Engine ratings during late WWII

[tomo pauk]

Ah, that's cheating a bit. Water injection (to cool the intake charge via evaporative cooling) and Nitrous Oxide (to add more oxygen and cool the intake charge even more) let you run more boost than the fuel can support without detonation.

Is the water-alcohol injection R-2800 on the P-47s, so they can make 2300-2600-2800 HP instead of 2000-2100 HP, also cheating?

As opposed to not running crap fuel that's 87 octane at best and running either more compression and/or more boost.

Having to use anti-detonation injection or nitrous oxide is cheating because as soon as you run out of either ADI or NOS you lose a massive amount of power.

Engineer in ww2: we've managed to make our engines do extra 15-20-30% HP more, with reasonable reliability.
21st century enthusiast: cheater!

Using 115/145 octane avgas and 15-20psi boost is 24/7 power.

Three things:
- there was no 115/145 oct avgas in ww2, at least not in combat service
- there was no engine capable to be over-boosted by even an hour in service, let alone 24/7
- a precious few engine types were using +15 to +20 psi boost in actual service (perhaps Merlin family being the only types to do it without the water-alc injection?); when they did it, having water-alcohol injection helped at a lot of them

 

[Artie Bob]

three things:
- there was no 115/145 oct avgas in ww2, at least not in combat service
- there was no engine capable to be over-boosted by even an hour in service, let alone 24/7
- a precious few engine types were using +15 to +20 psi boost in actual service (perhaps Merlin family being the only types to do it without the water-alc injection?); when they did it, having water-alcohol injection helped at a lot of them

 

[Artie Bob]

The R-2800, IIRC was run at 3000 HP for 100 hours on standard Av fuel but with water injection and overboost (not available on an A/c installation, but possible on the factory test stand. The maximum eventually was reached was 3800 HP and the test engineer felt he could possibly have reached 4000 hp. There was no practical value to these extreme ratings, except to show the mechanical strength of the basic engine design.

As to fuel ratings, during WW II, instead of octane, the standard in the USA became performance number ( PN) , it appears quite certain that 150 PN fuel was available to operational units

 

[Scott Kenny]

As to fuel ratings, during WW II, instead of octane, the standard in the USA became performance number ( PN) , it appears quite certain that 150 PN fuel was available to operational units

Right, because "100 octane" means "a fuel with detonation qualities equal to 100% iso-octane".

Once you get performance better than 100% iso-octane you need a different standard. The two avgas grades are 100/130 and 115/145, which gives performance numbers at full lean and full rich, respectively.

115 has a LOT of lead in it to suppress detonation.
Then there was a low-lead 100/130 equivalent, tagged as 100LL. (the older spec stayed as 100/130 or occasionally 100HL)
Now there's a completely unleaded 100/130 equivalent, tagged as 100UL.

 

[Dagger]

Three things:
- ....................
- ....................
- a precious few engine types were using +15 to +20 psi boost in actual service (perhaps Merlin family being the only types to do it without the water-alc injection?); when they did it, having water-alcohol injection helped at a lot of them

Note that the only reason the Merlin could run at those high boost values was because its piston engine had a CR of only 6.

German DB piston engines had a CR in the order of 7, the PW 2800 had a CR in the order of 6.7 so they were limited to lower boost values than the Merlin even if running on exactly the same fuel.

A CR of 7 i/o 6 means a higher fuel efficiency (at least in theory, provided all other factors equal) but it also means that a larger displacement volume is required to produce the same power due to lower allowable boost when running on exactly the same fuel.

 

[tomo pauk]

The R-2800, IIRC was run at 3000 HP for 100 hours on standard Av fuel but with water injection and overboost (not available on an A/c installation, but possible on the factory test stand. The maximum eventually was reached was 3800 HP and the test engineer felt he could possibly have reached 4000 hp. There was no practical value to these extreme ratings, except to show the mechanical strength of the basic engine design.

