An Allison V-1710 with a two-speed supercharger in 1942 (7 Viewers)

[drgondog]

Curiously enough, the USAAC contemplated a multi-engine/heavily armed/fast convoy escort fighter in the Air War Plans Division AWPD-1 delivered in September 1940. The only difference scheme wise from the future YB-40 was the stipulation that it be agile and fast enough to quickly respond to attacks on the escorted bombers from any position of escort.

The senior leaders of AAC/AAF understood that deep daylight strikes would require escort fighters capable of defeating enemy single engine pursuit but were still frozen in beliefs that a single engine escort with required fuel fractions to achieve the necessary range, would not be capable of competing one-on-one against enemy interceptors.

Candidly, that belief was certainly shared by senior RAF and LW leaders. However, the Japanese A6M certainly was a shock. Mitsubishi pioneered the future 7075 aluminum alloy combined with integral drop tank and ruthless stripping of weight, including self-sealing fuel tanks and pilot protection.

 

[Shortround6]

The senior leaders of AAC/AAF understood that deep daylight strikes would require escort fighters capable of defeating enemy single engine pursuit but were still frozen in beliefs that a single engine escort with required fuel fractions to achieve the necessary range, would not be capable of competing one-on-one against enemy interceptors.

They might have been right, from a theoretical standpoint.
But the P-51 had several advantages over the Defending German fighters. Better (newer) aerodynamics with lower drag and fuel that allowed powerplant with a better power to weight ratio. Didn't hurt that the Bombers they were escorting were not very streamlined either, so their fuel fraction didn't allow for much further flights while still carrying a large bomb load.

Candidly, that belief was certainly shared by senior RAF and LW leaders. However, the Japanese A6M certainly was a shock. Mitsubishi pioneered the future 7075 aluminum alloy combined with integral drop tank and ruthless stripping of weight, including self-sealing fuel tanks and pilot protection.

I think it was more of they didn't add the protection/extra weight at the same time other countries did.
Not that the P-40 was ever going to be a light weight but the P-40NL (no letter) gained around 500lbs empty by the time they got to P-40C and about 300lbs of useful load (a lot of that was extra ammo). P-40 may a worst case but that 1938-40 design period was tough, lots of changes in wanted features and for some countries not a lot of increases in engine power in the near future.

As far as fuel fractions and escort fighters goes?

What happens when they try to stuff 1154 US gallons inside the plane.
Use the same engine/s as the P-80.
It is not easy trying to design an escort fighter that fight with a short range fighter using the same technology.

 

[wuzak]

A Merlin XX, while attractive based on higher FTH, would probably have generated a better fighter for AAF doctrine but would not have been of much more benefit as a point interceptor and would have IMO required substantial and time-wasting changes for the benefits expected.

The Merlin was heavier than the Allison and the center of mass would have been further aft of the wing aerodynamic center plus required an updraft carb forcing a lower fuselage carb intake as well as radiator intake scoop below or more efficient ones imbedded in leading edge. Almost certainly the wing would have to move aft and a fuselage plug required to maintain static margins with existing empennage design.

In other words, even the concept of a Merlin XX was a non-starter due to War Production Board demands for production commitments. Additionally, Bell was never going to get a Merlin allocation from Packard production as AAF diverted their share of Packard Merlin 1650-1 to Curtiss. Very soon thereafter the priority for Packard Merlins was shifted to NAA for the new 1650-3 and the balance of 1943/Q1-44 for Packard production for AAF was divided for 1650-1 spares and 1650-3 for Mustangs.

The V-1710 was designed to be somewhat modular/

This allowed the P-39 concept to go forward - just remove the propeller reduction gearbox from the engine, use a remote gearbox connected by a drive shaft.

The Merlin's propeller reduction gearbox was partially integrated into the crankcase casting.

To use the V-1650-1 for the P-39 would have required modifying the airframe and the engine to suit (changing to a downdraft carby as well).

 

[Shortround6]

changing to a downdraft carby as well).

