What If: Pratt Whitney build/develop the Sabre under licence?

I think it would have proven unfruitfull for Pratt and Whitney. Developement likely would have tanken longer and been more expensive than R-4360 developement an engine which itself not only missed the war but was producing only around at its end 3000hp. The much vaunted B-36 had only lacklustre performance in its B-36A form and even B versions were not completely convincing.

While the sleave valve seems a good idea history shows that the poppet valve engine was able to keep up and was more practical to build.

Let's fast forward to Formulae 1 up to 1989, the BMW M12/M13 engine was able to produce 900hp out of 1.5 Litres swept volume at 3.8 ata in race trim on ordinary premium grade motor fuel as sold in European service stations. The 1986 engine was said to produce about 1,300 hp (969 kW) in qualifying, that being the most powerful figure of all the turbo-charged engines in Formula One.

A Merlin with the same specific power outout of 600hp/L would have been producing 15,600hp without WEP. This is almost 10 times the Merlins output.

R-3360's installed in Bearcats replaced the R-2800) and in Seafury's (replaced the sleave valve centaurus) both hold world speed records.

Even if the sleave valve breathes better than the poppet valve you just increase blower pressure a little and the same mass of air can be forced in.

The sleave valve seems to have promised smaller cylinder lenths (there were no overhead cams let alone rockers) and it seems to have promised higher and smoother running speeds.

Somehow the valve bounce issue on poppet valves was overcome, just as the hot spots causing preignition on the valve issue was overcome (with sodium cooled valves). I would suspect a sleave valve would have a smaller frontal area though, it couldn't be much.

I am not arguing it could have been a usefull engine, and might have made the war, but by the time it was ready it both the Jet engine and the turbo prop offered better performance in therms of speed or fuel economy (due to lighter weight). Pratt Whitney was a practical company of impressive abillity (it had a machine tool background). Unlike the greed managment of Curtiss Wright in the 1950s it invested in R+D and produced jet engines and so unlike Curtiss Wright didn't disappear.

There are also more than one way to skin a cat. The BMW 802 (It would have been called an R-3560 under the US classification system) for instance would have received variable valve timming on the exhaust thus increasing both power and efficiency something not easily possible on a single sleave. (Some British cars from the 20s/30s had not only sleave valves but duel sleave valves.

I have seen sleaves seriously proposed in automotive engines in conjunction with poppet valves, the sleave reduces friction saveing both fuel and engine wear but is not involved in aspiration.

Hence the wear issue may have turned out about the only serious advantage and possibly staved of the gas turbine a little longer. The Centaurus had a good reputation in this regard but few if any seem to survive.

I also don't think that the Germans would have had too much trouble keeping up with the performance of these sleave valves. The Jumo 213J was expected to produce 1912kW (2560hp) takeoff and about 10% more WEP, I've heard of up to 3150hp as hoped for. The DB603N about 2800hp. The Jumo 222E/F 2800hp with 3500hp eventually expected. As it was the competition gambled on jets engines.

Of course what would be done with these powerfull engines.

Siegfried said:

Hence the wear issue may have turned out about the only serious advantage and possibly staved of the gas turbine a little longer. The Centaurus may have had a good reputation in this regard but few if any seem to survive.

Click to expand...

R4360 and R2800 engines are in greater supply than Centaurus for the simple reason that both engines were kept in production and use for longer. 30 years ago you could buy an R2800 for little more than scrap value there were piles of them lying about at desert plane breaking yards. The Centaurus was built in penny packets in comparison and I doubt if any spares were available 50 years ago. If your plane has a Centaurus engine your going to limit use for the simple reason if you break it that could be it. No spares means no engine so its good sense to replace it with a P&W engine.

Shortround6 said:

A Bristol Centaurus was 3270 cu in, a replacement for the R-3350??

A Tempest II a year early might be interesting

Click to expand...