In the ww2 era for service use, the R-2800s on P-47Ds managed erstwhile 2300 HP with 130 grade fuel and water injection, and, very shortly after (but still in 1944) and with the upgraded w-i kit, 2600 HP. On 150 grade fuel, they managed 2800 HP.
The improved R-2800s on the P-47M and N were making 2800 HP with 130 grade and water (by mid-1945).

FWIW: table, taken from here

 

[sienar]

Note that the only reason the Merlin could run at those high boost values was because its piston engine had a CR of only 6.

German DB piston engines had a CR in the order of 7, the PW 2800 had a CR in the order of 6.7 so they were limited to lower boost values than the Merlin even if running on exactly the same fuel.

A CR of 7 i/o 6 means a higher fuel efficiency (at least in theory, provided all other factors equal) but it also means that a larger displacement volume is required to produce the same power due to lower allowable boost when running on exactly the same fuel.

from; https://www.jstor.org/stable/44438051

 

[tomo pauk]

Note that the only reason the Merlin could run at those high boost values was because its piston engine had a CR of only 6.

German DB piston engines had a CR in the order of 7, the PW 2800 had a CR in the order of 6.7 so they were limited to lower boost values than the Merlin even if running on exactly the same fuel.

Merlin was also very strong - a 27 liter engine that weighted same as the 33-35L German competition. It became even more suitable with the introduction of the 2-piece block by 1942, and with introduction of the strengthened drive for the S/C (ie. after the Merlin 61). Being a liquid cooled engine, it was better suited for the peak boost levels than what the radial engines of the day were able to sustain. Small cylinders were also easier to cool than the big cylinders. Having the poppet valves also helped, since the sleeve valves were unable to sustain the very high boost levels.

The experience that RR had with the 'R' engine during the interwar period was certainly of help with the Merlin.

 

[Dagger]

................ since the sleeve valves were unable to sustain the very high boost levels.

Didn't know that. Always thought that sleeve valve engines like the Napier Sabre, Bristol Hercules, Bristol Centaurus used low boost because they had a CR of around 7 (i/o 6 like Merlin and Griffon). But I suppose it could just as well be the other way around: due to low allowable boost they had to use higher CR than usual in other British engines to maximise power.

 

[tomo pauk]

Didn't know that. Always thought that sleeve valve engines like the Napier Sabre, Bristol Hercules, Bristol Centaurus used low boost because they had a CR of around 7 (i/o 6 like Merlin and Griffon). But I suppose it could just as well be the other way around: due to low allowable boost they had to use higher CR than usual in other British engines to maximise power.

The 'low allowable boost' was a result of the chosen CR (among other things), not the other way around

 

[Tonton-42]

It is good to recall some fondammentaux for our young audience ...

Wikipedia says :
The confusion is often made between the compression ratio and the pressure in the cylinders: the compression ratio is an invariable value because determined according to the respective volumes of the cylinder and the combustion chamber to the exclusion of any other data, whereas cylinder pressure is a value that varies continuously according to the load required by the engine, the presence or absence of a compressor or turbocharger, the engine temperature and its mechanical condition.

Thus a compression ratio of 11 does not mean that the pressure in the cylinders is 11 bars but that the volumetric ratio between the «low dead point» and the «high dead point» is 11. In addition, the compression ratio does not take into account the opening moment of the valves (intake and exhaust). It should therefore only be considered as a simple performance indicator.

It is possible to measure the compression pressure of an engine at a given rpm (the one obtained with the vehicle starter) by means of a pressure probe, sometimes called «compressometer4», mounted in place of the spark plug or injector. The compression pressure is measured in each cylinder in the absence of combustion. The reading must be above a minimum specified by the engine manufacturer to ensure minimum performance.

This measurement of the compression pressures makes it possible to determine the state of wear of the engine and the cause or causes of the wear, while the only knowledge of the compression ratio is of no use for such a diagnosis. Low compression pressures can be due to, among other things, a lack of tightness of the segments, the cylinder head gasket, the condition of the valves and their seats.

