A 'proper' way to have a 24 cylinder liquid-cooled aero engine for the ww2?

[Shortround6]

People tried with X, H and W layouts, some worked better than the others, even if the layout was the same. So how should the least troublesome, not too heavy or bulky, and very powerful 24 cyl engine looked like? We need it for the 1st version(s) to be in service when ww2 erupts, ie. in September of 1939.

Well, in service in 1939 pretty much means 87 octane for the British, and most everybody else.

This defines the upper range on BMEP

Then you can play with the cylinder size/s. Pretty much limited to the size cylinder's that will give you about 3000ft per minute piston speed. Some countries/engines may be lower.

Now the juggle act begins.

If you want high power you are not going to get light weight. Can't be done. The Bulky part as some leeway.

If you want cylinders at operate at the upper range of the BMEP then those cylinders need to be strongly constructed. Both the cylinders themselves and reciprocating parts.
Likewise if you want engines are running at high rpm (3000fpm) you need strong construction to keep the engine from breaking.
Please note that in 1939 the V-12s weren't that far apart in engine weight and power output. The DB 601 and the Jumo 211 traded big but slower running cylinders for the Merlin and Allison engines higher stressed, running cylinders for very close to the same weight at the close to the same power (talking about 5-10% or so). There was no magic about it. Just different choices by the different engineer teams.

Now you can split off into bespoke 24 cylinder engines or trying to use existing 12 cylinder engine parts to make big 24 cylinder versions. But those are just about always heavy and bulky. They offer the promise of easier development, not always realized in practice. They do offer more promise to higher power than the "medium" size 24 cylinder like the Sabre and Vulture (which is sort of a hybrid).
The Sabre passed a type test in August of 1939 at 1800hp. It didn't pass a 2050hp type test until June of 1940. The early production engine in 1940 were being removed for inspection after 25hrs.

 

[Howard Gibson]

People tried with X, H and W layouts, some worked better than the others, even if the layout was the same. So how should the least troublesome, not too heavy or bulky, and very powerful 24 cyl engine looked like? We need it for the 1st version(s) to be in service when ww2 erupts, ie. in September of 1939.

The Rolls Royce Griffon engine was as successful as it was because they kept it simple, and they used technology they understood, definitely including two stage superchargers. Napier's sleeve valves finally worked because Bristol stepped in and showed them how to do it. Bristol spent something over ten years figuring them out, fortunately, all before the war. They would not have helped a competitor under civilised conditions.

The advantage a liquid cooled engine has over a radial is its small frontal area, resulting in less pressure drag. There are all sorts of nifty things you can do to reduce drag from the radiators. When you go to X-24s and H-24s, you increase frontal area. At some point, the radial engine is a better idea.

Engines got more powerful during WWII primarily because they got bigger. The Merlin was a dinky little 27litres. It was successful because of the availability of high octane fuel and their willingness to develop two stage superchargers. Upgrading to the 36litre Griffon was a good idea. The Americans replaced their 30litre radials with 46litre radials. The Japanese did a phenomenal job of getting 350mph out of the 28litre Zero, but that was still not sufficient.

Instead of a liquid cooled V-12, how about a liquid cooled V-18? 50% more displacement! People were building V-16s back in the thirties.

 

[pbehn]

Instead of a liquid cooled V-12, how about a liquid cooled V-18? 50% more displacement! People were building V-16s back in the thirties.

Too long, as I understand it the crankshaft twists.

 

[tomo pauk]

Instead of a liquid cooled V-12, how about a liquid cooled V-18? 50% more displacement! People were building V-16s back in the thirties.

Thread was supposed to be about 24 cylinder engines. Not about V12s,V16s or V18s.

 

[pbehn]

There was no such nonsense before 1939, at least not from the British perspective.
RR can ditch the Exe and make a liquid-cooled engine instead of it. Say, a 40 liter H layout engine that uses Kestrel/Peregrine bore and some internals to speed up the development.

