The Zero's Maneuverability

[wrathofatlantis]

I have to laugh at your P-51B Me 109G-6 above. The Bf 109 has a lower stall speed than the P-51B, and fights quite well at lower speeds. No self-respecting P-51B pilot would try to get SLOWER against a Bf 109 unless he was VERY fast to begin with. Best corner speed was about 270 mph IAS at 8,000 pounds and 10,000 feet (not many P-51 fights at 10,000 feet. Some, yes). It changed as you got heavier. By "best", I mean 8-g at the slowest speed where it can GET 8-g, meaning smallest radius of turn, at least for a part of a turn.

Nobody flew 75 - 100 turns in combat. Assume the pilot is flying a good fighter. His 360 degree turn is about 25 seconds give or take a bit. 75 turns would take half and hour or more. Right ...

Have you even looked at my post I wonder?

You know better than World War II Mustang Ace Clayton Kelly Gross, 355thFS, 354th FG, I presume?



View: https://youtu.be/wkaTGSpRuJI?si=Oolu80vxuQFWKXvD

AT THE 1:20 MARK "Fortunately, those shells went just behind my tail. I made I think probably 75 to 100 circles. Whether the Mustang was that much better (makes a small pinching gesture) or I was that much better than him, or a combination of both, I was gaining on him."

I already explained why a circle fight "locks you in." You cannot roll out.


Sustained speed was more like 23 seconds in my estimation, due to the reduced radius, so 3 full circles was roughly 69 seconds.

90 circles was thus around 34.5 minutes. On two other occasions other pilots stated "We flew in a circle for half an hour."


Quote "Nobody flew 75 - 100 turns in combat."


Tell that to Clayton gross then... I know of 3 examples, plus one shorter, all involving the P-51B vs the Me-109G-6. One is a P-51D.

While rare, it is not at all unknown.

As I pointed out, one Oscar survived 1/2 hour of gunnery passes by 16 P-38s until they all ran out of fuel and ammo. He escaped unharmed.

The Ki-43-II is less than 15 second to a circle. 30 minutes would be around 120 circles.

Quote from one P-38 pilot who saw this: "All he did was loafing in a circle at reduced power with his hood open."

The general conception of what WWII combat looked like is based either on the jet era or video games. Add to this a few German aces against isolated radioless receiver-only Soviet pilots, or P-38 aces against literally radio-less Zero pilots. (Apparently an effect of magnetic conditions in the South Pacific, affecting the Japanese Navy radios more: Justin Pyke devotes a good part of his interview to this.)

Hit and Run mostly worked on a straight moving unaware target. Turns ruined hit and run approaches, or allowed a head to head. That is why WWII combat often looked like this. Turning both protected you and trapped a target.

1945 FW-190A-8 ace commenting on a painting of his aircraft wings level: "Our wings were never level in combat. We turned continuously to one side. Outnumbered as we were by then (1945), it was the only way to survive."

-Notes by QAZ, "Aviation Decisive Battle Weapons" P. 164-165, "Unknown Sword" P. 86-89 (Me-109E-7 vs
Ki-27): Oblt. Losigkeit: "I think the Bf-109 can shoot when the Ki-27 tries [begins] to turn. [However] the Ki-27 often
turns, so I cannot dive on it."


-From Osprey "Ki-43 'Oscar' aces of World War 2": P.50: (Sgt Toshimi Ikezawa, Ki-43 ace) "I heard Major Eto
had refused delivery of the Ki-84.--- Hayate (Ki-84) pilots would simply drop the nose, and be off in a flash... They
could not avoid an attack if it came from above however, because of the Ki-84's poor rate of turn. (Ki-43-II 540
km/h, Ki-84 660 km/h)
I think we owe our survival to the Ki-43, as the Ki-84 would have left you in a mighty tight spot if you were
attacked from above by P-51s. ---Skilled (Spitfire Mk VIII) enemy pilots such as flight leaders would pull out of
their dives when they realized they could not catch us [unaware]. New pilots would dive straight down on us,
leaving them vulnerable in a turning fight." (Mk VIII: 650 km/h)



-Iseo Mochizuki (Ki-61 pilot ace, 2009 interview, 90 years old): "In the case of the Hien, you
would make a high speed attack [makes dive and zoom hand gesture]. But then, because [of
the speed] the turn radius became wide, and the enemy pilot [target] could turn inside you.
Because of that [speed] the chances of being shot down was high."