 

[Nicknick]

Didn't know that. Always thought that sleeve valve engines like the Napier Sabre, Bristol Hercules, Bristol Centaurus used low boost because they had a CR of around 7 (i/o 6 like Merlin and Griffon). But I suppose it could just as well be the other way around: due to low allowable boost they had to use higher CR than usual in other British engines to maximise power.

Sleeve valve engines were more resistant to knocking, that was one of there main advantages, so I would highly doubt that. According to Harry Ricardo, the compression ratio could allways be about one point higher than in the valve engines (no hot exhaust valves in the combustion chamber).

 

[tomo pauk]

It is good to recall some fondammentaux for our young audience ...

Wikipedia says :
The confusion is often made between the compression ratio and the pressure in the cylinders: the compression ratio is an invariable value because determined according to the respective volumes of the cylinder and the combustion chamber to the exclusion of any other data, whereas cylinder pressure is a value that varies continuously according to the load required by the engine, the presence or absence of a compressor or turbocharger, the engine temperature and its mechanical condition.

Confusion was furthered by some British engineers that also called the "pressure ratio" (that is a measurement that 'belongs' to the compressors) as the "compression ratio". Eg. like in this doc, where the 'compression ratio' is stated in caption under the Fig.11, despite the graph saying 'pressure ratio'.

Sleeve valve engines were more resistant to knocking, that was one of there main advantages, so I would highly doubt that. According to Harry Ricardo, the compression ratio could allways be about one point higher than in the valve engines (no hot exhaust valves in the combustion chamber).

Not the 1st designer/engineer/author that talks highly about the device he spent a lot of time working on.
Calum Douglas notes that sleeve valves are not conductible to the very high boost levels due to the sleeves deforming under the great stress.

 

[Tonton-42]

Find out more about Sliver Jacket Engines and their aeronautical applications.

Moteurs à chemises louvoyantes

Page Moteurs à chemises louvoyantes

jnpassieux.fr

 

[Nicknick]

According to our forum member, the Sabre run on 1 bar boost pressure (15 psi), presumly with a 7:1 compression ratio:

Napier H-24 Sabre Aircraft Engine

The Naiper Sabre was the last H engine designed by Frank Halford. From a troubled start, the complex Sabre ultimately produced more power for its displacement than any other WWII aircraft engine.

oldmachinepress.com

Please note, that liquid cooled engines are generally better suited than air cooled ones.

 

[tomo pauk]

According to our forum member, the Sabre run on 1 bar boost pressure (15 psi), presumly with a 7:1 compression ratio:

British nomenclature included the + sign before the value of the boost, eg. +6.25 psi (well-known Merlin III on 87 oct fuel max boost), or +12 psi (again well known Merlin III on 100 oct fuel max boost at the beginning of ww2).
A lot about the Merlins can be read here.

Per British nomenclature, zero boost = normal atmospheric pressure at S/L and 15°C; boost was in negative values at very high altitudes.

Sabre was good for +7 psi on the 100 oct fuel in 1942, improving to +9 psi by 1943, and again improving to +11 psi (with strengthened internals and 150 grade fuel) by 1944. See the same website.
Merlins were pushing +16 psi by early 1942, and the 2-stage versions were doing +18 psi, +21 psi, then +25 psi with 150 grade fuel.

All the values for service engines in ww2 (ie. no prototypes).

Sabre VII was supposed to be pushing +17.5 to +20 psi with water-alcohol injection on the Fury, and the Merlin +30 psi with w-a injection on the P-51H.

Please note, that liquid cooled engines are generally better suited than air cooled ones.

For higher boost levels - certainly.

 

[Scott Kenny]

Calum Douglas notes that sleeve valves are not conductible to the very high boost levels due to the sleeves deforming under the great stress.

