F4U Corsair vs P-51 Mustang

[mig-31bm]

On a separate but related question: what is realistic dogfight altitude for WW II fighter?.I mean F4U-5 seem to be much better than P-51H above 25000 ft, but can they really dogfight at that altitude though?. What would be the turning radius?

 

[Shortround6]

Engine in the F4U-5

Two superchargers turning in parallel feeding the main supercharger through intercoolers (not shown) and this was a later engine than the "C" engine used in the F4U-4 which only used a normal two stage supercharger/intercooler set up.
This engine weighs about 150lbs more than the engine in the F4U-4 (without intercoolers) and makes the same power at 30,000ft that the F4U-4's -18 engine makes at 23,000ft.

This was NOT a simple modification or a tweak. F4U-5 first flew Dec 1945, 10 months after the first flight of the P-51H.

 

[Zipper730]

What's the CL max for the F4U and the P-51 and P-51H?

BTW: From what I remember, there were numerous changes to the P-51's wing. Somewhere along the way a kink was added which added a little extra area, I think the wing somewhere between the P-51/A and P-51B/C had the wing slightly repositioned (bizarre as that sounds)

 

[mig-31bm]

What's the CL max for the F4U and the P-51 and P-51H?

BTW: From what I remember, there were numerous changes to the P-51's wing. Somewhere along the way a kink was added which added a little extra area, I think the wing somewhere between the P-51/A and P-51B/C had the wing slightly repositioned (bizarre as that sounds)

If I recall correctly, P-51D has the kink which add a little bit of additional wing area, however, P-51H has a straight wing which reduce the wing area back to P-51B level again, beside it has less thick wing, which probably lead to lower CL

 

[mig-31bm]

Engine in the F4U-5
View attachment 795479
Two superchargers turning in parallel feeding the main supercharger through intercoolers (not shown) and this was a later engine than the "C" engine used in the F4U-4 which only used a normal two stage supercharger/intercooler set up.
This engine weighs about 150lbs more than the engine in the F4U-4 (without intercoolers) and makes the same power at 30,000ft that the F4U-4's -18 engine makes at 23,000ft.

This was NOT a simple modification or a tweak. F4U-5 first flew Dec 1945, 10 months after the first flight of the P-51H.

btw, so you know why P-51H top speed chart has a sharp reduction at middle altitude while F4U-5 has such smooth curve til high altitude?
p-47 chart is also like that. No sharp reduction in the middle

 

[drgondog]

If I recall correctly, P-51D has the kink which add a little bit of additional wing area, however, P-51H has a straight wing which reduce the wing area back to P-51B level again, beside it has less thick wing, which probably lead to lower CL

The P-51H wing was 235sq ft to P-51 (all through K) 233 sq ft. The Naca 66 2(18155) wing with slightly greater MAC had the same basic 'thickness' in actual dimension, but very slightly less Max Tc..

Spitfire 13%, F4U 18% FW 190 15.5%, P-47 14.6%

So tell me what you think CLmax was for each?

 

[Shortround6]

btw, so you know why P-51H top speed chart has a sharp reduction at middle altitude while F4U-5 has such smooth curve til high altitude?
p-47 chart is also like that. No sharp reduction in the middle

P-51s with Merlin engines used a two speed drive to the superchargers. Both impellers (stages) turned at the same speed and much like a car or bicycle with all gears removed but 2, you have either low gear or high and neither one is the best except at certain conditions.
F4U-4 has one speed for the engine supercharger and uses a 3 speed (or two gears plus neutral) for the Auxiliary supercharger. At low altitude the Aux supercharger is in neutral and all boost is provided by the main supercharger (and ram air in the intake, if present). After a few thousand feet the main supercharger cannot provide the needed boost and the aux supercharger clutch is engaged and the aux supercharger spins in low gear. After a certain point/altitude the this does not provide enough boost and the aux supercharger is shifted in high gear. This takes more power from the engine but provides more boost at higher altitudes.
F4U-5 has the main supercharger (not the same one as the F4U-4 ?) but the two auxiliary superchargers are powered by a variable dive system. They turn at the same speed as each other but the speed as which they turn is varied by the drive depending on the altitude the plane is at. No sudden change, just a gradual change is speed depending on altitude.

