British bomber development in regards to long range fighters.

[XBe02Drvr]

The NAA engineers carried out much research through the RAE at Farnborough and NACA, including the oft mentioned Meredith Effect, so it was no happy coincidence.

This makes the "drawings on a napkin to first flight in some ridiculously short (90 days?) time frame" mythology look rather dubious. Unless they were already into the research and this was just a fortuitous opportunity to try some of these things out?
Cheers,
Wes

 

[gingerbob]

The question I have is, did all the design choices... result from a coherent attempt to build a better P40, or were they a string of happy coincidences, where one choice more or less led to the next?
Also, how much American input was there to the basic design principles, (other than pulling them together into a single package) vs concepts developed from British R&D?

I suppose in aircraft design one choice always "more or less [leads] to the next". Or, put another way, there's seldom such a thing as making a simple choice in isolation. In my opinion, the Mustang represented an attempt at a state-of-the-art single-engine fighter. The P-40 wasn't a bad place to start, representing the prior (or current production) "state of the art", especially considering that using the Allison- though a new model- was a given. Bear in mind that Curtiss was doing the same thing, with the XP-46. (I see those two more in parallel than Curtiss' attempt influencing North American's, which is one reason NAA wasn't particularly interested in Curtiss data.) As for the XP-46, I think it was a wise decision for Curtiss to abandon that and go back to "building a better P-40".

Reducing drag, through airfoil selection and cooling installation, was highly influenced by NACA research, and NACA people talked directly with NAA people as the Mustang was being designed. Remember too, though, that prior to the war, the aerodynamic research was shared throughout the world, so German, British, US, and perhaps other research all got stirred into the pot. That NAA, like everyone else, still had a lot to learn can be seen in the ongoing cooling-system development, which changed significantly in the Allison models, more for the Merlin models, and was still being tweaked for the "lightweight" Mustang/ P-82. And I won't even mention that ridiculous canopy design...

It is also important to remember that NAA was designing a "general purpose" fighter, not a "bomber escort". The happy coincidence was the realization that what made a good general purpose fighter also could make for a good escort fighter. However, lest we be TOO critical of those steering the development of bomber escorts, the rapid growth in horsepower- and to a degree in airframe size- also allowed the newer fighters to carry more disposable load than was the case earlier.

 

[nuuumannn]

Unless they were already into the research and this was just a fortuitous opportunity to try some of these things out?

No, NAA had been researching such things for awhile.

Reducing drag, through airfoil selection and cooling installation, was highly influenced by NACA research

Let's not forget the Royal Aircraft Establishment at Farnborough, from whence the Meredith Effect was researched, after RAE aerodynamicist F.W. Meredith.

Meredith effect - Wikipedia

The whole back-of-a-napkin thing shows that the Mustang was subject to a dose of good ole propaganda to make a good story, like the designer versus the sceptical Air Ministry for a high speed unarmed bomber in the development of the Mosquito.

 

[Shortround6]

Never let the truth get in the way of a good story

 

[Zipper730]

If I'm to understand you the P51 BCD models could takeoff without drop tanks but with a full fuselage tank? I'm trying to determine fuels effects on C/G, or allowances.

I could be wrong here, but I don't think the drop-tanks had any real effects on CG. The reason they didn't have a full fuselage tanks unless they were carrying drop-tanks had to do with the fact that the center-tanks had the CG up on the aft limit.

Few would want to fly a plane that way, so most people from the air-group commanders on down to the pilots would rather use the drop-tanks to stretch range unless the range demands really needed it.

If I recall, to make room for the center-tank, they had to reposition some radio-equipment.

Actually, the "bombing the whoever into submission" was integral part of the Soviet military doctrine for quite long time. Massive fleet of TB-3 have been built not just for parades. Strategic bombing force of the USSR has been probably the largest in the world until the WWII.

I didn't know about the scale of their bomber-force, but they definitely supported the teachings of Giulio Douhet: The problem was the Soviet military was trained by Trotsky, and Stalin was a malignant narcissist (a terrifying combination of narcissist, psychopath, paranoid, and sadist all in one) who figured he'd be better off just rubbing them all out, rather than risk any of them turning on him.

The people who remained were either mostly proponents of battlefield support, or Stalin was and none of them felt like getting their brains blown out.

You know Wes, you could make a whole new thread on that last question alone! There are quite a few threads on this site that will answer those prickly Mustang questions for you. But to summarise, the airframe design was pretty much all NAA; no British input (apart from separate papers acquired from Farnborough) at all, except that it was to have an Allison engine in common with the P-40. The NAA engineers carried out much research through the RAE at Farnborough and NACA, including the oft mentioned Meredith Effect, so it was no happy coincidence.

