Radial engines more favored in Germany, 1935-45?

Ad: This forum contains affiliate links to products on Amazon and eBay. More information in Terms and rules

Well, i probably said this with other occasions, but what Germany really lacked was a mid-1930s 14 cylinder radial equivalent to say the GR14K, at about 36-37 litres or so. For this, instead of competing with BMW, Bramo would be in an ideal position to design a 14 cylinder radial instead of the Bramo-323. Let's call it Bramo-324. Initially it would be good for what, 1100-1200 PS? Then the later two-speed variant should be good for 1300-1350 PS? I see it primarily used in the Do-17 variants where it would give it some much needed extra power. FW-200 would benefit too, or Ju-90 and i'm sure other aircraft i can't think of right now.

Of course, in this ATL BMW and Bramo should never be merged. BMW should entirely focus on improving the BMW-132 instead of BMW-116/117 inlines, the 132 badly needs better 2-speed superchargers, and then work on the 14 cylinder BMW-139 of 1500 PS, but with 3 separate bearings from the start if it's to be usable. It will then entirely focus on improving it up to 2000 PS instead of wasting time with any 18 or 28 cylinder projects. Bramo would then at the same time work on an 18 cylinder of 2000 PS, the Bramo-300 or whatever it will be called. And for the duration of the war focus on improving it up to at least 2500 PS.

As far as mergers, Argus could be merged into BMW or Bramo for instance, their air-cooled inlines beyond the As-410 and other fancy projects of theirs are a waste of time so their capacities would be better used supporting something that already works. They did build BMW engines under licence in OTL anyway.
 
Last edited:
I do like this concept as well, and i would add have them borrow a bit from DB as well, and bore it out to 155mm like the later Jumo-213J for a bit more capacity. Should be good for what, 1675 PS at TO and say 1450 PS at 5500 m? Would be very useful if this is ready by 1942 so it could take full advantage of Jumo's surplus capacity. Note, i see this entirely replacing the OTL Jumo-213, which while eventually becoming an excellent engine, it took far too long to be available.
 
All big radials suffered so yes it was an inherent feature of the air cooled radial design. The inherent feature is that air cooling is more difficult than water cooling. This is why it has become almost extinct in large engines. The problem is exacerbated at higher altitudes because of the lower density and hence heat capacity of the air. The numbers of cooling fins on a late production radial is astounding.
The relatively small flat 6 engines produced by Continental and Lycoming still use air cooling but the more recently designed Rotax engines use liquid cooled heads.
 

Large engines were the air cooled types, not the liquid-cooled types. Air cooled engines were working just fine at high altitudes, too. Even without the forced cooling.
Even with the number of cooling fins accounted for, large radials were manufactured in huge numbers, including the many thousands of R-3350s or late R-2800s.
As for the suffering - the R-2800, Centaurus or Ha 42 were yet to know the suffering of the Wright R-2160, Jumo 222 or Sabre.
 
All big radials suffered so yes it was an inherent feature of the air cooled radial design.

I wouldn't be so sure about that conclusion. How can you pinpoint the reason to be the inherent features of a radial engine, vs., say, the P-51 being a very advanced low drag aero design for it's time and thus had sufficient performance without 115/145 gas?


I'm sure we all know inlines got much better power/volume figures, but we all also know that ultimately what actually matters is power/weight and power/drag. An engine with lower power/volume (say, due to air cooling limitations) can to an extent compensate by using lighter construction.

The relatively small flat 6 engines produced by Continental and Lycoming still use air cooling but the more recently designed Rotax engines use liquid cooled heads.

The Rotax 900 series is very interesting and perhaps indicative of the direction aero piston engines would have developed hadn't turbines killed them except for the smallest sizes.

But I'm not sure we can draw much conclusions about the potential capabilities of WWII era high power engines from the Rotax.


Yes, though to a large extent I think the dominance of radials in the post WWII era before jets took over is largely due to the industrial dominance of the US, and their historical preference for radials in bombers and commercial aircraft, rather than any inherent superiority of radials vs inlines.
 
Check out the British and the French post-war, who were buying more of Bristol's 2-row radials than RR or HS V12s, let alone the Sabre.
 
Check out the British and the French post-war, who were buying more of Bristol's 2-row radials than RR or HS V12s, let alone the Sabre.
Radials certainly make sense for commercial aircraft. Rolls Royce tried hard with Merlins but were ultimately unsuccessful. The ultimate performance offered by inlines had its proper application in high performance combat aircraft such as the Spitfire , P-51 , Mosquito, ME 109, FW 190D etc
 
Last edited:
The ultimate performance offered by inlines had its proper application in high performance combat aircraft such as the Spitfire , P-51 , Mosquito, ME 109, FW 190D etc
The ultimate performance offered by radials have had it's proper application in high performance combat aircraft like the P-47, Sea Fury, F4U, F7F, F8F and the like
 
Liquid radiators dump heat out into the airflow. Are radials really that bad?
 
