"Slosh" or "Whine", Monsieur....? (ALLIED vs. AXIS Supercharger Drives)

[xylstra]

A DISCUSSION ABOUT ALLIED vs. AXIS Super-Charger Drives: Recently I had several exchanges with an author of a recently published, well-researched book on WW2 aero-engine development. I was rather interested in his commentary on the merits of the use of the Föttinger hydraulic coupling employed by Daimler-Benz in the DB-600 series engines. It's rather obvious that an infinitely variable drive will offer smooth transitions in boost level to track changes in altitude in contrast to the seemingly archaic 'staircase' boost intervals produced by fixed-ratio, multi-speed gear drives. However, there is always a penalty to pay, namely poor transmission efficiency, leading to poor fuel economy on top of an already thirsty engine. He rather astounded me by stating that the hydraulic coupling saved 10% of engine power on take-off (altitude performance not stated). I remain somewhat sceptical. My understanding from what I've read of hydraulic couplings of the Föttinger type is that transmission efficiency barely rises above ~80% (see: https://journals.sagepub.com/doi/abs/10.1243/PIME_PROC_1948_158_015_02?journalCode=pmea) whereas a well designed meshed gear-train can and often does exceed ~90%. That's by no means the only issue, the early DB's were prone to over-heating of the hydraulic fluid although apparently solved in later marks. Usually this requires additional cooling capacity which in turn means more weight and incurring a parasitic drag penalty to boot. Moreover, the Föttinger super-charger drive still requires a step-up gear and it's efficiency losses are additional.
My understanding is that the R.A.E. extensively tested salvaged DB601's and I believe that ROLLS-ROYCE separately tested the DB601, not forgetting that, vice-versa the Germans also comparatively tested salvaged Allied aero-engines [Does anyone have copies of any of these reports?] I am given to understand that the British acknowledged the benefits of this form of super-charger drive yet ROLL's persevered with the change-speed gear drive. I assume that they did so due to the greater mechanical efficiency of this drive albeit prepared to suffer the additional mechanical complexity and the occasional 'crack-ups' (!).
But am I wrong, was the Föttinger in fact, actually more efficient? Intelligent analyses, please.
As if to further complicate the issue, it should not be over-looked that a variant of the American ALLISON V-12 also employed a hydraulically coupled super-charger drive but oddly, this was never extended to other variants in the range. Why was that? Did they encounter problems that were just too difficult or time-consuming to solve or were there other more prosaic reasons?

 

[TomcatViP]

It's a great question that would have to take into account a broader aspect in the context of a global war and the scarcity of resources. Yes, variable coupling is more efficient (more torque, hence less blade, usable with a wider chord).
Yet you have the need of a large production batch and the fast increase of performances as needed by the user. The easiest way (and cheaper at the end) is to increase material quality as on a cost/logistic perspective you are working on larger sample of resource with a minimal change of methods. This is for example the earlier road followed by civil engine manufacturer until CFD's R&D was the main source of improvement.
In the context of Nazi Germany, encroached in a fight with minimal resource at her disposition (the Madness of the Madmen), rapid improvement in material was a route that they could not follow in term of delivered quantities to the front-line. As the jet engine history told us (and the turbo impellers), improving the material tolerance to temperature and strain while providing better performances was out of reach for them.
The Fottinger route gave them a mean to follow the pace of engine performance growth with a minimal impact on production or the need for mass produced exotic resources. Such coupling's performances are dependent of geometric variables at the slight expense of an efficiency drop as you noted.

IMOHO, this should be one of the main reason to take into consideration dealing with such question.

Hope it helps..

 

[Dagger]

A DISCUSSION ABOUT ALLIED vs. AXIS Super-Charger Drives: ......................... He rather astounded me by stating that the hydraulic coupling saved 10% of engine power on take-off (altitude performance not stated). I remain somewhat sceptical. My understanding from what I've read of hydraulic couplings of the Föttinger type is that transmission efficiency barely rises above ~80% (see: https://journals.sagepub.com/doi/abs/10.1243/PIME_PROC_1948_158_015_02?journalCode=pmea) whereas a well designed meshed gear-train can and often does exceed ~90%. ....................................

My understanding:
At take-off the DB engine supercharger would run at lower speed (thanks to its hydraulic coupling) than the Merlin supercharger, and therefor use less power from the crankshaft than the Merlin. That outweighed the difference in transmission efficiency.
The only way for the Merlin to avoid too high boost pressure (avoid knocking) during take-off was to throttle the supercharger inlet.

 

[xylstra]

A DISCUSSION ABOUT ALLIED vs. AXIS Super-Charger Drives: ......................... He rather astounded me by stating that the hydraulic coupling saved 10% of engine power on take-off (altitude performance not stated). I remain somewhat sceptical. My understanding from what I've read of hydraulic couplings of the Föttinger type is that transmission efficiency barely rises above ~80% (see: https://journals.sagepub.com/doi/abs/10.1243/PIME_PROC_1948_158_015_02?journalCode=pmea) whereas a well designed meshed gear-train can and often does exceed ~90%. ....................................

