward reducing cost/improving efficiency...otherwise Allison is going bankrupt.

USAAF isn't paying more on the contract to pay French gov't royalties*, and they aren't paying Allison for engineering/analyst/new machinery, new employees, etc to incorporate 2 speed drive. They have spent $M on turbochargers. Last time I checked, American companies were in business to make a profit.

If, in '39, you were saying France would fall in '40 in 6 weeks, they would have had a padded cell for you. Then when France falls in '40, all the unpaid invoices are null and void. Companies often pay 60-90 days after receipt, so Allison was on hook for $M until UK/USAAF took over the contracts.

So, it is mid '40, at best before Allison could start designing the Farman 2 spd into their V-12 without having to pay French gov't. And the Farman 2 spd doesn't really improve power because the compressor is so inefficient. So, you need to find an American Stanley Hooker and get them to work with your existing engineers to redesign the rear accessory section and compressor. The Merlin XX doesn't come out until mid '40 either, and it going to be a few months (mid-'41?) before its compressor's advantages are known/UK is going to release information about it, even to allies.

*The patent royalties aren't cheap: Patents are a depreciating asset - people find way around them; they only last 19 years. So, if you have a patent a better mousetrap e.g. Farman 2spd drive, you make the most of it. So, the French gov't is charging thousand(s) of $/engine (The French people want money back for the nationalizing of Farman brothers company).

We do have the superchargers on the Pratt and Whitney engines, and if I'm reading this thread right, components for these are made by the same subcontractor that Allison is using for key supercharger parts, i.e. G.E.. Is that wrong?

Even a late 1941 start seems like enough time to get something into production by late 1942.

(While keeping in mind I don't even know how to change the spark plugs on my car, so I admit, I am very amateur regarding enignes)

 

[tomo pauk]

We do have the superchargers on the Pratt and Whitney engines, and if I'm reading this thread right, components for these are made by the same subcontractor that Allison is using for key supercharger parts, i.e. G.E.. Is that wrong?

Several things might be noted.
1 - By the time V-1710 was in service (ie. mid-1940 on), Allison was long designing & producing their own S/Cs;
2 - P&W superchargers were less efficient than the Allison S/Cs, with straight-bladed impellers and cluttered & squished intakes until 1944 - it took two stages of compression on the R-1830 to match just 1 stage on the Merlin XII and X, or the DB 601N; the V-1710 S/C with the 9.60:1 drive was in the ballpark, too.
3 - P&W superchargers on the Hornet from the early 1930s were actually with better layout (better intake elbow and with impeller blades that have curved front sections) than what they started doing by mid/late 1930s - perhaps these S/Cs were from GE? The impellers on the Hornet were small, though, under 7in diameter.

(BMW made the same mistake that P&W did - they went from the good layout from the Hornet to the bad intake elbow layout from the mid-range BMW 132s and on the BMW 801A/C/D, while also going a step behind with the design of the impeller blades; granted, the 13in impeller on the 801 was a far cry from the small impellers on the Hornet)

 

[don4331]

The other one I forgot: The Allison V-1710 is not a RR Merlin with a 0.1" greater diameter pistons.

The Allison has 6.65:1 CR while the Merlin has 6.0:1 CR, so, you can't exactly tack on a Merlin XX compressor, run it at the same ratio at the same altitude and get the same results.

The higher compression ratio makes the engine more "efficient", but at the same time it reduces the amount boost you can apply.​

Note: Running the bigger supercharger faster, further reduces the "efficiency" of the engine.​

And if you want to lower compression, you can't just reduce the piston height (like Detroit did in 70s), because that messes it the turbulence/quench in combustion chamber and you get pre-ignition - which you further reduce the compression ratio.*​

​

tomo pauk I owe you this from a while back:

We'll start with the following temperature and pressure: 29" Hg, 70°F, but we need to work from absolute 0, so ~460° + 70° ~530°R

2 psi boost 1 psi = 2.03" Hg (for our 33" Hg) because we can easily measure pressure. Now, some formula's:​

Pressure ratio = (standard pressure + boost)/standard pressure​

(Epsilon) Y = Pressure ratio^.283 - 1. (.283 is number for an ideal gas, 80% Nitrogen/19% Oxygen aka air is close enough) Will give us the temperature rise for adiabatic (no heat added/subtracted) compression.​