I have posted this before

This gets the Cyclone up to the 1200hp for take off version. The 1300-1350hp versions got A LOT MORE finning. They also ran at higher rpm.
A rather vague chart From P&W.

Rather generic in that they don't say which engines. We have to guess based on power and year (1925-27 is the R-1340 Wasp)
I fully expect another companies to follow the same progression. There are photos of Bristol sleeve valve cylinders showing the massive change in fins just from 1932 to about 1940.
Hercules went through a bunch of cylinder heads before they even got to the 100 Series.

I have said this before. It doesn't matter what the designers can think up/draw up on the drafting board. It also doesn't matter what the tool room/prototype shop can make a handful of. What matters if you can make hundreds or thousands of cylinders and heads per month and that depends on casting/forging techniques and the ability of the machine tool suppliers to come up with the machine tools that can make the desired cylinder barrels/heads in quantity without excessive scrap rates.

Are radials really that bad?
Actually they are.

In a good liquid cooled engine the liquid jackets can be rather close to the head surfaces and/or valve stem area and they can adjust for size/volume of the coolant flow and deal with the heat later (in the radiator) while the air cooled engine has to transfer the heat from the inner surfaces to the fins (though the solid parts of the head )


And lets remember that air at 20,000ft has only around 55-60% of the cooling power it does at sea level. Yes it is colder but you are getting about 1/2 the mass of air flowing over the fins for the same cubic feet of air.
And the real problem isn't high speed, it is climb.
 

I suspect that if turbines hadn't arrived and destroyed the prospects for even higher power piston aero engines, we'd have seen something like liquid cooled radials. Seems two row 18 cylinder engines were about the upper end of what could be reliably air cooled (the four row Wasp Major evidently suffering from overheating issues that were never properly resolved). And similarly for V engines, while V-12 is in many ways a sweet spot, those cylinders are getting awfully big when you scale the engine ever bigger.

Take banks of 6 cylinders (nice primary and secondary balance, suitably long) and put them around a common crankshaft. The ill-fated RR Vulture being an example of the approach, but why stop at only 4 banks. With, say, 9 banks and 6 cylinders in each bank you'd have 54 cylinders (though spare a thought for the poor bloke tasked with checking the spark plugs on that monster!).

Though I'm not sure whether such engines should be classified as inlines since they'd have banks of inline cylinders (presumably of monoblock construction), or are they radials since there's multiple banks arranged radially around the crankshaft? With the Vulture and similar engines being classified as 'X', maybe with more banks it would be a, uh, 'star' configuration?


To an extent, that problem is present for liquid cooling engines as well, in the sense that the designer wants to make the radiator as small as possible to minimize drag. If you size the radiator for a slow full power climb at 20k, well, then the radiator is going to be needlessly large and draggy in most other situations.
 
Wright tried to go down the liquid cooled radial route
 
In both types of engine you can adjust the cooling air flow, liquid cooled engines usually used a flap at the back of the radiator and radials tended to use a series of flaps around the rear of the cowl. Trickier methods were sometimes used like sliding shutters. Trying to balance the needed air flow with drag was often difficult.

A rather large air brake fitted around the engine. Part of the problem is that were usually few intermediate positions.
As noted in other places, the R-2800 C engines due to better fins needed about 10% less air flow than the R-2800 B series engines for the same power, however the "C" series engines were promptly rated at 2100hp instead of 2000 and post war engines were rated at 2300-2400hp for take-off using higher octane fuel and/or water injection in airliner use.

I would note that the cooling problems were the least of the R-2160s problems, a poster child of how not to design an aircraft engine. Using 42 cylinder to get 2500hp should have been the first clue
 

That is true, but not particularly relevant to the point I was making.

I would note that the cooling problems were the least of the R-2160s problems, a poster child of how not to design an aircraft engine. Using 42 cylinder to get 2500hp should have been the first clue

To the extent the liquid cooled radial is a amalgam of liquid cooled engine and radial engine technology, I think there would have been a somewhat long and painful learning curve for any engine company, considering the companies tended to be focused on one of the engine types and thus lack experience with the other. In reality, WWII and then turbines meant that there was never the time to properly work out all the issues and get a functioning design into use.

