What was the problem with the allison engine?

The Allison docs say it came from the -109.

If I have to choose between Allison and Vees for Victory, I choose Allison since they made the engines. In the end it doesn't matter since it never flew, even if it was interesting.

I think the V-1710 still had development potential, as we have gone over before, but the jet engine put that on hold forever. At least the Allisons Joe is delivering these days have -100 series internals and almost perfect balance. I'm sure that all helps a lot more than anything else.

Cheers.

wuzak said:

Practically speaking, using the carburettor between compressor stages allowed the core engine to be, essentially, the same as used in a turbocharged installation. The downside is that charge cooling could only come using ADI, and not an aftercooler.
I believe it also shortened the length of the whole engin without that massive carburettor hanging off the back.

Click to expand...

The common core engine arrangement is why I'd have thought they'd have gone with that configuration from the start, using the aux stage literally in place of a turbocharger (sans intercooler ... though, technically, an intercooler could have been installed as well, just as with a turbo).

I don't see why that arrangement would preclude use of an aftercooler, though. Use of a standard modular core engine would preclude it, yes, but the inter-supercharger carburetor placement seems perfectly compatible with aftercooling. (an intercooler in series with the aux stage and carb intake would be more compatible with the modular add-on arrangement though and technically COULD be of liquid cooled type and not air to air, and would probably be best located at the carb intake, minimizing bulk and length of glycol plumbing to only slightly more than the aftercooler arrangement does)

Mimicking the behavior of a turbocharger really seems like the best/most foolproof direction Allison could have taken and is probably why they discarded the fixed gear ratios in favor of a fluid coupling relatively early on. (even a single-speed fluid coupling would allow pretty useful performance, particularly with a full-neutral setting avoiding excessive oil/fluid heating at low altitudes) I wonder if examining the impeller size and RPM range of GE's turbos would have accelerated development more than building off Allison's own superchargers alone (or potentially outsourcing to Wright or P&W)

I know GE's supercharger designs were quite lacking in the 1920s and early 30s (leading to P&W and Wright investing in developing their own) but the added competition led GE improving their own hardware quite a bit from what I understand. (and even if the compressor design was inferior to what Allison was already working with, the mass flow, diameter, and operational RPM range would be useful for testing -including avoiding stall conditions between the two stages)

For that matter, running an auxiliary supercharger of identical construction to the integral stage, but driven totally independently (off some sort of supplemental powerplant or perhaps even electric motor) could have been used for initial trials and rather quickly written off that sort of pairing as unworkable, or established what range of operating conditions it was workable within and whether those were worthwhile. (without wasting time actually developing the 2nd stage's mechanical or fluid coupling mechanism)

wuzak said:

The change improved the altitude performance not because of the carburettor position, but because of the reduction in losses in the air flow for 1st stage to 2nd stage.

Click to expand...

This is counter-intuitive given the intermediate carb installation requires a longer duct with more twists and turns leading from the aux stage to the carb than is the case for the aux stage directly into the engine stage.

Reduction in intake losses from the carb-less aux stage intake manifold seems to be the big gain here, possibly in addition to charge cooling between the two stages being more efficient than at the intake of the aux stage.

Was there ever any development with a air/water intercooler in the turbo/Allison installation? It seems like it would have greatly simplified the ductwork, since you can simply install a fairly small air/water heat exchanger in the intake plumbing between the turbo/supercharger, rather than ducting that large volume of air to a big chin-mount core-type air/air heat exchanger (or wing leading edge duct), then back half the length of the boom. Much easier to route a couple small water lines. Not only would the plumbing be much simpler but the frictional losses in the intake ducting would be less, and pressure drop across the HE should have been lower. In addition the 2nd water/air cooler (to "dump" the heat from the water) could have been smaller, less drag inducing and easier to package. Correct me if I'm wrong, but weren't the 2-stage Merlins intercooled this way?

Some of the two stage Allisons had a Merlin style aftercooler.

The two stage Merlins were actually "after-cooled". The terms get confused at times. Especially when used in reference to cars.
An inter-cooler is used to cool the intake charge between supercharger stages. An after-cooler is used to cool the charge after all supercharging is done. Some engines have used both.