Yep. Remember, a sleeve valve engine has a cylinder with some cutouts in it as the intake and exhaust valves, and the piston goes up and down inside this sleeve. This makes a weakness in the sleeve.

Sleeve valves are quieter running, but do consume a little oil in the process. They're also a lot more mechanically complex (just look at all the gears in the front case of a Centaurus!).

In the 1920s and 1930s, there was a belief that poppet-valve engines could not reach 1hp per cubic inch (61hp/liter). We have since demonstrated that was complete BS, but it did take some special work to make it happen. Exotic alloys and even sodium filled valves.

 

[Nicknick]

Wouldn't neccessarily call it a weekness, the sleeve also serves the similar function like an iron liner in an aluminium crank case. The sleeve should be strong enough to bear to pressure without relying on the outer walls and should also have enough clearance in the opening area to not beeing pressed in the openings.

We are more used seeing complex valve trains than sleeve mechanism, I don't think it really matters, but the part number of a sleeve valve radial engine is lower that that of a corresponding valve train.

@tomo pauk : The Merlin had a much better supercharging system than the Napier Sabre (this is also written somewhere, but I forgott the source...) and this was presumingly the reason for the higher boost pressure. Ther Bristol Hercules also suffered from it's charging system.

 

[Scott Kenny]

Wouldn't neccessarily call it a weekness, the sleeve also serves the similar function like an iron liner in an aluminium crank case. The sleeve should be strong enough to bear to pressure without relying on the outer walls and should also have enough clearance in the opening area to not beeing pressed in the openings.

The corners in the cutouts are literally crack generators...

Also, the lack of material where the cutouts are is a place where the sleeve is easier to deform.

Finally sleeve valves are not as strong as a pressed-in engine sleeve, because the sleeve valve needs to float on an oil film not solid metal to metal contact.

We are more used seeing complex valve trains than sleeve mechanism, I don't think it really matters, but the part number of a sleeve valve radial engine is lower that that of a corresponding valve train.

Except for needing gears to drive each sleeve valve, each of which needs to be perfectly timed and is physically possible to get multiple teeth off time. While a radial engine with poppet valves has a much simpler valve timing requirement and is either correctly lined up on top dead center or not.

Poppet valves have the valve as a separate piece from the springs and spring cap and valve retainer (and valve seal), plus rockers in however many pieces and then pushrods and cam followers. So yes, physically more parts but harder to get messed up.

I can tell you I'd much rather work on a poppet valve type than a sleeve valve type!

 

[tomo pauk]

The Merlin had a much better supercharging system than the Napier Sabre (this is also written somewhere, but I forgott the source...) and this was presumingly the reason for the higher boost pressure. Ther Bristol Hercules also suffered from it's charging system.

The 2-stage superchargers of the Merlin 60, 70, 80 and 100 series were indeed excellent.
The 1-stage S/Cs as on the Merlin 20s, 30s, 40s and 50s were probably about as good as what the Sabre used were the 2nd league (but still very good).
The S/Cs used on the earlier Merlins, as well as most of the Hercules engines, were the the worse of these mentioned above; it took until the 100 series of Hercules (late 1944) to receive a S/C on par with the Merlin 20-50 series. However, for the late 1930s, S/C on the Merlin was still very good. Similar S/C on the Sabres of ww2 vintage (despite the 2-faced impeller on the Sabre).

Merlin was able to withstand up to +18 psi with the 'old' and improved 1-stage S/Cs and the 1-piece block engine, as on the Merlin 45M for example (and with 100/130 grade fuel). Engines with 2-piece block (like the later 20 series) and 150 grade fuel were supposed to withstand +25 psi (more here).

 

[Tonton-42]

The corners in the cutouts are literally crack generators...

Also, the lack of material where the cutouts are is a place where the sleeve is easier to deform.

Finally sleeve valves are not as strong as a pressed-in engine sleeve, because the sleeve valve needs to float on an oil film not solid metal to metal contact.



