In retrospect, were the BMW radial engine developments a mistake?

[z42]

So the RLM(?) decreed in the early-mid 30'ies that BMW should drop work on liquid cooled aero engines and instead focus on radials. In a way a sensible decision from the perspective of not putting all the eggs in one basked, considering Daimler and Junkers were already focused on inlines. But what if they had decided the other way around, to go all-in on inlines? As an aside, the Italians made the opposite choice, decreeing in the 30'ies that all future aircraft were to be designed with radials, and as a result inline R&D was put on hold for several critical years. Later on when they realized that inlines could have something to offer, they were far behind and didn't manage to produce a good inline engine during the war.

I'm not arguing this from a "radials are bad, boo!" perspective, but more from the perspective of rationalizing limited R&D resources. So as a result of this decree and BMW realizing they were starting from zero wrt radials, they licensed the P&W Hornet to kickstart their development.

Eventually they produced the 801, which was in many ways an excellent engine, and Kurt Tank's use of it in the FW190A with the low drag installation was genius. But if the 801 isn't available, then Kurt Tank would design the 190 from the get-go with an inline, as he eventually wound up doing anyway with the D series and the Ta-152.

However, like most new engines, the 801 had a rough start and it took a while before they got it running reliably. And ultimately it was limited by poor supercharging for high altitude work, and all the various experiments with multi-stage supercharging, turbocharging etc. didn't enter use. And similarly, none of the follow-on work on bigger radials (802/803/...) worked out either. So in a way it was a huge amount of resources on radial R&D that ultimately resulted in only one good design produced in numbers. And even that one had a relatively short lifetime due to the reliability issues when it was first introduced, to being superseded by the inline powered FW190D only a couple years later.

From a fuel supply perspective the 801 was also tricky, requiring C3 to develop decent power. And as C3 took much more resources to produce than B4, it further exacerbated Germany's critical situation wrt aviation fuel.

I think the critical question is, what would the resources not being spent on radial R&D have been used for instead? Turbojets, though it doesn't seem realistic to decree in the mid-30'ies that BMW should drop piston aviation engines to focus on jets? Auxiliaries, like supercharger and turbocharger development for Daimler/Junkers inlines? Kommandogerät for DB & Jumo engines? Move staff over to Daimler and Junkers to develop and produce their designs? Or license produce Daimler and/or Junkers inlines? Or would German aircraft designers have three families of inlines to choose from, if BMW would have developed their own inline engines?

 

[tomo pauk]

Eventually they produced the 801, which was in many ways an excellent engine, and Kurt Tank's use of it in the FW190A with the low drag installation was genius. But if the 801 isn't available, then Kurt Tank would design the 190 from the get-go with an inline, as he eventually wound up doing anyway with the D series and the Ta-152.

For the alt-Fw190 to materialize with a V12 in the nose, there needs an understanding in the RLM that there will be a suprplus of the V12s in the very near future, otherwise the FW fighter project has a good chance to be axed. So we'd need for BMW - they are big enough to pull this one off, especially after the Bramo was bought - to make a competitive engine (a good V12 for this scenario).

However, like most new engines, the 801 had a rough start and it took a while before they got it running reliably. And ultimately it was limited by poor supercharging for high altitude work, and all the various experiments with multi-stage supercharging, turbocharging etc. didn't enter use. And similarly, none of the follow-on work on bigger radials (802/803/...) worked out either. So in a way it was a huge amount of resources on radial R&D that ultimately resulted in only one good design produced in numbers. And even that one had a relatively short lifetime due to the reliability issues when it was first introduced, to being superseded by the inline powered FW190D only a couple years later.

Fw 190A was not superseded by the 109D after only a couple of years, but after more than 4 years. That is more than USA or USSR were directly involved in the ww2.
One good design > a few mediocree designs, IMO. Poor supercharging was just partially connected with the fact that BMW 801 was a radial engine. Jumping on the 801E by early 1943 would've improved things at high altitudes.
German engine companies, on aggregate, were rather blind to the fact that 2-stage superchargers are that better. BMW came up with solution to the corroding exhaust valves (due to lack of nickel; they started chrome-plating them, it took DB a full year to do it, but that delay is probably to blame on the RLM technical department).
Instead of the follow-up radials, I'd have BMW start work on jet engines.