No need.
Just Stick the radiators and oil coolers out under the wing or in the wing leading edge Fairley Firefly style

Adjust the wing forward or aft to suit center of gravity and stick the carb intake out the bottom. Stick good filter on it to stop FOD from the front wheel.
Never let reality spoil a good "What if"

 

[MikeMeech]

The thing with "trick" armament layouts is that they depend on the enemy to keep doing the same things.
Four guns in a power turret compared to single guns on unpowered mounts sounds like a good idea (it really wasn't) but it depends on your enemy to keep using single non-powered guns for defense 3-4 years (more?) in the future.
Meanwhile the British are sticking twin and quad gun power mounts in their newest bombers (Wellington a Whitley) and are installing twin non-powered guns in Hampdens.
French are putting 20mm cannon in power mounts in their better twin engine bombers.
That would be something to see, Defiants and LeO 451s exchanging broadsides like flying ironclads
For some reason the Germans, when it came to power turrets, got drunk in the pub, staggered down the dock, missed the ferry and fell in the harbor as the ferry steamed away.
Expecting your enemies to do that is poor planning.

It also tends to overlook the Bf 110, which was the most notable long range/heavy fighter of it's day but far from the only one. Assuming that your enemy would only have single engine short range escort fighters 3-5 years in future also seems like poor planning/wishful thinking.

110s actually worked when used as intended but then the plan also needed co-operation by the British. Get ahead of the bombers and attack the fighter fields as the interceptors were warming up or at low/mid altitude climbing (low energy). Radar warning tended to screw that one up. Making the 110s fly at the same speed as the bombers and nearly the same height (low energy) finished off any hope of the 110s acting as escort fighters.
Brisfits worked in WW I

Crew could touch each other and at 120mph had time to 'communicate', depending on the intercom and with things happening 3 times faster (300mph) and with no forward gun/s depending on crew co-ordination was a lot harder.
They knew in WW I that the DH 4 had problems with crew co-ordination compared to the Brisfits and DH-9s which did not stick a large fuel tank between the pilot and gunner.
British expected a lot from the intercom?

Hi
Yes the DH-4 had those communication problems, also experiments were undertaken into various means to overcome them if possible, for example the following report:

The information on this use of the Klaxon (along with another image) was included in my article on 'The RFC & RAF use of The Klaxon during World War One' in 'Cross & Cockade International' Summer 2017. Other methods such as a rear view mirror were also fitted to some DH.4 so the pilot could see the observer/gunner:

As for the Bf 110, which was designed to meet a 1934 requirement for "an aircraft capable of accompanying bomber formations and penetrating deep into enemy territory", was not a great success in that task despite it being quite a good aircraft. In the BoB formations of Bf 110s would form circles on occasions to defend each other against single seat fighters that out performed them, all rather like RFC FE.2bs during WW1 when coming up against German SS fighter formations. You could not protect the bombers like that though. This was the problem with long range fighter designs in the 1930s they believed they needed two engines and another crew member on the whole, and so would generally always be out performed by SS interceptor fighters.

Mike

 

[Geoffrey Sinclair]

Woah! Incredible!! You really nailed me here. I'm so ashamed, you caught me using quickly googled production numbers in my post from four pages back. What gave it away? Drat!! I should have known I couldn't pull one over on a swift you, and I therefore must clearly deserve your ponderous contempt!

Could it be when i wrote this bit ?

"Some production numbers here (just googling, so correct me if I'm wrong)"

The reaction to the corrections indicates the invitation to make them is rather like the company slogan "Customer is number one", getting the embarrassing bit out of the way in the title, rather than sincerity.

One minor correction though - America's Hundred Thousand, while a great resource (and which I like many here have on my shelf) which does refer to primary sources, is not itself a primary source.

Quite correct, but it is known several people reading have the book so can see production figures for themselves. Apparently the web search was preferred to the book, why?

Francis Dean in America's Hundred Thousand has a table of the 8 main WWII fighter types, by numbers produced but has differences with the USAAF Statistical digest.

The big difference is the P-40 production June to August 1942, Dean says 560 the Digest says 1,158. As a result mainly of this Dean's P-40 production grand total is around 600 less than other references at 13,143, versus 13,738 in the digest for example.

For the P-39 the period July to November 1942 Dean says 835 produced the digest says 784. Overall Dean says 9,547 P-39s built, the digest says 9,588.

