Jump to content

  • Log in with Facebook Log in with Twitter Log in with Windows Live Log In with Steam Log In with Google      Sign In   
  • Create Account

Photo
- - - - -

Air vs liquid cooled engines.


  • Please log in to reply
546 replies to this topic

#11 Flo

Flo

    Regular Member

  • Regulars
  • PipPipPip
  • 923 posts
  • Joined 4 Years, 7 Months and 24 Days
  • 46 topics

Posted 02 April 2012 - 07:00 PM

It's certainly possible that some other factor, such as a willingness to make multiple passes at a target, resulted in increased Thunderbolt losses.

Is it possible that the design of the aircraft was that factor?

Put another way, if it was the Typhoon who suffered significantly greater losses, would you look for external factors, or attribute them to the Typhoons reputed weaknesses; its' compressibility issues and the liquid cooled Sabre?

#12 NeoConShooter

NeoConShooter

    Banned

  • Banned
  • PipPipPipPip
  • 1,454 posts
  • Joined 3 Years and 12 Days
  • 5 topics

Posted 13 April 2012 - 07:37 PM

What's the consensus? Is an air cooled power plant less susceptible to battle damage? And if it is, does that make for a more survivable aircraft?

A few thoughts.

Firstly, 'survived heavy damage' means, by definition, 'was hit'. If hit by ground fire, this would almost certainly be from the front. Radials offer a large target area. The ability to fly with a cylinder shot off is rather impressive, but if a liquid cooled engine, with it's smaller profile, would have avoided the damage completely it wouldn't have to worry about it's vulnerable cooling system in the first instance.

The placement of that cooling system is therefore a major concern to the designer. Buried in the wings, away from the point of aim, may offer some protection. Sticking it behind the propeller, where incoming fire would already have inflicted heavy damage on the fighter, is more fatalistic. Placing it deep in the fuselage, surrounded by armour, is almost certain to defend it from most ground fire.

The ability to armour the engine more effectively in a liquid cooled power plant cannot be ignored. The cost, in weight, must be considered, but liquid cooled engines tended to offer more power, compensating somewhat.

So the question can be answered as either 'the liquid cooled engine is delicate, therefore more vulnerable' or 'the air cooled engine is bigger, therefore more vulnerable'.

Thoughts, gentlemen?

Some stats, to help out. Area/hp for Merlin 1200sq in/1500hp; R-1830 1809sq in/1200hp; Napier Sabre 1840sq in/2400hp; R-2800 2190sq in/2100hp.

Between July 25 and August 7, 1944 the 9th AF lost 80 aircraft, the majority Thunderbolts. Over the whole of those two months the 2nd TAF lost 151 Typhoons. The majority of losses to both were to ground fire. Typhoons and Thunderbolts were operating under very similar conditions, the Hawkers in direct support, the P-47s in close support, with considerable cross over in roles. During the Normandy operations the 9th AF flew 2891 sorties, the 2nd TAF 9896. The 9th AF claimed 2654 vehicles destroyed, the 2nd TAF 3597. The 2nd TAF claimed 257 of the 391 tanks destroyed by Allied aircraft, reflecting the slight difference in tasking between the two forces.
It's the closest example of air vs liquid cooled propulsion I can think of. Typhoon losses ran at about twice those of Thunderbolt equipped units, but they mounted three times the sorties. Just as well, since they made one third of the claims per sortie. It's possible to cut the figures to demonstrate various points- I fully expect Duncan and Lightning to do just that! :D- but I think they're close enough to refute claims of any superiority of either power plant.

I haven't mentioned air to air losses for a reason. There are a multitude of factors beyond the survivability of a single aircraft component involved in the loss of an aircraft. For example, late war Japanese fighters were much better protected than earlier models, but they suffered far greater attrition.

The figures above are largely derived from this pamphlet: http://www.ibiblio.o...DDay/index.html
I was going to use a Brit source, but then I remembered I may have let slip that I'm a bit of a Hawker fan! ;) I'm pretty sure Richard Hallion holds that air cooled engines are tougher, amongst other things. Any primary documentation countering anything I've scrawled above will be gratefully received. Fire away, guys...