The air to air setup vs the air to liquid (water alone may not work at high altitudes without anti-freeze) is hard to evaluate based on generalities. The US may have felt/believed that the air to air system was more resistant to battle damage. A few 8-13mm holes in the duct work (or even in the matrix, US system of matrix between supercharger stages meant there was only compressed air in the system, no fuel)) while reducing the performance will not lead to total failure of the system in a few minutes time. A liquid cooled system that is holed may reach a point where it is no longer providing any cooling. That may balance against the harder job of keeping large ducts air tight even during normal flying vrs keeping the liquid lines tight, a much easier job. The air to air system only has a few doors or flaps as moving parts vs adding a pump for the liquid cooled system which may still require a door or flap to control airflow through the remote (non-engine) heat exchanger.
Pressure drop across the intercooler is dependent on the relative airflows, the size and design of the matrix and to some extent the changes in duct size leading to and from the matrix.
Size of the Matrix (heat exchanger) is also dependent on the amount of cooling desired, and thus the airflows needed. I believe, but could very well be wrong, the the Merlin system was supposed to remove 40% of the heat added by the superchargers (both stages). The American turbo system intercoolers were measured in a totally different way (although performance of the matrix/heat exchanger could have been measured exactly the same) in that they were supposed to deliver the required air to the carb mounted on the engine supercharger at no more than 100 degrees Fahrenheit at sea level pressure (or close to it).
Trying to guess which system was better with so little information is difficult, and we are referring to the systems of the time, not which system is better in 1990 and later race cars, as air flows would be measured in pounds of air per minute not cubic feet as the system/s have to work at altitudes where the air density is 1/3 or less of what it is at sea level. They also have to work at climbing speed which is about 1/2 or less of level speed flight and so gets 1/2 or less of the air flow of full speed level flight.
Granted neither engine was using anywhere near it's full supercharger capacity at sea level or near it and so didn't need the full amount of charge cooling.
Both the Melrlin and Allison were developing about the same amount of power in the cylinders. 1650cu in vs 1710 at 3000 rpm and at 60-62in of manifold pressure means aside from the 4% difference in displacement the only other variables are the compression ratio and the actual intake charge density (which is dependent on the intake charge temperature). Pressure is an easy to measure data point but it is not mass airflow (pounds per minute) and it is mass airflow that governs power.
Large differences in power to the crankshaft are from differences in internal friction, pumps and such and how much power was being used to drive the superchargers.
I would note however that the Merlin system at altitude (over 23-24,000ft) is running air that is already compressed 5-6 times before it hits the cooling matrix and is very hot. The P-38 system is cooling air that is compressed only about 3 times or less.

The XP-51J used one cooling circuit for joint needs of the aftercoolers and the oil system, and another circuit for colling the water-gycol mixture. The joint aftercooler/oil system used heat exchanger to split the cooling further. Heat exchangers were also used on the Jumo 213 line of engines (coolant+oil in A, C and F versions, coolant+oil+aftercooler on E variant).
D.520 'Ameiloree' (sp?, means 'Improved') also used heat exchanger for oil coooling.

As far as susceptibility to battle damage of a air/liquid vs air/air system, I can see advantages and disadvantages to each. The air/liquid system has smaller liquid lines than the ductwork of the air/air system and should have a smaller heat exchanger. Meaning that it should be slightly less likely to take hits. A hit on the ductwork of the air/air system or the HE means that you are losing pressurised air, meaning reduced boost available to the engine. Whereas by eliminating most of the ductwork you reduce that concern, while opening yourself up perhaps more to a loss of inter/aftercooling. Pick your poison I guess.

As far as intercooling vs aftercooling, aftercooling would seem to offer a couple advantages. First you are cooling the intake air after all the compression is done, meaning that you should be able to supply a cooler air mass to the engine. Intercooled air is re-heated in the mechanical supercharger, limiting how cool it will be at the engine. In addition, by providing the 2nd stage of boost before cooling, the air charge is hotter. Heat exchanger efficiency increases with increased deltaT or temperature difference between the two "fluids" (air is a fluid for this purpose). So a given number of BTUs/min can be dissipated with a smaller HE and with a smaller volume of cooling air.

Intercooling reduces the power required to drive the second stage. A useful thing, as it increases power to the prop.