(Delta) ΔT ideal = Y * temperature in °R​

Compressor efficiency in %.​

ΔT actual = ΔT ideal / compressor efficiency​

​

Density ratio = standard temp/(standard temp + ΔT actual) * Pressure ratio​

​

Boost	2 psi	2 psi	10 psi	10 psi
Pressure ratio	(2*2.03 + 29)/29 = 1.14	(2*2.03 + 29)/29 = 1.14	(10*2.03 + 29)/29 = 1.41	(10*2.03 + 29)/29 = 1.41
Epsilon Y	1.14^.283-1 = 0.038	1.14^.283-1 = 0.038	1.14^.283-1 = 0.103	1.14^.283-1 = 0.103
ΔT idea	0.038*530 = 19.9 °R	0.038*530 = 19.9 °R	0.103*530 = 54.7 °R	0.103*530 = 54.7 °R
Compressor Eff.	40%	70%	40%	70%
ΔT actual	19.9/.4 = 49.7 °R	19.9/.7 = 28.4 °R	54.7/.4 = 136.7 °R	54.7/.7 = 78.1 °R
Density ratio	(530/(530+49.7))*1.14 = 1.04	(530/(530+28.4))*1.14 = 1.08	(530/(530+136.7))*1.41 = 1.12	(530/(530+78.1))*1.41 = 1.23
Improvement		4%		10%

At 10 psi, its not just the 10% improvement in density, the 50°R increase (almost hot enough to boil water) makes the air/fuel mixture much more sensitive to pre-ignition.

p.s. Wright engineer Kenneth Campbell should be remembered with the same reverence as Ellor and Hooker: He took the Cyclone supercharger from GE with efficiency in the 30-40s to over 75% for Wright's own. Unfortunately very late in the war, but valuable for the post war commercial use.

*The 350 Chev hotrod of 350hp 11:1 CR is suddenly a 160hp 8.3:1 CR dog (OK, the removal of lead reduces Octane rating of fuel, so 11:1 wasn't possible anymore). Detroit eventually figured out the error of their ways and increase CR to 9.5:1 and hp increased back to 245hp. My point being Allison would need to do other changes correctly to make the better supercharger work.

 

[Steamed_Banana]

The other one I forgot: The Allison V-1710 is not a RR Merlin with a 0.1" greater diameter pistons.

The Allison has 6.65:1 CR while the Merlin has 6.0:1 CR, so, you can't exactly tack on a Merlin XX compressor, run it at the same ratio at the same altitude and get the same results.

The higher compression ratio makes the engine more "efficient", but at the same time it reduces the amount boost you can apply.​

They apparently applied huge amounts of boost to the V-1710, and this capacity increased from the F3R to F4R, but then diminished somewhat with the higher ratio supercharger gear for the later Allison marks. Still, 57" Hg was plenty for good performance.

Note: Running the bigger supercharger faster, further reduces the "efficiency" of the engine.
And if you want to lower compression, you can't just reduce the piston height (like Detroit did in 70s), because that messes it the turbulence/quench in combustion chamber and you get pre-ignition - which you further reduce the compression ratio.*​

I don't understand why would you want to lower compression?

​

tomo pauk I owe you this from a while back:

We'll start with the following temperature and pressure: 29" Hg, 70°F, but we need to work from absolute 0, so ~460° + 70° ~530°R

2 psi boost 1 psi = 2.03" Hg (for our 33" Hg) because we can easily measure pressure. Now, some formula's:
Pressure ratio = (standard pressure + boost)/standard pressure
(Epsilon) Y = Pressure ratio^.283 - 1. (.283 is number for an ideal gas, 80% Nitrogen/19% Oxygen aka air is close enough) Will give us the temperature rise for adiabatic (no heat added/subtracted) compression.
(Delta) ΔT ideal = Y * temperature in °R
Compressor efficiency in %.
ΔT actual = ΔT ideal / compressor efficiency

Density ratio = standard temp/(standard temp + ΔT actual) * Pressure ratio
​

Boost	2 psi	2 psi	10 psi	10 psi
Pressure ratio	(2*2.03 + 29)/29 = 1.14	(2*2.03 + 29)/29 = 1.14	(10*2.03 + 29)/29 = 1.41	(10*2.03 + 29)/29 = 1.41
Epsilon Y	1.14^.283-1 = 0.038	1.14^.283-1 = 0.038	1.14^.283-1 = 0.103	1.14^.283-1 = 0.103
ΔT idea	0.038*530 = 19.9 °R	0.038*530 = 19.9 °R	0.103*530 = 54.7 °R	0.103*530 = 54.7 °R
Compressor Eff.	40%	70%	40%	70%
ΔT actual	19.9/.4 = 49.7 °R	19.9/.7 = 28.4 °R	54.7/.4 = 136.7 °R	54.7/.7 = 78.1 °R
Density ratio	(530/(530+49.7))*1.14 = 1.04	(530/(530+28.4))*1.14 = 1.08	(530/(530+136.7))*1.41 = 1.12	(530/(530+78.1))*1.41 = 1.23
Improvement		4%		10%