But yes, no need to limit yourself to small cylinders (unless you're explicitly going for a high rpm design). With Vulture sized cylinders, with 7 banks you might be able to hit somewhere around 3500 hp. Or if going with Griffon sized cylinders, with 9 banks something on the order of 9000 hp might be achievable. All moot of course thanks to turbines, but just as a thought exercise.
 
Of course, in this ATL BMW and Bramo should never be merged. BMW should entirely focus on improving the BMW-132 instead of BMW-116/117 inlines, the 132 badly needs better 2-speed superchargers,
Indeed, both 2-speed drive and a better S/C was needed.
One of the things that BMW, in it's infinite wisdom, did when moving away from the Hornet to the 132 (half-way through the 132 line, actually) was to replace the, at least, modern-looking impeller, that was with curved vanes, with an obsolete impeller with straight vanes*, while also replacing the direct entry intake with the 'squished' intake.
Go figure.
(yes, Hornet's impeller was very small, under 7 in diameter)
* granted, P&W did the same with the later engines...
BMW 132A (basically, Hornet by another name; note also the vanes in the intake elbow; red is 'fresh' oil, green is not):



Later siblings of the 132A (like the E, G, T or Z) have gotten the straight-vane impeller, but also a bit more refined intake elbow.
Cross section that shows the squished intake elbow, eg. as on the F, M, N; never mind the blue arrows :



I've tried to measure the impeller diameter on the pic, it is 155.5(=bore)/0.626262 = 248.3mm for these later engines, or ~9.77 in (don't quote me on this ). Newer versions were in aggregate much more capable then the 'old tech' versions, but even so they were badly behind the curve come 1939.

Revert to the old style intake, but with greater area required for the greater air flow, and a more modern impeller should've've pushed it to be a bit over what the later Pegasus versions (no fuel injection) or the Bramo 323 (a bit lower displacement) were capable for on 87 oct fuel.
A bigger impeller would've also been interesting, Bristol radials were regularly using impellers of circa 11 in diameter.
 
two speed engines do not actually make more power.
There is a difference between many US radials and many European radials (and I am throwing the British in with the Europeans) were used/rated in the 1930s.
The US favored lower altitudes for their engines ( P&W and Wright) than the Europeans. This may have been partly because the US market for engines was tipped to commercial aviation and not military. European companies sold a higher percentage of their engines to the military. The Military wanted more altitude performance and used higher supercharger gears.
2 speed engines sort of redistribute the power. We have to look at specific engines. A European engine (most of them) often had a FTH of 3-4000 meters and if/when fitted with a two speed supercharger they did get a boost in low altitude power. However this was due to 3/4 things, 1 was less power going to the supercharger and more going to the propeller, heat load in the cylinders was the same. 2 the lower power used in the supercharger heated the air less and thus the charge was denser which allowed for more power. 3 fewer pumping losses in the system. 4 the ability, if the fuel allowed it, for higher boost in the low speed setting but this often was not used as it required the pilots to use different boost limits and could lead to pilot errors.
In the early/mid 30s a two bearing 14 cylinder radial was a viable option. Since the 73 octane fuel limited pressures in the cylinder there was not the strain on the crankshaft that later engines would get. You also had a limited ability to put enough fins on the cylinder. This indirectly affected rpm. Doesn't do much good to run at 2200 rpm if you can't cool the cylinders at 2000rpm. The two bearing design saved a couple of hundred pounds of engine weight.
Now the big question is when did the 3 bearing design get viable. P&W used in in the early 30s, but the early R-1830s were good for around 800-900hp and were under 1200lbs.
They soon made more power and got heavier.
For the Germans getting into 14 cylinder two row engines too early may be counter productive. The GR14K was licensed/copied by a number of countries but it came with limitations for future growth. The GN14K maxed out at about 1100hp in Soviet hands as the M-88 (with 2 speed supercharger) and the weight was over 1400lbs. It took the Soviets about 5 years to reach that point.
Of course, in this ATL BMW and Bramo should never be merged. BMW should entirely focus on improving the BMW-132 instead of BMW-116/117 inlines, the 132 badly needs better 2-speed superchargers, and then work on the 14 cylinder BMW-139 of 1500 PS,
Only a few companies develop engines for which there is no market. BMW had been one of the few companies in Germany allowed to make aero engines in the late 20s and early 30s . If the German Air Force says they want V-12s (and inverted ones at that) does BMW tell them "No, what you really want are radials" ?. BMW was selling a crap load of old V-12s to the Luftwaffe but the Luftwaffe and put out the requirements for the "new" engines that lead to Jumo 210 and the DB 600 and the Jumo 211. The Luftwaffe didn't have any interest in high powered radials and the German domestic market (Ju 52) wasn't crying out for new improved engines in 1932-35.
The need for two speed superchargers didn't really show up until after 1935, sometimes closer to 1939-40.
You needed better fuel.
You needed better propellers.
You need airplanes that need to fly in 'teens' and not at under 10,000ft and that means military aircraft. Most airliners will not fly at altitudes bordering on needing oxygen masks.
Some will to get over certain mountain ranges but sometimes not all planes in the fleet will get the high altitude engines.