 

[Ivan1GFP]

When it comes to the Zero not having a high internal fuel capacity, I have to ask the question in return: Where did you find this information? The US analysis states an internal fuel capacity of 140-145 gallons which seems quite impressive to me for an aircraft with an engine that is in the 1000 hp class.

I'm not saying 2600 rpm 35" boost was used operationally. But those speed numbers are the most optimistic figures I have seen so far (as in from flight trial data) for the Zero. The engine data comes from a Japanese chart for the Sakae 12 engine that only goes up to 2550 rpm and in which I have extrapolated the effect of running the engine to 2600 rpm. From this I reverse engineer a Cdo to tune the simulation model. But if more credible data surfaces then I will for sure update the model. But as it is, the best I've seen so far (and which I use) is from the data Mike Williams and Neil Stirling generously have made available at WWII Aircraft Performance. On the other hand, even if the power or drag was to be slightly updated with new numbers, this will only have a marginal impact on the turn performance I posted earlier. So the general trend in those figures should still stand.

Then when it comes to the 60" boost for the Allison, this was actually cleared for operational use later in the war, and in a letter from the 12th​ December 1942, the Allison Division at the GMC complains that even though 60" of boost is now officially approved, some pilots are not content with this and even ran it at 70" for "prolonged periods of time".

The F4F-3 carried about 160 Gallons of internal fuel.
The typical P-40 from the C model onward carried around 147 Gallons except for the silly ones that deleted the forward tank.
Earlier than the P-40B, the P-40 could carry more fuel because it didn't have self sealing tanks.
Even the P-39 versions with self sealing tanks which were known for short legs carried 120 Gallons.

The Sakae 12 engine ran at 2550 RPM at Take-Off.
For Rated Power (Military?) it ran at 2500 RPM.
I haven't found anything listing a higher RPM though I am sure someone who doesn't know any better and without operating manuals might have tried it.
FWIW, 35,0 inches HG is actually slightly below Rated Power setting. It should be 35.83 Inches.

Regarding Allison engines:
The early Allisons V-1710-39 on a P-40E were serious screamers at LOW altitude.
In the test against Koga's A6M2, the fellow flying a P-39D with a very similar engine tried to take-off with 70 inches Hg of manifold pressure!
There were claims that early Hawk-87s were also running up to 70 inches Hg in the African desert. THIS is what Allison was responding to when they posted the memo to clear 60 inches Hg for operational use in the -39 and -73 engines. It was a number they could accept and still stand behind their product.
The problem is that supercharger on these engines didn't have the capacity to maintain boost to any great altitude.
Just for perspective, Allison didn't believe that 70 inches was plausible without a LOT of ram but 67 inches was plausible. Unfortunately, that 67 inches could not be maintained past about 2,000 Feet altitude.
I do not believe your estimate of altitude for 60 inches boost is reasonable for a P-40E. The quick reduction of performance with altitude was why the Merlin P-40 was seen as so much better than the early Allison P-40E/K.

 

[wrathofatlantis]

Hey wrathofatlantis,

re


Many aircraft in WWII had to ease-up on the stick and a few had to push it forward in high energy turns, in order to not involuntarily exceed AOA and/or G limits. This phenomenon was most commonly encountered while pulling out of high speed dives, but was not uncommon when pulling into high G instantaneous turns - this is part of what the prohibition against snap turns was intended to prevent. Less common, but occasionally encountered by some aircraft, was when there was a forward movement of the center of lift (as Ivan1GFP mentioned above), when the center of gravity moved to the rear (due to fuel or ammunition depletion), or a combination thereof. Depending on the aircraft design, this could happen anywhere in the envelope depending on the specifics of an aircraft designed to be marginally stable - or marginally unstable - in combat load conditions.

The reason you might end up actually pushing on the stick (as opposed to just easing-up) is that the forces on the elevators become so high that the control linkages transmit the force to the stick (overcoming the normal advantage the mechanical advantage of the lever/cam control linkage the pilot otherwise enjoys), whereupon the pilot has to use forward force to keep the stick from going fully to the rear and potentially tightening the turn to the point where the aircraft either stalls out or exceeds the allowable G load.