In that they do not include the cooling systems for LC engines. Once that is included, the vulnerable area to power ratio changes to favor air cooled engines. Also, the AC engine has a very much smaller vulnerable area in that any part of the LC engine that is hit by even a rifle caliber bullet will eventually down the plane. But radials require hits from powerful cannon to damage the engine and the resulting area is much smaller in that it either takes more than one/two hits to stop the engine and they must be in a much smaller location, IE the crank case as you point out, shooting even 3-4 of the 18 cylinders off of a R-2800 will not stop the engine, most times. In addition, RC bullets will not damage the heavy metal fins that are the cooling device on air cooled engines. A .22 LR will perforate the cooling jacket of any engine, or radiator, or hose, etc.

Your point about the location of auxiliary equipment is also fallacious in that the point of aim is the entire plane and dispersion of the bullets in the projectile stream mean that regardless of the exact point of aim, the very few hits that are made are truely random in nature. A RCMG will have a cone of fire between 2-3 1/2' per 100 M. A heavy MG with it's better mounting is ~40" / 500 M. About the same as a wing mounted cannon, but at least twice that of an engine mounted gun. A .50 caliber M2/3 HMG is 15" radius at 600 yards from a equivalent mount. A ground mounted, manually pointed cannon is lucky to keep 10% of it's shells with in 105' of the moving target at 1,000 M! Or 330' at 2000 M.

I know that these stats sound incredible, but they are best case examples! Because there are so many more RCMGs than there are HMGs and many times more HMGs than cannon, the smaller caliber weapons have a very deadly record. One ~8 MM slug in the radiator, oil cooler, water jacket, hose(s), pumps, holding tank(s), oil pan below the fill line and that engine WILL STOP running sooner than later! That is many square feet of vulnerable target area. The air cooled radial on the other hand does not have a pressurized coolant or oil cooler bathed in Prestone to kill it quick. The only point of vulnerability is the crank case below the fill line, the oil hoses and tank. Total target area about 1/3 to 1/4 that of the LC engine!

Instead of guessing about areas, why not take the two planes you mentioned and see what one random hit at a time would do to each plane in the same relative location. If you do that say ten times you should get an idea about the relative vulnerability of the two planes. Then you will see that the plane with the largest target area will be the most vulnerable!


#13 Wuzak

Wuzak

    Forum Guru

  • Forum Guru
  • PipPipPipPipPip
  • 3,720 posts
  • Joined 9 Years, 11 Months and 24 Days
  • 158 topics

Posted 14 April 2012 - 12:06 AM

Air cooled engines do need oil coolers, and they are bigger for an air-cooled engine with the same power as a liquid cooled engine.

Loss of a head or a hole in the crankcase will cuase the engine to lose oil, so it will sieze at some stage.

In the case of the R-2800 losing "3-4" of its cylinder heads the pilot will have been extremely lucky that the heads were taken off without causing damage to the pistons or cylinders, each of which could cause a catastrophic failure of the engine.

#14 Flo

Flo

    Regular Member

  • Regulars
  • PipPipPip
  • 923 posts
  • Joined 4 Years, 7 Months and 24 Days
  • 46 topics

Posted 14 April 2012 - 10:15 AM

But you're largely speculating and applying what you perceive to be common sense in your rebuttal.

In that they do not include the cooling systems for LC engines. Once that is included, the vulnerable area to power ratio changes to favor air cooled engines. Also, the AC engine has a very much smaller vulnerable area in that any part of the LC engine that is hit by even a rifle caliber bullet will eventually down the plane. But radials require hits from powerful cannon to damage the engine and the resulting area is much smaller in that it either takes more than one/two hits to stop the engine and they must be in a much smaller location, IE the crank case as you point out, shooting even 3-4 of the 18 cylinders off of a R-2800 will not stop the engine, most times. In addition, RC bullets will not damage the heavy metal fins that are the cooling device on air cooled engines. A .22 LR will perforate the cooling jacket of any engine, or radiator, or hose, etc.

Your point about the location of auxiliary equipment is also fallacious in that the point of aim is the entire plane and dispersion of the bullets in the projectile stream mean that regardless of the exact point of aim, the very few hits that are made are truely random in nature. A RCMG will have a cone of fire between 2-3 1/2' per 100 M. A heavy MG with it's better mounting is ~40" / 500 M. About the same as a wing mounted cannon, but at least twice that of an engine mounted gun. A .50 caliber M2/3 HMG is 15" radius at 600 yards from a equivalent mount. A ground mounted, manually pointed cannon is lucky to keep 10% of it's shells with in 105' of the moving target at 1,000 M! Or 330' at 2000 M.