From a fuel supply perspective the 801 was also tricky, requiring C3 to develop decent power. And as C3 took much more resources to produce than B4, it further exacerbated Germany's critical situation wrt aviation fuel.

Good call on the BMW 801 needing high-oct fuel. It was also very heavy, and expensive to produce (at least wrt. cost in manhours), even when compared with Hercules with it's sleeve valves.

I think the critical question is, what would the resources not being spent on radial R&D have been used for instead? Turbojets, though it doesn't seem realistic to decree in the mid-30'ies that BMW should drop piston aviation engines to focus on jets? Auxiliaries, like supercharger and turbocharger development for Daimler/Junkers inlines? Kommandogerät for DB & Jumo engines? Move staff over to Daimler and Junkers to develop and produce their designs? Or license produce Daimler and/or Junkers inlines? Or would German aircraft designers have three families of inlines to choose from, if BMW would have developed their own inline engines?

Me, I'd go with the a big V12 instead of the BMW 139/801, talk 40-45 liters. As a follow-up, start thinking about the jet engines.

 

[z42]

For the alt-Fw190 to materialize with a V12 in the nose, there needs an understanding in the RLM that there will be a suprplus of the V12s in the very near future, otherwise the FW fighter project has a good chance to be axed. So we'd need for BMW - they are big enough to pull this one off, especially after the Bramo was bought - to make a competitive engine (a good V12 for this scenario).

Me, I'd go with the a big V12 instead of the BMW 139/801, talk 40-45 liters. As a follow-up, start thinking about the jet engines.

Yes, absolutely. In this scenario, the 60000+ 801's produced during the war would have needed to be replaced with inlines, be it increased production of DB or Jumo ones, or new ones from BMW. As for a hypothetical 40-45L BMW V-12, yes, that makes sense. The ~35L category was already pretty crowded with DB 601/605 and Jumo 211/213, so it would have made sense for BMW to aim at a slightly different market segment.

Fw 190A was not superseded by the 109D after only a couple of years, but after more than 4 years.

My argument is that while the 190A entered service in August 1941, it took a while until the issues with the engine were resolved, and it wasn't until maybe 1942 it really hit its prime. And by early 1944 it was already no longer on par with the latest Allied fighters.

One good design > a few mediocree designs, IMO.

Indeed. So here the argument is that BMW had good in-house knowledge of inline aero engines going back to WWI, and of course they were also developing inline auto engines. I think they would have had a better chance of producing a good inline engine than by starting from scratch with radials. The UK and the USA, being the most industrialized and prosperous nations on the planet, could well afford to develop both radial and inline engines. But a post-Versailles Germany, suffering from the devastation caused by economical collapse of the Weimar republic, hyper-inflation and all that, and having grand plans for a quick reindustrialization and rearmament, would have needed to be much more focused in its R&D efforts.

Take for instance the BMW VI. A V-12 producing 550/750 hp (max continuous/takeoff) in 1926, with a weight of 510 kg. But they threw it away, licensed the P&W radial and produced an improved version, the BMW 132. Which admittedly was a successful engine, powering the Ju-52 among others. But in a time when aviation technology was developing rapidly, the 132 produced about as much power at about the same weight as the VI. So they had spent 7 years taking what, in terms of power/weight and max power, was a big sideways step rather than forward.

So what if they instead of going down the radial route had doubled down on developing the VI platform? Admittedly an inline engined Ju-52 would be pretty funny looking! Though the VI was a large volume engine, 47L, in the size class we're interested in, I don't think it could be adapted for the power levels required for a mid-late WWII fighter. They could have made another engine with the same volume (perhaps reusing some parts and tooling?) but much sturdier and made for high boost ratio, 4V/cylinder, fuel injection, higher rpm, and all that jazz.

And yes, then concentrate on improving that basic engine, and not flail away with a dozen unrelated concepts, except maybe for jet engines.