This is basic mechanics in engineering and physics (and hence in aerodynamics as well).

Modern combat fighters (F-16 is one example) that are designed to be inherently unstable in combat condition use the computerized fly-by-wire control system to overcome the inherent tendency to immediately go into a high G turn.

(I feel like I am leaving something out but cannot see what it is at the moment. )

Yes, high energy turns, but this is not the context of the Eric brown quote, since it was a trim condition that appeared going into the lower speed, which means the FW-190A gained extra forward lift when speed dropped below 355 km/h.


Duels in the Sky" Eric M. Brown, Naval Institute Press, 1988; "The change in trim could easily be gauged in turns. [I will add here, he specifies in turns as opposed to dive pull-outs, because in dives the prop is unloaded]. The FW-190 had a tendency to tighten up in a turn. Above [355 km/h] [I will add here: lower prop load at higher speeds, thus lower wing-prop longitudinal air compression tension], backward stick pressure was required."


It was a change in trim generated from the aircraft itself when the turn decelerated into lower speeds, not an excess of pilot input.

In my theory, the lower speed simply allowed, at 355 km/h. more air to bifurcate up between the prop and wing, since it is a point of meeting of air moving forward from below the wing (circulation theory) and the prop pushing it backwards. At 355 km/h in a 3G turn, the conflict between the the two airflows is resolved upwards, causing the pilot to push on the stick...

Furthermore, it shows an increase in lift AS SPEED IN THE TURN DECREASES.

This trim effect also only appears in turns and NOT in dive pull-outs, because in turns the prop is still highly loaded.

This explains the terrible performance of the FW-190A in dive pull-outs but not in turns (E. Brown: "Care must be taken on dive pull outs to not kill speed by sinking." Red Fleet: "Diving at 40 degrees for 1400 feet, the FW-190A will drop another 660 feet after reaching a nose level position. ........There is a moment where the FW-190A "hangs": It is then convenient to fire."), which can keep going down nose-up for hundreds of feet, crushing the pilot while performing no maneuvers.

This trim issue IN TURNS may also explain why they moved the whole engine forward another 6 inches from the A-5 model onward.

I will also add to a GrepP comment that "compression" here has nothing to do with "compressibility", but that should have been obvious...

 

[GrauGeist]

Notes by QAZ, "Aviation Decisive Battle Weapons" P. 164-165, "Unknown Sword" P. 86-89 (Me-109E-7 vs
Ki-27): Oblt. Losigkeit: "I think the Bf-109 can shoot when the Ki-27 tries [begins] to turn. [However] the Ki-27 often
turns, so I cannot dive on it."

So a Bf109E-7, flown by a German, was outclassed by an Imperial Japanese Army KI-27?

Odd, because the several Bf109E-7s shipped to Japan in 1941, were flown by Japanese pilots amd tested against the KI-43 (which had replaced the KI-27), the KI-44 and the KI-61.

So when and where did this rare event occur?

 

[Aeroweanie]

For the A6M2 Zero 21, I find two sources (Intelligence Summary 85 and USAAC memo) that say it had 54 gal of fuel in each wing and 37 gal in the fuselage. Another source (the RAAF flight test report, which gives imp gal) says 57 gal in each wing and 20 gal in the fuselage.

 

[GregP]

I have a design document for the Zeke 32 that says 134 U.S. gallons ... 20 gal in a fuselage tank just behind the firewall, and 57 in each wing tank. The exhaust is routed to the tanks, rendering the gas in the tank above the fuel level inert.

 

[NZTyphoon]

So a Bf109E-7, flown by a German, was outclassed by an Imperial Japanese Army KI-27?

Odd, because the several Bf109E-7s shipped to Japan in 1941, were flown by Japanese pilots amd tested against the KI-43 (which had replaced the KI-27), the KI-44 and the KI-61.

So when and where did this rare event occur?

Coincidentally, I have seen several identical comments by a "Gen Tlavok", who has often posted in Hyperscale Forums to push this particular barrow. The same pundit generally ignores flight test data in favour of anecdotal information from random sources. He also ignores factory production drawings in favour of photos to "prove" his particular pov regarding aircraft design.