At 10 psi, its not just the 10% improvement in density, the 50°R increase (almost hot enough to boil water) makes the air/fuel mixture much more sensitive to pre-ignition.

p.s. Wright engineer Kenneth Campbell should be remembered with the same reverence as Ellor and Hooker: He took the Cyclone supercharger from GE with efficiency in the 30-40s to over 75% for Wright's own. Unfortunately very late in the war, but valuable for the post war commercial use.

*The 350 Chev hotrod of 350hp 11:1 CR is suddenly a 160hp 8.3:1 CR dog (OK, the removal of lead reduces Octane rating of fuel, so 11:1 wasn't possible anymore). Detroit eventually figured out the error of their ways and increase CR to 9.5:1 and hp increased back to 245hp. My point being Allison would need to do other changes correctly to make the better supercharger work.

So is the gist of this simply that there is a random element in supercharger development? We don't know if Allison (or GE, or whomever actually does this part) can produce a high efficiency compressor in time, it's not predictable... so they might be able to do it in a year or they might not? Or are you saying that it's impossible or...

I.e. this is fascinating insight into how engines and compression work, but how does this bear upon the proposed question in the OP and central issue of the thread. Could they make a two-speed supercharger faster than we know they in fact did make the two stage supercharger for the P-63? Could they make something better and more reliable?

And how would a hydromatic speed supercharger ala DB 600 series compare with the two speed option?

 

[Steamed_Banana]

I would still assume having the P&W and Merlin XX superchargers to reverse engineer or study, would still be helpful even if you can't just make a direct copy (which I wouldn't assume was possible anyway) just because many of the technical problems they solved whether it has to do with mounting, cooling, supercharger gear teeth, special materials needed for strength or heat resistance, airflow distribution etc., can be learned from.

Of course you still do need skilled designers to do it. No doubt about that. I ain't saying I could do it...

 

[Steamed_Banana]

I would also note that the Japanese also figured out how to successfully design and manufacture two speed superchargers for several of their engines, (for example the Nakajima Ha 115 on the Ki-43-II) and fairly early in the war too, in the time frame we are discussing...

 

[MiTasol]

I don't understand why would you want to lower compression?

Too high a compression combined with a high inlet temperature to the cylinder resulting from high boost pressure results in detonation and that takes only a very short time to destroy an engine - usually by burning holes in the pistons. Pistons with holes in them do not produce much power and have very short lives.

 

[tomo pauk]

I.e. this is fascinating insight into how engines and compression work, but how does this bear upon the proposed question in the OP and central issue of the thread. Could they make a two-speed supercharger faster than we know they in fact did make the two stage supercharger for the P-63? Could they make something better and more reliable?

IMO, they could've made a 2-speed supercharged V-1710 before the 2-stage version.

And how would a hydromatic speed supercharger ala DB 600 series compare with the two speed option?

The hydromatic drive would've made the power curve much smoother. Thus the engine will be giving close to ideally possible max power at the ~6000 to ~15000 ft altitude band, where the engine with the 2-speed drive experiences a bit of loss in power since were away from the ideal altitude for 1st and second gear. Shortcoming of the hydromatic drive is that it heats the oil in transfer case when there is a lot of slip (= at lower altitudes), so the oil system will need to grow in size.

I would also note that the Japanese also figured out how to successfully design and manufacture two speed superchargers for several of their engines, (for example the Nakajima Ha 115 on the Ki-43-II) and fairly early in the war too, in the time frame we are discussing...

Japanese were making and selling their engines to the Government and the other costumers years earlier than Allison did that. 'Early in the war' for the Americans is 'mid-war' for The British and Germans

 

[Steamed_Banana]

Too high a compression combined with a high inlet temperature to the cylinder resulting from high boost pressure results in detonation and that takes only a very short time to destroy an engine - usually by burning holes in the pistons. Pistons with holes in them do not produce much power and have very short lives.

I recognize this is one of the difficult needles that engine (and compressor / supercharger) designers have to thread, but it does seem like a problem which was solved. Multiple times and by multiple people.