 

[tomo pauk]

My argument is that while the 190A entered service in August 1941, it took a while until the issues with the engine were resolved, and it wasn't until maybe 1942 it really hit its prime. And by early 1944 it was already no longer on par with the latest Allied fighters.

Okay, roger that.

Take for instance the BMW VI. A V-12 producing 550/750 hp (max continuous/takeoff) in 1926, with a weight of 510 kg. But they threw it away, licensed the P&W radial and produced an improved version, the BMW 132. Which admittedly was a successful engine, powering the Ju-52 among others. But in a time when aviation technology was developing rapidly, the 132 produced about as much power at about the same weight as the VI. So they had spent 7 years taking what, in terms of power/weight and max power, was a big sideways step rather than forward.

Truth to be told, the latest 132s were capable for 1000 HP (and for another 100 on high-oct fuel). But at any way, by 1939 the BMW 132 was 4th best German engine (at least the Bramo 323 have had a 2-speed supercharger already by 1938), and something much better is needed.

So what if they instead of going down the radial route had doubled down on developing the VI platform? Admittedly an inline engined Ju-52 would be pretty funny looking! Though the VI was a large volume engine, 47L, in the size class we're interested in, I don't think it could be adapted for the power levels required for a mid-late WWII fighter. They could have made another engine with the same volume (perhaps reusing some parts and tooling?) but much sturdier and made for high boost ratio, 4V/cylinder, fuel injection, higher rpm, and all that jazz.

We can take a look on what Mikulin's team did.
While looking at the legacy V12 engine (licence produced BMW VI), they developed modern, big V12s with AM-35 and -38 (not that these were perfect, but still). At the end of the day, BMW would've emulated that, topping the range with the AM-39 equivalent. With no 'radial excursion', it is not a long shot that their 'BMW-35' would've been available for the BoB...

 

[Shortround6]

I get a little lost when people talk about "in house Knowledge" of a particular type of engine.

This thing was two WW I cylinder banks mounted on a common crankcase. Pretty up to date in 1918.
In 1933????
Germany had been prohibited or severely restricted in aircraft and and aircraft engine development during the 1920s.
The Russians had spent a number of years fighting each other in the 1920s and destroying what little infrastructure they did have. It took them several years in the late 20s/early 30s just to get back to 1917.

There are some physical limits to big cylinders.
A major one is flame speed. How fast the flame front moves across the top of piston. Doesn't matter what kind of fuel you use or the shape of the cylinder head (within reason) you need a certain amount of time for the flame front to travel a certain distance. This is one of the big factors in how much RPM you can use with a big cylinder. If you use a high rpm the combustion is taking place with the piston lower in the cylinder, wasting some of the power and venting the gases into the exhaust before you have 'captured' the work.

The vast majority of aircraft engines used fixed ignition timing unlike cars, boats, motorcycles, etc.

The BMW VI peaked at 1650rpm?

Stress in the reciprocating and rotating parts go up with the square of the speed.

AM-35 gained 300kg and ran at 2050rpm.

27 liter engine running at 3000rpm or 47.7 liter engine running at 2050rpm?
Or 36.7 liter engine running at 2750rpm?

 

[z42]

Obviously the VI wouldn't cut it as a fighter engine in WWII. But, as a 1926 design it was on par with their 1933 radial in power/weight. I'm suggesting they might have been better off developing a series of increasingly powerful 47L V-12's rather than pivoting to radials.

As for big cylinders going fast, the 35L Jumo 213J reached 3700 rpm. If we take that as an upper limit wrt flame propagation, a 47L engine should be able to reach substantially higher than the 2050rpm of the Mikulin.

 

[ThomasP]

I have always liked the idea (from a manufacturing efficiency & continuity viewpoint) of having 2 (or more) companies producing the same or closely related system. An example of this is the multi-plant production of aircraft like the P-51 or P-47. A particular airframe variant could be produced at one plant, while the next variant could be introduced into production at the second plant - keeping losses in production to a minimum.

What if BMW produced a variant of the DB601 initially, then switched to the DB605 - allowing DB to allocate more resources to other engines like the DB603? Not that DB would stop producing the DB605, just maybe not as many. Or maybe it would be better for BMW to act in such a manner for the Junkers Jumo engines?