 

[SaparotRob]

Coincidentally, I have seen several identical comments by a "Gen Tlavok", who has often posted in Hyperscale Forums to push this particular barrow. The same pundit generally ignores flight test data in favour of anecdotal information from random sources. He also ignores factory production drawings in favour of photos to "prove" his particular pov regarding aircraft design.

Professor Propwash wishes to learn from such a learned personage.

 

[Ivan1GFP]

I have a design document for the Zeke 32 that says 134 U.S. gallons ... 20 gal in a fuselage tank just behind the firewall, and 57 in each wing tank. The exhaust is routed to the tanks, rendering the gas in the tank above the fuel level inert.

Hello GregP,
Here is a little better information from a discussion a few years back:

Aerodynamic Drag Properties of the A6M

Hello Jetcal1, Actually all the A6M external tanks were droppable even though the fairings might have been a bit different. The problem was that when flying the longer missions, the A6M pilots did not have the option of dropping the tanks because they carried a substantial portion of the total...

ww2aircraft.net

So the actual total volume works out to 480 Liters or 126.8 US Gallons.

 

[GrauGeist]

Coincidentally, I have seen several identical comments by a "Gen Tlavok", who has often posted in Hyperscale Forums to push this particular barrow. The same pundit generally ignores flight test data in favour of anecdotal information from random sources. He also ignores factory production drawings in favour of photos to "prove" his particular pov regarding aircraft design.

 

[Aeroweanie]

I make the same estimate as you for the Bf-109, but where did you find a low speed Clmax as high as 1.5 for the Mustang?

The best I've seen so far is 1.4 from the NACA full scale testing documented in NACA report 829, and which is what I currently use in my simulation modelling.

In addition, I have a hard time understanding how the Mustang could have such a high Clmax as 1.5 on aircraft level? Both the Spitfire at 1.36 and the Bf-109 at 1.4 come in lower, and comparing the wing profiles on these aircraft (looking at things like camber lines and nose-radiuses) I would expect a lower, not higher, Clmax on wing profile level for the Mustang.

There is also a German report (FB 1712 by Doetsch) in which they only get a Clmax of 1.22 for the Mustang. And this is only on profile level, so I would expect a Clmax significantly higher than 1.4, or else it's difficult to explain that the Mustang is able to generate the 1.4 on aircraft level NACA recorded. But I suspect this may be due to Re effects, since the Germans only ran the tests at Re=2.7M. But I know from a paper you have done that you have studied the P-51's aerodynamics in detail, so maybe you have an explanation?

I calculated the P-51B CLmax value from stall speeds given in the flight manual, assuming zero measurement error.

I make the same estimate as you for the Bf-109, but where did you find a low speed Clmax as high as 1.5 for the Mustang?

The best I've seen so far is 1.4 from the NACA full scale testing documented in NACA report 829, and which is what I currently use in my simulation modelling.

In addition, I have a hard time understanding how the Mustang could have such a high Clmax as 1.5 on aircraft level? Both the Spitfire at 1.36 and the Bf-109 at 1.4 come in lower, and comparing the wing profiles on these aircraft (looking at things like camber lines and nose-radiuses) I would expect a lower, not higher, Clmax on wing profile level for the Mustang.

There is also a German report (FB 1712 by Doetsch) in which they only get a Clmax of 1.22 for the Mustang. And this is only on profile level, so I would expect a Clmax significantly higher than 1.4, or else it's difficult to explain that the Mustang is able to generate the 1.4 on aircraft level NACA recorded. But I suspect this may be due to Re effects, since the Germans only ran the tests at Re=2.7M. But I know from a paper you have done that you have studied the P-51's aerodynamics in detail, so maybe you have an explanation?

I checked and it turned out that for expediency, I used CLmax values from P-51D stall data for the P-51B. I got a P-51B manual and reduced the stall speeds to a CLmax. At a Mach number of about M=0.12, I get CLmax values ranging from 1.65 to 1.72. There is a problem though - the airspeed correction tables don't go below 150 mph and the stall speeds range from 87 to 96 mph, depending on GW. For this reason, I am wary of these CLmax values.

I also went through several P-51D manuals and find a range of stall speeds and correction tables. Depending on the manual and weight, I find values from 1.34 to 1.44, at Mach numbers from M=0.13 to M=0.15. It should be noted that these are all power off.