I know that these stats sound incredible, but they are best case examples! Because there are so many more RCMGs than there are HMGs and many times more HMGs than cannon, the smaller caliber weapons have a very deadly record. One ~8 MM slug in the radiator, oil cooler, water jacket, hose(s), pumps, holding tank(s), oil pan below the fill line and that engine WILL STOP running sooner than later! That is many square feet of vulnerable target area. The air cooled radial on the other hand does not have a pressurized coolant or oil cooler bathed in Prestone to kill it quick. The only point of vulnerability is the crank case below the fill line, the oil hoses and tank. Total target area about 1/3 to 1/4 that of the LC engine!

Instead of guessing about areas#, why not take the two planes you mentioned and see what one random hit at a time would do to each plane in the same relative location. If you do that say ten times you should get an idea about the relative vulnerability of the two planes. Then you will see that the plane with the largest target area(#2) will be the most vulnerable!


We've all heard survivor stories, but be wary of drawing broad conclusions from narrow results. As Wusak posted, not every cylinder hit is survivable. By the same token, not every hit to a cooling system is fatal. George Unwin recalls surviving machine gun hits to his Spitfire engine with little effect, that doesn't mean I'd be willing to claim that Merlins were particularly resilient to mg fire! ;)
Either way, as I posted above, there are different options for placement of cooling systems. But reguardless of whether I'm right or wrong in my assumptions, my question stands- does the choice of cooling system negatively effect the survival prospects of an aircraft? If you believe so, can you provide figures to back your proposition? Not common sense arguments, or thought experiments, but specific, real world instances?

#We've already done this in the UK. It was what persuaded us to develop rapid fire cannons in the first place. :D

#2 This is why I linked to a pro P-47, American written article, comparing aircraft of similar size, performance and role, operating alongside one another, for loss figures. The 9th and 2nd trained on the same bombing ranges, with the same staff, prior to D-Day, so even their aircrew training was directly comparable. I cannot think of a clearer example of where the only major difference between aircraft performing the same mission against the same foe was the type of engine employed.
That difference doesn't seem to have increased the survivability of the air cooled fighter.
Of course, my friend Ickysdad has a quite extensive library. I'm rather hoping he'll join us. Perhaps he can provide some more precise figures for the ninths P-47 losses. They might demonstrate the relationship you appear to favour. I'd take that as a definitive answer to my question, unless someone could post convincing evidence otherwise. Would you accept the most obvious conclusion if they do not?

Edited by Flo, 15 April 2012 - 12:44 AM.


#15 NeoConShooter

NeoConShooter

    Banned

  • Banned
  • PipPipPipPip
  • 1,454 posts
  • Joined 3 Years and 12 Days
  • 5 topics

Posted 19 April 2012 - 06:25 AM

Air cooled engines do need oil coolers, and they are bigger for an air-cooled engine with the same power as a liquid cooled engine.

Loss of a head or a hole in the crankcase will cuase the engine to lose oil, so it will sieze at some stage.

In the case of the R-2800 losing "3-4" of its cylinder heads the pilot will have been extremely lucky that the heads were taken off without causing damage to the pistons or cylinders, each of which could cause a catastrophic failure of the engine.

Do a Google search for T-Bolt pics with damage. There are literally dozens with several entire cylinders gone that flew back to england after getting hit.

#16 NeoConShooter

NeoConShooter

    Banned

  • Banned
  • PipPipPipPip
  • 1,454 posts
  • Joined 3 Years and 12 Days
  • 5 topics

Posted 19 April 2012 - 06:48 AM

But you're largely speculating and applying what you perceive to be common sense in your rebuttal.



We've all heard survivor stories, but be wary of drawing broad conclusions from narrow results. As Wusak posted, not every cylinder hit is survivable. By the same token, not every hit to a cooling system is fatal. George Unwin recalls surviving machine gun hits to his Spitfire engine with little effect, that doesn't mean I'd be willing to claim that Merlins were particularly resilient to mg fire! ;)
Either way, as I posted above, there are different options for placement of cooling systems. But reguardless of whether I'm right or wrong in my assumptions, my question stands- does the choice of cooling system negatively effect the survival prospects of an aircraft? If you believe so, can you provide figures to back your proposition? Not common sense arguments, or thought experiments, but specific, real world instances?