I would note that neither system, as used in WW II, actually cooled the air down to anywhere near ambient temperatures. The R-R system was supposed to reduce the intake charge temperature by about 40%. I have no figures for a two stage Merlin but a Merlin XX supercharger raised the temperature by 148 degrees centigrade for 9-10lbs of boost. A two stage supercharger would heat the air a bit less if keeping the boost the same but will heat the air more if higher boost is used or 9-10lbs boost is used at higher altitudes than the MK XX supercharger could provide it. A MK XX Merlin in a Hurricane doing 340mph at 20,000ft including the ram effect was compressing the air about 3.64 times, a Merlin 61 in a Spitfire could compress the ambient air 6.2 times at 33,000ft (including ram) for 9lbs boost or 5.7 times at 27,000ft for 15lbs boost.
A heat exchanger that lowered the intake temp by 40% would lower the MK XX intake charge by about 60 degrees C. leaving it at 88 degrees higher than the ambient air it took in, leaving out heating in the intake duct (yes a ram air intake does heat the incoming air even if only by 10 degrees or so) and cooling by fuel evaporation. Obviously the Merlin 61 even with the 2 stage supercharger is going to heat the air a whole lot more.
IF you want a bigger temperature reduction you need a bigger heat exchanger and a higher flow of cooling medium, wither air or liquid.
However after around a 40-45% reduction you start getting into diminishing returns. For WW II Air to air inter-coolers/heat exchangers a 40% reduction could be had for an air flow equal in mass to the intake charge. a 1.5:1 ratio got around a 53/54 % reduction, a 2:1 ratio got around a 62% reduction while a 3:1 ratio only got a 70% reduction.
The US system/s (at least the turbo units) tried to get the inlet temperature of the engine supercharger down to 90-100 degrees Fahrenheit under worst case scenario. To do this they had to reduce intake charge temperatures of around 170-180 degrees Fahrenheit (at 25,000ft) down to the under 100 degree limit. This is using ambient air of about -30 degrees Fahrenheit. (rise in temperature due to the turbo compressor was around 220 degrees F.
I would note that a 50 degree rise in intake temperature carries through the entire engine. That is 50 degrees higher in the intake duct means 50 degrees higher in the intake manifold, it means 50 degrees higher in the combustion chamber and it means 50 degrees higher exhaust gas temperature.
As has been mentioned the US system did make the the second stage more efficient but at the cost of greater bulk.
Without more specific information it is very hard (involves too much guess work) to pick which was better.

AS far as minor battle damage goes (a few small caliber bullets) the effect on the air to air system is going to be minimal.
At 99 degrees C an 8mm hole is going to flow about 30cu ft a minute at a pressure differential of 30lbs inside and 5lbs outside air pressure or less than 3% of the air flow of an engine making around 1100hp to the prop. That would be a hole in the intake duct not a hole in the cooling air duct.

Why would intercooling reduce the power required to drive the second stage? The air-fuel mixture is not burning in the intercooler, it is only getting cooled. If the after cooler brings the charge to the same temperature before getting injected into the cylinders, then the power produced would be the same.

Now if the aftercooler doesn't quite get the charge cooled down to the same temperature as without the intercooler, then adding an intercooler while retaining the aftercooler would lower the charge temperature and possibly produce slightly more power, but it doesn't change the power required to drive the seconds stage in any way.

It may make the 2nd stage more efficient. Better results for the "same" power input. Since the cooler air is denser the supercharger may be able to flow more air at the same impeller rpm and pressure. However due to the denser air it may take more power to turn the impeller so saying the intercooler reduces power may not be quite right either.
You have pressure and volume/mass flow on one side and power required and efficiency on the other and rarely are all four put together.


This compressor map gives us pressure, air flow and efficiency but does not give us the power required. It will take more power to turn the impeller/s at 35,000rpm and flowing 60lb/min than it will at 35,000rpm and flowing 30lb/min.

I would also note that the difference in efficiency is pretty much related to excess heating of the intake charge. A supercharger that is 65% efficient is using 65% of the input power to actually compress the air. the other 35% is pretty much turning into excess heat over and above the heat of simple compression. A supercharger doing the same work but running at 72% efficiency requires slightly less power but more importantly heats the intake charge less. With an aircraft supercharger using 100hp or more for the type engine/s we are talking about 5-10 hp more or less to drive the supercharger is almost too small to measure or worry about but 5-10hp turning directly into heating the intake charge to higher temperatures may have more impact on actual power developed due to the difference in density of the intake charge and running closer to the detonation limits.

GregP said:

Why would intercooling reduce the power required to drive the second stage? The air-fuel mixture is not burning in the intercooler, it is only getting cooled. If the after cooler brings the charge to the same temperature before getting injected into the cylinders, then the power produced would be the same.

Now if the aftercooler doesn't quite get the charge cooled down to the same temperature as without the intercooler, then adding an intercooler while retaining the aftercooler would lower the charge temperature and possibly produce slightly more power, but it doesn't change the power required to drive the seconds stage in any way.

Click to expand...

These answers might help:
what is the purpose of inter cooler in multistage compressor

Inter-coolers are provided between successive stages of a multi-stage compressor to remove the heat of compression hence reduces the work of compression (power requirements). The work of compression (power requirements) is reduced by reducing the specific volume through cooling the air. Thus inter-cooling affects the overall efficiency of the machine.

Click to expand...

I believe that only if we assume the first compression gets the mixture to a point where it will be excessively hot when compressed the second time. If they used an aftercooler only and it got to the same temperature as the intercooled and aftercooled unit, then I'd doubt it very seriously. Still, it could be.