It sounds like the other issue is just how well does it get solved, do you end up with a high efficiency supercharger or middling or low. They didn't have computer design in this period so i guess it's somewhat of a crap shoot and it took a while to determine the merits of a given design. Which means there is a random element, effectively.

 

[Steamed_Banana]

IMO, they could've made a 2-speed supercharged V-1710 before the 2-stage version.

That is also my theory.

The hydromatic drive would've made the power curve much smoother. Thus the engine will be giving close to ideally possible max power at the ~6000 to ~15000 ft altitude band, where the engine with the 2-speed drive experiences a bit of loss in power since were away from the ideal altitude for 1st and second gear. Shortcoming of the hydromatic drive is that it heats the oil in transfer case when there is a lot of slip (= at lower altitudes), so the oil system will need to grow in size.

Good point about the smooth curve, interesting detail about the oil system. This would also increase weight and create a bit more vulnerability to the engine.

Japanese were making and selling their engines to the Government and the other costumers years earlier than Allison did that.

US firms were developing multi-speed superchargers pretty early.

'Early in the war' for the Americans is 'mid-war' for The British and Germans

I'm well aware. I did say relatively, and by that I mean, relative to the end of the war. The war pivoted in mid 1942 through 1943. So that in one sense is the middle, for all of the major participants - Soviets, British, Germans, Japanese, US, Italians.

US industry, including specifically aircraft and engine makers, were supplying weapons in large numbers to other countries since 1939. Notably to France and then Britain, and to China and then the Soviets, before the US even got into it.

 

[Shortround6]

Allison started development of the two-stage engine in 1938.
I have no idea if that was the first memo or pencil sketch on the back of an envelope
Actual hardware didn't show up until 1940 but things ran very slowly.
Engine was first run in Feb 1942 in the Allison Altitude chamber. At this time the Aux supercharger was using a 9-1/2in impeller and was using an 8.00:1 gears in a single speed mechanical drive. Feb 1942 also development testing of a hydraulic drive and it was in July 1942 that the decision to switch to the hydraulic drive was made.
In April 1942 2000 V-1710-47 engines were ordered for the P-39E (21in fuselage stretch) with more orders following but things did not go well in testing.

From a business standpoint the P-39E was a prototype for the P-76 which was supposed to use the Continental I-1430-1 but since that was not ready for flight the Allisons were a substitute. Army still wanted the Continental I-1430-1 engine but it wasn't ready yet. Army had spent several million dollars on a new factory for the I-1430 but the factory never made any. They built radial engines for trainers instead.

Allison started redesigning the -47 into the -93 engine in fall of 1942 and testing started in Dec 1942 but it was a long and troubled path with full model test not completed until Nov 27th 1943 after several hundred P-63s were already built and flying under certain restrictions.

The Continental I-1430 was the Army's fair-haired child. Much of the original design was by army officers at Wright Field with fabrication, assembly and test done by Continental. Final design decisions were made by the Army. The Allison was sort of the red headed stepchild. Unfortunately the I-1430 had a bone disease at birth and would never be strong enough to really run and it took several years for the Army parents to accept that.

Think of Allision and GM, by late 1942/early 43 they had orders for over 10,000 2 stage engines but they knew that the Army still wanted the I-1430 and if Continental fixed the problems the huge order could be slashed. Turns out that Allison had problems with two stage engine and orders were cut considerably. Aside from prototypes, during WW II the two stage engines were only fitted into about 3300 P-63s although there were a fair number of spare engines in crates.

Now to compare our hypothetical 2 speed engine...................The two stage V-1710-47 engine flying in the P-39E prototypes was supposed to be good for 1150hp at 21,000ft and 1325hp for take-off. The -93 was rated at 1150hp at 22,500ft. Its low altitude WEP was 1825hp but that was part of the development problem. They needed new pistons, new piston rings, new valve springs and some other new internal parts to stand up to the 1800hp power levels (also needed water injection).

Maybe Allison did screw up by not working on a 2 speed engine. But the window of opportunity was small and the desire to jump to two stage was large. Turns out it was too much of a jump. Allison arrived late and with too large an engine to fit in existing fighters.

 

[don4331]

So is the gist of this simply that there is a random element in supercharger development? We don't know if Allison (or GE, or whomever actually does this part) can produce a high efficiency compressor in time, it's not predictable... so they might be able to do it in a year or they might not? Or are you saying that it's impossible or...

Think of it as a Rubik's cube (but with a few more faces).

