Time Machine ENGINE consultant!!!

[gorizont]

It is different running an engine in a test cell than it is on an airplane, but doing these tests establishes reliability numbers for the engine and accessories as well as allowing the manufacturer to establish overhaul criteria.

Yes, You are right. Including the note "It is different running an engine in a test cell than it is on an airplane"/

From Pratt and Whitney R-2800 Double Wasp - USA :
" Perhaps the most outstanding example was the great R-2800 Double Wasp, which went into production in 1940 for the B-26 Marauder at 1,850 hp and by 1944 was in service in late model P-47 Thunderbolts (and other aircraft) at a rating of 2,800 (experimental) hp on 115-grade fuel with water injection. Of course, all engines naturally grow in power with development, but a major war demands the utmost performance from engines fitted to aircraft, whose life in front-line service was unlikely to exceed 50 hours' flying time over a period of only a month or two.

In peace time, the call was for reliability over a period of perhaps a dozen years. "

 

[FLYBOYJ]

Yes, You are right. Including the note "It is different running an engine in a test cell than it is on an airplane"/

From Pratt and Whitney R-2800 Double Wasp - USA :
" Perhaps the most outstanding example was the great R-2800 Double Wasp, which went into production in 1940 for the B-26 Marauder at 1,850 hp and by 1944 was in service in late model P-47 Thunderbolts (and other aircraft) at a rating of 2,800 (experimental) hp on 115-grade fuel with water injection. Of course, all engines naturally grow in power with development, but a major war demands the utmost performance from engines fitted to aircraft, whose life in front-line service was unlikely to exceed 50 hours' flying time over a period of only a month or two.

In peace time, the call was for reliability over a period of perhaps a dozen years. "

In peacetime, ailiners that operated R2800s had between 1,000 and 1,500 hour overhaul periods assigned to them by the manufacturer and the FAA.

 

[gorizont]

When you mess with R-2800, you mess with me

As you wish, sir

When we calculate in the cooling system, weight difference is negligible. We could talk about drag, yet P-47M -N were every bit as fast as anything powered by piston engine.

You forget the calculate weight for wider cowling for a radial, its cowl shutter and its mechanization, for P-47 weight of turbo-charger and its tubes. So...
And, you know, cooling system of one aircraft differes from cooling system of another one. As you remerber, for instance, F4U-1 had an interesting system. Intakes for its intercoler and oil cooler were mounted in the roots of wings. And we have some extentions of their pipesto the engine in that aircraft.

Then we could talk about resilience to battle damage, and radials come 1st there.
As for development speed and money invested, it was liquid-cooled/inline engines that were suffering big-time setbacks (Jumo-222 and a plethora of US engines - Chrysler and big Allisons for example).

Crysler and Allison never had such goverment funding.
What about resilience to battle damage - I'm not going to argue this argument, but l suppose that degree of superiority of radials over inline engines to some degree overstated.
Volume of oil system of large radials is as a rule greater than of inlines. one of the reasons - at large radials oil system played a role of cooling subsystem. And this role of oil system for radials is greater than for inlines.
Look at pipe net on R-2800 - http://www.aviation-history.com/engines/pr-2800-2.jpg
And you know, what happened to an engine if its oil system is damaged.

 

[Shortround6]

Don't forget. for every "green" pilot who gets shot down/crashes on his first flight there has to be another plane that goes just over 100hrs to bring the average to 50hrs

Over building and over testing ensures there is also a margin for less than optimum conditions, like North African dust and sand.

Look at the testing that was done and then read the combat reports/squadron diaries that mention engine problems and ask yourselfs if building engines with a 50 hr life would have really been smart.

 

[gorizont]

Continuing.
I should say that in my notes I mean militiry air engines not for commerchial usege and moreover engines for fighters.
To simplify sketch let's put aside the matter concerned super- and turbo-chargers.

Soif now we put aside details and look at general ways - we have three ways to increase power output of the aero-engine :
1) to give more boost (this connected with a) super- and turbocharger construction, b) quality of fuel and c) strength of engine details)
2) to increase compression ratio (connected with quality of fuel and strength of engine details)
3) to increase RPM.

If you look through performance of large radial of WWII you'll see that limit of RPM for them is 2800 (for instance it's max RPM for R-3350, the limit for R-2800 is lower - just 2700, as far as I know).
We have two exceptions, about first I will mention later, about second - it is Homare 23, which lack of reliability is widely known.