 

[Engineman]

I think the premise that the "BMW 801" was a bad use of resource and a failure is not the whole story.
Looking only at the results, the engine was proposed from about 1939 and had a short development before production. There seems little doubt that the engine was underdeveloped before its introduction to service. However, in the Fw 190 it gave the RAF and the Allies a big problem, right from the start. The initial development timescale was too optimistic.
The 801 series became far better over a few years and it was, IMO, a very successful engine, becoming the saviour of many aircraft programmes.

Eng

 

[Engineman]

Kommandogerät for DB & Jumo engines?

There is a common misconception that BMW had a monopoly of ability with their engine control systems. However, DB and Junkers had their own automatic systems.
In fact, Daimler-Benz chose to keep the elements of their automatic engine controls as separate units on some engines too keep the complexity down and keep serviceability
and maintainability higher in service.

Eng

 

[tomo pauk]

The BMW VI peaked at 1650rpm?

Stress in the reciprocating and rotating parts go up with the square of the speed.

AM-35 gained 300kg and ran at 2050rpm.

27 liter engine running at 3000rpm or 47.7 liter engine running at 2050rpm?
Or 36.7 liter engine running at 2750rpm?

I'd say that nobody is under illusion that the 'no free lunch' rule is abolished The 'BMW-35' will certainly not be as light as the BMW VI, same as Griffon was much heavier than Buzzard, or Jumo 213 than 211.
We can recall that 44.5L DB 603 run at 2700 rpm, while AM-38F run at 2350 - IOW, not everything is bleak wrt. RPM vs. capacity.

I get a little lost when people talk about "in house Knowledge" of a particular type of engine.

This thing was two WW I cylinder banks mounted on a common crankcase. Pretty up to date in 1918.
In 1933????
Germany had been prohibited or severely restricted in aircraft and and aircraft engine development during the 1920s.

BMW was moving on. They designed the small V12 (BMW 116) and bigger one (BMW 117), but were ordered by the RLM to stop the liquid cooling engine projects, and make air-cooled engines from now on.

Looking only at the results, the engine was proposed from about 1939 and had a short development before production. There seems little doubt that the engine was underdeveloped before its introduction to service. However, in the Fw 190 it gave the RAF and the Allies a big problem, right from the start. The initial development timescale was too optimistic.

BMW 801 was more of a redesign of the BMW 139, than a blank paper project. BMW gotten it right with a number of details on the 139, the Achiles heel being the set-up of the main bearings - all three bearings were one next to another, between the crank throws. Benefit was shorter and lighter engine, shortcoming was unreliability. The 801 reverted to the 'normal' layout of bearings - one in front of the front crank throw, another between the crank throws, and last behind the aft crank throw. Enigine - the BMW 801 as we know it - was heavier and longer than the 139, but it worked better.

 

[Shortround6]

I'd say that nobody is under illusion that the 'no free lunch' rule is abolished The 'BMW-35' will certainly not be as light as the BMW VI, same as Griffon was much heavier than Buzzard, or Jumo 213 than 211.

Well, the BMW VI didn't have a reduction gear and didn't have a supercharger (sea level was about as good as the power got) so there was going to be a weight gain no matter what.

We can recall that 44.5L DB 603 run at 2700 rpm, while AM-38F run at 2350 - IOW, not everything is bleak wrt. RPM vs. capacity.

There were other things that limited RPM but the point about the way gasoline burns remains.
Another point is piston speed. Commonly accepted that about 3000ft/per/minute was upper limit. A bunch of things entered into this one. Again the stress of the pistons as rods reversing direction so quickly and the wear/friction of the piston rings/piston skirts.
Every Piston engine was a series of compromises. Some designers stuck with what they knew and branched out slowly. Sometimes they stuck with an old engine too long.
Do you gamble on a new engine or spend increasing amounts of time and money on small improvements to an old engine?
The Soviets were somewhat stuck. They had a very small aero industry before 1917 and then it collapsed. Many of the best and brightest fled the country either at the time or in several years of civil war that followed. The Soviet leadership wanted to modernize and the fastest way to do that was to import foreign technology will the domestic workers/engineers were educated/gained experience. Trouble was that some of the technology was constrained by what countries would deal with them and when (although the did wind up buying quite a few designs). They were also somewhat limited in materials.
You can also trade performance for engine life. Which comes at a cost of needed more engines for a given number of missions, assuming your aircraft last long enough need spare engines. Perhaps a good gamble for short range fighters or tactical aircraft, not a good gamble for long range strategic bombers.