Interestingly the P-38, 47, 51, 61, 63, Buffalo, Wildcat, Hellcat and Corsair breakdowns by model type all are totalled on the page, and the totals agree with the table at the front of the book. The P-39 and P-40 breakdowns by model do not have a total, indicating somebody noticed a problem.

P-38, Dean includes the production of F-4 and F-5 (unarmed reconnaissance versions of the P-38) under P-38. His monthly figures are identical to the digests except for August 1944 when he is 2 less. He also has the 1945 yearly total include 3 aircraft not accounted for in the monthly totals. There were 500 F-4 and F-5 produced, plus other P-38s were converted.

P-47, identical

P-51, as with the P-38 the reconnaissance version of the P-51, the F-6 is included in the totals, but not the A-36 variant. The two sources are identical to the end of 1944.

Wildcat, the digest gives only yearly totals, and they are consistent with Dean as are the Hellcat figures, once you know the 1942 Hellcat (10) and 1943 Wildcat (100) production at Grumman.

P-40, until June 1942 the figures are close to identical (difference of 2 in January 1941, 1 in May 1942, out of around 4,770 built to end May 1942). Then comes the upheaval, table is date, Dean, Digest,

6/42 / 282 / 347
7/42 / 135 / 421
8/42 / 143 / 390
Totals / 560 / 1158

Normal service then resumes in September, with Dean being 1 less than the digest, same total in October, Dean 3 more in November, same total in December 1942. The figures are then identical until deliveries end in November 1944.

P-39, until July 1942 the figures are close to identical (difference of 1 in April 1942 out of around 1,425 built to end April 1942). Then comes the upheaval, table is date, Dean, Digest,

7/42 / 255 / 170
8/42 / 309 / 60
9/42 / 0 / 132
10/42 / 3 / 145
11/42 / 268 / 277
Totals / 835 / 784

Normal service resumes in December, Dean being 3 less. The figures are then identical until deliveries end in August 1944.

Corsair, apart from the Statistical Digest's yearly figures Barrett Tillman in The F4U in World War II and Korea gives monthly figures until December 1945, Dean gives monthly figures until December 1944. It is interesting that the figures he gives in the table are not consistent, the monthly figures do not always add up to the yearly totals, the yearly figures are also different from those in the Corsair chapter.

Dean's yearly totals from 1940 to 1945 in the table are given as,

1, 0, 178, 2293, 5380, 3578

Putting the monthly numbers into a spreadsheet the totals are

1, 0, 178, 2287, 5440, 3578 (there is only a yearly total for 1945)

Go to the Corsair chapter and the numbers are

1, 0, 178, 2298, 5372, 3578.

Barrett Tillman's figures are

0, 0, 178, 2298, 5372, 3567 (3385 January to August 1945) (so he misses the 1940 delivery of the prototype).

The USAAF digest's figures are

0, 0, 178, 2293, 5380, 3387 (to end August) so around 3569 for 1945, Using Tillman's figures for September to December 1945.

Following my analysis, detailed below, I make the figures as

1, 0, 2293, 5380, 3385 (to 8/45) or 3567 for all of 1945 with the note
the 1945 figures are the least certain and Dean claims another 11
Corsairs were built in 1945.

So the overall totals are Dean table grand total 11,484, the table's yearly figures 11,430, adding up the monthly figures 11,484, but a typo of 60 aircraft reduces this to 11,424, Tillman 11,415, Digest 11,420 using the Tillman figures for the final 4 months of 1945.

Differences between Dean, Tillman

10/42 Dean 1 more
11/42 Dean 1 less,
8/43 Dean 6 less, assume Dean in error, as this means the totals then add up to the Digest total Naval fighters figure for the month. It also means the Dean yearly total for 1943 equals the sum of the monthly totals instead of being 6 more.

1/44 Dean 1 more, assumed correct as the numbers then add up to the Digest total.

2/44 Dean 60 more, this appears to be a typo, the Hellcat production figure copied into the Corsair column, using the Tillman figure means the totals then add up to the Digest figure.

Dean's monthly figures stop in December 1944.