According to Wiki
the R-3350-C18-BA had a diameter of 55.78 inches with a swept volume of 3347 inches. It opperated at 2800 rpm.
the Bristol Centaurus had a diameter of 55.3 inches wth a swept volume of 3370 inches. It opperated at 2900 rpm.

The spec for the R-3350 is for a very early version the ones still swalling valves in 1943, by 1944 they were producing 2500hp takeoff.

The engines would be very interchageable. I also question what the advantage of the Sleave valves was apart from 0.4% more swept volume and 0.9% less diameter.

Siegfried said:

I think it would have proven unfruitfull for Pratt and Whitney. Developement likely would have tanken longer and been more expensive than R-4360 developement an engine which itself not only missed the war but was producing only around at its end 3000hp. The much vaunted B-36 had only lacklustre performance in its B-36A form and even B versions were not completely convincing.

Click to expand...

Pratt Whitney did built, test and run a Sabre like engine (actually a series of engines) - the X-1800/XH-2600/XH-3130/XH-3730. I am suggesting replacing those progarms with a Sabre program - which would essentially be a productionising process rather than a design and development project.

The R-4360 started after the X-1800 was cancelled.

Siegfried said:

Let's fast forward to Formulae 1 up to 1989, the BMW M12/M13 engine was able to produce 900hp out of 1.5 Litres swept volume at 3.8 ata in race trim on ordinary premium grade motor fuel as sold in European service stations. The 1986 engine was said to produce about 1,300 hp (969 kW) in qualifying, that being the most powerful figure of all the turbo-charged engines in Formula One.

A Merlin with the same specific power outout of 600hp/L would have been producing 15,600hp without WEP. This is almost 10 times the Merlins output.

Click to expand...

Fuel was not pump fuel. It was specially formulated for the turbo engines, and required extra special handling methods. The qualifying engines were also basically good for one lap at max boost (around 90s), and as they were in circuit racing cars spent only about half the lap at full throttle, let alone max power.

I also believe that sleeve valves were banned in F1 by then.

Siegfried said:

I also don't think that the Germans would have had too much trouble keeping up with the performance of these sleave valves. The Jumo 213J was expected to produce 1912kW (2560hp) takeoff and about 10% more WEP, I've heard of up to 3150hp as hoped for. The DB603N about 2800hp. The Jumo 222E/F 2800hp with 3500hp eventually expected. As it was the competition gambled on jets engines.

Click to expand...

As the Jumo 213J was a similar size to the Sabre It may have been possible. The DB603 was 20% larger in capacity.

I doubt those numbers are practical performance numbers, though.

fastmongrel said:

R4360 and R2800 engines are in greater supply than Centaurus for the simple reason that both engines were kept in production and use for longer. 30 years ago you could buy an R2800 for little more than scrap value there were piles of them lying about at desert plane breaking yards. The Centaurus was built in penny packets in comparison and I doubt if any spares were available 50 years ago. If your plane has a Centaurus engine your going to limit use for the simple reason if you break it that could be it. No spares means no engine so its good sense to replace it with a P&W engine.

Click to expand...

About the only civil aircraft that used the Centaurus was the Neville Shute designed Airspeed Ambassador. It was fine aircraft, pressurised, quiet on the inside however it entered service in 1952 and by 1955 was testing the Proteous and Dart turboprops that would help make it obsolete.

Within the big heavy inlines that would been 'deserved' to be produced in USA, maybe just Allison V-3420 can candidate? Two-stage versions were some 20% heavier than Sabre. The military rating was typically 2300 HP at 20000 ft, maybe 50% more than Sabre was capable to do?
Another good choice might've been Griffon, though the US production would've started perhaps too late for war.

wuzak said:

Pratt Whitney did built, test and run a Sabre like engine (actually a series of engines) - the X-1800/XH-2600/XH-3130/XH-3730. I am suggesting replacing those progarms with a Sabre program - which would essentially be a productionising process rather than a design and development project.

