Radial engines favored for powering the tanks & AFVs, 1935-45

[marathag]

We can take a look at the Valentine tank. The engine bay was long enough for the 6-cyl diesel to fit, that went at 79 in long? The R-670 was under 35 in of lenght. Let's be conservative and have the radial engine of good power being 40 inches shorter - a full meter. In the time when tanks were 5+- meter long, that is a major thing. Even saving of 50cm leaves a lot of weight allowance to be used on other stuff.

Valentine Engine Compartment

 

[tomo pauk]

Me neither. Just pointing out that engines, even from the same manufacturer, had wildly varying costs. So I don't think in general you can make the argument that a radial would be cheaper than a liquid-cooled inline, all else being equal.

I've tried to make a comparison between a 9 cyl radial (often used on the US tanks/AFVs) vs. a 12 cyl liquid cooled engine (standard on German, Soviet and British tanks/AFVs worth speaking about) tanks, not just any radial vs. any liquid cooled engine.
In this comparison, my opinion is that the radial is a less costly choice, while also offering the knock-on effects wrt. tank weight/price/armor etc.

I don't think you can attribute that 500kg (rough difference between BMW 132 and Maybach HL 230, if we're going by wikipedia numbers?) solely to cast iron vs aluminum. Or radial vs inline, for that matter.

I'm not sure that I've simply attributed to the materials chosen the weight difference between the two engines. Iron/steel was possibly 'guilty' for 200+- kg weight creep on the HL 230 (that went to 1300 kg)?
1-row radials were very light for the power they were making, though. The BMW 132 was far lighter than the HL 230, possibly up to 900 kg when we also account for the cooling system. Germans can also use the Mercury engine made in Poland as the base for their tank engine for the 1940s (Mercury was with a smaller diameter). Or make their own 'Mercury equivalent' with the Bramo 323 or BMW 132 base.

Maybe? But maybe you instead lose by needing a higher hull, which adds weight by itself as well as being an easier target. If you look at the volume of the powerplant instead of focusing on any one particular dimension, I'm not sure the radial wins by that much, if at all, anymore. Also keep in mind the radial will need a large fan with assorted ducting for cooling, which will increase the length somewhat (of course the liquid cooled engine will need radiators and fans as well, but has a little more flexibility how these are placed). With a long and narrow engine like a V-12, maybe you can compensate with a lower hull, and by putting things like fuel tanks beside the engine bay?

Seems like that Valentine tank, that used the humble inline 6 engine, was tall to the roof same as the M3 light. It was also much wider. German tanks were also not known for their small height no width, especially the 'big cats'.
We can also see that a bit smarter packaging, like what the people that designed the M-18 Hellcat did, keeps the height within the modest proportions. It used the same engine as many of the M3s/M4s, yet it was a lot lower because there was the gearing added that kept the propeller shaft low.
In case that gearbox is at the back, as it was common on the French, British and Soviet tanks, even that gearing is not needed.

 

[marathag]

In this comparison, my opinion is that the radial is a less costly choice, while also offering the knock-on effects wrt. tank weight/price/armor etc.

The biggest downside was low torque output at low RPMs, that was overshadowed by the great Horsepower to weight ratings.

Aircraft engine sit at cruise RPMs for most operations, where that low torque just never came into play.

The way to avoid all that, is adding the complexity of electric or even hydraulic drive, or what the Germans did with their high rpm, low torque V-12s, add more gears.

 

[tomo pauk]

A table from the Russian book about the tanks, FWIW:

 

[Shortround6]

M3A3 light tank.

M5A1

Two V-8s side by side.
Added 1250lbs to the tank to max out at 33,000lbs. Also includes the Hydro-matic transmission.
You may have to look at situation on case by case basis. Blanket always good or always bad might not apply.

 

[z42]

I've tried to make a comparison between a 9 cyl radial (often used on the US tanks/AFVs) vs. a 12 cyl liquid cooled engine (standard on German, Soviet and British tanks/AFVs worth speaking about) tanks, not just any radial vs. any liquid cooled engine.
In this comparison, my opinion is that the radial is a less costly choice, while also offering the knock-on effects wrt. tank weight/price/armor etc.