Everyone makes mistakes "Geoffrey", as you have done repeatedly in this thread. It's ok to admit when you do, just own up! But you always ignore corrections, deny numbers that you don't like, and blatantly, ponderously, fudge numbers, omit data that (it turns out after you are called on it) you clearly had access to, simply because you are trying to paint a narrative that you prefer over whatever the data shows.

Yes it is the attempt at truth by repeating the same phrases over and over, evidence is irrelevant. And again accusations of making up data, with hiding data now added. Think of it this way, convince yourself the bad guys will rape, loot and murder, which makes them real bad, so in the name of the greater good the good guys can lie, cheat and steal, which makes them real good. So the accusation is not including all the data deemed relevant, from the person who complains my messages are too long.

At least this keeps some junk history from being posted, though plenty of people clearly think the price is too high.

This is why there is really no reason to ever read your posts.

Says someone who did a line by line reply and really needs selective reading.

If the claimed Axis losses in the MTO are correct the allied fighters were collectively poor performers, disguised by major over claiming.

Well I heard that British Commonwealth pilots were worse shots than the French. Seems legit, right?

One of the anecdotes Christopher Shores liked to tell is the shock even experienced RAF fighter pilots felt when taking a course at the RAF Central Gunnery School. The one constant in the Mediterranean Air War series is the daily listings of combat losses and claims, makes it easy to see over claiming.

Meantime a bunch of campaign loss figures were provided which, if correct, indicated major problems with allied fighter claims, the figures were given once without attribution and apparently must not be seen again, as the answer above shows. They also contradict sources like the USAAF Statistical Digest.

If you think finding accurate production data is hard try losses. So were the loss figures provided
1) Due to enemy action?
2) On operations?
3) By combat units?
4) Include write offs?
5) Include the whole air force?
6) non combat losses?

In mid 1943 the allies in the Mediterranean had the 9th, 12th, Strategic, Coastal and Desert air forces, plus Malta Command and Middle East Command, with probably some Coastal Command operating out of Gibraltar. So were they all included?

Similar for the axis, the forces in East Africa, the Western Desert, Greece, Italy, later Tunisia and France.

Other topics,

Some of the 15th Air Force 1945 monthly reports have Engine Change Reports, graphs of average running time per aircraft engine changed. Generally speaking the R-1830 radials in the B-24 and the R-1820 radials in the B-17 had at least an average of 300 hours running when changed, few had 400, with the R-1820 ahead, most having 350 or more hours. Overhauled engines usually had around half the average hours of new engines. The V-1650 Merlins and the V-1710 Allisons usually averaged over 200 hours, few made it to 300 until March/April. Overhauled Allisons were around half the new engine hours, overhauled Merlins more like a quarter.

Defiant, prototype to F.9/35 two set day and night fighter, turret armament amidships, speed similar to single seat fighters, provision for small bombs. Hawker Hotspur prototype also built, Hotspur ordered from Avro in 1936 but order cancelled.

First 5 Defiant built in September 1939, with 16 by end 1939, monthly production hit double figures in March 1940.

264 squadron received its first Defiant in December 1939, while 141 replaced its Gladiator and Blenheim If with Defiant April/May 1940

January 1940 trials with 2 squadron, army co-operation
mid February 1940 264 squadron night flying began
February/March 1940 multiple trials against bomber formations, a single one on one with a Spitfire, Defiant won convincingly.

6 May 1940 Fighter Command first reports Defiant on strength, 1 squadron, 16 aircraft, though 264 squadron is reported to have become operational mid March.
4 June 1940 Fighter Command reports 2 Defiant squadrons on strength, with 25 serviceable aircraft between them.