I didn't know they needed research at the RAE to help design the radiator correctly, but from what I remember: They were working on fighter concepts as early as 1935 (it was a two-seat fighter along the lines of the PB-2), and around 1939-1940 they were working on a fighter design.

I would assume they'd have loved to sell it to the USAAC, but regardless of their exact interests, they were definitely planning to export it. When the British came along, they'd just as well develop their own design over the P-40. The British were okay with this and proceeded along.

The original P-51's actually had a variable-area radiator that (if I recall correctly) extended out at low speeds where more airflow was needed, and retracted in for high-speed where sufficient airflow was present or cooling. It did around 382 mph, which was faster than the P-40 on the same engine; When the P-51A came along (the British might have inspired the radiator design), the earlier configuration (which had a small splitter to remove some of the turbulent flow) was replaced with a fixed geometry configuration that had a much larger splitter (probably a larger divergent angle for the radiator intake), and proved to be more effective (and mechanically simpler): Speed increased to around 409 mph.

 

[Dimlee]

I didn't know about the scale of their bomber-force, but they definitely supported the teachings of Giulio Douhet: The problem was the Soviet military was trained by Trotsky, and Stalin was a malignant narcissist (a terrifying combination of narcissist, psychopath, paranoid, and sadist all in one) who figured he'd be better off just rubbing them all out, rather than risk any of them turning on him.

The people who remained were either mostly proponents of battlefield support, or Stalin was and none of them felt like getting their brains blown out.

RKKA (Red Army) was created by Trotsky. He lost control over the army in 1925-1926. Internally exiled in 1928. Left USSR forever in 1929. When rapid growth of VVS (Air Force) began, he was forgotten and condemned.

Battlefield support strategy prevailed ultimately due to changes in Kremlin's political doctrines and priorities. You don't need to destroy another country's industry and infrastructure if you plan to use them for your own benefit - soon.

 

[Shortround6]

The original P-51's actually had a variable-area radiator that (if I recall correctly) extended out at low speeds where more airflow was needed, and retracted in for high-speed where sufficient airflow was present or cooling. It did around 382 mph, which was faster than the P-40 on the same engine; When the P-51A came along (the British might have inspired the radiator design), the earlier configuration (which had a small splitter to remove some of the turbulent flow) was replaced with a fixed geometry configuration that had a much larger splitter (probably a larger divergent angle for the radiator intake), and proved to be more effective (and mechanically simpler): Speed increased to around 409 mph.

Have you got some sources for this??

The P-51A was the 4th version of the Mustang (if one counts the A-36), 620 Mustang Is, 150 P-51/Mustang IAs, 500 A-36s and then 310 P-51A/Mustang IIs.
With over 1200 Mustang/Apaches built before the P-51A there should be some sort of pictures of this radiator arrangement you talks about.

Speeds also changed because the planes used different engines with different supercharger gears and different allowable maximum boosts.

 

[tomo pauk]

Have you got some sources for this??

The P-51A was the 4th version of the Mustang (if one counts the A-36), 620 Mustang Is, 150 P-51/Mustang IAs, 500 A-36s and then 310 P-51A/Mustang IIs.
With over 1200 Mustang/Apaches built before the P-51A there should be some sort of pictures of this radiator arrangement you talks about.

Initial arangement, sporting the 'droppable' intake lip was used from XP-51 to P-51. The A-36 lost the lip and gained boundary layer splitter. Oil radiator was 'surrounded' by coolant radiator.
Next iteration was used when P-51 got 'merlinized', the intake gaining even more prominent drop away from fuselage; size and shape of radiators also changed, so did the location of oil cooler; intercooler radiator was also added. Check out the pages 360-363 of the 'America's hundred thousand', as well as Gruenhagen, especially the pg. 79.
The 'lightweight Mustangs' used a further modified radiator set-up, as did the XP-51J.

Speeds also changed because the planes used different engines with different supercharger gears and different allowable maximum boosts.

Very true.

 

[Shortround6]

So the intake lip was adjustable. Not that the radiator itself was movable as many aircraft had resorted to in the 30s.

 

[Zipper730]

So the intake lip was adjustable. Not that the radiator itself was movable as many aircraft had resorted to in the 30s.

That's correct...