Fair enough. So limiting ourselves to single row radials, for instance, in 1944 a R-1820 (9-cyl radial) cost $8920, and a V-1710 cost $10565. So yes, the radial is slightly cheaper. OTOH the V-1710 makes a lot more power, so on a price/hp comparison I'd guess the win goes to the Allison. Of course both of these are way too big for a tank, but just as a data point. So no, I don't think it's at all clear that a radial will automatically be cheaper.

I'm not sure that I've simply attributed to the materials chosen the weight difference between the two engines. Iron/steel was possibly 'guilty' for 200+- kg weight creep on the HL 230 (that went to 1300 kg)?
1-row radials were very light for the power they were making, though. The BMW 132 was far lighter than the HL 230, possibly up to 900 kg when we also account for the cooling system.

My point is, aero engines in general were pretty light, both radials and inlines. But this did come at a cost. In a non-aero application, you're probably quite ready to sacrifice some weight for lower cost, general sturdiness, and lower maintenance requirements.

Seems like that Valentine tank, that used the humble inline 6 engine, was tall to the roof same as the M3 light. It was also much wider. German tanks were also not known for their small height no width, especially the 'big cats'.

A low profile engine is a necessary but not sufficient condition for creating a low profile hull. There are certainly many other ways to fail!

We can also see that a bit smarter packaging, like what the people that designed the M-18 Hellcat did, keeps the height within the modest proportions. It used the same engine as many of the M3s/M4s, yet it was a lot lower because there was the gearing added that kept the propeller shaft low.
In case that gearbox is at the back, as it was common on the French, British and Soviet tanks, even that gearing is not needed.

I think I've mentioned in multiple threads that IMHO, gearbox in the rear is the correct(TM) design for a tank, so no argument here.

 

[tomo pauk]

So limiting ourselves to single row radials, for instance, in 1944 a R-1820 (9-cyl radial) cost $8920, and a V-1710 cost $10565. So yes, the radial is slightly cheaper. OTOH the V-1710 makes a lot more power, so on a price/hp comparison I'd guess the win goes to the Allison. Of course both of these are way too big for a tank, but just as a data point.

Something was rotten in the R-1820 land. It was the only engine that was to cost less before 1941 (8814 US$) than in 1944 (8920 US$). Any other engine saw a price drop of 30+- percent, comparing before 1941 vs. 1944. The 2-row radial R-1830 costed less in 1942-44 than a 1-row R-1820, despite staring out 'before 1941' at 11151 US$.
In 1941, AAF was paying 10070 US% for the R-1820, and R-1670 (Hornet of 875 HP) was 8500.

The next 1-row radial worth mentioning, the R-1340 (probably 600 HP), was 6322 US$ in 1941.
Engines that have had versions for tanks/AFVs, the R-975 and R-670, were at 6000 and 2500-3000, respectively, what the AAF was paying. R-985 was similar in price to the R-975.
The R-775 was thereabout with the R-670; both were 7 cyl engines.

The air-cooled V-770 V12 was pricey, no doubt due to the small series made; in 1942, it was 12100+ US$ a pop.

tl;dr - I'd say that going for an 1-row radial is cheap enough to matter when compare with a V12, especially if that engine is with 7 cylinders; add the savings in tank size and weight (both direct and indirect), and there is a lot going for these engines as tank engines.

My point is, aero engines in general were pretty light, both radials and inlines. But this did come at a cost. In a non-aero application, you're probably quite ready to sacrifice some weight for lower cost, general sturdiness, and lower maintenance requirements.

A low profile engine is a necessary but not sufficient condition for creating a low profile hull. There are certainly many other ways to fail!

Agreed.