Case History of Fighter Plane Range Extension Program, Parts I and II," USAF Historical Research Document 202.2-11, 22 February 1945, Enclosure 1." Note lack of fuel reserves. AFHRA Reel A2069 Page 1658 dated 6 August 1943, calculations, not flight tests.
P-39N 7,700 pound take off, 120 gallons internal, 52 Gallons for 5 minutes War Emergency Power & 15 minutes Military Power, 40 Gallons used during Warm-up, Take-off, and Climb, 38 miles Horizontal Distance on Climb to 25,000 feet, range 156 miles

Note lack of fuel reserves. AFHRA Reel A2069 Page 1621 dated 20 November 1942, at 16,000 feet, calculations, not flight tests.
P-39N, 120 gallons internal fuel, 75% power, 331 mph TAS, range 600 miles
P-39N, 120 gallons internal fuel, 50% power, 267 mph TAS, range 760 miles

 

[MikeMeech]

Are you a supporter of the Defiant contract then, or do you see that the concept was critically flawed and a waste of important effort?
The points you raise are simplistic ones. Despite the apparent bonus of a four-gun turret against single guns positions, the multiple bomber aircraft out gunned the turret fighter. Furthermore, aiming the unstabilised turret was difficult and the Defiant had an 80% restricted field of fire! The whole concept of flying in formation with a group of armed bombers and slugging it out was wrong. Remember, "The Bomber will always get through"? That was another fine example of muddled thinking.
The great advantage of a high performance agile fighter with heavy forward firing armament was that it could attack with manoeuvre and fast closure that made aiming defensive guns against it difficult, an important fact that the RAF almost missed and some argued against.

Eng

Hi
Why should I be a "supporter" of the Defiant contract or not come to that? After all only two RAF squadrons were equipped with them and they were really only used briefly in daylight during the BoB, however, the contract with BP was not going suddenly to end in mid-1940 for the obvious and well known reasons of the situation at the time. Equipment that was due replacement was kept in production as it was not known if there was going to be a 'free' 1941! Decisions and ideas of the 1930s could only be based on the experience of WW1 where Britain had to deal with German 'terror bombing' attacks on London and other places by airships and large bombers, and the fact that technology was changing rapidly during the 1930s and the 'powers that be' had to find ways of overcoming and using it. You really need to read Sinnott's book, including this little extract:

The Hampden was armed like the German bombers of the period and did not have power turrets, this tactic would not have worked with Wellingtons. However, in this we have the 110 using its rear gun with 109s waiting to swoop if the formation broke up. The Defiant was not going to be used in isolation there were also going to be SS fighters involved in attacking the bomber formations. Would it have worked if the Germans could not use their SS fighters for escorts, I don't know.
Baldwin stated that the bomber would always get through in November 1932, things changed by 1940, however, bombers still did get through, despite losses, there were enough bombsites in the UK and Europe by the end of the war to prove that point, not to mention graves.
It is also of interest to look at the technology of detecting aircraft during the inter-war period and the problems that arose with increasing bomber speeds, this book is useful for that:

All I want to get across is that everyone should look at the contexts during the inter-war period before condemning this or that decision, they had to try and foresee the problems and how the technology may or may not progress. They did not have hindsight, even in 'What if?' scenarios you have to base it on the reality of the period.

Mike

 

[Shortround6]

All I want to get across is that everyone should look at the contexts during the inter-war period before condemning this or that decision, they had to try and foresee the problems and how the technology may or may not progress. They did not have hindsight, even in 'What if?' scenarios you have to base it on the reality of the period.

This true but sometimes (a lot of times?) there were conflicting points of view at the time. Only time showed who was right and who was wrong.
You also have instances of one side not talking to the other side. The Fairley Battle, Blenheim and Hampden take a lot of explaining when it comes to the defensive armament. OR none, they just repeated the armament they were using in DH bombers of WW I 20 years later with a few small upgrades. Using a fixed forward firing gun for forward defense in planes that were supposed to fly in formation take some explaining, if anybody thought to ask. At times this was brushed off by noting how maneuverable the lighter British bombers were, but if you don't allow them to maneuver it takes a mighty unlucky enemy to fly directly in front of a fixed gun. The Battle has some excuse, where do you put a flexible gun in a single engine plane firing forward? Gets a lot harder answer for the Blenheim.
The Hampden really needs the excuse machine to be revved up to full throttle. While it may have looked OK compared to some foreign bombers it was well behind British bombers of it's time. The fixed gun in the nose, the single Vickers K guns above and below the fuselage that were manually aimed? At least the Blenheim had a powered mount that offered a decent amount of traverse. They later double the number of guns fitted but didn't do much for the covered arcs.