 

[Milosh]

 

[Reluctant Poster]

As noted there were problems with P-51's original duct intake. Ray Wagner quotes Edgar Schmued in his book "Mustang Designer":
"The British Air Ministry was extremely helpful. Among others they sent us Dr. B. S. Shenstone [who arrived February 25, 1941], to assist us in some of the airflow problems into the radiator. The radiator, as we had it, consisted primarily of a fairing, which started at the bottom of the fuselage and enclosed the radiator. Dr. Shenstone advised us to provide an upper lip on the radiator housing, which was about 1 ½ inches below the fuselage contour. By doing this, we got a much better pressure distribution in the air scope."
Beverly Shenstone was the chief aerodynamicist on the original Spitfire.
On the other hand the Ed Horkey, the P-51 aerodynamicist, gives the credit to Irving Ashkenas.

North American had problems again with the original Merlin P-51 (duct rumble) which was resolved by the distinctive slant to the intake.

The Mark III Spitfire was to have a boundary layer splitter but this was not implemented in later marks. The Me 109F had a splitter but this was abandoned in the 109G. Perhaps the complication was not worth it in the constricted space of the wings.

One of the points of contention in the P-51 story is the aerodynamic data North American purchased from Curtiss. According to some Curtiss partisans NA stole the P-51 design from Curtiss, while on the other side some North American protagonists claim the data was never looked at. I have always been intrigued by the attached NACA report on the aerodynamics of US aircraft cica Oct 1940. Figure 14 shows the radiator installation of Airplane 11, which must be the Curtiss P-46. The installation looks very similar to the original P-51. The entire report is an interesting read.

 

[KiwiBiggles]

As noted there were problems with P-51's original duct intake. Ray Wagner quotes Edgar Schmued in his book "Mustang Designer":
"The British Air Ministry s extremely helpful. Among others they sent us Dr. B. S. Shenstone [who arrived February 25, 1941], to assist us in some of the airflow problems into the radiator. The radiator, as we had it, consisted primarily of a fairing, which started at the bottom of the fuselage and enclosed the radiator. Dr. Shenstone advised us to provide an upper lip on the radiator housing, which was about 1 ½ inches below the fuselage contour. By doing this, we got a much better pressure distribution in the air scope. Beverly Shenstone was the chief aerodynamicist on the original Spitfire.
On the other hand the Ed Horkey, the P-51 aerodynamicist, gives the credit to Irving Ashkenas.

North American had problems again with the original Merlin P-51 (duct rumble) which was resolved by the distinctive slant to the intake.

The Mark III Spitfire was to have a boundary layer splitter but this was not implemented in latter marks. The Me 109F had a splitter but this abandoned in the 109G. Perhaps the complication was not worth it in the constricted space of the wings.

One of the points of contention in the P-51 story is the aerodynamic data North American purchased from Curtiss. According to some Curtiss partisans NA stole the P-51 design from Curtiss, while on the other side some North American protagonists claim the data was never looked at. I have always been intrigued by the attached NACA report on the aerodynamics of US aircraft cica Oct 1940. Figure 14 shows the radiator installation of Airplane 11, which must be the Curtiss P-46, it looks very similar to the original P-51. The entire report is an interesting read

Shenstone gets nowhere near the praise he deserves, in general. A lot of his best work was around the airflow at the fuselage/wingroot area, so the idea that he was the source of the redesigned P-51 radiator intake rings true to me.

 

[drgondog]

I will devote a couple of pages in the Appendix to the evolution of the Mustang cooling system. Before making any other comments, Shenstone was key to the first evolution from X73 to mid NA-73 config of lowering the intake scoop from the wing. The 'alligator jaw' requirement was strictly for low speed/climb/taxi cooling requirements. The radiators also evolved from square with oil cooler at the top of the Radiator with the lower 80% devoted to engine coolant to round with oil cooler matrix in the center - as the design progressed from P-509 to X73 to NA-73. The A36/P-51A were the first 'fixed' intake scoops - with an upper intake duct conforming to the fairing covering the CL at the wing attach region, and wing dihedral.

The XP-51B was a Huge change as the cooling requirements for both the engine and supercharger, including aftercooling the fuel-air charge temp due to the two speed/two stage Merlin supercharger. The R-R team placed a separate aftercooler under the engine, used the Mustang I Radiator/Oil cooler basic design but moved the oil cooler core in front of the Radiator and increased radiator size. The R-R team also experimented with greater scoop intake areas and exit gate adjustable areas.