 

[Shortround6]

Engine prices are all over the place. As a very, very rough rule price was related to horsepower. But this depends a lot on engines being similar. back in 1930 in the US a set of reduction gears for an aircraft engine could cost from 1000 dollars to almost 4,000 dollars depending on the engine.
It also dependent on other accessories and the quantity ordered under one contract.
Had anybody wanted Allison engines for a large tank they would have been ordering an engine with no reduction gear and no supercharger. But they would need a different intake manifold and different carburetor/s and flywheel.
The US turned away from radial engines in tanks during the war. It might not have been cost alone but for a better utilization of resources. They had more factories that could build large truck engines (and tank type engines) than they had factories that could build aircraft engines. Even 200-450hp trainer engines.

 

[GrauGeist]

So no, I don't think it's at all clear that a radial will automatically be cheaper.

You need to also take into consideration the weight of the engine, the radiator(s) and related plumbing plus the armor to protect the cooling system.

 

[EwenS]

We can also see that a bit smarter packaging, like what the people that designed the M-18 Hellcat did, keeps the height within the modest proportions. It used the same engine as many of the M3s/M4s, yet it was a lot lower because there was the gearing added that kept the propeller shaft low.
In case that gearbox is at the back, as it was common on the French, British and Soviet tanks, even that gearing is not needed.

As far as height goes, you need to remember that the M10, M36 & M18 were not tanks. They were tank destroyers. All had open topped turrets which also meant less of an issue accommodating the breech end of the gun as it was depressed.

 

[tomo pauk]

As far as height goes, you need to remember that the M10, M36 & M18 were not tanks. They were tank destroyers.

I don't think that I've ever specified the height of these vehicles, but the way that M18 specifically have had a device that lowered the propeller shaft.

All had open topped turrets which also meant less of an issue accommodating the breech end of the gun as it was depressed.

Okay. Even though I can't see that the roof being open or not had any bearing on the layout of the prop shaft.

 

[Shortround6]

Okay. Even though I can't see that the roof being open or not had any bearing on the layout of the prop shaft.

It doesn't really. It may have an bearing on the overall height of the vehicle.

A lot of Pre war and WW II vehicles had really horrible ergonomics. Drivers sometimes sat on a cushion on the floor with their legs bent with high knees and with back bent forward to reach levers. The Hull was small (Low) but such vehicles were really tiring to drive for even a few hours. Tall engines may force designers to use taller hulls that allow more room for the crew. However the turret crew needs room to work which means a certain height for seated crewmen or a combination of seated and standing (crouched over?) with loaders being a prime example of standing crewman. How much does he need between roof and turret floor or hull floor? Some tanks just ran the drive shaft though the crew area with varying degrees of guards to keep the crews feet from tangling with the shaft. Or tripping over the shaft housing as the turret rotated. Some designers opted for the rear drive to allow the loader to stand. This varies with time as sitting down and slinging 37-47mm shells around is not that hard. Get to the high velocity 75mm and bigger and the loader has to start using his back and legs.
Turret height is governed by depression of the gun, and not just how far the gun breech goes up but if the designer/tank committee wants the gun to be loaded at maximum depression. This also depends on the gun and ammunition. Is the ammo longer than the gun's recoil distance? You do need a high enough roof to keep the breech block from hitting the roof and then you need enough distance for a 1/2 meter (or longer?) shell to be loaded into the breech, unless you decide to raise the gun for loading?

What is decided when the tank is undergoing preliminary design may not be the way things end up if the gun or some other feature is changed 1/2 through the design process.

Open topped vehicles could be shorter because they expected the crews to fight with the tops of their heads above the top level of the armor, at least a lot of the time, and only ducking when needed. Not fighting (or being in a danger area) for several hours.

 

[marathag]

Open topped vehicles could be shorter because they expected the crews to fight with the tops of their heads above the top level of the armor, at least a lot of the time, and only ducking when needed. Not fighting (or being in a danger area) for several hours.

The M36 'Jackson' that the crews only called 'Slugger' should have been done this way

The M10 or M10A hull with the twin Diesels or Ford V8 was just too tall for what it was to do, carry a gun.