The R-4360 started after the X-1800 was cancelled.

Fuel was not pump fuel. It was specially formulated for the turbo engines, and required extra special handling methods. The qualifying engines were also basically good for one lap at max boost (around 90s), and as they were in circuit racing cars spent only about half the lap at full throttle, let alone max power.
It would be a good idea if the Sabre promised more power than the PW 2800, however I do not believe any mark of Sabre outperformed the PW-2800 in service. In the long run the Sabre may have had more potential but by then the CW R-3350 might have been reliably producing power in the 3500hp class as well.

Modern F1 fuel is merely a tightly specified version of road fuel. In times past special formulations were used but they were never allowed to have ingredients that were not found in normal road fuel. The turbo era did lead to use of fuels with high content of tolulene which would make these fuels have a lean RON of 115 (since tolulene is rated 116) so about the same rating as US 115/145 fuel or the 100/150 (which was really 110/150). Without this fuel the BMW 12/13 might have suffered a 16% or so drop in output.

I also believe that sleeve valves were banned in F1 by then.

As the Jumo 213J was a similar size to the Sabre It may have been possible. The DB603 was 20% larger in capacity.

I doubt those numbers are practical performance numbers, though.

Click to expand...

It would be a good idea if the Sabre promised more power than the PW 2800, however I do not believe any mark of Sabre outperformed the PW-2800 in service. In the long run the Sabre may have had more potential but by then the CW R-3350 might have been reliably producing power in the 3500hp class as well.

Modern F1 fuel is a tightly specified version of road fuel. In times past special formulations were used but they were never allowed to have ingredients that were not found in normal road fuel. The turbo era did lead to use of fuels with high content of tolulene which would make these fuels have a lean RON of 115 (since tolulene is rated 116) so about the same rating as US 115/145 fuel or the 100/150 (which was really 110/150). Without this fuel the BMW 12/13 might have suffered a 16% or so drop in output. But you take my point, poppet valves had plenty of room to manouver.

It's conceivalbe that PW might have fixed the Sabres problems: poor serviceabillity and Napiers quality control issues in time pump out larger numbers of tempests. It might have powered Warwicks, B-26's and a low altitude version of the P-47.

The swept volume of the DB603 v Jumo 213 I think has little to with power output which is constrained by the amount of air processed. What the Jumo 213 lacked in swept volume it would make up with RPM and boost levels.

Siegfried said:

It would be a good idea if the Sabre promised more power than the PW 2800, however I do not believe any mark of Sabre outperformed the PW-2800 in service. In the long run the Sabre may have had more potential but by then the CW R-3350 might have been reliably producing power in the 3500hp class as well.

Click to expand...

The only R-3350s that made 3500hp were the turbo-compounds. IIRC the best the R-3350 could do otherwise was 2800hp, and during the war years it was 2200hp.

The Sabre made 2000hp when the R-2800 could only manage 1850hp (A-series). When the 2000hp R-2800s were available the Sabre IIB was running at 2200-2400hp. By the end of the war, when a few R-2800s could make 2800hp the Sabre was rated (in WEP) at over 3000hp for use in Tempests.

Siegfried said:

Modern F1 fuel is a tightly specified version of road fuel. In times past special formulations were used but they were never allowed to have ingredients that were not found in normal road fuel. The turbo era did lead to use of fuels with high content of tolulene which would make these fuels have a lean RON of 115 (since tolulene is rated 116) so about the same rating as US 115/145 fuel or the 100/150 (which was really 110/150). Without this fuel the BMW 12/13 might have suffered a 16% or so drop in output. But you take my point, poppet valves had plenty of room to manouver.

Click to expand...

Those restrictions have been applied since the turbo era. A year or so ago Ferrari did a publicity stunt for Shell and ran one of their F1 cars on pump fuel - it was only slightly slower than the race fuel.

40 years of development helped. I wonder what could have been achieved with the sleeve valve with similar amounts of development?