#We've already done this in the UK. It was what persuaded us to develop rapid fire cannons in the first place. :D

#2 This is why I linked to a pro P-47, American written article, comparing aircraft of similar size, performance and role, operating alongside one another, for loss figures. The 9th and 2nd trained on the same bombing ranges, with the same staff, prior to D-Day, so even their aircrew training was directly comparable. I cannot think of a clearer example of where the only major difference between aircraft performing the same mission against the same foe was the type of engine employed.
That difference doesn't seem to have increased the survivability of the air cooled fighter.
Of course, my friend Ickysdad has a quite extensive library. I'm rather hoping he'll join us. Perhaps he can provide some more precise figures for the ninths P-47 losses. They might demonstrate the relationship you appear to favour. I'd take that as a definitive answer to my question, unless someone could post convincing evidence otherwise. Would you accept the most obvious conclusion if they do not?


P-47
Total Sorties = 423,435
Bomb tonnage = 113,963
Lost in combat = 3,077
Enemy AC Kills = 3,082
Destroyed Grnd 3,202
Combat loss/Sortie 0.7%

Just some numbers to add fuel to the fire. I do not know the Typhoon numbers. If there is a significant differance between the two Plane's Mission loss and bomb tonnage rates, then I would say there might be something to this????

Just off of the top of my head... I think that when you look at the bomb tonnage per mission you will find that the T-Bolt was top of the heap in Ground Support? I will also go out on a limb and think the T-Bolt will have a lower total loss per mission ratio. I could be wrong because I really do not have a clue what the Typhoon's numbers are.
Cherry picking any small section of days might not yeild a true picture of the values?

#17 Ricky

Ricky

    Moderator

  • Moderators
  • 2,510 posts
  • Joined 12 Years, 2 Months and 7 Days
  • 127 topics

Posted 19 April 2012 - 10:25 AM

Just some numbers to add fuel to the fire. I do not know the Typhoon numbers. If there is a significant differance between the two Plane's Mission loss and bomb tonnage rates, then I would say there might be something to this????

Just off of the top of my head... I think that when you look at the bomb tonnage per mission you will find that the T-Bolt was top of the heap in Ground Support? I will also go out on a limb and think the T-Bolt will have a lower total loss per mission ratio. I could be wrong because I really do not have a clue what the Typhoon's numbers are.
Cherry picking any small section of days might not yeild a true picture of the values?


There will be a significant difference between bomb tonnage rates, because Typhoons used rockets more often than P-47s did (though both used both).
If you want the best tactical aircraft to be decided by bomb tonnage per sortie than the Lanc wins (they were used in a tactical role during Goodwood) followed by the B-17 or B-24 (whichever one was used in Cobra).

It is very hard to work out the effectiveness of ground-attack aircraft - claims are always massively inflated, and generally hard to substantiate, even when you do subsequently conquer the territory. And it is pointless dropping 15000 lbs of bombs (or firing 15000 rockets) if they all miss...

Loss rates are a better comparison, but even here there are differences. Mission profiles, ordnance used, theatre of operation, target types... all have an impact on loss rates. That was why Flo selected a very limited date and place range, because within that window you have P-47s and Typhoons flying the same mision type, against the same targets, from the same bases, using the same or similar tactics. Given that as many variables as possible have been removed, the total figures for that period should be our best bet for figuring out which plane was more vulnerable. All we need are the full P-47 figures.

#18 NeoConShooter

NeoConShooter

    Banned

  • Banned
  • PipPipPipPip
  • 1,454 posts
  • Joined 3 Years and 12 Days
  • 5 topics

Posted 05 June 2012 - 01:55 AM

But you're largely speculating and applying what you perceive to be common sense in your rebuttal.

Either way, as I posted above, there are different options for placement of cooling systems.It seems that the P-39 has something to say for this and the best L/sortie rate of all AC in WW-II? 0.4%/ sortie??? But reguardless of whether I'm right or wrong in my assumptions, my question stands- does the choice of cooling system negatively effect the survival prospects of an aircraft? If you believe so, can you provide figures to back your proposition? Not common sense arguments, or thought experiments, but specific, real world instances?