My understanding:
At take-off the DB engine supercharger would run at lower speed (thanks to its hydraulic coupling) than the Merlin supercharger, and therefor use less power from the crankshaft than the Merlin. That outweighed the difference in transmission efficiency.
The only way for the Merlin to avoid too high boost pressure (avoid knocking) during take-off was to throttle the supercharger inlet.

I think that's a little simplistic and needs to be scrutinised more closely. Hydraulic couplings exhibit high-slip when first started from a stand-still but this characteristic exists for mere moments unless the control system and impeller geometry are deliberately designed to achieve a stable rate of slip in this take-off mode. Has this been authoritatively studied and described anywhere? ....whereas the SAE have numerous papers on slippage in fluid converters in automotive transmissions.

 

[Hobbes]

Yes, the DB supercharger was designed to allow a variable rate of slip (by varying the amount of oil in the coupling).

The efficiency loss in the Merlin was the result of throttling back, which increases the pumping losses by restricting the airflow.

 

[xylstra]

Yes, the DB supercharger was designed to allow a variable rate of slip (by varying the amount of oil in the coupling).

The efficiency loss in the Merlin was the result of throttling back, which increases the pumping losses by restricting the airflow.

Sure, but I think you're making my point about the relative inefficiency of the Fottinger coupling. Only when it is completely full of fluid and absent of any empty voids or air pockets can it ever hope to obtain a high efficiency which is almost never. Most of the time the fluid cavity will be partially empty while operating at some intermediate boost pressure. Being effective does not equate to being efficient. The two are mutually exclusive.
Also, some of the DB series engines also employed a form of intake manifold throttling so pumping losses arising are not exclusive to ROLLS ROYCE all of which still leaves a fundamental question unanswered: if the Fottinger supercharger drive was so efficient and effective then why didn't R-R copy it - and also JUNKERS for that matter?

 

[Hobbes]

you have to look at the efficiency of the whole engine.
The Merlin supercharger required in the region of 400 hp of drive power- even at low altitude. To prevent overboosting, they had to throttle back, making the supercharger pull against a vacuum in the inlet tract.
In the DB compressor, supercharger drive power could be reduced at low altitude. So instead of sending 400 hp to the supercharger, they could send e.g. 200 hp. As long as that savings is larger than the losses incurred in the hydraulic coupling, you're coming out ahead.

if the Fottinger supercharger drive was so efficient and effective then why didn't R-R copy it

what impressed me throughout Calum's book was RR's willingness to stick to a proven concept. They maximized the cost:benefit ratio of their R&D by ruthlessly dismissing lines of development that had promise, but would also be complex and take a long time to perfect.

 

[xylstra]

It still leaves out any clear determination of the numerical mechanical efficiency of the hydraulic coupling. I'm starting to conclude that no-one can actually present these numbers for specific examples for say, the DB-series supercharger whereas there are plenty of sources from the automotive world that attest to little more than 80% (did you actually see the graph in the link I posted?) can be expected, why else do transmission manufacturers add mechanical complexity by producing lock-up torque-converters in order to improve their underlying inefficiency?
You also forget that it's not just about ROLLS-ROYCE; what about Allison and Junkers who had already invested R & D and yet did not extend hydraulic coupling driven super-chargers across their engine range. Why??

 

[Hobbes]

From this forum:

Geoffery Wilde at Rolls did a study on the coupling for Stanley Hooker, as he had worked
previously with these couplings and knew their workings. Because the Merlin was a smaller
engine they knew they would have to boost it considerably higher than the DB in order to
have a chance. This means higher compressor air power, hence in a Merlin with a higher
pressure ratio compressor, the power lost to the coupling will go up and up.

Geoffery worked out that a Merlin equipped with a fluid drive (there are layout drawings
and a complete theoretical set of curves), that with projected boost developments,
that the drag of the required oil cooler would almost nullify the advantage of the
coupling (which is basically that it "fills-in" the area under the "saw-tooth"
pattern of a gearbox equipped supercharged aero engine).

 

[xylstra]

That's along the lines of my intuition; that any benefit would be eroded and off-set to some degree by the need to maintain the coupling fluid temperature between safe operating limits which translates to extra installation weight, parasitic drag and sheer plumbing complexity. To some extent though it does make the boost control architecture easier to integrate into a KommandoGerat type of control system.

 

[iverson]

<snip>
You also forget that it's not just about ROLLS-ROYCE; what about Allison and Junkers who had already invested R & D and yet did not extend hydraulic coupling driven super-chargers across their engine range. Why??

Allison did, in fact, develop a hydraulically driven auxiliary stage supercharger for the late-series V-1710s used on the P-63 and P-82. They weren't used more extensively because Allison had to develop them at its own expense. The company's main customer, the USAAF, was only interested in turbosuperchargers. The company studied but did not develop a two-stage gear-driven superhcharger for the same reason. See the book <i>Vee's For Victory!: The Story of the Allison V-1710</i>.