There were reasons that a lot of engines developed the way that they did.
For the BMW 132 vs the R-1820 you had several things going against the BMW.
One is that you have only 93% of the displacement.
Two is a big one, the 1200hp R-1820 used 100 0ctane fuel.
Three is that the 1200hp rating was only good to 4100ft. The benefit if the 2 speed supercharger.
At 14,000ft it was good for 1000ft in high gear and we can perhaps compare that to the 840hp that the Bristol Mercury had at about the same altitude.

Now go back to post #70 with the charts on the Cyclone 9 cylinder finning, in addition the cylinder fin changes the Cyclone 9 to reach the 1200hp was on it's 4th (or more?) different crankcase and crankshaft. Not minor changes either. They were on their 2nd different forged steel crankcase after the early engines had use Aluminum alloy crankcases.
Now in 1938 Wright had built over 8000 Cyclone 9s since the late 20s and most of them were for the commercial market. That is what paid for the continual development. Note that this is before the 1200hp version even shows up.
P&W had sold over 13,500 engines from 1925 to Aug 1938 of 7 different types. And there were few experimental failures (not counted) and 2 commercial failures (one around 1930 and in in the 1938 period).

We can use the retrospectoscope to see what might have gone different but it needs very careful focus and a number of things to go our way.
BMW licensed the P&W Hornet to get into the radial market fast and easy. However the rise of the Nazi's in Germany tended to cut the Germans off from other nations in terms of technical support/R&D from the parent companies.
And like I keep saying, you need to able to make the engines.

Ford is often overlooked in the development of the R-2800 engines. They did not change much of anything in the engine itself, they did make changes in how it was made.
IN 1941, very soon after production started, they figured out a way to make cast cylinders using centrifugal casting instead of using forgings. These cylinders were stronger, required less machining and required much less labor. Also freed up forging equipment for other uses.
The US did not overwhelm the axis by size alone. They used part of their industrial capacity to provide tools/processes that were more efficient which increased the rate/s of production over many of the axis companies.
 
Me, I'd still make the BMW 132 with 2-speed S/C, together with a better S/C. Even Bramo did it, the 323 ending up as a better engine than the 132, despite the smaller displacement of the 323.
Both bombers and transports need ever greater power for take off as well as for better altitude performance (= you can have your cake, and eat it, too). Engine-out situation with an 1000-1100 HP engine still running is much better than if a 850-900 HP engine is still running. Can offer to Junkers for the Ju 52, so their Ju 52/3m can become 52/2m, that is more affordable for the users (both military and commercial).

From the German point of view, a BMW 132 that can do 1000 HP for take off on 87 oct fuel has the benefit of avoiding the need for 100 octane fuel to be used for overboosting in order to gain 1000 HP, as it was the case with historical 132s used on military Fw 200s. Or, still use 100 oct for take off, but now it is 1100-1200 HP.
 
Last edited:
From the German point of view, a BMW 132 that can do 1000 HP for take off on 87 oct fuel has the benefit of avoiding the need for 100 octane fuel to be used for overboosting in order to gain 1000 HP,
It would be benefit to the Germans. The problem is how do you get there?
You have two choices but both require more cylinder finning (including head finning to keep from typing so much).
You can increase cylinder pressure (increase compression and/or increase boost) and use higher octane fuel.
You ca increase engine rpm and keep the same fuel.
But both methods burn more fuel per minute in the cylinders and so need better cooling.
The BMW 132 K & M were rated at 960hp for take-off. Problem was that they used a 7.0 supercharger gear to do it and altitude performance went into the toilet compared to the 132 N that used a 10.14 gear set. Just use a a 2 speed supercharger to get both ratings or use the 11.4 gears from the F & J to get 890hp at 11,500ft.


Using water injection will help.

Both methods usually result in modified engines compared to earlier engines, The increased boost method requires stronger construction due to the greater loads of each power stroke.
The higher rpm method also needs stronger construction but in different areas of the engine.