I was joking, if RR could have farmed out the development of the Vulture or kicked out anything not related to the Merlin maybe there could have been Vultures in production in numbers to have Hawker Tornados in the BoB the first prototype flew in Oct 1939. However as an engine it must have taken more man hours and effort to produce than a Merlin or Griffon. It gives you a more powerful S/E fighter but what would it be in a bomber? A Lancaster with 4 Merlins has more power than a Manchester, a Lancaster with 4 vultures would be an even bigger re design.

 

[Howard Gibson]

Thread was supposed to be about 24 cylinder engines. Not about V12s,V16s or V18s.

I question the need for 24 cylinders. The Pratt and Whitney R-1830 Twin Wasp was two seven cylinder rows resulting in fourteen cylinders. The Pratt and Whitney Double Wasp was two nine cylinder rows, and eighteen cylinders total. The cylinders were bigger and it was still a good engine.

In one of his books Bill Gunston states that all the important piston engines of WWI were developed in the 1930s. They took a lot of development. There were a number of cool designs that would have been good engines if they had kept at them, but resources were finite. Pratt and Whitney tested a liquid cooled X-24, but let's face it, the Double Wasp worked.

The British and the Americans anticipated the need for 2000HP engines back in the thirties, and I don't think anybody else did. Napier got their H-24 working with a lot of effort. Rolls Royce and Pratt and Whitney just make bigger cylinders.

 

[don4331]

OK, I missed the XH-2470

H-2470 has same stroke as Napier but is just over 10" wider in horizontal configuration (its 10" longer too, but at least 1.5" of that is the larger cylinder bore. It's also over 15" thinner but part of that is because the propeller shaft if offset 8-3/4" (8.747" if we are splitting hairs) from engine line. The biggest question is: Does your crystal ball show that before you can overcome the lubrication issues with sleeve valves, better materials for poppet valves will have been developed making sleeve valve development a waste of time??? A flat H-24 Kestrel...I'll have to work it up. Companies license products to their competitors all the time - they just demand royalties. If you're willing to pay, you can use it.

R-2600/BMW 801/Hercules. BMW 801F produced 2,400hp by end of war - development versus building what was good enough stopped a bunch of progress during war.

I assume you mean radial individual rows having odd number of cylinders for firing order of 1-3-5-7-2-4-6 for 7 cylinders, i.e. every 2nd cylinder in order. There are lots of twin row radials with even number of cylinders. Again, I'm not aware of even number of cylinder 4 strokes. But am interested if there are any (I'm aware of even cylinder row 2 strokes and a couple with "boxer" cranks which allow the engine to look like it has even number of cylinders but it is really 2 rows superimposed).

Thanks for the actual bore spacing for the Vulture - my tape measure comparison of exhaust ports of both Merlins and the only Vulture that I know still in existence (well at least 20 years ago when I was last in England) had both at ~6-1/8", to me it just confirmed what I had read. I've never laid my hands on actual blueprints of either. I have Vulture as being more/less 48" square (again, hand held tape measure), so it is a little greater frontal area than the "H" engines, but way smaller than radial of same power. RR has intake manifold, magnetos, carburetor, water pumps, and engine mounts filling most of the open area in the "v". And the RR H Merlin has more space between cranks than the Lycoming does.

Given I have offset ground a crankshaft but 0.100" (for 0.200" additional throw) along with making offset rod bearing inserts to keep compression height correct, and machinists at RR were able to reassemble a camshaft after the designer got the firing order wrong, offsetting the Vulture manufacturing equipment wasn't rocket science. Modifying by 1/2" or more...not so possible.

Information I have on Vulture was it was making 1,800hp@3k rpm with 9psi boost. And was getting 180 hours between overhauls at that level when it was discontinued which is pretty much same as Merlin of same period. And I know it made 2,500hp in lab, but Merlin made >2,100hp in Speed Spitfire. Neither of those aren't engines you want to fly to Germany and back on daily basis.