 

[Nik]

Was there any hybrid prototype that was geared plus hydraulic ??

 

[Dagger]

The German DB601 engine superchargers were geared between engine and hydraulic coupling, so as to increase the SC speed far above that of the engine, just like in the Merlin.
The hydraulic coupling transferred this high speed to the SC, with the possibility to reduce the high speed somewhat so as to limit the boost at take-off and altitudes below critical altitude.

 

[edwest4]

The next book by the author of The Secret Horsepower Race, Calum Douglas, will be Turbocharger Design. Contents: https://www.calum-douglas.com/contents/

 

[xylstra]

<snip>
You also forget that it's not just about ROLLS-ROYCE; what about Allison and Junkers who had already invested R & D and yet did not extend hydraulic coupling driven super-chargers across their engine range. Why??

Allison did, in fact, develop a hydraulically driven auxiliary stage supercharger for the late-series V-1710s used on the P-63 and P-82. They weren't used more extensively because Allison had to develop them at its own expense. The company's main customer, the USAAF, was only interested in turbosuperchargers. The company studied but did not develop a two-stage gear-driven superhcharger for the same reason. See the book <i>Vee's For Victory!: The Story of the Allison V-1710</i>.

It's the "development" side of the issue that interests me Iverson: what efficiencies did they obtain from this drive versus a fully-geared drive?

 

[xylstra]

The German DB601 engine superchargers were geared between engine and hydraulic coupling, so as to increase the SC speed far above that of the engine, just like in the Merlin.
The hydraulic coupling transferred this high speed to the SC, with the possibility to reduce the high speed somewhat so as to limit the boost at take-off and altitudes below critical altitude.

Counter-intuitive way to do. Would have made more sense to put the gears after the coupling to avoid frictional heating of the fluid which I understand to have been one of the operational problems experienced by pilots.

 

[Dagger]

In a DB engine the supercharger and hydraulic drive both run at about 10 (ten) times the speed of the crankshaft.
What matters with both is the tip speed. Running them at a lower shaft speed would require a larger diameter to achieve the same tip speed.

If the hydraulic coupling would have run at the speed of the crankshaft (with the gears for the supercharger aft) its diameter would have to be too big to fit in a streamlined engine.

Moreover the supercharger in a DB engine is on the side of the engine, not at the back like with a Merlin. Presumably that was done to have room at the back for the cannon that fired through the propeller shaft. In such configuration gears between hydraulic coupling and supercharger would be impractical.

 

[xylstra]

A DISCUSSION ABOUT ALLIED vs. AXIS Super-Charger Drives: ......................... He rather astounded me by stating that the hydraulic coupling saved 10% of engine power on take-off (altitude performance not stated). I remain somewhat sceptical. My understanding from what I've read of hydraulic couplings of the Föttinger type is that transmission efficiency barely rises above ~80% (see: https://journals.sagepub.com/doi/abs/10.1243/PIME_PROC_1948_158_015_02?journalCode=pmea) whereas a well designed meshed gear-train can and often does exceed ~90%. ....................................

My understanding:
At take-off the DB engine supercharger would run at lower speed (thanks to its hydraulic coupling) than the Merlin supercharger, and therefor use less power from the crankshaft than the Merlin. That outweighed the difference in transmission efficiency.
The only way for the Merlin to avoid too high boost pressure (avoid knocking) during take-off was to throttle the supercharger inlet.

I think that's a little simplistic and needs to be scrutinised more closely. Hydraulic couplings exhibit high-slip when first started from a stand-still but this characteristic exists for mere moments unless the control system and impeller geometry are deliberately designed to achieve a stable rate of slip in this take-off mode. Has this been authoritatively studied and described anywhere? ....whereas the SAE have numerous papers on slippage in fluid converters in automotive transmissions.

Here's a couple of links from the automotive world that deal with efficiencies of hydraulic torque-converters:

Performance Analysis and Improvement of Flat Torque Converters Using DOE Method - Chinese Journal of Mechanical Engineering

Automotive torque converters have recently been designed with an increasingly narrower profile for the purpose of achieving a smaller axial size and reducing weight. Design of experiment (DOE) and computational fluid dynamics (CFD) techniques are applied to improve the performance of a flat...

cjme.springeropen.com

....and:

Greenlings: Why do automatic transmissions now get better fuel efficiency than manuals?

Not so long ago, it was common for automatic transmissions to be referred to as slushboxes, since that's how they often behaved. Rather than use a mechanical clutch, traditional automatic transmissions use a fluid coupling between the engine and the gear-sets to transmit drive torque. This...

www.google.com

I don't doubt the performance and mechanical benefits of a variable rpm hydraulically driven supercharger but I remain sceptical on the issue of efficiency.

 

[Dagger]

Sure, the hydraulic drive will have lower a efficiency than a gear, but when comparing Merlin and DB power at take-of (what this topic started with) the energy waste by the Merlin supercharger is much bigger than the difference between gear and hydr. coupling.
It is only when the aircraft climbs close to the critical altitude (rated altitude, full throttle altitude), and above it, that the Merlin waste becomes less than the DB hydr. coupling waste.