BMW was moving on. They designed the small V12 (BMW 116) and bigger one (BMW 117), but were ordered by the RLM to stop the liquid cooling engine projects, and make air-cooled engines from now on.

Well, part of the problem here was duplication of effort. The BMW 116 was in direct competition with the Jumo 210 and the BMW 117 was in competition with both the DB 600/1 and the Jumo 211. Perhaps Germany got it wrong but they did not have an inexhaustible supply of engineers. Perhaps BMW should have been put to work on a BMW 118 to compete with the DB 603?
The BMW 801 fulfilled a need, it was a way to get a 1500-1600hp engine into service for larger aircraft than the DB 601 and Jumo 211 could power (at the same point in time) the extra cylinders meant that its roughly 20% large displacement would make that easier, the air cooling would make it a bit harder.
The Allies got a big leg up with better fuel which meant they could just use more boost (and stronger engine parts) instead of building bigger engines. They did do that also but it took some of the pressure off and bunch off the American "newer" engines went nowhere because they weren't needed. P & W never built a new cylinder larger than 2.55 liters in a production engine after building the R-2800, they just used more or less of them.
P & W brings up another point (as does Wright) It was very hard to get more power out of air-cooled engines once you had designed them. You had to modify/increase the cooling capacity every time you wanted more power. They were much more sensitive than liquid cooled engines.

 

[tomo pauk]

Every Piston engine was a series of compromises. Some designers stuck with what they knew and branched out slowly. Sometimes they stuck with an old engine too long.
Do you gamble on a new engine or spend increasing amounts of time and money on small improvements to an old engine?

Have the BMW make a new engine (we can recall that BMW designed 3 separate engines in the late 1930s).

Well, part of the problem here was duplication of effort. The BMW 116 was in direct competition with the Jumo 210 and the BMW 117 was in competition with both the DB 600/1 and the Jumo 211. Perhaps Germany got it wrong but they did not have an inexhaustible supply of engineers. Perhaps BMW should have been put to work on a BMW 118 to compete with the DB 603?

There is certainly no point in the 116. The 117 makes better sense, especially if it can be had by the time the DB 601 is available. However, the a big V12 (~45L) gets my vote, indeed a competition for the DB 603.

The BMW 801 fulfilled a need, it was a way to get a 1500-1600hp engine into service for larger aircraft than the DB 601 and Jumo 211 could power (at the same point in time) the extra cylinders meant that its roughly 20% large displacement would make that easier, the air cooling would make it a bit harder.

This is why the big V12 gets my vote - to bring the level of power that Jumo 211 or DB 601 are likely to struggle to deliver.

 

[Shortround6]

There is certainly no point in the 116. The 117 makes better sense, especially if it can be had by the time the DB 601 is available. However, the a big V12 (~45L) gets my vote, indeed a competition for the DB 603.

I am not sure about a 3rd 34-35 liter V-12. If you need more engines build another factory for the DB 601 or Jumo 211 and simplify spare parts and training.
The British had the Hercules and Merlin to compete with each other. The Sabre, Vulture and Centaurus were the next step. (and Deerhound)

This is why the big V12 gets my vote - to bring the level of power that Jumo 211 or DB 601 are likely to struggle to deliver.

Part of the problem is when, the Jumo 211 and DB 605 were already the size of the Griffon (or close) so how far can you go with the 'simple' V-12 before you get into diminishing returns. This was the reason for all the 24 cylinder engines. They gave the promise of higher rpm which the big V-12s could not match. Promise and delivery are different.