Siegfried said:

The swept volume of the DB603 v Jumo 213 I think has little to with power output which is constrained by the amount of air processed. What the Jumo 213 lacked in swept volume it would make up with RPM and boost levels.

Click to expand...

True.

Siegfried said:

It's conceivalbe that PW might have fixed the Sabres problems: poor serviceabillity and Napiers quality control issues in time pump out larger numbers of tempests. It might have powered Warwicks, B-26's and a low altitude version of the P-47.

Click to expand...

Probably the biggest problem with the Sabre was the manufacture of the sleeves. That was fixed by the previously mentioned centreless grinders redirected from P&W, allowing Napiers to make the sleeves to an acceptable roundness tolerance. The life of teh sleeves was extended using Bristol's finishing techniques. Certianly I think P&W could have done the former, whether they could figure out the latter is another matter.

Why a low altitude P-47? Wouldn't you try to hook up the turbo that the P-47 already had?

Replacing an R-2800 with a Sabre (an H-2240?) sounds like a whole lot of work for very little result. The Sabre is heavier in dry weight, a two stage R-2800 weighs about what a single stage Sabre does, not including the inter-coolers. The R-2800 doesn't need 5-600lbs worth of radiators and coolant, although it's cowling is a bit bigger and heavier. Even if the Sabre's nuts and bolts had been a bit more accessible the R-2800 is still easier to work on, needs 12 fewer spark plugs at plug change time and so on.

At the end of the war the Sabre offered very little that the big radials could not do.

Leaving peak power aside for a moment and comparing "rated" power or max continuous power we find that the various engines (using single stage superchargers in high gear) were good for:

R-2800 "C" series..... 1500hp/2600rpm/17,500ft.
R-4360-4............... 2200hp/2550rpm/13,000ft.
R-3350BD............... 1800hp/2400rpm/13,000ft.
Centaurus 57.......... 2000hp/2400rpm/16,000ft.
Griffon VI................1340hp/2600rpm/14,800ft.
Sabre VII................1960hp/3700rpm/18,250ft.

a Merlin 130 was good for 1325hp/2850rpm/20,750ft with it's two stage blower.

Take-off power for the engines were:

R-2800 "C" series..... 2400hp/2800rpm/+13.0lbs boost and ADI
R-4360-4............... 3000hp/2700rpm/+10.5lbs boost dry
R-3350BD............... 2800hp/2800rpm/+13.0lbs boost and ADI
Centaurus 57.......... 2800hp/2700rpm/+11.5lbs boost and ADI
Griffon VI................1820hp/2750rpm/+15lbs boost dry
Sabre VII................3000hp/3850rpm/+17.25lbs boost and ADI

Data from the 1946 edition of Wilkinson's "Aircraft Engines of the World"

Shortround6 said:

Replacing an R-2800 with a Sabre (an H-2240?) sounds like a whole lot of work for very little result. The Sabre is heavier in dry weight, a two stage R-2800 weighs about what a single stage Sabre does, not including the inter-coolers. The R-2800 doesn't need 5-600lbs worth of radiators and coolant, although it's cowling is a bit bigger and heavier. Even if the Sabre's nuts and bolts had been a bit more accessible the R-2800 is still easier to work on, needs 12 fewer spark plugs at plug change time and so on.

At the end of the war the Sabre offered very little that the big radials could not do.

Leaving peak power aside for a moment and comparing "rated" power or max continuous power we find that the various engines (using single stage superchargers in high gear) were good for:

R-2800 "C" series..... 1500hp/2600rpm/17,500ft.
R-4360-4............... 2200hp/2550rpm/13,000ft.
R-3350BD............... 1800hp/2400rpm/13,000ft.
Centaurus 57.......... 2000hp/2400rpm/16,000ft.
Griffon VI................1340hp/2600rpm/14,800ft.
Sabre VII................1960hp/3700rpm/18,250ft.

a Merlin 130 was good for 1325hp/2850rpm/20,750ft with it's two stage blower.

