Which country designed the best engines for WWII?

[Shortround6]

Just so we are talking about the same things.

DB624

Like a lot of fancy multi-stage superchargers/intercoolers the volume of the supercharger system approaches (or exceeds) the volume of the engine itself.
Please note there is no way in this universe that this thing is going to fit into a F190/Ta 152 airframe. Me 410 is going to look like WHAT with a pair of them???
Granted a production version might have been better packaged but a lot of these "paper/test" engines leave an awful lot to be desired when compared to engines that actually saw service.

From Daimler-Benz Aircraft Engines

 

[Old Wizard]

 

[Trilisser]

Snowygrouch mentioned that Jumo 211 enjoyed a 10 - 15 % superiority in lowest sfc over the Merlin III. I have checked a few sfc curves, including for the Jumo 211F/J and DB 605A, and in neither does the curve fall below (as far as I remember) 200 g/hp/h. However, e.g. the R-1820 and R-1830 can achieve figures around 190 g/hp/h.

As for the for the best engine designing country, much depends on how much durability and reliability are valued. If those are considered significant, then both the British and the Amis were way above the rest.

 

[MiTasol]

If your definition of best is based on which is most likely to get you home when badly shot up the Allison V-1710 has to be at or near the top of the list based on several aircraft which were able to remain in the air for over an hour after the cooling system was shot out.

 

[tomo pauk]

Snowygrouch mentioned that Jumo 211 enjoyed a 10 - 15 % superiority in lowest sfc over the Merlin III. I have checked a few sfc curves, including for the Jumo 211F/J and DB 605A, and in neither does the curve fall below (as far as I remember) 200 g/hp/h. However, e.g. the R-1820 and R-1830 can achieve figures around 190 g/hp/h.

There was several times on this forum a comparison between the likes, eg. the Packard Merlin V-1650-1 and DB 601E, showing the DB using less fuel than the Packard Merlin for same HP developed. And one of the closest comparable V-1710, the -81, consuming even more than V-1650-1, on every setting compared.
Eg. for 15 min rating, the V-1710-81 will consume 135 US gals/hr and make 1125 HP at ~15000 ft, the V-1650-1 will consume 120 US gals/hr and make 1120 HP at 18500 ft. The DB 601E will consume 116 US gals (440 l)/hr and make 1300 HP (1320 PS) at 15750 ft.

As for the for the best engine designing country, much depends on how much durability and reliability are valued. If those are considered significant, then both the British and the Amis were way above the rest.

This is probably the truth.

 

[Shortround6]

The fuel injected engine will make more power for less fuel at the higher power settings. The difference is much less marked at economical cruise.
Allisons could get into the mid 0.40s lbs per HP hour at low level cruise (1950rpm 25 in boost for instance) or about 205grams per hp hour. Exact power, altitude and model engine could vary. At military power they were running around 0.70lbs per hp hour (around 315g/HP/HR) Again give or take altitude and exact model and boost rating.
Extra fuel was being used for detonation suppression and internal engine cooling.
Once the carbureted engines go into lean settings there isn't a whole lot to choose between the the fuel injected and carbureted engines (Merlins being a bit worse due to lower compression).

 

[tomo pauk]

For making 828 HP (840 PS) at 16730 ft (5.1 km), the DB 601E will consume 63.4 US gals (240 L) per hour.
The V-1710-81 will make 760 HP at 16600 ft while consuming 63 US gal/hr.
V-1650-1 will make 758 HP at 16000 ft while also consuming 63 US gal/hr.

The DB 601E looks like having a 8-9% better specific consumption in these cruise settings.

 

[Shortround6]

Again, a lot depends on altitude and desired speed. The Allison at the power level you describe is at maximum lean cruise in the power chart (2300rpm and 31in) . However in the Flight operation chart (which does cover a wide range of weights and so is somewhat of an approximation) 2300rpm and 31 in can use anywhere from 58 gallons an hour at sea level up to 70 gal an hour at 15,000ft. Speed at 15,000ft is given as 294mph (true) and range on 130 gallons is around 545 miles. Slowing the plane down to 277mph at 15,000ft and using 2300rpm and 29in cuts the fuel burn to 56 gallons an hour and adds 95 miles to the "range" on the same amount of fuel.
Slowing down even further gets the P-40 down to 232mph ( rather useless for flying over occupied territory) while running at 1800rpm and 25in and burning 39 gallons an hour.
I would note that the difference between 0.45lbs/hp/hr and 0.50lbs/hp/hr is about 10% and that 0.45lbs/hp/hr is 204grams/per hour and 0.50lbs/hp/hr is 227grams per hour.