 

[tomo pauk]

I am not sure about a 3rd 34-35 liter V-12. If you need more engines build another factory for the DB 601 or Jumo 211 and simplify spare parts and training.
The British had the Hercules and Merlin to compete with each other. The Sabre, Vulture and Centaurus were the next step. (and Deerhound)

Part of the problem is when, the Jumo 211 and DB 605 were already the size of the Griffon (or close) so how far can you go with the 'simple' V-12 before you get into diminishing returns. This was the reason for all the 24 cylinder engines. They gave the promise of higher rpm which the big V-12s could not match. Promise and delivery are different.

You can note that I advocate for a really big and simple V12 in this thread, ie. something much bigger than either the 601 or 211.

 

[marathag]

I think the critical question is, what would the resources not being spent on radial R&D have been used for instead? Turbojets, though it doesn't seem realistic to decree in the mid-30'ies that BMW should drop piston aviation engines to focus on jets? Auxiliaries, like supercharger and turbocharger development for Daimler/Junkers inlines? Kommandogerät for DB & Jumo engines? Move staff over to Daimler and Junkers to develop and produce their designs? Or license produce Daimler and/or Junkers inlines? Or would German aircraft designers have three families of inlines to choose from, if BMW would have developed their own inline engines?

Radials isn't a terrible choice, but getting any radial to run in a close ultra streamlined cowl is hard business. US failed on many attempts..

BMW working on a smaller displacement 18 cylinder, something like Nakajima's Homare, only 46.5" in diameter, but 2000hp at 3000rpm on 92 octane, would have been very helpful, than the way the 802 engine turned out to be, overly heavy and complicated for not much more power.

 

[Shortround6]

You can note that I advocate for a really big and simple V12 in this thread, ie. something much bigger than either the 601 or 211.

My view is that that the really big and simple V12 is not quite as simple as it appears.

Griffon, 36.7 liters...........168mm stroke.............2750rpm....................3025ft/min piston speed.
DB 605, 35.7 liters...........160mm stroke............2800rpm.....................2940ft/min piston speed
DB 603, 44.5liters............180mm stoke..............2700rpm.....................3188ft/min piston speed
AM-38, 46.7liters.............190mm stroke...........2350rpm.....................2929ft/min piston speed, not counting the longer stoke in the other cylinder bank.

If you have to reduce the rpm you are not getting quite the power the increase in displacement suggest.
Yes the Griffon cheats by using higher octane fuel but but the Griffon is 39.2% larger than the Merlin but then cuts the rpm to 91.66%

I have said many times that power to weight is more important than power per liter and perhaps the big engines show up better this regard.

I would note that small engines that perhaps could run faster (and still keep the 3000ft/min) piston speed can do as well as the 27-37 liter engines per liter but they tend to run higher friction losses.
The German and Soviet engines had problems due to raw material shortages so it is difficult to figure out engine life and some other problems. The Big V-12s had problems.
The smaller 24 cylinder engines had problems of their own. Vibration problems with many/most/all? of these engines were much greater than anticipated in 1939-40.
P & W with the R-4360, take the R-2800 cylinders, only put 7 on a row, use four rows, what could go wrong???
4 years later and few dozen test engines and..........................
Wright with the R-3350, Take the R-2600 and add two cylinders per row, easy peasy right.....................................

 

[tomo pauk]

My view is that that the really big and simple V12 is not quite as simple as it appears.

Griffon, 36.7 liters...........168mm stroke.............2750rpm....................3025ft/min piston speed.
DB 605, 35.7 liters...........160mm stroke............2800rpm.....................2940ft/min piston speed
DB 603, 44.5liters............180mm stoke..............2700rpm.....................3188ft/min piston speed
AM-38, 46.7liters.............190mm stroke...........2350rpm.....................2929ft/min piston speed, not counting the longer stoke in the other cylinder bank.

If you have to reduce the rpm you are not getting quite the power the increase in displacement suggest.
Yes the Griffon cheats by using higher octane fuel but but the Griffon is 39.2% larger than the Merlin but then cuts the rpm to 91.66%

Thank you.
I choose a ~45L V12 for BMW to make.

I have said many times that power to weight is more important than power per liter and perhaps the big engines show up better this regard.