P-47 (and all US turbo-chargers) was also variable speed. The engine drove the engine supercharger at a fixed ratio from the crankshaft but the exhaust turbine in the supercharger got a different amount of exhaust gas at full throttle depending on the altitude. At low altitude most of the exhaust gas was vented out before it could act on the turbine. As the plane climbed into thinner air the vent/waste gate closed until at around 25,000-28,000ft the vent/waste gate was closed and just about all the exhaust gas was acting on the turbine.
A P-47D-10's turbo was spinning at 7,200rpm at sea level at high speed. At 31,000ft the Turbo was turning at 22,000rpm (max speed for the turbo). So the increase in pressure and thus speed was gradual. Please note that the compressor in the turbo needed power in proportion to the square of the speed so at 31,000ft the turbo required about 9 times the power it needed at sea level.

 

[mig-31bm]

P-51s with Merlin engines used a two speed drive to the superchargers. Both impellers (stages) turned at the same speed and much like a car or bicycle with all gears removed but 2, you have either low gear or high and neither one is the best except at certain conditions.
F4U-4 has one speed for the engine supercharger and uses a 3 speed (or two gears plus neutral) for the Auxiliary supercharger. At low altitude the Aux supercharger is in neutral and all boost is provided by the main supercharger (and ram air in the intake, if present). After a few thousand feet the main supercharger cannot provide the needed boost and the aux supercharger clutch is engaged and the aux supercharger spins in low gear. After a certain point/altitude the this does not provide enough boost and the aux supercharger is shifted in high gear. This takes more power from the engine but provides more boost at higher altitudes.
F4U-5 has the main supercharger (not the same one as the F4U-4 ?) but the two auxiliary superchargers are powered by a variable dive system. They turn at the same speed as each other but the speed as which they turn is varied by the drive depending on the altitude the plane is at. No sudden change, just a gradual change is speed depending on altitude.

P-47 (and all US turbo-chargers) was also variable speed. The engine drove the engine supercharger at a fixed ratio from the crankshaft but the exhaust turbine in the supercharger got a different amount of exhaust gas at full throttle depending on the altitude. At low altitude most of the exhaust gas was vented out before it could act on the turbine. As the plane climbed into thinner air the vent/waste gate closed until at around 25,000-28,000ft the vent/waste gate was closed and just about all the exhaust gas was acting on the turbine.
A P-47D-10's turbo was spinning at 7,200rpm at sea level at high speed. At 31,000ft the Turbo was turning at 22,000rpm (max speed for the turbo). So the increase in pressure and thus speed was gradual. Please note that the compressor in the turbo needed power in proportion to the square of the speed so at 31,000ft the turbo required about 9 times the power it needed at sea level.

So as I understand it, since Spiteful XVI use Griffon 101 fitted with three-speed supercharger that mean its speed graph would have somewhat similar shape as the F4U-4, is that correct?
P/s: I tried to plot Spiteful speed graph with available data point, but it seem like it should have the sharp reduction in the middle?

 

[mig-31bm]

The P-51H wing was 235sq ft to P-51 (all through K) 233 sq ft. The Naca 66 2(18155) wing with slightly greater MAC had the same basic 'thickness' in actual dimension, but very slightly less Max Tc..

I couldn't find data for wing area of P-51H, so what I did was basically work back the wing area from the summary of P-51 test. The wing loading at 9450 lbs is 40.5 lbs/ft^2, so I work out the wing area to be 233.3 ft^2. Whereas P-51D wing area is 235 ft^2

Spitfire 13%, F4U 18% FW 190 15.5%, P-47 14.6%

So tell me what you think CLmax was for each?