The first major difference, aside from the Harrison Radiator with aftercooler core residing in the top ~ 1/4, was moving the oil cooler forward and then fairing over it to 'smooth out' the flow to the Harrison radiator. This design provided a 'divided duct'. The top surface of the oil cooler intake 'lip' was deviated from the top surface of the plenum extending to the radiator only a little bit before expanding after the back of the oil cooler. The forward upper lip of the oil cooler was about 6" aft of lower intake scoop lip. The front face of the intake scoop is still perpendicular to the airflow. Through the first several iterations at NACA did not appreciably alter the Plenum X-section

The NAA Horkey/Ashkenas team began with a modified P-51A scoop that was 'slightly bent' forming a shallow V conforming to the wing dihedral and close to the CL wing bolt fairing. The next iteration was to drop the upper scoop lower, and straighten it so that an increased gutter was created at the edges. The suspected root cause for the continued Rumble was believed to be caused by multiple boundary layer separations at both the upper and lower plenum

The first iteration of duct rumble 'solution' was to drop the upper lip and 'straighten' it. Improved but the duct rumble continued to be severe.

The next iteration was to extend the upper lip forward but not change the geometry of the Plenum, specifically the upper surface of the oil cooler duct. Still Heavy Rumble with exit gate in 'closed position but very little at medium speed and wide open exit gate.

The final two iterations for the B/D production design a.) moved the upper surface of the oil cooler duct all the way forward, and b.) provide upper oil cooler duct with small gradual change in slope further aft then opening up to intersect at the base of the Harrison radiator. This change resulted in minimal BL separation (as proved by Lednicer in 1999 paper that I put into this forum some 10 years ago), but BL separation still exits a little too far in front of the upper line of the Harrison radiator. Gruenhagen has the test table and images in pages 78-79 of his Mustang Book.

BTW as you may imagine, the process ultimately yielded the closest approach to achieve net zero internal pressure drag of the plenum/radiator system designed per Meredith Effect principles. Personally, I believe that net zero internal cooling drag was achieved at top speeds as Horkey and Ashkenas stated in the P-51B/D and H Performance Analysis Reports - Lednicer postulates ~ 85% recovery of Cooling Drag recover at M=.5 at 15,000 feet. That said, he did not run his CFD models for the .57M tests by NACA during the duct rumble testing for 430mph at SL.

 

[drgondog]

As noted there were problems with P-51's original duct intake. Ray Wagner quotes Edgar Schmued in his book "Mustang Designer":
"The British Air Ministry was extremely helpful. Among others they sent us Dr. B. S. Shenstone [who arrived February 25, 1941], to assist us in some of the airflow problems into the radiator. The radiator, as we had it, consisted primarily of a fairing, which started at the bottom of the fuselage and enclosed the radiator. Dr. Shenstone advised us to provide an upper lip on the radiator housing, which was about 1 ½ inches below the fuselage contour. By doing this, we got a much better pressure distribution in the air scope."

TRUE
On the other hand the Ed Horkey, the P-51 aerodynamicist, gives the credit to Irving Ashkenas.

Partially true for the final iterations to the intake, divided duct and change to radiator intake plenum geometry

North American had problems again with the original Merlin P-51 (duct rumble) which was resolved by the distinctive slant to the intake.

No. As may be observed and pondered upon for the XP-51F/G/J, P-51H and XP/F 82 - all perpendicular to airflow.

The Mark III Spitfire was to have a boundary layer splitter but this was not implemented in later marks. The Me 109F had a splitter but this was abandoned in the 109G. Perhaps the complication was not worth it in the constricted space of the wings.

The concept of Splitter to achieve less separated BL flow is important, but the internal planum/single radiator and oil cooler 'deeply' within the outer surface of the airframe, and exit plenum long enough to achieve desired exit flow properties are all necessary.

One of the points of contention in the P-51 story is the aerodynamic data North American purchased from Curtiss. According to some Curtiss partisans NA stole the P-51 design from Curtiss, while on the other side some North American protagonists claim the data was never looked at. I have always been intrigued by the attached NACA report on the aerodynamics of US aircraft cica Oct 1940. Figure 14 shows the radiator installation of Airplane 11, which must be the Curtiss P-46. The installation looks very similar to the original P-51. The entire report is an interesting read.

The Curtiss folks are simply wrong. The P-509 progenitor to the X73 per the 1620 and conceptually started with 'Meredith Effect' big ass radiator aft of the wing with extended plenum front and back in the mid 1939 timeframe, using funds from the cancelled NA-53 project. Objectively, one really needs to look at the airframes and compare the X-73 to either the XP-40 or XP-46 or XP-60, including plan views and radiator/plenum designs.