It can be done and it was done. But for the R-1820 you had

engine model...........................power..................rpm...................Weight..............fuel
G....................................................1000.....................2200.................1200....................87
G100............................................1100......................2350.................1275....................91
G200............................................1200......................2500.................1310*.................95-100

There were a lot of models and some variations. I tried to use engines with 2 speed superchargers. Wright used a low weight 2 speed mechanism.
They also figured out how to make their 2nd steel crankcase used in the G200 about 50lbs lighter than the one used in the G100 while being stronger.
There was quite a variety in the fuel between the engines even in each model of engine. The actual fuel used did NOT usually affect the max power of the engine. What it did do was increase the height the the specified power could be used at. Like a version of the G100 that ran on 100 octane was still limited to 1100hp at take-off but could hold the pressure needed for 1100hp level several thousand ft higher.

Differences between the US engines and German engines was that the later German engines used direct fuel injection and the German played more games with the supercharger gear ratios. They often traded power at altitude for take-off power. Wright did that for commercial customers, Military engines were usually 7 to 1 for single speed engines and 7.14 and 10 to 1 for the two speed engines (you will see very minor variations like 10.04) and this is for all three series of the Cyclones.

To me this indicates that the Germans didn't actually modify their engines that much. Maybe my sources are bad, but the the higher power versions of the BMW 132 were the same basic engine but just used a lower supercharger gear to get more power at take-off and traded power at altitude to get it.

P&W used 4 different cylinders on the R-1830, the 1st may never have made it out of the experimental shop? Some engines rated at around 950hp are noted as having "B" cylinders, Then came the "C" cylinders which took them to 1200hp. The later cylinders with the aluminum cooling muff took them to 1350hp. For some reason they didn't call them the 'D"?

I am sure the Germans could have made similar improvements to the Americans. However the improvements often meant changes to tooling and manufacturing processes. You got more powerful engines but they cost a bit more. The US was also spreading the cost out over a lot more engines.

One source claims about 5500 Bramo 323s made through 1944. 21,000 BMW 132s ?
Wright built
G...................4092
G100...........5319
G200...........4155 commercial
G200........81838 Military (including licenses)
C9HC..........9063 Military (1300-1350hp engines) end of WW II.

You can do better than the Germans did, but you need to spend a lot more money to do it.
 
Am I really required to suggest for the 3rd time, in a same thread, that a 2-speed drive for an improved S/C would've mean a better 132?

You can do better than the Germans did, but you need to spend a lot more money to do it.
The fellow member suggested that BMW does not make the 116 and 117, so there is more money to spend.
 
Am I really required to suggest for the 3rd time, in a same thread, that a 2-speed drive for an improved S/C would've mean a better 132?
sorry to keep picking at this, we can figure out the power of the "better" 132 with a 2 speed supercharger pretty well by using the power figures from the existing 132 engines. Doesn't require redoing the engine. If you want more than that then we have to get really tricky.
The fellow member suggested that BMW does not make the 116 and 117, so there is more money to spend.
This is the ideal world.
I have no idea why BMW didn't get a contract for one engine or the other. Like if the BMW engines were really not as good as the Jumo or DB engines or if there was some sort of politics/favoritism going on. Or if like England the air ministry decided that they only needed a certain number of engine makers of one type of engine and other companies (Fairey and Alvis need not apply).
BMW wound up making over 9,000 of the old VI type V-12s during the 1930s and licensing them to the Soviets and Japan. Having BMW voluntary give up on liquid cooled engines in the early 30s to contrate on radials????
And somethings happened when they did for reasons. Extra research/development in 1932-34 may or may not give better engines in 1938-39.
P&W had two oops's in the late 20s and early 1930s.

and the R-1860 Hornet.

And an experiential R-2270 14 cylinder two row. one built.

All of which helped cement P&W to the R-1830 project.

I will note that this part of the Wiki entry is in error.
"Although a technically competent design, the enlarged Hornet B engine was not a commercial success. Customers preferred to buy the R-1830 Twin Wasp instead, which in time became the most numerous aircraft engine ever produced."

The R-1860 engine started design in 1928, it was first run in 1928. They made 446 of them.
The R-1830 was started in Dec 1929, first run in April 1931. I am having trouble tracking down the actual first use (not company test flights) but it seems to be 1934-35. 1st commercial use was the Martin 130 Clipper at the end of 1935 (design started much earlier) and first Military use may have been the XB-14 which was a converted Martin B-10 and this seems to have been converted to a B-12 when the engine tests were done.
P&W had developed the R-1690 Hornet to make more power than the R-1860 Hornet B by 1933-34 and that may have been what killed off the Hornet B, which had some cooling issues and lacked development potential due to rpm, at least in P&W's eyes.

It does not appear that the R-1860 Hornet B and the R-1830 Twin Wasp actually overlapped. Hornet Bs were still flying but there was around a 2 year gap in installations.
 

Users who are viewing this thread