Fan Cooling of Radials

[MIflyer]

The BMW engine of the FW-190 used fan cooling. I have wondered why this was not more widely used in WWII.

 

[tomo pauk]

The BMW engine of the FW-190 used fan cooling. I have wondered why this was not more widely used in WWII.

Cooling fan was using engine power to turn. For BMW 801, at sea level and 2700 rpm, it needed 70 HP, at 5.7 km (~18500 ft) it used 50 HP - the 'no free lunch' rule applies as ever.
The power/weight ratio of the BMW 801 was not that special, the engine was as heavy as 1-stage R-2800 and 15% heavier than Bristol Hercules. The p/w ratio of similar Japanese engines (Ha 109, Kasei) was especially much better than of the BMW 801.

 

[MIflyer]

Well, sure they were using engine power to turn the fan - and gearing to turn the fan at a different rate than the prop (sorry, Monogram). But was it worth it? The FW-190 does have a nice tight cowl, probably not equaled until the F8F and XP-72.

But what comes to mind is battle damage. Radials are forgiving of damage but if that fan gets knocked askew, I think it is all over. I recall reading of an FW-190 pilot who was shot down over France by P-38's soon after 6 Jun 1944, made it back to his airfield by hitchhiking, and then decided to take a damaged FW-190 up instead of risking a ride in a truck with P-47's overhead. Well, guess what? It overheated and he crashed again.

 

[tomo pauk]

Probably the main consequence of having the fan was that it allowed for tight cowling to be used, thus shaving some drag?

 

[swampyankee]

I think it's somewhat telling that very few service airplane other than the FW190 used a fan-cooled radial. I also don't think many liquid-cooled aircraft used a radiator fan.

I don't believe this indicates that FW's designers were smarter or more competent than those at Grumman, Republic, Douglas, Boeing, Bristol, Vought, Nakajima, etc. I think it means that they had more trouble getting a radial engine that was properly baffled. The Cd0 values I've seen for the FW190A were roughly the same as for the Corsair (and less than that of many models of the Bf109). If you tabulate Cd0 data for all the single-engine propeller fighters produced between 1940 and 1945 and do a chi-squared test, you'll probably find the difference between radial and V-12 (and H-24) aircraft is not statistically significant, especially when you throw in the fact Cd0 would get changed noticeably by a different paint job.

 

[pbehn]

The object of the exercise is to cool the engine, a fan uses power. Machining of fins on many US aircooled engines was a work of art that cost a lot of time and no doubt money to develop. I think there was also an improvement in the metals used with better thermal conductivity. These may have been more expensive solutions overall but didn't use horsepower in the air.

 

[Tom Fey]

Cooling fans require considerable power to drive them. The R-4360's in the Republic XF-12 Rainbow had two-speed cooling fans geared to the engine. Under certain combinations of air speed and altitude, when the high speed cooling setting was engaged it bogged down the engine to a significant degree, loosing power to the propeller, but provided the require cooling. Going to the lower cooling fan speed returned the engine power, but didn't provide adequate cooling. So under these conditions the engine was unable to deliver the expected performance regardless of fan speed. Scylla and Charyabdis.

 

[yosimitesam]

The BMW engine of the FW-190 used fan cooling. I have wondered why this was not more widely used in WWII.
View attachment 599828View attachment 599829View attachment 599830

The XP-47J was tightly cowled and fan-cooled. It reached 505mph (supposedly a record for prop-driven level flight) using the R2800 with a bigger supercharger. Never went into production, though, as the end of the prop driven era and WWII were both in sight.

 

[MIflyer]

Well, cooling fans do require power but then again the drag associated with wider cowlings takes power as well. It has been pointed out that in the 1930's when the USAAC wanted Curtiss fighters powered by inline engines and the USN wanted radial engines on almost the same airframes, there was not much difference in performance. Drag varies by the square at lower speeds and the cube for higher speeds, so fan cooling is far more useful at the high end.

 

[tomo pauk]

...
It has been pointed out that in the 1930's when the USAAC wanted Curtiss fighters powered by inline engines and the USN wanted radial engines on almost the same airframes, there was not much difference in performance.
...

Biplanes or monoplanes?

 

[MIflyer]

Biplanes, P-6E versus FC11

 

[tomo pauk]

Than you.
With radial engine, there is no many ways for the designer to mess up the engine installation. With liquid cooled engine, if the designer does not use what that engine type offers (fuselage can be both of lower height and width, radiator can be either half-burried and/or faired in the wing, ejector exhausts invented yet?), swap to the in-line will not provide any performance advantage.
We also have the thing of a 'dirty' biplane negating most of not any perspective advantage of an in-line - a reason why RR purchased He 70 Bitz in Germany so they can test installation details.

 

[MIflyer]

Don Berliner said that Curtiss screwed up the radiator intake on the short nosed P-40's, making it too big, so that air spilled out the front, ran down the fuselage, and resulted in the need to extend the tail. He left the company over that argument.
It s true that in the time of the Curtiss Hawk biplanes there were not many options for cowls, mainly either none or a ring cowl. On the next generation of biplanes, the Grumman F2F and F3F, they did try for a lower drag cowl.

 

[tomo pauk]

Don Berliner said that Curtiss screwed up the radiator intake on the short nosed P-40's, making it too big, so that air spilled out the front, ran down the fuselage, and resulted in the need to extend the tail. He left the company over that argument.
...

Hilarious. It is always easier to blame someone else.

 

[basil]

It seems that not all contributors of this blog have read the report in post #1. It describes in great detail why cooling by a (geared) fan may be advantageous in total horsepower consumption compared to a non-fan installation of a radial engine.

MIflyer, is it possible to post the missing part of the article, too (page 260)?

 

[Maty12]

I seem to recall that the PBM-3D Mariner had engine cooling fans to help with overheating in tropical climate. This was not retained in the PBM-5, which replaced the earlier R-2600 with an R-2800 and a longer, more streamlined cowling. Not sure what effect the fan or its removal had on performance

 

[silverangel]

Came upon this thread looking for the performance improvements for the FW 190s cooling fan details, for a class I am taking for fun. And I did some digging to finish the article that was posted here. Turns out he kept quite a lot going for the 2nd issue in Aug. I got the lossless copies (linked below, Archive.org) and a smaller PDF attached here. I found the RAWs (*.jp2) that can be opened in Adobe Acrobat Reader. Hopes this helps. 1st part: Aviation Week 1944-07-01 : Aviation Week : Free Download, Borrow, and Streaming : Internet Archive , 2nd Part: Aviation Week 1944-08-01 : Aviation Week : Free Download, Borrow, and Streaming : Internet Archive

 

[MIflyer]

Here is a little more on the FW-190 cooling.

 

[Deleted member 68059]

This British wartime study by the RAE in December 1942, may provide some answers. Despite the power required to run the fan, it seems that (providing it runs contra-rotationally to the prop) a net gain in thrust is achieved.

 

[Shortround6]

I seem to recall that the PBM-3D Mariner had engine cooling fans to help with overheating in tropical climate. This was not retained in the PBM-5, which replaced the earlier R-2600 with an R-2800 and a longer, more streamlined cowling. Not sure what effect the fan or its removal had on performance

I believe the earlier PBMs had 1700hp R-2600s and the fans were used on a version giving 1900hp for take-off. Along with much different cooling fins on the cylinders.
The R-2800 offered a bit more power with less cooling problems.