As the M36 started to get used like tanks, crews got hurt. this was done to protect the crew for 1945 production

removing the slight reduction in turret height

Israel was on the this track for lowering height, they went the hull for these test mules

Chrysler thought of what those would become, big gun, short hull, but wasn't approved.

Lowering the hull was missed opportunity for the whole Sherman line after the Radials without the dropped driveshaft were no longer in production


I think this is my quotas for images, so will stop for now

 

[don4331]

The biggest downside was low torque output at low RPMs, that was overshadowed by the great Horsepower to weight ratings.

Aircraft engine sit at cruise RPMs for most operations, where that low torque just never came into play.

The way to avoid all that, is adding the complexity of electric or even hydraulic drive, or what the Germans did with their high rpm, low torque V-12s, add more gears.

Why is everyone saying the German V-12 have low torque??

a. hp = torque (ft.lbs.) * rpm/5252...so if you are making 500hp at 2,600 rpm , you are making ~1,000 ft.lbs. of torque not matter if it is R-9/V-8/W-12 or V-12. That's pure physics.​

b. The Maybach HL230 is making peak torque at 2,100 rpm which compares favourably with the 2,200 rpm for say the Ford GAA. Especially when compared to the peak power (3,000 rpm vs 2,600 respectively). Maybach has 10% torque rise - which equates to "lugging" performance...i.e. the loss of rpm hp) is compensate by the increase in torque and you "power" thorough; Ford only has ~4%.​

​

Now, the overall performance (hp/ton) favours the Allied tanks but that isn't the raw performance of the engine.

Radial engines require a supercharger - it might not need to create much boost, but without one, the lower cylinders run rich/upper ones run lean as gravity causes gasoline in intake to fall. And 99% of radials use a centrifugal supercharger; a centrifugal supercharger makes boost at rpm^2. Given the supercharger has to have capacity to ensure proper mixture at 'cruise' rpm, it makes significantly more at 'peak' power rpm. Using GAA as example, impeller is making 40% more boost at 2,600 rpm vs 2,100.

So, the torque peak and the power peak are the same in a centrifugally supercharged engine. As a result, they have zero lugging ability. As soon as you start losing rpm, you're lost both torque and power. Therefore, a radial engine is a poor powerplant choice for an automotive vehicle, well at least without a CVT.

The Germans looked at electric* and hydraulic drive because they were lacking the materials (Manganese/Nickel/Chromium/Molybdenum) and tooling to produce large/quality gears/bearings. If you don't have the materials to make the final drive, where are you getting the materials to make a box to lower the driveline?

* Germany was equally short of copper, so I don't understand Porsche's fascination with electric drive - while it works very well, if you don't have the materials to make it there's no point designing it. A decent engineer knows better than to design something that can't be built. ​

 

[marathag]

Why is everyone saying the German V-12 have low torque??

V-12 Engines of the WWII era

Maybach HL230 1457 cubic inches 3000 rpm(2600 rec) 700HP Torque 1,364 ft. lbs@2100rpm

A-65 1568 cubic inch V-12 SOHC from Chrysler, 650HP@2600rpm Torque 1458 ft.lb. at 1600rpm

V-12 Hall-Scott 2181 cubic inches 575 BHP@2100 RPM brings ,1500 torque lb. ft.

Kharkiv V2 500 HP@1800 for 2368 cu. inches 1600 lbs/ft torque

Ford GAC V-12 1649 cubic inches max rpm 2800 Max torque 1560@1600

Lower RPMs for peak torque is better than higher 'Area under the Curve'.

German V-12 needed higher RPMS for top HP and torque than the others. To stay in the peak powerband, you need more gears.

 

[don4331]

I think I've mentioned in multiple threads that IMHO, gearbox in the rear is the correct(TM) design for a tank, so no argument here.

I've stayed away from commenting previously about the proper(TM) design for a WWII tank gearbox being in the front as I felt this was aircraft forum, not armour.