At higher power settings the German engines were much more efficient than the allied engines. At cruising speeds the differences had a lot more to do with the compression ratios used and other engine features/characteristics than the difference between fuel injection and carburetors.

 

[tomo pauk]

Re. altitude - I've tried to give the respective altitudes that would've been in th ballpark. For the DB 603E, the quoted 828 HP is the lowest power level I've able to find in short amount of time.
It was probably the combination of fuel injection and high compression ratio that was giving the DB engine the edge in sfc vs. the comparable Allied V12s, since even at low power levels there is still the issue of some cylinders (closer to the S/C) receiveng more fuel than needed so the other cylinders (further away from the S/C) will receive just the needed amount of fuel.
The other benefit from having fuel injection is that backfire screens can be removed (= less intake losses), and having a carb anywhere in the intake ducting will mean increased intake losses vs. the fuel injected engine.
On the other hand, too high the CR will significantly reduce the allowable boost, and production of fuel injected engines vs. carburated will probaly means that respective engines will be more expensive and it will take a bit a longer time to produce them.

 

[Trilisser]

Tomo, at least at low down e.g. a Twin Wasp (and low altitude comparison is a fairer comparison of the basic engine design) has considerably lower economical cruising sfc than the DB 605A (and from available data, the D), despite not inconsiderably lower compression ratio. What is more, in some conditions it was possible to lean manually further. Not possible with the DB.

While I would prefer direct injection, there seems to be some basic design features (flaws) in the DB and Jumo. E.g. the DB is very frail in regards of overspeeding as the manual states that if the take-off rpm is even momentarily exceeded by more than 2 %, the engine must be sent to the factory. This when e.g. the V-1710 tolerated 40 % momentary overspeeding.

 

[Shortround6]

When comparing specific fuel consumptions we have to be careful that we are comparing the right things. Tomo did a good thing trying compare at an altitude as close to each other as he could get figures for.
However there are a number of things that affect SFC. On aircraft engines there is the supercharger and it can be a big variable.
This is where altitude can make a big difference. Merlin XX (V1650-1?) used 219 hp to drive the supercharger at 15,000ft at 3000rpm and 9lbs boost. It needed 225hp at 20,000ft. This is using high gear 9.49:1. using low gear (8.15:1) at an altitude that allows the throttle plate to be wide open or close to it would require about 74% of the power to get the same 9lbs boost.
The Allison was using a smaller diameter impeller but the late model engines (like the -81) used 9.60 gears.
Granted cruise power doesn't need as much power going to the supercharger.
The Wright engines sometimes used 7.00 and 10.0 gears in two speed drives. With the supercharger needed power in proportion to the square of the speed of the impeller the Wright engines using those gears need just about 1/2 the power to drive the supercharger in low gear. This is one of the reasons for the very low FTH of some Wright engines in low gear. The P & W R-1830 used a 7.15 supercharger gear for most of it's versions, most single speeds and as the low gear on most 2 speed engines. Military power (1200hp) was somewhere between 3700ft and 4900ft depending on fuel grade. high gear on the 2 speed engines was usually 8.47 which required about 40% more power to drive.
These commercial derived radials offered low rpm in cruise (low internal friction) and low power going to the supercharger to help maximize the % of power going to the propeller which makes for a very good SPC in the advertising brochure.
However performance at higher altitudes wasn't so good (they often had to shifted into high gear at a lower altitude than the V-12s) and higher drag meant more power had to be used to get the same speed as the V-12s.

The DB and Jumo engines also ran a bit slower than the Merlin and Allison. Granted piston ring tension and a few other things could affect internal friction but friction losses also vary with the square of the speed the engine is running at.

There is little doubt that the German Fuel injected engines offered much better SFC at high power outputs but over the course of most flights their actual advantage in range/endurance was probably under 10%. The Merlin having one of the worst SFC fuel consumption's of a commonly used engine at many rpm and boost settings, it may have had a few sweet spots. The Allison could be around 8% better than the Merlin at times. This last is one of the reasons that the Merlin P-38 didn't make it off paper, and may be more important than the conspiracy theories about GM lobbing congress to prevent it.

 

[Trilisser]

SR6, that is exactly why I stressed the importance of low altitude comparisons: to minimize the effects of supercharger design on the performance of the basic engine.

Second, there seems to be evidence to suggest that neither the DB nor the Jumo liked to be cruised at the most efficient low rpm/high boost combination.

 

[Shortround6]

Low altitude benefits engines that use a low level of supercharging. Like the American radials. Finding figures for low level Merlins is a bit difficult. Like the MK VIII with a 6.313 supercharger gear on the full sized impeller or the MK 30/32 which used a reduced size impeller (not quite as small as a cropped Merlin 45).