Note moreover that the purpose of a supercharger is only to increase the air mass flow through the piston engine.
Apart from that it is an energy waster, unless it would be driven by a turbocharger.
The Merlin was only 27 liters with a compression ratio of only 6
The DB601 was 34 liters with a compression ratio of 6.9 , later DB engines had an even higher compression ratio.
In theory: the higher the compression ratio the higher the efficiency of a piston engine.
Note that compression ratio is a volumetric ratio. The associated pressure ratio is higher.
Consequently the DB's needed less boost to achieve the same air mass throughput and the same piston pressure at TDC than the Merlin. Less boost means: less energy waste by the mechanical supercharger.

In my opinion the main advantage of the later Merlins was the intercooler (aftercooler), much less all the other stuff that one reads about the Merlin.

 

[xylstra]

Sure, the hydraulic drive will have lower a efficiency than a gear, but when comparing Merlin and DB power at take-of (what this topic started with) the energy waste by the Merlin supercharger is much bigger than the difference between gear and hydr. coupling.
It is only when the aircraft climbs close to the critical altitude (rated altitude, full throttle altitude), and above it, that the Merlin waste becomes less than the DB hydr. coupling waste.

Note moreover that the purpose of a supercharger is only to increase the air mass flow through the piston engine.
Apart from that it is an energy waster, unless it would be driven by a turbocharger.
The Merlin was only 27 liters with a compression ratio of only 6
The DB601 was 34 liters with a compression ratio of 6.9 , later DB engines had an even higher compression ratio.
In theory: the higher the compression ratio the higher the efficiency of a piston engine.
Note that compression ratio is a volumetric ratio. The associated pressure ratio is higher.
Consequently the DB's needed less boost to achieve the same air mass throughput and the same piston pressure at TDC than the Merlin. Less boost means: less energy waste by the mechanical supercharger.

In my opinion the main advantage of the later Merlins was the intercooler (aftercooler), much less all the other stuff that one reads about the Merlin.

Take-off performance is such a small proportion of the overall mission profile of any piston-engined fighter that efficiency advantages are irrelevant for this brief interval. It's the overall efficiency for the bulk of operational time that has significant impact on fuel consumption.
This website has some of the best and authoritative technical reports you're likely to find including an especially good one on the R-R Merlin:

WWII Aircraft Performance

 

[Dagger]

What do you mean by "this website" ? www.secretprojects.co.uk ?

I am familiar with websites www.wwiiaircraftperformance.org and ww2aircraft.net

 

[xylstra]

What do you mean by "this website" ? www.secretprojects.co.uk ?

I am familiar with websites www.wwiiaircraftperformance.org and ww2aircraft.net

My original message included the link to the site (now amended). Just one of the vagaries of site technical issues/network issues I'm afraid.

 

[iverson]

Take-off performance is such a small proportion of the overall mission profile of any piston-engined fighter that efficiency advantages are irrelevant for this brief interval. It's the overall efficiency for the bulk of operational time that has significant impact on fuel consumption.

This website has some of the best and authoritative technical reports you're likely to find including an especially good one on the R-R Merlin:

WWII Aircraft Performance

I'm not an internal combustion expert by any stretch of the imagination. But I suspect that efficiency was simply not an important design metric for most combat engines. Such engines were not designed to produce maximum power efficiently. They were designed to maintain a militarily adequate power level throughout a range of tactically significant operating altitudes. The "critical altitude" was the part of this range where maximum power was required and thus the altitude for which the engineering solutions were optimized.

How the above goal was achieved was a matter of constraints and trade offs.

The Merlin had the gear-driven supercharger because it was a known quantity, perfected on the Kestrel and R and well-suited to Rolls-Royce's manufacturing and development approaches, its understanding of fuel chemistry, and Britain's access to high-octane American fuel. After a rocky start, it worked and proved amenable to a lot of development. RR was able to develop multispeed gears to maintain power over the critical altitude and multiple stages and intercoolers to improve supercharger efficiency at ever higher altitudes. But the supercharger was complex, heavy, and pretty much built-in to the engine. Fitting the gear train was a lot harder than bolting a turbo (or an auxiliary stage) to an Allison.

The latter mattered to Allison. The Allison V-1710 had an allegedly better combustion chamber and a modular design suited both for volume production and for customization to the customer's requirements. It didn't have a lot of gear-driven supercharger experience because of the USAAF's insistence on turbochargers, which were third-party products. It favored the hydraulically driven auxiliary supercharger over a gear-driven type because it was a modular, bolt-on solution rather than an integral part of a dedicated engine design (as in the case of the Merlin).

The Daimler Benz and Junkers engines had hydraulic superchargers. But, to complicate things, they were also larger, had direct fuel-injection, had less access to high-quality fuel, and used water-injection to control detonation and nitrous oxide for boosting power. All of these features would, I suspect, reduce the need for highly efficient supercharging.