I would assume that the CL value of
F4U > P-47 > Spitfire> FW-190
But of course, FW190 and P-47 are very heavy compared to spitfire which leading to much higher wing loading value, so in term of turning capability would think that
Spitfire > F4U > P-47 > FW-190

 

[drgondog]

I couldn't find data for wing area of P-51H, so what I did was basically work back the wing area from the summary of P-51 test. The wing loading at 9450 lbs is 40.5 lbs/ft^2, so I work out the wing area to be 233.3 ft^2. Whereas P-51D wing area is 235 ft^2
View attachment 795644


I would assume that the CL value of
F4U > P-47 > Spitfire> FW-190
But of course, FW190 and P-47 are very heavy compared to spitfire which leading to much higher wing loading value, so in term of turning capability would think that
Spitfire > F4U > P-47 > FW-190

All Mustangs XP-51/Mustang I thru P-51K wing area =233.2 sq ft.
All Lighytweight XP-51F thru XP-51J wing area = 235 sq ft with same span, slightly longer MAC

 

[Zipper730]

So as I understand it, since Spiteful XVI use Griffon 101 fitted with three-speed supercharger that mean its speed graph would have somewhat similar shape as the F4U-4, is that correct?
P/s: I tried to plot Spiteful speed graph with available data point, but it seem like it should have the sharp reduction in the middle?
View attachment 795643

Usually you get a "saw-tooth" shape in the power-curve owing to the fact that you usually will reach critical altitude, have power fall off a bit before clutching into the next-gear.

 

[mig-31bm]

Usually you get a "saw-tooth" shape in the power-curve owing to the fact that you usually will reach critical altitude, have power fall off a bit before clutching into the next-gear.

So how about the XP-72?.
It seem to have no turbo charger like P-47, and however it had a R-4360-13 engine equipped with a two-stage variable speed mechanical supercharger, I would assume that the speed chart is also a smooth like similar to P-47 or F4U-5 instead of saw tooth shape like F4U-4 or P-51 right?
btw, if variable speed drive is so beneficial to top speed, why fighter only use 2-3 speed drive?
Also it seem like turbocharger is quite unpopular compared to supercharger, is there a particular reason for that ?

 

[Zipper730]

So how about the XP-72?.
It seem to have no turbo charger like P-47, and however it had a R-4360-13 engine equipped with a two-stage variable speed mechanical supercharger, I would assume that the speed chart is also a smooth like similar to P-47 or F4U-5 instead of saw tooth shape like F4U-4 or P-51 right?

I'm not sure if both supercharger stages are variable speed, but if the second-stage only was, I'd imagine you'd get the following: You'd see a normal power-curve when at low altitude since the auxiliary blower would be offline and the main-stage would be running as normal. At critical altitude, you'd see a fall off in performance before it was time to kick-on the second stage. At this point you'd see a fairly steady curve through the RPM range (there's usually a range of gear-ratios these systems work through) with a steady increase in speed up until the critical altitude would be reached at which point a normal fall-off would occur.

 

[GregP]

Kinda off topic, but does a game like Il Sturmovik really portray the Mustangs manuverability accurately?

Or does it hype it up?

I do like all the stall characteristics in Sturmovik, even though it can be annoying when your plane crashes from too tight a turn. But still, it helps make the planes feel more realistic.

and again, I know it's just a game.

The real P-51 has a benign stall. Once it gets into a full spin, though, it can take 10,000 feet to recover or it can pop out in 1 turn. In a fully-developed spin, the nose oscillates from about 80° below the horizon to above the horizon ... or it can recover quickly and easily, seemingly at random.

Several people I know who fly it practice stalls and stall approaches, but do not allow a spin to develop. They release back pressure at the first sign of a yaw break and it seems to do just fine. Go look at a flight manual and stalls are described as mild.

Spins are described and "uncomfortable" at best.

The moral of the story is don't get it into a spin, but there is nothing wrong with stalling it in coordinated flight. At least, that's what the pilot's manual says as well as a few P-51 pilots I know.

I have heard it described as "easy to fly, but with enough power and spin idiosyncrasies to kill you if you aren't careful at stall and when applying power at low speeds."

Not too sure about the spin characteristics, but MOST mid-war and later WWII fighters have enough power to torque-roll at low airspeeds if you apply power too quickly, so that's not unique to the P-51. As for spins, some are better at it than others, but a spin is not exactly a combat maneuver anyway. Might be useful if you find yourself in the soup above your airfield and some on the radio says the ceiling is 1,500 above ground level. Then, you could just spin down and recover when you get out of the soup.