The absolute best proof is not just a physical examination and comparison - but look to the flight testing between the two ships.

As an aside, I know that I would have been interested enough to open the boxes of the XP-40 and 46 wind tunnel reports, as well as the data supplied for empennage design, wing fillets, engine cooling and other data requested by Atwood on April 11th from Wright. That said, you can not point to one thing in the XP-40 or both XP-46 airframes as 'same or nearly same' on the X73. The FINAL Design Release from Schmued's PD group was July 1941 and very closely followed the P-509 Side Profile drawing completed on 3/10/1940 and submitted to BPC on or about 3/18-20 by Atwood in NYC.

 

[gingerbob]

The other factor is that the real (potential) value of the Curtiss data for NAA would have been NACA analysis, but since NAA was already talking directly to NACA, they didn't need to get it via Curtiss. The deal was also a government-blessed payoff to Curtiss for the wasted effort of the XP-46.

 

[Reluctant Poster]

I certainly agree that Curtiss was not involved in the design of the P-51. The question is are the similarities in the radiator design an example of convergent evolution or do the designs share the same roots (ie NACA).

 

[stona]

"At least he was a realist. Whilst he eventually got his wish for a central gunnery school, and the RAF under Harris began to improve accuracy with the addition of advanced navigation and bombing aids - another of Ludlow-Hewitt's bugbears about the contemporary pre-war BC, he then added the following, which was, promptly ignored:

"Experience in China and Spain seems clearly to indicate that with the aircraft in use in these theatres of war at present , Fighter Escorts are considered absolutely essential for the protection of bomber aircraft. So far as I am aware this policy runs counter to the view long held by the Air Staff."

I don't know when Ludlow-Hewitt wrote that. I suspect he made that comment in the late summer of 1938 when he suggested a reconsideration of Air Ministry policy on escorts, though he admitted to not knowing what a suitable type might look like. He seems to have envisaged a bomber force operating out of France (as in WWI) and aided by heavily armed escorts.

He may have been referring to a 1938 report by Wing Commander Goddard, Chairman of the Joint Intelligence Sub-Committee on Spain, which was reprinted and circulated again in 1939. He had been invited on a fact fighting mission to Spain by the Republican War Minister. Goddard reported, "The escort of bomber formations proceeding to and from their objective by double, or more than double, their number of fighters, has been found by both sides to be a necessity, notwithstanding the ability of the bomber to shoot down fighters." Goddard did qualify this bold statement pointing out that bombers could defend themselves with speed and their forward and rearward gun defences, "Bomber crews were confident in their ability to bring down fighters and many successes by bombers were claimed."
Most British observers in Spain argues that the short ranges of the missions meant that such 'lessons' from Spain did not apply.

The reason that the British conception of an escort tended towards a heavily armed bomber rather than a fighter was because, in on going discussions through the 1930s, they could not see any way an escort fighter could be made fast enough, manoeuvrable enough or sufficiently long ranged to accompany bombers on deep penetration raids into Germany.

Cheers

Steve

 

[nuuumannn]

I don't know when Ludlow-Hewitt wrote that. I suspect he made that comment in the late summer of 1938 when he suggested a reconsideration of Air Ministry policy on escorts, though he admitted to not knowing what a suitable type might look like.

According to Max Hastings in Bomber Command, it was "a few months before the outbreak of war..."

 

[stona]

According to Max Hastings in Bomber Command, it was "a few months before the outbreak of war..."

I think this would have been in July 1939, certainly after he would have read Goddard's report. This was around the time he told the Air Ministry that if Bomber Command did carry out its declared intention to launch a counteroffensive against German aggression his entire fleet of medium bombers would be lost in less than a month, and the heavy bombers (by the then current definition) in less than two! I bet that went down well. It led to the severe restrictions on British bombing imposed by Newall a few weeks before the war began.

To be fair to Ludlow-Hewitt he did harbour serious doubts about the self defending bomber from the late 1930s, that came to a head in 1940 when he strongly opposed a plan to attack the Ruhr's power plants on the grounds that the losses would be catastrophic.

He wasn't exactly a Cassandra, others did listen to him and discussions on escort aircraft (not specifically escort fighters) came up regularly, like a hardy perennial as one writer (can't remember who) described them. The problem was as I wrote above, nobody knew how such an aircraft might be produced.

Cheers

Steve