Take-off power for the engines were:

R-2800 "C" series..... 2400hp/2800rpm/+13.0lbs boost and ADI
R-4360-4............... 3000hp/2700rpm/+10.5lbs boost dry
R-3350BD............... 2800hp/2800rpm/+13.0lbs boost and ADI
Centaurus 57.......... 2800hp/2700rpm/+11.5lbs boost and ADI
Griffon VI................1820hp/2750rpm/+15lbs boost dry
Sabre VII................3000hp/3850rpm/+17.25lbs boost and ADI

Data from the 1946 edition of Wilkinson's "Aircraft Engines of the World"

Click to expand...

30% more maximum continuous at altitude and 25% more takeoff power for the Sabre VII compared to the R-2800 'C' series doesn't sound like "very little result".

The Centaurus and the R-4360 are better for max continuous than the Sabre VII, but at lower altitudes. The R-3350 has slightly less max continuous, but at a much lower altitude. The R-4360 matches the Sabre for takeoff power, but is heavier than the Sabre even when coolant and radiators are taken into account.

Of ocurse if you put a Sabre in an airfarme designed for a radial you probably won't gain much, if any, aerodynamic advantage.

Just adding some extra data.

Sabre VII, maximum continuous, in low gear (MS gear) 2,235hp @ 3,700rpm @ 8,500ft.
Sabre VII take off 3,000hp @ 3,800rpm, +17.25psi, 3,500hp @ 3,800rpm, +20psi and ADI.

Centaurus 57, maximum continuous, in low gear (MS gear) 2,160hp @ 2,400rpm, +6.5psi @ 5,000ft, in high gear (FS gear) 1,975hp @ 2,400rpm, +6.5psi @ 15,750ft.

Centaurus 57 take off 2,470hp@ 2,700rpm, +9.5psi.

Data from Lumsden, all with 100/130 fuel.

There is no doubt that Sabre looks like a more suitable engine than R-2800 at the 1st glance. Looking deep, the categories where Sabre is better are power levels at lower levels and (depending on installation) streamlining. R-2800 comes ahead with power above 20000 ft, reliability and availability. In order for P&W/Sabre combo to work, they need to make a flawless Sabre clone (= more reliable) as early as 1942, mass producing from early 1943, to make a difference - that would've been quite the achievement Mating it with a turbo looks great, but so does the fan cooled R-2800 (real thing, delivering 2800 Hp in XP-47J).
The advent of the R-2800-18W means that gap remains, for high altitude work with non-turbo instalation. Sabre again better down low.

The performance of the single stage R-2800s at altitudes would appear less than the Sabre's - as shown by Shortround's numbers.

The superior altitude performance of the R-2800 only comes about because of two stage superchargers or turbochargers. If P&W developed the Sabre, it would have probably developed two stage supercharging at some stage, and the USAAF may have stuck a turbo on one.

As for the production prospects, if the Sabre replaced the X-1800 program it would have been worked on since 1938. The major reliabiliy issue was the roundness of the sleeves. As P&W were more used to quantity production and their quality control was much better I have little doubt that the reliability issues would have been solved earlier.

As for the R-2800 in the XP-47J, 2800hp was no doubt achieved with PN150 fuel and a lot of ADI. The Sabre VII achieved 3000hp with PN100/130 fuel without ADI, and some 3500hp with ADI (for takeoff).

Indeed, I was refering to the two stage R-2800
The two-stage Sabre really sounds as a great idea, too bad Napier did not have that ready for Tempest in early 1944.

The XP-47J was maintaining 2800 HP at 30000 ft, no ram, in 1944 (1st flight in Nov. 1943), and the Sabre that can beat (only under 12500 ft, but the supercharger was not forte of Napier) was produced in 1945. Care to toss in some info about Sabre doing 3500 in service, I was able to found 'only' 3055 HP, 1945?