I would not want to try that in a P-51, though. In a Hurricane? Sure.

 

[Shortround6]

Also it seem like turbocharger is quite unpopular compared to supercharger, is there a particular reason for that ?

It might be because while turbos are easy in theory they are/were hard in practice. GE tried to keep the temperature of the gases going into the turbo at 1750 degrees F.

Polished sheet/mirror reflecting the underside. Red is hot and blue is cold (inlet).

If you could make a WW II turbo you could make the turbine section of a jet engine. You needed the high temperature alloys and the manufacturing technology to build hundreds of turbos per month and thousands of those blades per month.
Or you have to figure out alternatives like hollow air cooled blades and how to make them and get the air flow through the blades at 18-22,000rpm.

btw, if variable speed drive is so beneficial to top speed, why fighter only use 2-3 speed drive?

Again, cost and weight.
Also variable speed doesn't actually give better speed at the altitude/s the individual gears are best at. It gives better speed at the in-between speeds.
A bit like car with a 2 speed manual transmission and a car with a automatic transmission ( a very crude old automatic) or torque converter.
The Torque converter "slips" with the output shaft running around 1/2 of the speed of the input shaft, this gives the "low" altitude performance. as the plane flies higher more oil is pumped into the torque converter (or the drive and driven parts are moved closer?) until they two shafts achieve parity or darn close. Fluid couplings do add to the cooling load depending on the conditions.
The Fluid drive/variable speed was usually heavier, bulkier and had more parts (cost) and may have required more maintenance. The Allies had shot down/crashed examples to look at in 1940 if not before. Allison got theirs into production at the end of 1943 (for the P-63) and P & W didn't get theirs into service until mid 1945 (?)

 

[don4331]

M mig-31bm :
To answer a couple of your questions in a little more detail:

Why did the F4U Corsair need more fuel: The wing is the thing. (And I'm taking a shot at drgondog CLmax question).

Plane	Root chord	Tip Chord	CLmax	Cdmin
Spitfire	NACA 2213	NACA 2209.4	~1.7	.006
F4U Corsair	NACA 23018	NACA 23009	~1.6	.007
Fw.190	NACA 23015.3	NACA 23009	~1.7	.0065
P-47	Republic S-3 11%	Republic S-3 11%	~1.2	.005
P-51 Mustang	NACA 45-100* (16.5%)	NACA 45-100* (11.2%)	~1.2	.004**

Thanks to Aeroweanie for the base information; CLmax numbers for NACA profiles are from Theory of Wing Sections.

As I understand it, the Mustang wing profile started as a 45-100, but was 'massaged' to get the characteristics they wanted. (I'm not great at the higher math's they used). I'm using Reynolds numbers in the 6 x10^5 range (which is important; you get very different results when you drop number down to 5x10^5 <say for your 1/10 RC scale plane>).

My question for drgondog : Does the P-51 Mustang airfoil have what I consider the classic 'laminar bucket', where Cd is only very low for a narrow range of CL or is it more like the 22xx and 230xx airfoils which are a smooth curve.

So, the much lower wing drag during cruise means the Mustang needs less fuel (and that's not getting into fuselage and cooling drag). It all gets extremely complicated.

For why air cooled engines can't operate at as high of boost, I would have said there were a couple reasons.

1. RR developed an extremely efficient supercharger, which meant there was less temperature rise. RR also cools their mixture after both stages of compression (when its at its hottest, getting most benefit from intercooling), while P&W cools their mixture between 1st and 2nd stages. A hotter mixture is more susceptible to detonation.​

2. Liquid cooling has the advantage short term of boiling. While, it only takes ~4.2 joules to heat 1g of water 1* C, it takes >2200 joules to boil 1g of water (already at 100*C). While, we don't want to boil all the cooling liquid, allowing a small fraction to boil allows for short term "over boost". (I hope I have my units correct, been a long time).​

​

So, while the maximum boost pressure for liquid cooled is higher, the sustained pressures are much closer.

The mechanic supercharger in a WWII fighter aircraft takes a lot of power and is turning quite rapid, but the powerplant (engine and supercharger) needs to be compact.