So, when talking of efficiency, what do we mean? The gear-driven Rolls-Royce supercharger may or may not have required less horsepower to drive it than a hydraulic equivalent would. But was it as "weight-efficient"? Or as cost efficient (all those high-precision hand-fitted gears)? Or as efficient at maintaining power over a range of altitudes?

The bottom line is, I think, that efficiency of any kind is pretty much irrelevant in wartime. These engines were designed and developed out of necessity at a time of looming international crises and constrained by economic, military, political, and industrial realities that did not favor ideal design. Engines with widely differing concepts succeeded (Merlin, R-2800, DB601/3, VK105PF, etc.) or failed (HS12Y and HS12Z, GR14N) based more on chance, time, and place than on inherent design excellence.

 

[Thos9]

Pratt & Whitney adopted hydraulic variable-speed supercharging around war's end. The Goodyear F2G Corsair used the single-stage variable speed R-4360-4 and the later Grumman F8F-2 Bearcat used the R-2800-30W, also single-stage. The Vought F4U-5 used the two-stage R-2800-32W with dual "sidewinder' hydraulic first stages and a mechanical second stage.

Same with the Packard V-1650-19. I've seen a chart (can't find it now unfortunately ) showing how well the variable-speed supercharger filled in the "saw teeth" of the two-stage V-1650 Merlin's output.

Junkers came round too. IRRC both the Jumo 213J and 222E/F employed two-stage superchargers with hydraulic first- and mechanical second-stage drive. In 1943 BMW successfully tested a 2000 PS 801E with single-stage hydraulic variable-speed supercharger but for some reason it was not adopted.

The Japanese (Aichi for the Navy and Kawasaki for the Army) copied DB techology. The USSR never tried it during the war.

AFAIK the variable-speed supercharger was never contemplated for adoption in Britain.

The "efficiency" of supercharger drives needs to be judged taking account of the mode of fuel insertion, whether by direct injection (which supports full cylinder scavenging and internal cooling through wide valve overlap) or by carburetion (which doesn't). Direct injection favours lower sfc.

 

[Hobbes]

But I suspect that efficiency was simply not an important design metric for most combat engines.

it may not have been the top priority, but since fuel efficiency impacts range and combat load, it's still a consideration. Certainly for the Allies in WW2, which struggled to build fighters that had enough range to escort bomber missions.

 

[iverson]

But I suspect that efficiency was simply not an important design metric for most combat engines.

it may not have been the top priority, but since fuel efficiency impacts range and combat load, it's still a consideration. Certainly for the Allies in WW2, which struggled to build fighters that had enough range to escort bomber missions.

I agree that range mattered. But I don't see that it was much of a consideration for the mainly fighter engines that we have discussed. The only engines where I am certain that efficiency was a prime objective are diesels conceived, in most cases, for commercial service (Junkers, Guiberson, Clerget) and the gigantic, turbocharged Studebaker H-9350 designed for intercontinental bombers.

Range was clearly not a priority for most combatants. European types like the Spitfire, Hurricane, Bf109, Fw190, and Yak 1/7 were all conceived as interceptors or tactical fighters and were thus relatively short-ranged. Even US types like the P-38 were intended to be fast-climbing interceptors, not escorts. When their engines were in development, the US and Britain thought gun turrets would be enough to let unescorted bombers fight their way through to their targets unescorted. So speed, ceiling, rate of climb, heavy armament, and, thus, power at altitude were the priorities, not efficiency.

Besides, with more power, fighters could always carry more fuel. Prewar, fighters intended as escorts, like the Bf110, were typically given twin engines for lifting large fuel loads. With a sufficiently powerful engine, good aerodynamic, and efficient structural design design, single-engined fighters could carry large quantities of fuel internally, as on the A6M and P-51, and in drop tanks. These aircraft could reduce fuel consumption by cruising to their targets at lower power settings (and more bomber-like speeds) and then shed their drop tanks and go to full power for combat.

In short, design was a matter of tradeoffs and priorities. In war time, when fuel was plentiful and its cost no object, a powerful engine could offset poor economy by carrying more and still be fit to fight after burning the extra. But a fuel-efficient engine could not make up for even comparatively modest reductions in peak combat power and combat performance.

 

[xylstra]

But I suspect that efficiency was simply not an important design metric for most combat engines.

it may not have been the top priority, but since fuel efficiency impacts range and combat load, it's still a consideration. Certainly for the Allies in WW2, which struggled to build fighters that had enough range to escort bomber missions.

I quite agree. Given the number of Merlin's produced for bomber use why would they want a mechanically inefficient hydraulic (ie, fuel-hungry) supercharger drive given that the flight profile would largely consist of long-range cruising at largely fixed altitudes since they wouldn't be expected to engage in aerial dogfighting with BF 109's! I also don't agree with some posters who think that an hydraulic drive would be too expensive to develop. Why would anyone in their right mind want to re-invent the wheel?: just simply copy the DB's Föttinger coupling with local adaptations. Probably less diifficult than developing a fuel injection system from scratch.