1,175 hp for developed Merlin on 87 octane fuel (RRHT No.19, p.148) was the expectation in '33 when engine is 1st run. That that Merlin III can make 1,300hp at lower altitude with 100 octane by tweaking the pressure regulation, just says RR had a little factor of safety in their design. My apologies, its is 1,800hp as practical limit for 100 octane fuel for Merlin, from same source.

The Griffon I was still borne as by the time it was designed, Merlin had passed it (so RR buried it along side the Grumman XF4F-1 biplane). The Griffon II was design for 1,800hp out of the box (taking over from where Merlin leaves off on 100 octane) with plans for 2,400hp. So, your redesigned Vulture needs to start at 2,400hp to avoid overlap with the Griffon (and to compete with the Sabre).

IMHO, Peregrine is discontinued more because you can get already get 1,200hp from a Merlin, so why spend R&D to develop the Peregrine for just one airplane. Vulture/Manchester sort of falling into same category - Tornado not withstanding. Unfortunately, the 300lbs additional per Merlin (+ bigger propeller) was more than Whirlwind could accommodate. I still stand by the fact that the Vulture production line could make Merlins with only minor updates to manufacturing equipment was significant factor in decision to cancel the Vulture.

RR was paying royalties to Farman for the 2 speed drive and to Hamilton Standard via de Havilland for the constant speed propellers. It really rubbed RR the wrong way that when you looked a Spitfire, you saw the de Havilland decal on the prop, not the RR emblem on the valve cover (as it was under the cowl). Hence, all the work on Rotol constant speed propellers.

 

[tomo pauk]

I question the need for 24 cylinders. The Pratt and Whitney R-1830 Twin Wasp was two seven cylinder rows resulting in fourteen cylinders. The Pratt and Whitney Double Wasp was two nine cylinder rows, and eighteen cylinders total. The cylinders were bigger and it was still a good engine.

24 cylinders can get you a lot of power on technology and fuel of the day.
What would you suggest for a layout (W, X, H, something else; valvetrain?) of a 24 cyl engine for service in 1939?

 

[Howard Gibson]

24 cylinders can get you a lot of power on technology and fuel of the day.
What would you suggest for a layout (W, X, H, something else; valvetrain?) of a 24 cyl engine for service in 1939?

The thing that increased power during WWII was increased displacement. You can do this by more cylinders, and/or by bigger cylinders. Complexity is bad.

If I were to re-think liquid cooled engines for WWII, I would look into a 40litre L-12. This would put out 2000+HP, it would reduce the frontal area of the aircraft, and it would provide a better view to the front and down.

The Germans, the British, the Italians and the Americans all built air cooled V-12s. Generally, these were too small to power WWII fighter aircraft even by 1939 standards. A number of them were extensively and successfully used on non-combat aircraft.

I am fascinated by the fact that all of Bristol's radial engines at 5.75" bores. Their Centaurus engines manufactured into the fifties I think, used pistons designed in 1919. There must have been all sorts of other standard hardware.

 

[pbehn]

The thing that increased power during WWII was increased displacement. You can do this by more cylinders, and/or by bigger cylinders. Complexity is bad.

What about fuels and boost pressure? The bore and stroke of most engines was about the same, with the Merlin being small.

 

[Howard Gibson]

What about fuels and boost pressure? The bore and stroke of most engines was about the same, with the Merlin being small.

The axis did not deploy the high octane fuels the Americans and British did. This was a huge advantage for the Americans and British. Higher boost pressures come from high octane fuel, and from methanol water injection. Extensively used WWII V-12s were anywhere from 27 to 44.5litres.

 

[pbehn]

The axis did not deploy the high octane fuels the Americans and British did. This was a huge advantage for the Americans and British. Higher boost pressures come from high octane fuel, and from methanol water injection. Extensively used WWII V-12s were anywhere from 27 to 44.5litres.

You didnt mention axis before. I believe the Germans had a high octane equivalent fuels but its in a book I got for Christmas.

 

[Howard Gibson]

You didnt mention axis before. I believe the Germans had a high octane equivalent fuels but its in a book I got for Christmas.