For the 1st stage the power demands/speeds are reasonable so you can get away with only one layshaft (like in a car manual transmission)​

​

However, for high ration the power/speed get large and little deflections of the shafts cause large issues, so RR used 2 lay shafts (like in a class 8 truck manual transmission)​

​

For the 3 speed Griffon, they used 1 at 12 o'clock (low speed), 1 at 6 o'clock (medium speed) and 2 at 3 and 9 o'clock and the supercharger drive area was filled to the brim. In order to have more ratios, RR would have needed to make the engine longer. (my diagram has space between the input and output gears for clarity, in real life, they are as close as the manufacturing can cut them).​

Then there is the shifting of the rations. Its not like your car where you can let off the gas to ease the shift, the shifts are happening at full power (what I would call powershifting, you push in the clutch, while pulling the shifter into next gear, never lifting off on the throttle). And that was a problem for the F4U-5 side winder - the load was so high, it broke things. So, rather than use clutches, P&W used more/less torque convertors. And "hydraulic connection" of torque convertor smoothed the acceleration of the auxiliary stage from neutral to low and from low to high (and back down again). Once the supercharger impeller had been accelerated, the connection speed was more/less constant. So, the R-2800-32W has a geared drive supercharger, it just has a fluid coupler for shifting, so it have the same jagged profile as the Merlin would.

Fun fact: For the 1st version of the R-2800-32, they put on 2 identical impellers in the auxiliary stage; and the P&W engineers couldn't figure out why it was so inefficient. Then, with red faces, they replaced the one impeller with one designed to turn the other direction and it worked as expected. One of those things, no one lets you forget.

For the XP-72, the auxiliary stage is actually a pump driving a motor with the motor running from ~1:1 to 6.375:1 (max) of the engine. Allowing smooth curve from ground level to 25k' (critical altitude). The challenge with a hydraulic drive - all the pressure you put into the oil to turn the supercharger creates heat. And last thing we want is more heat in the engine. (Luckily the adiabatic compression of oil doesn't add nearly as much heat as compressing air. And you multiply the efficient of the pump but the efficiency of the motor to get the efficiency of the system...so a 90% efficient pump * a 90% efficient motor gets you an 81% efficient system and the other 19% is heat (In contrast the gears in the Merlin are about 95% efficient). (The engine stage of the R-4360 is running 6.08:1 of engine for reference)

So, as a designer, you need to decide is it better to have just one speed = simple and light, couple/three speeds (e.g. take off and mid climb) or smooth curve. Can your country make the required parts? If you are missing critical materials to make great gears and/or turbocharger or are other solutions attractive.

 

[mig-31bm]

Again, cost and weight.
Also variable speed doesn't actually give better speed at the altitude/s the individual gears are best at. It gives better speed at the in-between speeds.

Ah, I see, that explain the speed variation between P-51H and F4U-5, they just overlapped back and forth

 

[mig-31bm]

M mig-31bm :
To answer a couple of your questions in a little more detail:

Why did the F4U Corsair need more fuel: The wing is the thing. (And I'm taking a shot at drgondog CLmax question).

Plane	Root chord	Tip Chord	CLmax	Cdmin
Spitfire	NACA 2213	NACA 2209.4	~1.7	.006
F4U Corsair	NACA 23018	NACA 23009	~1.6	.007
Fw.190	NACA 23015.3	NACA 23009	~1.7	.0065
P-47	Republic S-3 11%	Republic S-3 11%	~1.2	.005
P-51 Mustang	NACA 45-100* (16.5%)	NACA 45-100* (11.2%)	~1.2	.004**

Thanks to Aeroweanie for the base information; CLmax numbers for NACA profiles are from Theory of Wing Sections.

Damn, while I expected the CLmax of F4U and FW190 to be better than P-51, I didn't expect the different to be that big. In dogfight, F4U would be far better than P-51 even at same wing loading value

I also always thought that CLmax of P-47 wing would be bigger than P-51, it quite a surprise that their number are similar.