#We've already done this in the UK. It was what persuaded us to develop rapid fire cannons in the first place. :D

#2 This is why I linked to a pro P-47, American written article, comparing aircraft of similar size, performance and role, operating alongside one another, for loss figures. The 9th and 2nd trained on the same bombing ranges, with the same staff, prior to D-Day, so even their aircrew training was directly comparable. I cannot think of a clearer example of where the only major difference between aircraft performing the same mission against the same foe was the type of engine employed.
That difference doesn't seem to have increased the survivability of the air cooled fighter.
Of course, my friend Ickysdad has a quite extensive library. I'm rather hoping he'll join us. Perhaps he can provide some more precise figures for the ninths P-47 losses. They might demonstrate the relationship you appear to favour. I'd take that as a definitive answer to my question, unless someone could post convincing evidence otherwise. Would you accept the most obvious conclusion if they do not?


1. My uncle RTB'd twice with the P-47 shot all to hell and back, once with 22 20 mm, two 30 MM and >200 MG hits. I've never even heard of anything even remotely like that for any other plane.:eek:
2. From page 232 of American combat planes by Ray Wagner;
The P-47 flew 423,435 missions, dropped 113,963 tonnes of bombs, lost 3,077 in combat, shot down 3,082 and got 3,202 more on the ground and had a loss rate per Sortie of 0.7%.
P-51's numbers are; 213,873 sorties, 5,688 tonnes of bombs, 2,520 lost, 4,950-4,131 kills and losses per sortie of 1.2%.
The P-38 was; 129,849 sorties, 20,139 tonnes of bombs, 1,758 lost and 1,771 and 749 kills, for 1.4% lost/sortie.
The numbers for all other planes in US service Except the P-39 were not as good and they all flew a much lower %age of missions A2G.
I wonder what the other Allied and Axis planes mentioned here did for the entire war, instead of for a small percentage of the war? Say at least 50-100K sorties?

#19 NeoConShooter

NeoConShooter

    Banned

  • Banned
  • PipPipPipPip
  • 1,454 posts
  • Joined 3 Years and 12 Days
  • 5 topics

Posted 05 June 2012 - 02:01 AM

Loss rates are a better comparison, but even here there are differences. Mission profiles, ordnance used, theatre of operation, target types... all have an impact on loss rates. That was why Flo selected a very limited date and place range, because within that window you have P-47s and Typhoons flying the same mision type, against the same targets, from the same bases, using the same or similar tactics. Given that as many variables as possible have been removed, the total figures for that period should be our best bet for figuring out which plane was more vulnerable. All we need are the full P-47 figures.


I agree with all you said. It's just that bombs or rockets we used both at the same time sometimes tend to show how many times you were moving dirt. In the case of the P-47 that is about 1/4 of the time. If you are using bombs or rockets, you are probably not shooting things down??? So thses numbers would be veri indicative of who did what to whom?:rolleyes:

#20 Flo

Flo

    Regular Member

  • Regulars
  • PipPipPip
  • 923 posts
  • Joined 4 Years, 7 Months and 24 Days
  • 46 topics

Posted 05 June 2012 - 11:08 AM

Is that any attempt to 'cut' your figures is likely to show variation based on a great many factors. To get to a figure set that only reveals the difference between one distinct factor-power plants, say- you have to discount as many of those other factors as possible. Simply lumping every mission ever flown into a set won't tell you a great deal about why an aircraft performed the way it did. Ask Kiwi or Wuzak to explain the difference between descriptive and inferential statistics or what constitutes a statistically viable sample, if my layman's understanding seems off.

"I wonder what the other Allied and Axis planes mentioned here did for the entire war, instead of for a small percentage of the war? Say at least 50-100K sorties?"
What they did was follow a completely different development cycle, adapting to completely different circumstances.
Your own figures demonstrate that. Note that the ground attack specialist (P-47) had far fewer aerial 'kills' than the escort fighter (P-51), but much more ordinance delivered.
Note too that using only those figures we could confidently demonstrate that ground attack was less hazardous than escort work...
In reality the figures only demonstrate that the aircraft were employed differently. More accurate conclusions would require more precise data.
One way would be selecting an equivalent sample of P-47s and P-51s performing the same mission, against the same opposition, at the same time. This approach would offer a far better insight into the capabilities of the two airframes than pointing at '50-100K sorties' and trying to distil some sense out of it.

Edited by Flo, 05 June 2012 - 11:47 AM.





0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

AVIATION TOP 100 - www.avitop.com Avitop.com