There are a few other engines (or more than few) that had vibration problems at certain speed (rpm) ranges that would fall into a useful cruise speed/power output area of performance, forcing higher RPM that desirable.

 

[tomo pauk]

Tomo, at least at low down e.g. a Twin Wasp (and low altitude comparison is a fairer comparison of the basic engine design) has considerably lower economical cruising sfc than the DB 605A (and from available data, the D), despite not inconsiderably lower compression ratio. What is more, in some conditions it was possible to lean manually further. Not possible with the DB.

Shortround6 covered more or less the above. The military engines needed to satisfy several requirements, like this list by SR6 for example:
1. power
3. size (or weight)
5. weight (or size)
7. consumption

WIth 2, 4 and 6 being reliability.

A military engine that has a sizable S/C will have worse sfc than a civil engine with smallish S/C, where the list of requirements might start with reliability, then goes the consumption, etc.

While I would prefer direct injection, there seems to be some basic design features (flaws) in the DB and Jumo. E.g. the DB is very frail in regards of overspeeding as the manual states that if the take-off rpm is even momentarily exceeded by more than 2 %, the engine must be sent to the factory. This when e.g. the V-1710 tolerated 40 % momentary overspeeding.

I think you're giving the V-1710 too much credit re. over-revving. The max over-revving for it was 120 rpm atop the 3000 rpm mark (3120 rpm total obviously), ie. about 4 percent.
Fuel injection would've been a boon for the V-1710, that endured several redesigns of intake manifold in the lifetime, and needed backfire screens just to be on the safe side. OTOH, the fuel injection on the radial engines is not such a boon IMO, since the intake manifolds are of about equal length and shape for each cylinder.

SR6, that is exactly why I stressed the importance of low altitude comparisons: to minimize the effects of supercharger design on the performance of the basic engine.
Second, there seems to be evidence to suggest that neither the DB nor the Jumo liked to be cruised at the most efficient low rpm/high boost combination.

The 1st sentence is away from the 'purpose' of this forum & thread - we should be comparing military engines, and supercharger system is a crucial for those engines. Then, we should be comparing likes to the likes - V12s vs. V12s, just like the British compared in ww2 per above posts.
2nd sentence will need someone to prove a negative. Perhaps you could re-phrase it, and/or post the numbers?

 

[wuzak]

Fuel injection would've been a boon for the V-1710, that endured several redesigns of intake manifold in the lifetime, and needed backfire screens just to be on the safe side. OTOH, the fuel injection on the radial engines is not such a boon IMO, since the intake manifolds are of about equal length and shape for each cylinder.

Direct injection was proposed for the V-1710 (and X-3420) early in development but abandoned.

As for radial engines, the R-3350, at least, was converted to direct injection because of poor mixture distribution.

 

[Shortround6]

Allison didn't really abandon fuel injection as they never really started it. The Army proposed it and Allison asked which other project the Army wished to drop as Allison lacked sufficient engineering staff to work on ALL the Army's proposals. (Allison had fewer than 25 employees in the engineering dept at the the time, including the the guys that ran the blueprint machine). Army decided that fuel injection wasn't high on the list of priorities.

Aircraft engines are much more often an integrated system or approach than car engines. The Merlin got poor SPC in part because it used the lowest compression of any major WW II engine, however that allowed it to use more boost on the same grade of fuel for more peak power. However that fact is masked by the different fuel situations in the US and Britain in the late 30s and 1940/41 and part of 42. At some point in 1942 the US and Britain were issuing joint fuel specifications but by that time waaaayyyy to much time, money and tooling had been invested in the existing engines to try to change very much. The US going from 80 to 87 octane and then to 91 and on to 100 all within a few years. The British going from 77 to 87 to 100 but their 100 was different than the US 100. Nobody seems to know if the 87 octane was any different. Few engines were running enough boost to really tell any difference.

The LARGER German engines (DB 600 and Jumo 211) are larger only in displacement. In physical size and more importantly in weight they are with a few percent of the Merlin and Allison and just represent a different approach to getting power form 87 octane fuel. Large displacement but light low stressed engine (low rpm) vs small displacement and similar weight engine that is higher stressed ( high rpm).

With 87 octane fuel the Germans had little choice but to increase rpm for power as they couldn't increase cylinder pressure much. Higher RPM called for more weight.