There is some good P-51D data in NACA TR-1219 "Measurement and Analysis of Wing and Tail Buffeting Loads on a Fighter Airplane". This plot is of CNB, which is the normal force coefficient for the onset of buffet:

The report that throws a monkey wrench in the works is NACA TN-2525 "The Effect of Rate of Change of Angle of Attack on the Maximum Lift Coefficient of a Pursuit Airplane" (also NACA RM A8I30). This shows big variation in CLmax, depending on stall entry rate. I should note that this data is all power off too.

 

[Ivan1GFP]

This may be a silly question but are you calculating CLmax for the P-51 with flaps up or down?

 

[Holtzauge]

The F4F-3 carried about 160 Gallons of internal fuel.
The typical P-40 from the C model onward carried around 147 Gallons except for the silly ones that deleted the forward tank.
Earlier than the P-40B, the P-40 could carry more fuel because it didn't have self sealing tanks.
Even the P-39 versions with self sealing tanks which were known for short legs carried 120 Gallons.

The Sakae 12 engine ran at 2550 RPM at Take-Off.
For Rated Power (Military?) it ran at 2500 RPM.
I haven't found anything listing a higher RPM though I am sure someone who doesn't know any better and without operating manuals might have tried it.
FWIW, 35,0 inches HG is actually slightly below Rated Power setting. It should be 35.83 Inches.

Regarding Allison engines:
The early Allisons V-1710-39 on a P-40E were serious screamers at LOW altitude.
In the test against Koga's A6M2, the fellow flying a P-39D with a very similar engine tried to take-off with 70 inches Hg of manifold pressure!
There were claims that early Hawk-87s were also running up to 70 inches Hg in the African desert. THIS is what Allison was responding to when they posted the memo to clear 60 inches Hg for operational use in the -39 and -73 engines. It was a number they could accept and still stand behind their product.
The problem is that supercharger on these engines didn't have the capacity to maintain boost to any great altitude.
Just for perspective, Allison didn't believe that 70 inches was plausible without a LOT of ram but 67 inches was plausible. Unfortunately, that 67 inches could not be maintained past about 2,000 Feet altitude.
I do not believe your estimate of altitude for 60 inches boost is reasonable for a P-40E. The quick reduction of performance with altitude was why the Merlin P-40 was seen as so much better than the early Allison P-40E/K.

About the Sakae 12: AFAIK the mil rating was +150 mm 2500 rpm, while take-off was +250 mm 2550 rpm. However, is there any data suggesting that this (+250 mm 2550 rpm) was/could be used for anything else but take-off? And if so, for how long was it available as WEP? And about the US test: I can only conclude they ran it at 2600 rpms because that is what the report says. As to why they did it we can only speculate.

About the maintaining of 60" in my estimate being unreasonable: Remember that the speed data I posted is including RAM: In the climb chart I will post in my paper, in that figure the 60" boost is not even maintained up to 1000 m, i.e. not even half of that in the speed chart.

Then about the Zero's internal fuel capacity: I have three different Allied reports that come to three different conclusions: One states 2x55+37=148 gallons, another 2x54+37=145 and also 2x51.5+38=141 gallons. I found that impressive when compared to Spitfire's 85 gallons imperial and the Bf 109's 400 l that's all.

 

[Holtzauge]

I calculated the P-51B CLmax value from stall speeds given in the flight manual, assuming zero measurement error.


I checked and it turned out that for expediency, I used CLmax values from P-51D stall data for the P-51B. I got a P-51B manual and reduced the stall speeds to a CLmax. At a Mach number of about M=0.12, I get CLmax values ranging from 1.65 to 1.72. There is a problem though - the airspeed correction tables don't go below 150 mph and the stall speeds range from 87 to 96 mph, depending on GW. For this reason, I am wary of these CLmax values.

View attachment 799169View attachment 799170

I also went through several P-51D manuals and find a range of stall speeds and correction tables. Depending on the manual and weight, I find values from 1.34 to 1.44, at Mach numbers from M=0.13 to M=0.15. It should be noted that these are all power off.