But other thing about inlines - for instance max RPM of Jumo-213 was a figure of 3250.
DB engineers was going to increase power output of DB-603 besides the rest of the measures by increasing RPM from 2700 to 3000.
So You'll see that for inlines to increase power output engeneers can use three ways but for radials - only two.
What is the reason? The main reason - Valvetrain.
It lead to two limitations :1) Limited engine speeds or RPM — OHV engines have more valvetrain moving parts, thus more valvetrain inertia and mass, as a result they suffer more easily from valve "float", and may exhibit a tendency for the pushrods, if improperly designed, to flex or snap at high engine speeds. Therefore, OHV engine designs cannot revolve ("rev") at engine speeds as high as OHC

2) Limited cylinder head design flexibility — overhead camshaft (OHC) engines benefit substantially from the ability to use multiple valves per cylinder, as well as much greater freedom of component placement, and intake and exhaust port geometry. Most modern OHV engines have two valves per cylinder, while many OHC engines can have three, four or even five valves per cylinder to achieve greater power. Though multi-valve OHV engines exist, their use is somewhat limited due to their complexity and is mostly restricted to low and medium speed diesel engines. In OHV engines, the size and shape of the intake ports as well as the position of the valves are limited by the pushrods

(taken from Overhead valve - Wikipedia, the free encyclopedia)

So look at R-2800 especially at http://www.enginehistory.org/P&W/engi0000t.jpg

Now about first exception. It is such a radial as Bristol Hercules. It was forced to 1735 HP at Hercules VII, and its RPM was of 2900.
But - it has sleeve valves not pushroads and rocker arms!

 

[Shortround6]

Soif now we put aside details and look at general ways - we have three ways to increase power output of the aero-engine :
1) to give more boost (this connected with a) super- and turbocharger construction, b) quality of fuel and c) strength of engine details)
2) to increase compression ratio (connected with quality of fuel and strength of engine details)
3) to increase RPM.

# 2 is out. For any given fuel, say 100 octane, any ONE engine is going to have a limit on the allowable compression ratio/ boost combination it can use. Different engines have differnt limits on the same fuel. Going back to our "test" engine and the 100 ocatne fuel, if you raise the compression you have to lower the boost limit. The result is better fuel economy but less max power. On an existing engine you might be able to LOWER the compresion ratio, accept the worse fuel economy and raise the boost to get more power than the original engine.
Improving the fuel will let you raise one or the other or a little bit of both but higher boost will always give you more power than raising the compression.

So You'll see that for inlines to increase power output engeneers can use three ways but for radials - only two.
What is the reason? The main reason - Valvetrain.
It lead to two limitations :1) Limited engine speeds or RPM — OHV engines have more valvetrain moving parts, thus more valvetrain inertia and mass, as a result they suffer more easily from valve "float", and may exhibit a tendency for the pushrods, if improperly designed, to flex or snap at high engine speeds. Therefore, OHV engine designs cannot revolve ("rev") at engine speeds as high as OHC

Nope. The reason was all those cylinders acting on one crankpin. Or to put it another way, trying to increase revolutions on the big master rod and 6/8 link rods rapidly overloaded the Bearing. Forces/loads acting on the bearing go up with SQUARE of the engine speed.
The commonly quoted piston speed specification is not really related to piston ring failure but gives an idea of the loads on the crankshaft bearings. There is a formula for corrected piston speed that helps take into account piston weight. Under square engines get a slight reduction in piston speed while over square (large bore) engines get an increase in the rating.
V-12 engines, in general, Spread the load out over more bearings than a radial but had problems of their own. They are however, much more capable of higher rpm operation than a radial. They are also heavier than a radial of equel displacement.

2) Limited cylinder head design flexibility — overhead camshaft (OHC) engines benefit substantially from the ability to use multiple valves per cylinder, as well as much greater freedom of component placement, and intake and exhaust port geometry. Most modern OHV engines have two valves per cylinder, while many OHC engines can have three, four or even five valves per cylinder to achieve greater power. Though multi-valve OHV engines exist, their use is somewhat limited due to their complexity and is mostly restricted to low and medium speed diesel engines. In OHV engines, the size and shape of the intake ports as well as the position of the valves are limited by the pushrods

All the Bristol poppet valve radial engines used 4 valves per cylinder. The Jupiter was licensed to around 17 countries? All WW II Mercury and Pegasus engines use 4 valves per cylinder.

Some overhead cam engines used truely terriable intake and exhaust ports and passages, see the Hispano V-12s

Now about first exception. It is such a radial as Bristol Hercules. It was forced to 1725 HP at Hercules VII, and its RPM was of 2900.
But - it has sleeve valves not pushroads and rocker arms!

It also has a piston speed of 3,142fpm which is rather high for an aircraft engine. Hercules engines from the 1950s were pushed to around 2100hp but they kept the same rpm even though the the crankcase was redesigned with larger roller bearings to handle the increased loads.
A P&W R-2800 has a piston speed of 2800fps at 2800rpm.
the R-3350 has a piston speed of 3050.8 at 2900rpm.

THe old Pegasus had one of the highest at 3250fps at 2600rpm but then few other engines used a 7.5in (190mm) stroke. One that did was the Russian AM-38 engine used in the IL-2 but then they didn't rev that engine very high did they, inspite of the overhead cam and 4 valves per cylinder.