So consider that:
Performance of the XP-72 .
Sea level speed: 405mph (651 km/h)
Top speed: 490mph (788.5 km/h) at 25,000 feet
CLmax 1.2
Wing loading: 239.9 kg/m2
Power to weight: 0.515 hp/kg (0.233 hp/lbs)

Performance of Spiteful XVI (F.16)
Top speed: 494 mph (795 km/h) at 27,800 ft (8,473 m).
Sea level top speed: 409 mph (658 km/h)
CLmax: 1.2
Wing loading: 225.65 kg/m2
Power to weight: 0.536 hp/kg (0.24 hp/lbs)

Performance of the P-51H
Sea level speed: 413 mph ( 664.6 km/h)
Top speed: 474 mph (762.8 km/h) at 22,700 ft
CLmax: 1.2
Wing loading: 198.4 kg/m2
Power to weight: 0.517 hp/kg (0.234 hp/lbs)

Performance of the F4U-5
Sea level speed: 347 knots = 399 mph (642 km/h)
Top speed: 408 knots = 469.5 mph (755.6 km/h) at 27,000 ft
CLmax: 1.7
Wing loading: 200.5 kg/m2
Power to weight: 0.47 hp/kg ( 0.213 hp/lbs)

That mean in a turning fight:
F4U-5 > P-51H > XP-72 > Spiteful XVI

In term of boom/zoom fighting
Spiteful XVI > P-51H > XP-72 > F4U-5

 

[Holtzauge]

When looking at Clmax it's important to separate the Clmax on wing profile level with that on aircraft level.

And since the Clmax for the Fw-190 is given as 1.7 in the table above, this means that this is the Clmax for the profile, because that is ballpark what the NACA 230-series wing profile delivers provided the Reynolds number is high enough.

But what you really need to determine stall and turn performance is the Clmax on aircraft level, which is a completely different ballgame since this will depend on the wing's planform, profile variation with span and the washout etc.

The Clmax for the P-51 is elusive in that there are a lot of numbers floating around out there. You can find really low numbers for the wing profile Clmax like the 1.2 in that table, and I've seen those both from NACA, and from a German wartime report. However, it's important at what Re the testing is done at, and on a laminar profile also the surface roughness has a big impact. And it's unclear at what Re the German test was done (FB 1712 by Doetsch), but I suspect it was at a too low Re (my copy of report missing this info). And the low NACA number I've seen, is connected to a test with NACA "standard" roughness" which is more like sandpaper and while in-service aircraft may not have been polished, even the painted ones should have a better finish than NACA's "standard roughness". However, there is also another NACA report (by Abbot & Underwood Jan 1943) done at full-scale Re (6-13M) which pegs the profile Clmax for the XP-51 at an "intermediate" wing station at around 1.7 as well. However, note that on the P-51, the wing profile varies along the span, and that the profile becomes much more symmetrical towards the tip.

In addition, on the airplane itself the wing profile lift is not maintained at the tip and is lowered around the fuselage due to interference so the Clmax on aircraft level is much lower than the wing profile lift. But looking at Clmax on aircraft level, for the Fw-190 it's around 1.35, for the Spitfire 1.36 and the Bf 109 about 1.4 with the two latter probably as close as it gets since they are from RAE flight trial measurements using a trailing pitot system. For the P-51, there is actually a NACA report (No 829) which covers Clmax on full-scale aircraft in a large NACA wind tunnel which pegs the P-51's Clmax on aircraft level to around 1.4 as well.

But on top of all these numbers comes another caveat: These are the power off (with power on the Clmax is much higher) stall and turn Clmax numbers at low altitudes, and if you throw the effects of Mach into the mix as well (which comes into play at higher altitudes because you're going faster in turns), or look at instantaneous turns starting at higher speeds, then the Clmax numbers will be even lower due to buffeting effects.

Figure below from NACA report 829 (Mustang designated "Airplane 1" in report).

 

[Holtzauge]

Interestingly enough the F4U Corsair is in NACA report 829 only attributed a Clmax on aircraft level of 1.17 when in service condition, while with the wing faired and sealed, it got 1.26 which certainly is a bit better, but still far below the others and what I would have expected.

Figure below is from NACA report 829 (Corsair designated "Airplane 6" in report).