This arrangement for the flexible forward gun might not have instilled much confidence unless you could get the German fighters to co-operate.
Not a big surprise that the British shifted over to night bombing. 10-15mph edge over the early Wellingtons was not enough to show much improvement in losses.
What was known in 1938? Not much but what could be guessed at or figured out with some exercises?
We may be able to excuse the Wellingtons but at least they had power turrets with the hoped for 4-5 times improvement accuracy/effectiveness.

There were budget problems and production problems, but many late 1930s British bombers were flying with fewer guns than WW I bombers did. Granted the Vickers K fired faster than a Lewis gun but in 1938 they weren't shooting at linen covered wood framed fighters anymore.
The British were calling for 8 fast firing machineguns, The French were calling for a 20mm gun and 2 or 4 machine guns. Germans were putting 3-4 machine guns in the 700hp 109s they sent to Spain. What were they going to put in the DB 601 powered versions?

For some reason some people in the Air Ministry had decided that fighter pilots could not react quick enough at high speeds to aim effectively. British decided not to even try and came up with turret fighter, the zero allowance gun layout (amount of tilt?) and the "shotgun" (my name, aim each of the eight guns to a different impact point relative to the sight/s so that at least one gun had a chance of hitting). Maybe they were trying to improve actual gunnery at the same time for a 4th option?
See this web page for a short look at zero allowance shooting and the amount of effort/design work that went into it.

A description of no allowance sighting and shooting.

dingeraviation.net

I am sure that somewhere, somebody was trying to improve the normal gun sights. How much did this zero allowance 'stuff' divert resources?
A problem with zero allowance is that restricts the attackers to certain attack profiles. When it works it may work well, but the distances are off, or the speeds of the target and the firing aircraft are not closely matched it doesn't. Getting several "ranging" bursts while the target does nothing in response may not happen all that often either.

 

[Engineman]

Hi
Why should I be a "supporter" of the Defiant contract or not come to that? After all only two RAF squadrons were equipped with them and they were really only used briefly in daylight during the BoB, however, the contract with BP was not going suddenly to end in mid-1940 for the obvious and well known reasons of the situation at the time. Equipment that was due replacement was kept in production as it was not known if there was going to be a 'free' 1941! Decisions and ideas of the 1930s could only be based on the experience of WW1 where Britain had to deal with German 'terror bombing' attacks on London and other places by airships and large bombers, and the fact that technology was changing rapidly during the 1930s and the 'powers that be' had to find ways of overcoming and using it. You really need to read Sinnott's book, including this little extract:
View attachment 880604
The Hampden was armed like the German bombers of the period and did not have power turrets, this tactic would not have worked with Wellingtons. However, in this we have the 110 using its rear gun with 109s waiting to swoop if the formation broke up. The Defiant was not going to be used in isolation there were also going to be SS fighters involved in attacking the bomber formations. Would it have worked if the Germans could not use their SS fighters for escorts, I don't know.
Baldwin stated that the bomber would always get through in November 1932, things changed by 1940, however, bombers still did get through, despite losses, there were enough bombsites in the UK and Europe by the end of the war to prove that point, not to mention graves.
It is also of interest to look at the technology of detecting aircraft during the inter-war period and the problems that arose with increasing bomber speeds, this book is useful for that:
View attachment 880605
All I want to get across is that everyone should look at the contexts during the inter-war period before condemning this or that decision, they had to try and foresee the problems and how the technology may or may not progress. They did not have hindsight, even in 'What if?' scenarios you have to base it on the reality of the period.

Mike

Well, you can see where the Hampden crews went wrong in that story.
You do not need hindsight to realise that the turret fighter concept and the mass order of the Defiant was a waste of critical resources, supported by some who should have known better. Unfortunately, I do not feel that a modern opinion of "it could have been a good idea, but Hey, it didn't work!" is satisfactory when the real modern analysis should be, "it was a misguided idea, based on an incorrect understanding of airborne gun-fighting reality at that time".
I think that almost exclusively in WW2, turret or remotely aimed aircraft guns were only ever of significant use defensively.