 

[xylstra]

Pratt & Whitney adopted hydraulic variable-speed supercharging around war's end. The Goodyear F2G Corsair used the single-stage variable speed R-4360-4 and the later Grumman F8F-2 Bearcat used the R-2800-30W, also single-stage. The Vought F4U-5 used the two-stage R-2800-32W with dual "sidewinder' hydraulic first stages and a mechanical second stage.

Same with the Packard V-1650-19. I've seen a chart (can't find it now unfortunately ) showing how well the variable-speed supercharger filled in the "saw teeth" of the two-stage V-1650 Merlin's output.

Junkers came round too. IRRC both the Jumo 213J and 222E/F employed two-stage superchargers with hydraulic first- and mechanical second-stage drive. In 1943 BMW successfully tested a 2000 PS 801E with single-stage hydraulic variable-speed supercharger but for some reason it was not adopted.

The Japanese (Aichi for the Navy and Kawasaki for the Army) copied DB techology. The USSR never tried it during the war.

AFAIK the variable-speed supercharger was never contemplated for adoption in Britain.

The "efficiency" of supercharger drives needs to be judged taking account of the mode of fuel insertion, whether by direct injection (which supports full cylinder scavenging and internal cooling through wide valve overlap) or by carburetion (which doesn't). Direct injection favours lower sfc.

Great post Thos9. Would be nice to see some of the design studies used in the course of laying down these variants.

 

[Calum Douglas]

Pratt & Whitney adopted hydraulic variable-speed supercharging around war's end. The Goodyear F2G Corsair used the single-stage variable speed R-4360-4 and the later Grumman F8F-2 Bearcat used the R-2800-30W, also single-stage. The Vought F4U-5 used the two-stage R-2800-32W with dual "sidewinder' hydraulic first stages and a mechanical second stage.

Same with the Packard V-1650-19. I've seen a chart (can't find it now unfortunately ) showing how well the variable-speed supercharger filled in the "saw teeth" of the two-stage V-1650 Merlin's output.

Junkers came round too. IRRC both the Jumo 213J and 222E/F employed two-stage superchargers with hydraulic first- and mechanical second-stage drive. In 1943 BMW successfully tested a 2000 PS 801E with single-stage hydraulic variable-speed supercharger but for some reason it was not adopted.

The Japanese (Aichi for the Navy and Kawasaki for the Army) copied DB techology. The USSR never tried it during the war.

AFAIK the variable-speed supercharger was never contemplated for adoption in Britain.

The "efficiency" of supercharger drives needs to be judged taking account of the mode of fuel insertion, whether by direct injection (which supports full cylinder scavenging and internal cooling through wide valve overlap) or by carburetion (which doesn't). Direct injection favours lower sfc.

Great post Thos9. Would be nice to see some of the design studies used in the course of laying down these variants.

As is mentioned in my book, the Merlin came extremely close to ending up with a Föttinger coupling, Geoffrey Wilde did the layout for it and was actually given a prize for coming up with the idea. This was all in a memo at RRHT written by Wilde called "People Power and Pumping, my first year at Rolls-Royce". Many of the sludging problems it suffered with in the war in German use, were due to oil quality issues. Wilde`s first scheme had it on the end of the crank running at 1:1, with the gears afterwards, but you needed a HUGE coupling, so the next step he was told to do was to redesign it to be speeded up and hence miniaturised. This was what was being done when all the work was interrupted by the invasion of Poland, when, seeing what was going to come soon, all non-essential work was "canned", including Wilde`s hydraulically driven Merlin supercharger.

 

[xylstra]

Pratt & Whitney adopted hydraulic variable-speed supercharging around war's end. The Goodyear F2G Corsair used the single-stage variable speed R-4360-4 and the later Grumman F8F-2 Bearcat used the R-2800-30W, also single-stage. The Vought F4U-5 used the two-stage R-2800-32W with dual "sidewinder' hydraulic first stages and a mechanical second stage.

Same with the Packard V-1650-19. I've seen a chart (can't find it now unfortunately ) showing how well the variable-speed supercharger filled in the "saw teeth" of the two-stage V-1650 Merlin's output.

Junkers came round too. IRRC both the Jumo 213J and 222E/F employed two-stage superchargers with hydraulic first- and mechanical second-stage drive. In 1943 BMW successfully tested a 2000 PS 801E with single-stage hydraulic variable-speed supercharger but for some reason it was not adopted.

The Japanese (Aichi for the Navy and Kawasaki for the Army) copied DB techology. The USSR never tried it during the war.

AFAIK the variable-speed supercharger was never contemplated for adoption in Britain.

The "efficiency" of supercharger drives needs to be judged taking account of the mode of fuel insertion, whether by direct injection (which supports full cylinder scavenging and internal cooling through wide valve overlap) or by carburetion (which doesn't). Direct injection favours lower sfc.

Great post Thos9. Would be nice to see some of the design studies used in the course of laying down these variants.