Except for needing gears to drive each sleeve valve, each of which needs to be perfectly timed and is physically possible to get multiple teeth off time. While a radial engine with poppet valves has a much simpler valve timing requirement and is either correctly lined up on top dead center or not.

Poppet valves have the valve as a separate piece from the springs and spring cap and valve retainer (and valve seal), plus rockers in however many pieces and then pushrods and cam followers. So yes, physically more parts but harder to get messed up.

I can tell you I'd much rather work on a poppet valve type than a sleeve valve type!

My master in materials knowledge (IPETA Moulinard) had been a mechanic on an airplane. He told us on 1976 that on a Bristol-SNECMA Hercules it took 100 man-hours to reassemble and adjust the distribution ... A real piece of cake !

 

[Scott Kenny]

My master in materials knowledge (IPETA Moulinard) had been a mechanic on an airplane. He told us on 1976 that on a Bristol-SNECMA Hercules it took 100 man-hours to reassemble and adjust the distribution ... A real piece of cake !

100 hours for just the valve gears? That's honestly less than I expected!

 

[tomo pauk]

In the 1920s and 1930s, there was a belief that poppet-valve engines could not reach 1hp per cubic inch (61hp/liter). We have since demonstrated that was complete BS, but it did take some special work to make it happen. Exotic alloys and even sodium filled valves.

To return to this a bit.
The wish to reach the 1hp per cubic inch was a doom for the US Army funds in the lean years before ww2. The 'hyper engines' never bore the actual fruit as the working engines that are better than the 'normal' types already in production.
The 1hp per cu in was a metric that was probably very irrelevant to an aircraft designer, since that type of 'power density' meant the engines were very heavy (since they needed to be very strong), and the demands on cooler systems for the coolant and oil are still there to be catered for (more power = bigger radiators = greater drag and weight).

US Army was better to spend their funds on perfecting the R-2600, R-2800 and V-1710 instead of trying to show the engine companies how the engine should be designed, while burning through the money and time.

 

[Scott Kenny]

To return to this a bit.
The wish to reach the 1hp per cubic inch was a doom for the US Army funds in the lean years before ww2. The 'hyper engines' never bore the actual fruit as the working engines that are better than the 'normal' types already in production.
The 1hp per cu in was a metric that was probably very irrelevant to an aircraft designer, since that type of 'power density' meant the engines were very heavy (since they needed to be very strong), and the demands on cooler systems for the coolant and oil are still there to be catered for (more power = bigger radiators = greater drag and weight).

US Army was better to spend their funds on perfecting the R-2600, R-2800 and V-1710 instead of trying to show the engine companies how the engine should be designed, while burning through the money and time.

The engines you specify come close to or exceed the 1hp-per-cubic-inch goal by the end of the war.

Merlins were close to that by 1943. (1500hp out of 1650 cubic inches is 0.91hp/in^3)

 

[tomo pauk]

The engines you specify come close to or exceed the 1hp-per-cubic-inch goal by the end of the war.

V-1710 and R-2800 did that indeed, the engines being strong, and with help of high octane fuel and water-alcohol injection.

Merlins were close to that by 1943. (1500hp out of 1650 cubic inches is 0.91hp/in^3)

Merlin 63, 65, 66 etc. have beaten that by winter of 1942/43, with 1700+ HP.
Again, the engine being very strong by now (RR made the switch to the 2-piece block), S/C of 2 stages and with intercooler, 130 grade fuel = success.
Even the 1-stage engines, like the 45M and 50M, plus the late 20 series, were making 1600 HP by 1942. Not enough to beat the bar bet, but still an excellent power to displacement ratio.

 

[Tonton-42]

100 hours for just the valve gears? That's honestly less than I expected!

There is no valves when French people speak about sleeve valves ! Problem of vocabulary between our nations ... English words "sleeve valves" isn't correct in France because they replace the valves (soupapes in French) by rotating and translating "shirts" (chemises in French) where the pistons go in their alternative movement ...