The 2800hp rating for the R-2800 was done with 100/130 and a lot of ADI and was achieved without a fan on P-47Ms and Ns. These engines weighed 2315lbs without the turbo. Something to consider when comparing the Sabre to the R-2800. The Sabre may have offered more power but it weighed more, a lot more when you throw in the radiator.

The chances of P&W licensing the Sabre prewar are about zero. Why pay license fees/ patent royalties to Napair when they had their own patents for many of the features? They also had no existing manufacturing plant for large liquid cooled engines. No casting facilities (or not enough) to handle the large casting needed for crankcases and cylinder blocks. The radial crankcases being much smaller, and individual cast cylinder heads being almost minuscule in size compared to a Sabre crankcase.
We still don't know the cost of the engines. If you can build 2 R-2800s for the cost of 1 Sabre the Sabre really looks like a bad deal no matter how technically advanced it may be.
The Sabre didn't offer a big advantage over the R-2800 in 1939-40 and while it may have gained over the R-2800 during the war years it didn't really offer anything over the other big radials by the end of the war.

tomo pauk said:

Care to toss in some info about Sabre doing 3500 in service, I was able to found 'only' 3055 HP, 1945?

Click to expand...

The Sabre VII may only have been installed in two airframes. A long suffering Folland test bed and a single Hawker Fury prototype.

Good documentation on the Sabre is hard to find. Charts can be found for Sabre IIA using 12lbs boost at 3700rpm and 150 fuel. Books can be found showing 3750 rpm. Sabre IIB s are supposed to be good for 3850 rpm. Boost ratings jump up and down a bit. Sabre IIAs were being installed in the first tempests let alone rather late Typhoons. The Sabre V was fitted to the Tempest VI but I don't think they mad the war and I am not sure they got anywhere near 3000hp, could be wrong on that though.

Well, this is the 1st link when one googles 'hawker tempest':

The Hawker Tempest Page

There is the 3055 HP figure to be found, along with 2820 HP one. RPM up to 3850.

tomo pauk said:

Indeed, I was refering to the two stage R-2800
The two-stage Sabre really sounds as a great idea, too bad Napier did not have that ready for Tempest in early 1944.

The XP-47J was maintaining 2800 HP at 30000 ft, no ram, in 1944 (1st flight in Nov. 1943), and the Sabre that can beat (only under 12500 ft, but the supercharger was not forte of Napier) was produced in 1945. Care to toss in some info about Sabre doing 3500 in service, I was able to found 'only' 3055 HP, 1945?

Click to expand...

The only reason teh R-2800 could hold 2800hp @ 30,000ft was because of the turbo.

Add a similar turbo to the Sabre and it could give the 3055hp @ 30,000ft.

3500hp was for takeoff only, with +20psi boost - data from Lumsden.

Earlier marks of the Sabre were capable of 3000hp too. I believe the VA, which was installed in the Tempest, was good for 3000hp.

Those 2800 turbo HP were available in 1944, and no matter what supercharger we add, Sabre in 1944 was not capable for that. It was between 2400 HP (from the chart at ww2aircraftperformance) and 2600 HP. To achieve 3055 HP, we need to go to 1945.
When one says 'earlier marks', what year is that? Sabre VA is in 1945, and I'd like to know if the V and VII were ever in service, even in that year. One thing is to have a superb engine for testing, available in single digits, other thing are the engines produced in thousands.

I do maintain that R-2800 have had the edge in power at altitude, reliability, availability, maintainability, while being less susceptible to battle damage.
Sabre (as is) was offering a better performance at lower altitudes, and (in theory) better streamlining. Looking at R-2800 without turbo or two-stage supercharger is like taking into account only 1-stage Merlins.
For Sabre to beat R-2800 as a complete package, many things would've needed to come along, so the license production of R-2800 in UK makes far more sense than Sabre produced in USA.