As is mentioned in my book, the Merlin came extremely close to ending up with a Föttinger coupling, Geoffrey Wilde did the layout for it and was actually given a prize for coming up with the idea. This was all in a memo at RRHT written by Wilde called "People Power and Pumping, my first year at Rolls-Royce". Many of the sludging problems it suffered with in the war in German use, were due to oil quality issues. Wilde`s first scheme had it on the end of the crank running at 1:1, with the gears afterwards, but you needed a HUGE coupling, so the next step he was told to do was to redesign it to be speeded up and hence miniaturised. This was what was being done when all the work was interrupted by the invasion of Poland, when, seeing what was going to come soon, all non-essential work was "canned", including Wilde`s hydraulically driven Merlin supercharger.

 

[xylstra]

Hi Callum, you know, I'm a victim of 'bad timing'! I have a close friend whose father worked in RR's engine bench testing department in WW2 (post-war he became one of New Zealand'd leading track racers) but of course he's now long dead so unless, otherwise by seance he's not around to interview. Then, those many decades ago when I was a university student we received a visit from an engineer who helped design the Merlin and who presented a lecture on its design. To this day I bitterly regret being a consciencous student by foolishly attending my scheduled engineering lab class instead of the RR lecture. I was well aware that RR had assessed the DB's hydraulic supercharger drive and acknowledged it's benefits so I specifically wanted to ask him why RR didn't copy it. Robbie Coltrane also got close to answering this question in his TV series but just stopped short of the answer. Your research on the subject seems to have provided the answer to this mini-mystery so perhaps this particular chapter can finally be closed.

 

[Nicknick]

when regarding the efficiency, you should keep in mind, that throtteling is required in a fixed stage, mechanical drive train for the superchargers. It s better to have the enegry losses in the coupling, where the heat can be transfered to the cooling system, than by additional gas work (by throtteling and supercharging) which increases the charge temperature.

 

[edwest4]

Gentlemen,

On offer is a candy store here: http://www.wwiiaircraftperformance.org/

 

[tomo pauk]

You also forget that it's not just about ROLLS-ROYCE; what about Allison and Junkers who had already invested R & D and yet did not extend hydraulic coupling driven super-chargers across their engine range. Why??

Late to the party ....
Never the less. Supercharger(s) that have impellers driven via hydraulic coupling have considerable advantage over the impellers(s) driven via 1-speed gearing at lower altitudes. Advantage is less pronounced vs. impellers driven via 2-speed gearing, since the additional low-speed gearing will mean that less of engine power is 'sacrificed' to drive the supercharger's impeller. Junkers have had a 2-speed supercharged Jumo 211A by the time DB made the 601A (and the 2-speed S/Ced Jumo 210 versions even earlier), while RR had the Merlin X and were preparing the Merlin XX for production.
Allison have had the hydraulic-driven impeller of the auxiliary S/C stage in production by 1943 for the P-63.

We can note that DB 601/603/605 also employed throttling, especially under 2 km of altitude. Thus the take off power in practice was still not better than on comparable Jumo engines, or vs. the 2-speed supercharged RR engines.
Also, and probably more important: improvement in supercharger system beats the presence of hydraulic coupling for the S/C. Here the German engines started lacking vs. RR engines as war dragged on.

 

[xylstra]

You also forget that it's not just about ROLLS-ROYCE; what about Allison and Junkers who had already invested R & D and yet did not extend hydraulic coupling driven super-chargers across their engine range. Why??

Late to the party ....
Never the less. Supercharger(s) that have impellers driven via hydraulic coupling have considerable advantage over the impellers(s) driven via 1-speed gearing at lower altitudes. Advantage is less pronounced vs. impellers driven via 2-speed gearing, since the additional low-speed gearing will mean that less of engine power is 'sacrificed' to drive the supercharger's impeller. Junkers have had a 2-speed supercharged Jumo 211A by the time DB made the 601A (and the 2-speed S/Ced Jumo 210 versions even earlier), while RR had the Merlin X and were preparing the Merlin XX for production.
Allison have had the hydraulic-driven impeller of the auxiliary S/C stage in production by 1943 for the P-63.

We can note that DB 601/603/605 also employed throttling, especially under 2 km of altitude. Thus the take off power in practice was still not better than on comparable Jumo engines, or vs. the 2-speed supercharged RR engines.
Also, and probably more important: improvement in supercharger system beats the presence of hydraulic coupling for the S/C. Here the German engines started lacking vs. RR engines as war dragged on.

Your reply, in common with many others alludes to the underlying preconception that leads to the gears versus

You also forget that it's not just about ROLLS-ROYCE; what about Allison and Junkers who had already invested R & D and yet did not extend hydraulic coupling driven super-chargers across their engine range. Why??

Late to the party ....
Never the less. Supercharger(s) that have impellers driven via hydraulic coupling have considerable advantage over the impellers(s) driven via 1-speed gearing at lower altitudes. Advantage is less pronounced vs. impellers driven via 2-speed gearing, since the additional low-speed gearing will mean that less of engine power is 'sacrificed' to drive the supercharger's impeller. Junkers have had a 2-speed supercharged Jumo 211A by the time DB made the 601A (and the 2-speed S/Ced Jumo 210 versions even earlier), while RR had the Merlin X and were preparing the Merlin XX for production.
Allison have had the hydraulic-driven impeller of the auxiliary S/C stage in production by 1943 for the P-63.