Cheers

Eng

 

[MiTasol]

No need.
Just Stick the radiators and oil coolers out under the wing or in the wing leading edge Fairley Firefly style
View attachment 880592
Adjust the wing forward or aft to suit center of gravity and stick the carb intake out the bottom. Stick good filter on it to stop FOD from the front wheel.
Never let reality spoil a good "What if"

Except that the Merlin intake would be in the area of maximum amount of debris thrown up off the runway by the prop so unless you had an excellent intake air filtration system the aircraft would not be able to operate off wet or Marston matting or coral or ....

Engine failures and other defects on RAAF Spitfires operating off coral was a big problem.

 

[Jugman]

I hardly think so. The big trade off, especially for a fighter, was DRAG. Think about what it took to get air through that air-to-air intercooler. The liquid intercooler/aftercooler just had to be smaller than the air-to-air one. The other big and largely underappreciated factor was temperature control. Very easy to do with liquid cooling. Challenging with air-to-air. Undercooling was the problem with the P-38's up through the H model, which used leading edge intercoolers, which were great for drag but a challenge for manufacture and inadequate for cooling. The H used the same engines as the J but had less power available. Then with the beard type intercoolers on the J and L overcooling was the major problem, especially in the ETO.

I agree that liquid after/intercoolers can offer lower drag but this is do mostly to packaging flexibility. (A great example would be the XP-39. A liquid intercooler could have eliminated the bulge on the port side of the fuselage. Add in an improved waste gate setup and the XP-39 could have been a realistic 400mph fighter. Well it will make its poor longitudinal handling slightly worse.) Yes water is more efficient but you still need to get the waste heat into the atmosphere. That's the bottle neck. You pretty much need the same amount cooling air. If you can bury your heat exchanger, the drag differences boil down to the drag from the extra weight vs the extra cooling drag. In practice there are other factors to consider as well such as cost and availability.

It is interesting to me that P&W independently came up with the same supercharger and aftercooler configuration at the time as RR but chose not to pursue liquid aftercooling for their R-2800 aftercooler and R-4360VDT projects. Nor did any aircraft manufacturer choose liquid over air late or post war. The copper-brass fin-tube aftercooler core of a Packard Merlin weighed in at ~35-40lb dry and the aluminum housing add ~25-30lb. By comparison the copper fin-plate intercooler core of the F4U-1 had an average weight of 62lb and maximum of 74. The aluminum core version weighed 26lb. The later aluminum Airesearch charge-through tube intercooler for the B-17 weighed ~32lb. The original brass Harrison fin-plate type probably weighed around 80-100lb. Radiator dry weight should be ~60-80lbs for a copper-brass core construction and an aluminum one should weigh ~30-50lb.

For the RR power plant concept the weight should be ~125-200lb. wet. For a remote installation like the P-51 add 15lb for piping and extra coolant. I don't have the exact dimensions of the P-38's D shaped charge across tube intercooler but its frontal area was ~ 220-230sq.in. and length was less than 15in. If constructed from copper or brass weight should be ~110-135. If made from aluminum ~35-50lbs. The inlet area to supply this intercooler was only 80sqin. The P-51's aftercooler radiator core had an area ~150sq.in. and a depth of ~12-13in. The Packard aftercooler core had a volume ~900cu.in.(8.5X8.8x12) The Airesearch core for the B-17 had a volume ~2820cu.in. (10.25x20.75x13.25)and Harrison core 2555cu.in. (9.5x20.375x13.2)

Ultimately I think liquid intercoolers weren't a good tradeoff before ~1940 given relative engine power versus airframe performance. This is especially true for large multi-engine aircraft were powerplant drag is lower proportion of total drag. Realistically if liquid-cooling couldn't dominate the best case scenario (primary coolant), I don't see how it would the preferred choice for most applications.

On the P-38 intercoolers. From "The Alisson Time Bomb"
"While this story is about the major midwar crisis that confronted the V-1710 it should be clear that there
were three problems. One was resolved by the new intake gas pipe, the second by redesign and proper rigging of the P-38J turbo controls, and the third by introducing pressurized coolant, fuel and improved ignition systems as fitted on the P-38L"

Did the use of an air intercooler make the situation worse? Probably a little. Considering everything else was having some problems with spark plug fouling at this time, I don't see how the intercoolers can be blamed. If the P-38 had liquid intercoolers they would likely need to mount the radiators in the chin to minimize drag.