The primary reason being - it allows basically the length of the tank for debris (mud, rocks, flora) to be ejected from the track before the drive sprocket. Heck even in 2024, Caterpillar puts the drive sprocket up in all its large bulldozer, so at least some of the debris is ejected before it gets to the drive sprocket.​

Secondary reason - for manually shifted transmission, having short shift linkage is major advantage. >1/2 the recruits won't have seen an automotive vehicle - expecting them to operate/maintain a remote linkage/automatic transmission in the field is asking too much. Post war when you can provide proper training/users already have experience, it is whole different story.​

Additional reason - once you get beyond 37mm guns, the turret needs to be powered - when you're sitting on hill side the weight of gun and turret large enough for gun + armour against equal guns of opposition + crew are excessive for fast traverse manually. And if you're providing power forward to the turret, you might as well bring it all the way to the transmission.​

 

[marathag]

Secondary reason - for manually shifted transmission, having short shift linkage is major advantage. >1/2 the recruits won't have seen an automotive vehicle - expecting them to operate/maintain a remote linkage/automatic transmission in the field is asking too much. Post war when you can provide proper training/users already have experience, it is whole different story.

Shermans had all gears but 1st and Reverse synchronized, so new drivers didn't have to worry about double clutching. Soviet Tankers didn't get that Bourgeois luxury unless they were in Lend Lease.
same for most German Panzers, synchronization.
British also went for even more luxury, pneumatic pre-selector shifting- almost automatic.

 

[EwenS]

The M36 'Jackson' that the crews only called 'Slugger' should have been done this way
View attachment 810252

The only new build M36 were the 187 M36B1 built Oct-Dec 1944 on M4A3 hulls, to make up the required contract numbers.

Fisher built 300 turretless M10A1 hulls Dec 1943-Jan 1944 which were used to produce the first M36 with the addition of a new turret with the 90mm gun. Attention then turned to reconditioning M10A1 TD retained with training units in the USA and replacing their turrets. And finally as M10A1s began to run out attention turned in May 1945 to reconditioning diesel engined M10s as the M36B2.

So no opportunity to redo the the M36 drivetrain without spending a lot of money.

 

[don4331]

V-12 Engines of the WWII era

Maybach HL230 1457 cubic inches 3000 rpm(2600 rec) 700HP Torque 1,364 ft. lbs@2100rpm

A-65 1568 cubic inch V-12 SOHC from Chrysler, 650HP@2600rpm Torque 1458 ft.lb. at 1600rpm

V-12 Hall-Scott 2181 cubic inches 575 BHP@2100 RPM brings ,1500 torque lb. ft.

Ford GAC V-12 1649 cubic inches max rpm 2800 Max torque 1560@1600

Lower RPMs for peak torque is better than higher 'Area under the Curve'.

German V-12 needed higher RPMS for top HP and torque than the others. To stay in the peak powerband, you need more gears.

Let's throw out the diesel as its tough to compare very different fuels...

Then basically what your examples show - the Maybach is too small...

If you add 15% to the Maybach, to make it the same displacement as the Ford the torque would basically be identical. And if you supply better fuel to the Maybach, so it could run the same 7.5:1 CR (instead of 6.8:1) it would be better than the Ford!​

​

Higher rpms just means I need a lower final drive. E.g. If I make peak power at 3,000 rpms, I need 3:1 final gears* versus 2.8 final gear for an engine peaking at 2,800

*Tanks need a lot lower final gear ratio than 3:1, but I was just making the math simple.​

​

German tanks need more gears because their engines are smaller. Now, was that a good choice? The physical size difference between a Merlin and a Griffon wasn't much, but a 1/3 more power is significant.

On the other hand 1/3 more torque means instead of more gears, I need bigger and/or stronger gears/transmission/clutch/driveshaft/etc.​

Yes, the smaller engine slows down going up hill/across field, but its more economical to run on road.​

It doesn't change - radials aren't the right choice for AFVs.

 

[SaparotRob]

...
British also went for even more luxury, pneumatic pre-selector shifting- almost automatic.

And a tea kettle, right?