The Messerschmitt Bf 109 in Action Part 2 has a photo of Adolph Galland's Bf109F with a 100 octane sticker on it. They had high octane fuel, but not very much of it.

The Japanese extensively used 92 octane.

 

[pbehn]

The Messerschmitt Bf 109 in Action Part 2 has a photo of Adolph Galland's Bf109F with a 100 octane sticker on it. They had high octane fuel, but not very much of it.

The Japanese extensively used 92 octane.

What was the swept volume of his Bf109?

 

[tomo pauk]

The thing that increased power during WWII was increased displacement. You can do this by more cylinders, and/or by bigger cylinders. Complexity is bad.

One wonders how bad is the unwillingness to play withing the boundaries of the topics listed in the clear English language here.

 

[Howard Gibson]

What was the swept volume of his Bf109?

34litres. Twelve cylinders. They used two of the engines to make the Daimler Benz DB606 with the required twenty four cylinders, and it gave nothing but trouble.

 

[pbehn]

34litres. Twelve cylinders. They used two of the engines to make the Daimler Benz DB606 with the required twenty four cylinders, and it gave nothing but trouble.

The Griffon was 36.7 L

 

[Howard Gibson]

One wonders how bad is the unwillingness to play withing the boundaries of the topics listed in the clear English language here.

I would go for the X or W layout with everything on one crankshaft. The H-24 with two crankshafts is more complicated and less likely to work. A bunch of people developed X-24s, and the Rolls Royce Vulture is as close as anybody got to serious production.

As I am sure you know, Napier actually (and eventually) got their H-24 working, and it was a remarkable engine. It was a surprisingly small 37litres displacement, and it revved fast at 3800rpm. What would have happened if Napier had designed a 44litre V-12 instead? I am working on a project here at home and I have assembled a table of WWII engines. This is partial.

Engine​	Cylinders​	Displacement (l)​	Power (HP)​	Rotation (rpm)​	Mass (kg)​	BMEP (kPa)​
Merlin 61	12​	27.0​	1570​	3000​	815​	1730​
Griffon 65	12​	36.7​	2040​	2750​	978​	1810​
Sabre VA	24​	36.7​	2850​	3800​	1170​	1830​
Centaurus VII	18​	53.6​	2520​	2700​	1330​	1560​
R-2800	18​	46.0​	2100​	2700​	1170​	1520​
DB 603A	12​	44.5​	1730​	2700​	930​	1280​

I got cylinders, bore and stroke, power, rotation and mass from which ever version Wikipedia chose to document. I worked out displacement and BMEP. BMEP gives you a good idea of what fuel they were using. Assume the bottom three engines get more powerful with better fuel. Power is linear with BMEP. For the R-2800, power can be 2100/1520*1810=2500HP. For the DB603A, 1730/1280*1810=2450HP. I am surprised by the low weight of all the DB engines.

On the Griffon, the bore times stroke is 6"x6.6". For the Sabre, its 5"x4.75". I am not sure what limited the revs on these engines. The pistons? The strokes? The Sabre was an expensive engine to build.

On my complete chart, it appears that engine power correlates more closely with mass than with displacement. This is something I need to check.

The W layout works well in automobiles. Buyers of Bentleys and Bugattis are not concerned about cost. The frontal area of an automobile is controlled by the passenger space. You want a short engine if possible.

The frontal area of a WWII piston engined fighter is controlled by the engine. The 60° V layout looks to me like a good design compromise. If you are liquid cooled, long, skinny engines are better than short, fat ones. If you have a fat engine, you need to consider air cooling, and we all know there that leads.

 

[pbehn]

Griffon peak power was at 2,750 RPM, Merlin at 3,000 RPM and Sabre 3,800 RPM. From what I know the maximum piston speed is the limiting factor and that is a function of stroke and RPM. Forces and piston speeds increase exponentially with RPM and stroke and actually producing power is limited by the speed of the flame front/ gas expansion etc.

 

[PAT303]

Too long, as I understand it the crankshaft twists.

And vibration issue's.