View attachment 799171
View attachment 799173

There is some good P-51D data in NACA TR-1219 "Measurement and Analysis of Wing and Tail Buffeting Loads on a Fighter Airplane". This plot is of CNB, which is the normal force coefficient for the onset of buffet:

View attachment 799168


The report that throws a monkey wrench in the works is NACA TN-2525 "The Effect of Rate of Change of Angle of Attack on the Maximum Lift Coefficient of a Pursuit Airplane" (also NACA RM A8I30). This shows big variation in CLmax, depending on stall entry rate. I should note that this data is all power off too.

I think the greatest difficulty when using stall speed as a base is to get the position error correction right. And this seems not only to vary hugely between different aircraft, but be quite substantial as well: For example, the stall speed POC for the Bf-109E is in the order of 20 mph. So for Clmax values for the Spitfire MkI and Bf-109E, I find values derived from flight tests with trailing pitot systems much more reliable. And when the RAE performed such tests on these aircraft, they yielded a Clmax of 1.36 and 1.4 respectively.

I have seen NACA TN 2525, but that report covers transient Clmax as far as I can understand. And these type of tests usually result in a much higher Clmax than under gliding conditions (I've seen NACA data with similar high transient Clmax values measured for the P-47 as well). So for sustained turn calculations I would not rely on these since as far as I understand it, they are the result of a short lived leading edge vortex which forms at rapid angle of attack changes but then quickly dissipates.

In addition to the full scale test of the P-51 which resulted in a low speed Clmax of 1.4 (see attached excerpt from NACA report 829), there is also NACA TN 1044 which also captures the effects of Mach and altitude. And this report does seem a bit more optimistic since at M=0.15 such as in low altitude turning, it suggests 1.45. However, the Clmax seems to drop off much faster than I would have expected with a Clmax even below 1.25 at M=0.3. In addition, the NACA TN 1044 results are with power on which tends to increase the Clmax so there is that to consider as well.

 

[tyrodtom]

I've been shot at in a chopper, actual combat, and it's pretty stressful.
I've also circle track raced myself from 1996-2016, helped and spotted ( radio) for other racers at NASCAR, ARA, and other tracks from 1990-to present.

I won't say racing is as stressful as combat, maybe a 6 on a scale of 10.

But I've noticed I can't or any other racer I've been around can keep track of what lap they're on without outside help, like seeing the lap number on the score board, you're exsisting in the second, using every sense you have just to keep the vehicle under control till the next second, you can plan ahead, to some extent, but your mind is in survival mode, not record mode.
When it's over you remember some, but jumbled sometimes, out of order.

One of my early sponsors was a guy that videoed every race. He gave me videos of every race I was in.
Later I would watch those videos, and realize sometimes I didn't remember things in the way they actually happened.
And that's not unique to me alone.
For years afterward we also had GOPRO recorders in the car also.

The human mind is just not equipped to record accurately under stress.
They're not cameras, they're not recorders, they're not computors, they're humans beings.
Maybe rare individuals can, but they're very rare.
So rare I don't think I've ever met any.

Some of you seem to believe perfect recall, of events that happen 50-60, maybe even 70 years before.

I notice the pilot said he circled PROBABLY 75-100 circles. A big variation, and then there is that "probably".
I know most aircraft of that era had clocks.
How many pilots do you think wrote down the time when they entered a dogfight, and again wrote down when it ended?
I think we all know it's just a guess.

I read hundreds of books too, on aerial combat, watched many videos, but I've always realized they're just memories of a human mind.
My dad, and two of my uncles were WW2 Marine veterans, all fought in the Pacific.
We had some interesting family get togethers, this was in the 50's early 60's, I'd try to soak up all they had to say.
I was in my early teens, they were in their late 30's or 40's, about 15-20 years since the war.
Even then I could see that some of their recall of events getting meshed, re-arranged, that they didn't have perfect recall.
They were my heroes , but they were still human.

 

[tyrodtom]

Maybe the pilot was using his after mission debriefing notes that most pilots had to write asap after they landed.

But with him saying " probably 75-100 circles" shows that he didn't count them, and was just guessing, even just after the mission.

 

[Juha3]

Let me ask you this: Why does Eric Brown say that below 220 mph, in a turn with the FW-190A, you had to push on the stick to keep the nose from going up too high on its own?

Yes, ALL FW-190A pilots will tell you this: In a low speed turn that you made as tight as possible, you were pushing on the stick during the turn.

What would your aerodynamics professors say to that?


Still waiting for a Spitfire gaining turns on a FW-190A below 20 000 feet.