Getting one face correct is pretty easy then add the center square; 40% efficient GE did that and supplied to Allison, GE (for their turbochargers), P&W and Wright.
Getting the 2nd row correct on the cube might mean you have to do some of the 1st face to progress - Ellor does that for RR
Getting the whole cube correct needs someone like Hooker or Campbell to solve it; and then they have to publish the formula.​

Only when you have the formula you can "easily" design the supercharger for your application (Having super computers and multi-axis CNS machines helps in 21st century)​

There is only a very small window where a centrifugal compressor is in its "sweet spot". Trying to build too much pressure at too low of airflow results in compressor stalling and you get no boost (or surging), too much/too little airflow and your efficiency is low.

Unfortunately, WWII aircraft only measure pressure. If one simply heats a pot high enough (pressure cooker), you can achieve 57" Hg without adding any additional air. Obviously, the Allison supercharger wasn't that bad, but you have to know the temperature rise to get the actual density improvement and its the increased density that is making the additional power.

Redesign of an existing component/system is often far more difficult than the original design (The numbers I have: Original Mustang - 78,000 hours; redesign for the Merlin: 223,00 hours. Strengthening the crankcase, of the V-1710 is not just pour a little more aluminium in the model and send it.

I.e. this is fascinating insight into how engines and compression work, but how does this bear upon the proposed question in the OP and central issue of the thread. Could they make a two-speed supercharger faster than we know they in fact did make the two stage supercharger for the P-63? Could they make something better and more reliable?

The issue is who is paying? And why? Before '42, USAAF isn't at war and they don't see need to pay RR for better supercharger technology/Farman brothers for 2 speed mechanism. From '41 to '43, Allison has its hands full meeting demand. USAAF doesn't have the equivalent of a Whitley (twin bomber powered by V-1710) for which low supercharge ratio means the difference between getting off the runway with reasonable bombload or crashing into the trees at the end of the runway.

If USAAF pays Farman for their patent, pays Allison to design the 2 speed into their engine, design a more efficient supercharger, build a satellite factory and equip it, then you will get your 2 speed solution.

Note: German/Japan/Russia can careless about American/French/UK patents...they belong to the enemy.

And how would a hydromatic speed supercharger ala DB 600 series compare with the two speed option?

The challenge of a hydromatic drive is:

a. Your supercharger compressor needs 300hp at 20k' to supply 15psi boost to a V-1710 at full power....US industry isn't making very many 300hp hydraulic motor of aircraft weight (a 3,000lb industrial motor for 300hp isn't going to cut it).​

b. The efficiency of a gear drive is around 98-99% as your step up drive will have 2 set of gears - 300hp/(.98*.98) = 312hp driving the input gear to the supercharger drive​

The efficiency of a hydraulic gear motor is around 80% (same for a gear pump): 300hp/(.80*.80)= 468hp driving the input gear to the supercharger drive. And we have to cool the 168 hp not used by the supercharger out of the oil before the next round or the oil burns.​

​

So, is your hydromatic drive providing more power than a 'restricted' gear drive. RR pre-war calculations said no; but then they got ahold of DB and came to the conclusion of ....maybe in certain circumstances (i.e. heavy bombers for which take off power was all important)....

 

[Steamed_Banana]

Before I continue the banter, I just want to say to thanks to you and the others who contributed their very deep and specialized knowledge to this and the bomber thread.

I feel like I've doubled my (very limited) knowledge about WW2 aircraft engines and superchargers, and the industries which built them, since the thread started. and had a pleasant reminder of the many ways that history always surprises us by being more complex and more interesting than the high level summaries can leave you to assume. So once again, to everyone, thank you!

I feel these two threads are a good example of how even a silly "What If" type hypothetical question can tease out a lot of very useful information, thanks to all of you folks who chimed in.

 

[BarnOwlLover]

That's something I've discovered with this subject (among other things I've studied and followed) is that you need to have an open mind and have assertions challenged. I've gained an all new respect for the whole P-51 Mustang family as a result. Namely that each version was easily capable of better performance (especially speed and climb) than what a lot of books quote. An aircraft family that I respected has now become my favorite from this period as result.

Not to mention that it's expanded my book collection and got me interested in World War II aircraft photo archives for collection and study.

 

[Simon Thomas]

It's worth noting that the Merlin 2 stage s/c was adapted from the RR Vulture. It was rather fortunate that RR had a number of previous s/c designs in their box of tricks, such that one was close enough to work. At that stage of the war, the time factor was more important than getting the results precise to 8 decimal places.

How many employees did Allison have working in the engineering department in the 38/42 timeframe? I recall seeing a number on this site a few year back, and being stunned at how so much work was being done by so few people.