No need to preach to a convert

I would note that small engines that perhaps could run faster (and still keep the 3000ft/min) piston speed can do as well as the 27-37 liter engines per liter but they tend to run higher friction losses.
The German and Soviet engines had problems due to raw material shortages so it is difficult to figure out engine life and some other problems. The Big V-12s had problems.
The smaller 24 cylinder engines had problems of their own. Vibration problems with many/most/all? of these engines were much greater than anticipated in 1939-40.
P & W with the R-4360, take the R-2800 cylinders, only put 7 on a row, use four rows, what could go wrong???
4 years later and few dozen test engines and..........................
Wright with the R-3350, Take the R-2600 and add two cylinders per row, easy peasy right.....................................

I'm not sure what these radials ad 24 cyl engines had to do with V12s.

 

[Shortround6]

I'm not sure what these radials ad 24 cyl engines had to do with V12s.

Just showing that things were not a simple as many people believed at the time. They thought they had a easy way to increase power using proven components, just add more.

If you want to stick with V-12s look at the Hispano and what it took to get more power out of it.
Actually getting more power is easy, just like a race car, it is making more power and finishing the race that is the trick.
Every time you increase the RPM you can wind up with a number of new vibration patterns

Simply scaling things up can also set lose all sorts of vibration problems. The Vibration problems can show up as all kinds of fatigue cracks/breakages hours into the test runs.

 

[tomo pauk]

Just showing that things were not a simple as many people believed at the time. They thought they had a easy way to increase power using proven components, just add more.

If you want to stick with V-12s look at the Hispano and what it took to get more power out of it.
Actually getting more power is easy, just like a race car, it is making more power and finishing the race that is the trick.
Every time you increase the RPM you can wind up with a number of new vibration patterns

Simply scaling things up can also set lose all sorts of vibration problems. The Vibration problems can show up as all kinds of fatigue cracks/breakages hours into the test runs.

Any insight on why in late 1930s Germans should look at Hispano - of all engines - as an inspiration on how to make big (and bigger) V12 engines?

 

[z42]

My view is that that the really big and simple V12 is not quite as simple as it appears.

Griffon, 36.7 liters...........168mm stroke.............2750rpm....................3025ft/min piston speed.
DB 605, 35.7 liters...........160mm stroke............2800rpm.....................2940ft/min piston speed
DB 603, 44.5liters............180mm stoke..............2700rpm.....................3188ft/min piston speed
AM-38, 46.7liters.............190mm stroke...........2350rpm.....................2929ft/min piston speed, not counting the longer stoke in the other cylinder bank.

And then we have the previously mentioned Jumo 213J with a 165mm stroke and 3700 rpm for an average piston speed of 20.35 m/s, or 4006 ft/min if you prefer funny units. Then again, the 213J never entered service so hard to say what the practical reliability and longevity of it would have been.

Practically all(?) WWII aviation engines were undersquare, except Napier Sabre. Going for an oversquare design could be a way to have a large volume and high rpm while keeping the piston speed reasonable. Then again, that would exacerbate issues wrt flame propagation speed in big cylinders, so IDK if it would have worked out in the end.

I have said many times that power to weight is more important than power per liter

Obviously true, though easily overlooked. I guess the obsession with power/liter comes from things like auto racing where the volume is fixed, and thus power/liter is the metric to aim for. Also, over here back in the day car tax and maybe insurance premiums were partially based on the cylinder volume.

The Big V-12s had problems.
The smaller 24 cylinder engines had problems of their own. Vibration problems with many/most/all? of these engines were much greater than anticipated in 1939-40.
P & W with the R-4360, take the R-2800 cylinders, only put 7 on a row, use four rows, what could go wrong???
4 years later and few dozen test engines and..........................
Wright with the R-3350, Take the R-2600 and add two cylinders per row, easy peasy right.....................................

Indeed. The V-12 is kind of a sweet spot in terms of balance (well, inline six too but not relevant for high performance aviation engines), and if you want more cylinders all the options have more or less serious drawbacks, as does increasing the cylinder volume too.

And it seems for radials it's a bit similar, that a two row with 7 or 9 cylinders per row is the sweet spot, seems all the radials with >2 rows had serious problems with cooling the back rows as well as vibration issues.