AS far as durability goes, the Germans MAY have been facing materials shortages, I don't know one way or the other but am willing to give benefit of the doubt that their designs could have had better overhaul life if they had unrestricted access to the materials the Western Allies did. Or maybe the Western Allies had better lubricating oil or some other factor like different foundry techniques (casting/forging)
I will note that the Bristol Hercules seemed to depend on ball or roller bearings imported from Sweden at great trouble and expense. (Blockade runners) to a considerable extent. Hercules reputation with all British bearings????

This shows that design and execution is not always the same thing. Designers were always trying to push the envelope and sometimes pushed further than the state of the art allowed in manufacture. Late war R-2600 and R-2800 radials depended on manufacturing techniques that didn't exist in 1940-41 and mass manufacture depended on even more different techniques/processes.

Designer can put whatever he wants on paper, turning into reliable equipment at a feasible manufacturing cost is the trick.
Allison lucked onto an Aluminium casting process used by a team of sculptors that allowed for slightly lighter, slightly stronger block castings that required 10% less finish machining at some point in 1941(?) that was different than any other casting process in use by any major aluminium producer in the US.

I would note that the history of aviation fuel in the US/Britain has to be pieced together from different books and articles let alone the history in Germany or the other major countries.
The history of engine manufacture is even harder to find and lubricating oil is a total unknown. maybe it didn't change and maybe it did, I don't know one way or the other. Engine oil certainly changed post war.

 

[Trilisser]

I think you're giving the V-1710 too much credit re. over-revving. The max over-revving for it was 120 rpm atop the 3000 rpm mark (3120 rpm total obviously), ie. about 4 percent.

No, I am not overcrediting the Allison. Read Dan Whitney's bible. He states that later Allisons with the 12-counterweight crankshaft tolerated well over 4000 rpm.

 

[Trilisser]

SR6, I have seen all 4 side-by-side in a museum (Jumo 211F, DB 605, Merlin and Allison), and I have no doubt that both the Merlin and the Allison can be fitted "tighter". Especially the Allison is much more compact that the DB 605A.

As for the "only" choice for the Germans being increased rpm, I disagree. They could have adopted the Merlin route. However, for some peculiar reason, German engines were apparently unable to benefit from the "rich mixture response" to boost anti-knock limits. The classic book by Schlaifer & Heron (I had it as s stack of photocopies) states that German engines did not behave as expected when tested with rich mixtures. It is also quite odd that DB chose the increase the CR to the extreme when it was even then very clear that increasing CR is the most inefficient way to increase power with regards to anti-knock requirements.

As for the durability, years ago I talked to a Finnish aviation historian, who at that time was also the head of the FAF museum. He mentioned that even before the war German transport aircraft engines had substantially shorter TBO than comparable British and US engines. What is more, to me the mode of failure is very indicative of what is causing the problem. And the mode of failures in German engines (instead of rapid wear, parts broke) tend to suggest that something was wrong with the basic design, especially in the DB 600 series. I think it is worth noting how different paths DB and Jumo took: to DB all major power increases came though increased displacement whereas Jumo concentrated on refining the design. For example, the cylinder liner and block design of the DB is atrocious with its dry liners.

I do agree that someone with the proper engineering and industry credentials should do serious research for the fuels and lubricants of the major combatants.

 

[Trilisser]

By the way, it would be interesting to know what kind of TBOs were achived by the post-war French S.F.E.C.M.A.S. 12H (Jumo 213).

 

[tomo pauk]

No, I am not overcrediting the Allison. Read Dan Whitney's bible. He states that later Allisons with the 12-counterweight crankshaft tolerated well over 4000 rpm.

By the time we have V-1710 aproach close to 4000 rpm, there was several important design changes incorporated to the engine. You've mentioned the 12-counterweight crankshaft, I'll add the change in the technology of finnishing the crankshaft and other moving parts, like the shot-peening and nitriding (pg. 409 of the said Bible), new piston rings, new & stronger crankcase introduced in 1942, new reduction gearing (from internal spur to external spur). We won't have V-1710 C15 making 4000 rpm in combat more than once before it breaks.
BTW - Tex Johnson's racing Cobra II went to 'just' 3400 rpm, stock engines doing 3200 rpm of 1945+ era.
On the other hand, the DB 601A was rated to 2400 rpm, then itwas uprated to 2600 rpm by late 1940, and again uprated to 2800 rpm in 1941. That is without new crakshaft, new reduction gear, new piston rings.
Again on the claimed 4000 rpm - it is one thing to push the engine beyond limits on the bench, another thing is to push it when flying in service. The R-2800 passed 3500 HP on the bench already in 1944, the service machines never went beyond 2800 HP, and that is for the turboed engines. The V-1710 that went way above 3000 rpm and break will cost some Money when on the bench, the service engine will make the pilot loosing his life in combat that way.