We can note that DB 601/603/605 also employed throttling, especially under 2 km of altitude. Thus the take off power in practice was still not better than on comparable Jumo engines, or vs. the 2-speed supercharged RR engines.
Also, and probably more important: improvement in supercharger system beats the presence of hydraulic coupling for the S/C. Here the German engines started lacking vs. RR engines as war dragged on.

Well written post but see my earlier comments on fighter versus bomber use.

 

[tomo pauk]

Well written post but see my earlier comments on fighter versus bomber use.

I think that we're mostly in agreement. Plus, British and Germans were installing the engines with two-speed superchargers in their bombers by the time DB 601A materialized, it's supercharger drive offering, in theory, an infinite number of speeds between the min and max ratios. Advent and wide use of 100 octane fuel by the RAF meant that Merlin III, XII and 45 (engines usually used on fighters) were still useful engines at lower altitudes, too, despite having just 1 speed for their respective superchargers. Hi-oct fuel allowing for greater boost pressures (and thus power) without the detonation despite the throttle being open under the rated height.

As for the V-1710 with hydraulic coupling, that was used to drive the impeller on auxiliary S/C of engines used on P-A/C/D and P-82E. The engine-stage S/C still used 1-speed drive. Auxiliary S/C + engine-stage S/C = 2-stage supercharging. A few thousand of those engines were produced.

 

[Nicknick]

I’m pretty sure, that German engines preferred the hydraulic coupling because of their automated engine control system (Kommandogerät). This device needs steady functions for the air and fuel flow and switching would have been an issue which couldn’t be solved by atomization. Later on, the Americans used hydraulic coupling for the supercharger in the Bearcat (also single staged) in combination with something of an American “Kommandogerät” or let’s say engine control unit. There is a strong connection between these two technics.

 

[HoHun]

Hi Nicknick,

I’m pretty sure, that German engines preferred the hydraulic coupling because of their automated engine control system (Kommandogerät). This device needs steady functions for the air and fuel flow and switching would have been an issue which couldn’t be solved by atomization.

Would you care to elaborate what gave you that idea? The Jumo 213 had an automatic engine control system, but no hydraulic coupling, so I don'see any reason to assume these were mutually exclusively.

Regards,

Henning (HoHun)

 

[HoHun]

Hi Xylstra,

But am I wrong, was the Föttinger in fact, actually more efficient? Intelligent analyses, please.

Something which I believe that hasn't been pointed out explicitely in this thread yet is that the answer is a function of altitude.

There's a clear answer for each altitude on whether a hydraulic coupling or fixed gearing is more efficient.

It's also important to realize that it's only a fraction of the power of the engine that is transmitted through the hydraulic coupling, of which only a fraction is lost ... in total horse power, that's not all that much.

The clear answer I mentioned can be arrived at by comparing the power loss induced by the hydraulic coupling to the power loss induced in a fixed-gearing engine by the necessity to throttle the intake.

Near and above the full throttle height of the engine, the fixed gearing wins, at altitudes a bit below the full throttle height, the hydraulic coupling wins. If you're comparing an engine with a two-speed supercharger drive to one with a single-speed hydraulic-coupling, it gets a bit more complex, but basically the same idea applies - at altitudes near the two-speed equipped engine's full throttle engine that one wins, and at lower altitudes, the hydraulic coupling is more efficient.

Of course, if the ratio between the two speeds of the two-speed equipped engine is greater than the speed between the highest and lowest (sensible) speed of the hydraulic-coupling equipped engine, that's where you have to consider additional trade-offs, and the comparison gradually drifts into apples-vs.-oranges direction - but of course the engineers have to think about these tradeoffs.

There certainly was a trend towards the use of hydraulic couplings, as they were adopted for use with the V-1710, R-2800 and R-4360, so I'd say the American aeroengine builders must have considered the technology worthwhile. However, I believe that post-war the greatest efficiency was achieved with turbo supercharging and turbo compounding. Of course, the thus-equipped engines were primarily used on civilian airliners, where the design priorities probably were a bit different from those that lead to the DB 60x series engines

Regards,

Henning (HoHun)

 

[Nicknick]

I don’t know every version of the Jumo 213, but the version I see in a picture does have an hydraulic coupling. The gear shifts can be much smoother and better controlled by hydraulic couplings than by clutches.

The hydraulic coupling is producing heat in the oil system which isn’t critical for engine knock, whereas throttling will lead to higher charge temperatures which is especially critical when not using an aftercooler (German planes didn’t use aftercoolers as far as I know, but sometimes they used intercoolers).

Edit: Take a look in Colums book (page 419), there you can see a three stage version (1 couling, 1 torque converter) and one four stage version (two coulings)