​

F4F-3​

F4F-3 2​

F4F-4​

F4F-4 2​

F4U-1​

F4U-1D​

F4U-4​

F4U-5​

F6F-3​

F6F-5​

F6F-3​

YP-38​

P-38J​

P-47D​

P-47D-25​

P-51B/C​

P-63A-10​

P-47B-Cry​

P-60A​

P-60A-Cry​

Engine Section​

294​

298​

335​

335​

307.4​

308.2​

359.7​

485.1​

411.6​

426.2​

404.9​

445.3​

471​

379​

383​

274​

​

425​

461​

425​

Engine​

1565​

1540​

1568​

1568​

2459.2​

2476.5​

2537​

2697.5​

2469​

2455​

2444​

2536.7​

2730​

2265​

2283.2​

1670​

1620​

2000​

1325​

2000​

Accessories​

242​

242​

242​

242​

278.9​

259​

285.5​

291.3​

314​

318​

318.3​

661.2​

321.3​

954​

977​

​

105.3​

960​

639​

960​

Engine Controls​

22.5​

23​

25​

25​

45.1​

49.2​

60.7​

77.9​

36.5​

30.8​

30.8​

81​

80.8​

57.8​

57.8​

30​

40.7​

71​

35​

35​

Propeller​

314​

312​

315.5​

315.5​

493.7​

490.9​

644.6​

645.4​

485.2​

481.4​

480.4​

692.5​

827.3​

541.3​

659.3​

483​

460​

575​

527​

575​

Starting System​

43​

43​

43​

43​

60.8​

39.6​

29.7​

37.3​

57​

43.4​

64​

81​

82.3​

56​

56​

25​

57.6​

69​

50​

69​

Lubricating System​

30​

30​

35​

35​

143.2​

128.6​

175.8​

144​

144.3​

148.5​

150.8​

161.9​

194.1​

117.9​

138.1​

101​

134.9​

75​

131​

180​

Cooling​

​

​

​

​

​

​

​

​

​

​

​

863.6​

1065.1​

​

​

663​

347.4​

450​

332​

495​

Water Injection​

​

​

​

​

​

37​

36.3​

54.9​

​

41.5​

46​

​

​

​

​

​

50​

​

​

​

Supercharging​

​

​

​

​

​

​

​

​

​

​

​

​

613.5​

​

​

​

​

​

​

​

Sub Total​

2510.5​

2488​

2563.5​

2563.5​

3788.3​

3789​

4129.3​

4433.4​

3917.6​

3944.8​

3939.2​

5523.2​

6385.4​

4371​

4554.4​

3246​

2815.9​

4625​

3500​

4739​

Fuel​

233.5​

242​

264.5​

264.5​

286.8​

266.9​

255.3​

270.1​

458.5​

467.5​

429.9​

161.9​

505.8​

332.3​

422.2​

320​

342.7​

330​

267​

267​

Total​

2744​

2730​

2828​

2828​

4075.1​

4055.9​

4384.6​

4703.5​

4376.1​

4412.3​

4369.1​

5685.1​

6891.2​

4703.3​

4976.6​

3566​

3158.6​

4955​

3767​

5006​

Notes
weights in pounds
The P-51 accessories are included with the engine. I suspect the aftercooler unit is included with the cooling group
The P-63 engine section group is accounted for in the fuselage group.
The same goes for the wrapper shrouds and belly cowling P-47. Total weight of these is 220-228lb
The IV-2220 powered P-47B had an air intercooler and a liquid aftercooler.

I don't have figures for the P-61 but power plant weight is likely in the same ball park as the F4U and F6F per engine. The P-61C likely weighs 300-500lbs more per engine. The very rough breakdown is this:
CH-5 +330lb this includes the cooling cap, oil system, oil, and type B-3 controls
propellers +30-50lbs
engine -90-190lbs
Turbo snout and other changes +50-200lb