One logical supposition is that they would not have fitted an intercooler had it not been needed in the first place, so perhaps you are right. It may be that fitting both an intercooler and an aftercooler actually did produce results that justified the extra weight and complexity. I'd latch on to that only if we could see comparative testing that showed it, though. Interesting avenue of investigation that I may take some time to look into later. I can believe it was possible, but would want some proof of same.

Thanks for suggesting it. Don;t have time now due to getting ready to reach new (to me) classes, but will get back to it along the way.

Some of the car turbos are getting intercooled even with only 6 - 8 pounds of boost, so it is easy to believe a lot more boost (+18 or more) on a bigger engine could use it. Some cars are even getting aftercooled, but these are mostly exotics. Not too many for the everyday driver. So far, I haven't noticed one that is both intercooled and aftercooled in cars. Maybe in Formula 1, but I haven't checked that and so don't claim it to be so. I really like the F1 Mercedes turbo that runs the shaft through the engine to be away from the exhaust heat. It removes the necessity for some of the cooling by removing some of the heat from the surrounding environment.

with the cars we are getting back to confusion in terminology. Very few, if any, cars are using two stage supercharging for 6-8lbs boost.
Many times what is called an inter-cooler on a car is really an after-cooler. Cars also are using much less compression to get 6-8lbs of boost than aircraft unless that car can get 6-8lbs at the top of Pike's peak. 7.5lbs boost at sea level only calls for compressing the air 1 1/2 times at sea level. 7.5 lbs boost at 18,000ft calls for compressing the air 3 times.
Most cars (single seat race cars excepted) are nowhere near as volume limited as fighter planes and the weight/drag of an inter-cooler/aftercooler is relatively insignificant on street cars or race cars derived from street cars.

I can't think of any cars that use two stages of compression.

There are many with twin turbos - but that is usually one for a set of cylinders and a second for another set. Less commonly used is a series set up where there are two different sized turbos, one for low speed boost and one for high speed boost, but they don't work as a first and second stage compressor.

Another case is a supercharged and turbocharged car, a few of which exist. These also operate with one (the supercharger) for low speed and the other (turbocharger) for high speed.

The current F1 turbos have only one stage of compression per the rules. They are after-cooled.

Road cars have higher compression ratios than WW2 aircraft and lower boost, generally, but have less exotic fuels to use but better combustion chamber designs.

Also, the Rolls-Royce two stage engines featured water passages around the supercharger housing to achieve a small amount of inter-cooling.

The intercooler helps the power in two ways. It reduces the power required to drive the engine stage supercharger and it lowers the temperature at the outlet of the second stage.

The aftercooler does the latter, which enables more boost or higher compression to be run and more fuel.

Maybe BMW's quad turbo diesel.

Maybe I can help here. I've had a bit of experience in designing and spec'ing twin turbos for the purpose of high output diesels.

Shortround provided a good bit of info already, let's see if I can "compress" this answer.

The twin turbos I work with are sequential in that we have a large (primary) turbo that takes ambient air (this can be described as "high-volume, low pressure turbo) and compresses this air and feeds to smaller (secondary) high pressure turbo that compresses the air once more and feeds the engine.

Most commonly, there is only one Charge Air Cooler (CAC- more commonly described as intercooler, or as above "aftercooler"). The CAC is almost always air to air and the goal is to get the compressed air something close to ambient but is usually closer to 70-100*F over ambient.

The air coming out of a twin turbo designed as I describe above can compress air and create temps that exceed 600*F! This can often exceed the capabilities for a CAC to properly cool the air. An air to water CAC of much smaller dimension is used after the primary turbo that can often cool the air to below ambient before entering the secondary turbo making for a more efficient air charge and many more pounds/min of air (or often described as more oxygen molecules in the CFM that is moving). The drawbacks to the system is complexity, weight, space and air restriction. The restriction comes in as there is necessary dwell time of the air within the CAC on either end to be able to exchange the heat. The benefit is obvious. High air temps in a gasoline engine is undesirable as this can cause detonation. In any case, thoughtful Charge Air Cooling means that a super/turbocharger can then compress and move a much higher volume of air than otherwise possible due to excessive air temps. The hotter air is, the more it expands and the fewer oxygen molecules are available for combustion for a given air volume.

In my personal truck, (and yes you will read this correctly), my twin turbos have given me up to 100 psi (that's PSI, not inches of mercury), of boost. That is a diesel with special aerospace headstuds to keep the head down under those extreme cylinder pressures.

And that is a 5.9 liter engine that produces 1150 hp and pushes a 7800lb truck to low 11's in the 1/4 mile. That's what boost can do for you!

That's impressive, not just from European perspective
Any pictures you can post, either in this thread or elsewhere?

Sure. I'll take a couple this weekend and post them.

and of course water injection does an equivalent job by using it's latent heat to soak up the adiabatic heating of a compressed charge by turning the atomised water spray into steam.