You say something was happening ALL the time, with a huge margin of advantage, but won't provide ONE example. Curious.

How about this one

 

[GregP]

Hello GregP,
Here is a little better information from a discussion a few years back:

Aerodynamic Drag Properties of the A6M

Hello Jetcal1, Actually all the A6M external tanks were droppable even though the fairings might have been a bit different. The problem was that when flying the longer missions, the A6M pilots did not have the option of dropping the tanks because they carried a substantial portion of the total...

ww2aircraft.net

So the actual total volume works out to 480 Liters or 126.8 US Gallons.

I didn't mention external tanks, Ivan. The 134 gallons above is all internal. It is also A6M3 Model 32.

I can certainly find out the fuel capacity for the A6M5 Model 52 since we fly one, but I'd not bother the people about it until the airshow was over ... which it was 2 weeks ago. So, this weekend is a good time to ask and we HAVE the information if only the correct people are around on Saturday.

 

[GregP]

Maybe the pilot was using his after mission debriefing notes that most pilots had to write asap after they landed.

But with him saying " probably 75-100 circles" shows that he didn't count them, and was just guessing, even just after the mission.

Nobody flew 100 circles in continuous combat. The engine would grenade after 1/2 hour of combat power in most cases and there would be no point for another.

If you can't figure it out after 3 - 4 turns, you are 90% toast anyway since the other guy is as good or better than you are.

 

[Ivan1GFP]

About the Sakae 12: AFAIK the mil rating was +150 mm 2500 rpm, while take-off was +250 mm 2550 rpm. However, is there any data suggesting that this (+250 mm 2550 rpm) was/could be used for anything else but take-off? And if so, for how long was it available as WEP? And about the US test: I can only conclude they ran it at 2600 rpms because that is what the report says. As to why they did it we can only speculate.

About the maintaining of 60" in my estimate being unreasonable: Remember that the speed data I posted is including RAM: In the climb chart I will post in my paper, in that figure the 60" boost is not even maintained up to 1000 m, i.e. not even half of that in the speed chart.

Then about the Zero's internal fuel capacity: I have three different Allied reports that come to three different conclusions: One states 2x55+37=148 gallons, another 2x54+37=145 and also 2x51.5+38=141 gallons. I found that impressive when compared to Spitfire's 85 gallons imperial and the Bf 109's 400 l that's all.

Regarding the use of settings above their "Military" rating, we have the claim by Mr. Saburo Sakai of the A6M2 being able to achieve 345 MPH on "Overboost" setting.
We also have corroborating evidence in the third of the reports posted by R Leonard (SWP Report).
Read the comments section.
The Sakae was apparently a well built and tough engine as later TAIC reports about "Flash Performance" from Oscar and Mr. Sakai's encounter with a squadron of Hellcats would seem to indicate.

This particular report and most others that show some detail on Koga's rebuilt A6M2 seem to indicate that all was not right with the carburetor.
There are comments showing a bit of surprise when thing appear to be running correctly. There was also a mention of the automatic mixture control not actually working. I believe this was by Corky Meyer. The pilot found that after a bit of climbing, the engine was starting to misbehave until it was leaned out.
This particular engine had spent a month with part of it underwater. When rebuilt, it had to have a couple cylinders resurfaced. I do not know whether the carburetor was also underwater.
In many of the test reports on Koga's A6M2, the critical altitude seems a bit too low and for some reason they are running the engine as if they believe 2550 RPM is the proper limit.
I don't recall ever seeing a test run at 2600 RPM, so if you know where I can find that, I would appreciate it. There was speculation that 2600 RPM was the limit but I haven't actually found a test report showing this was done in a test flight.

Regarding 60 inches of boost on the P-40E at altitude. I believe you are being VERY optimistic. This level of performance was more characteristic of later Allison engines that did not run out of breath quite so low.

Regarding fuel loads: Spitfire and Me 109 both had just over 100 US Gallons of internal fuel. Both were also known for having very short ranges. What I found interesting was that Hurricane wasn't that much better at only 116 Gallons. You wonder why US was complaining about the Airacobra with 120 Gallons.
Of the reports, the 141 Gallon internal fuel load for A6M2 is closest to correct. It should really be about 141.3 US Gallons or 480 Liters nominal.