Small radials that reved high were the Bristol Taurus and the Gnome-Rhone "M" series. The Taurus hit 3100rpm, in part due to it's short stroke (143mm) which held the piston speed to 2,906. The Gnome-Rhone 14M hit 3030rpm but again it's short stroke held piston speed to a mere 2258fpm.

The Napier Sabre engine, a real Champ when it comes to RPM at 3850 kept it's piston speed down to 3,048fps due the 120mm stroke.

I will conclude this by noting that the Merlin had a piston speed of 3000fps and the Griffon had a piston speed of 3025fps.

 

[Vincenzo]

REMEMBER -- NO engine is EVER going to run more than 80 to 150 hours before a major overhaul, so testing them for 3000 hours IS STUPID.

What about test of R-2800-63 (P-47 engine) run at WEP (2,700rpm / 2,600hp) for 7-1/2 hours when it wasn't to be used for more than 5 minutes.

it's a test and it's common long endurange and are 7 and half hour in 3 days (around 5 minutes for hour)

 

[tomo pauk]

...
You forget the calculate weight for wider cowling for a radial, its cowl shutter and its mechanization, for P-47 weight of turbo-charger and its tubes. So...

The inlines had longer cowling, so that evens things out. Radiators also had cowl shutter mechanization, again to even the weight.
P-47 put extra weight to a goo use, so nothing to put radials to the disadvantage there.

And, you know, cooling system of one aircraft differes from cooling system of another one. As you remerber, for instance, F4U-1 had an interesting system. Intakes for its intercoler and oil cooler were mounted in the roots of wings. And we have some extentions of their pipesto the engine in that aircraft.

Yes, I know

Crysler and Allison never had such goverment funding.
What about resilience to battle damage - I'm not going to argue this argument, but l suppose that degree of superiority of radials over inline engines to some degree overstated.

With one important system less to be hit, damaged, destroyed, radials had were in advantage.

Volume of oil system of large radials is as a rule greater than of inlines. one of the reasons - at large radials oil system played a role of cooling subsystem. And this role of oil system for radials is greater than for inlines.
Look at pipe net on R-2800 - http://www.aviation-history.com/engines/pr-2800-2.jpg
And you know, what happened to an engine if its oil system is damaged.

Hmmm...Engine stops?

.

 

[davebender]

The Me-109 had two wing radiators. Either of these could be isolated by the pilot in case of damage. This should have helped as you will have half your cooling system remaining even when damaged.

The cowl mounted radiator of the Fw-190D9 is another way to mitigate cooling system vulnerability. Hitting the radiator means hitting the engine also. Once the engine is shot full of holes the radiator no longer matters.

Wth well thought out designs like these I suspect liquid cooled aircraft engines aren't much more vulnerable then air cooled engines.

 

[Bronc]

Moderator note - you did not start this thread so you will not hijack with your own agenda. Please refer to the first post.

http://www.ww2aircraft.net/forum/aviation/time-machine-engine-consultant-21643.html#post585013

If you want to start another tread with your agenda, go right a head

 

[gorizont]

# 2 is out. For any given fuel, say 100 octane, any ONE engine is going to have a limit on the allowable compression ratio/ boost combination it can use. Different engines have differnt limits on the same fuel. Going back to our "test" engine and the 100 ocatne fuel, if you raise the compression you have to lower the boost limit. The result is better fuel economy but less max power. On an existing engine you might be able to LOWER the compresion ratio, accept the worse fuel economy and raise the boost to get more power than the original engine.
Improving the fuel will let you raise one or the other or a little bit of both but higher boost will always give you more power than raising the compression.

It's true just for an engine improved up to the limits of its design. But almost all the late-war miltary aircraft engines (especially for fighters) didn't reach their limits (at least to the end of war, some of them were upgraded during after-war years), IMHO - the era of turbo-jets started earlier of that event.
Look at a couple of examples. First - BMW-801. BMW-801D has higher boost and higher compression, and has a higher power output and lower fuel consamption on the power regime than BMW-801C. But it's not a pure example because it consumed C3 fuel instead of B4 (for BMW-801C).
But we have a pure example - Homare 23.
Both Homare 23 mod.12 and mod.21 consumed fuel with octine number "92" (with ADI). But for mod.12 compression ratio was 7.0, boost - 45.7", max. power output - 1840 HP; for mod.21 compression ratio - 8.0, boost - 49.6", power ouput max. - 2050 HP (according to TAIC). Figures of fuel consumption is unknown to me.

Nope. The reason was all those cylinders acting on one crankpin. Or to put it another way, trying to increase revolutions on the big master rod and 6/8 link rods rapidly overloaded the Bearing. Forces/loads acting on the bearing go up with SQUARE of the engine speed.
The commonly quoted piston speed specification is not really related to piston ring failure but gives an idea of the loads on the crankshaft bearings. There is a formula for corrected piston speed that helps take into account piston weight. Under square engines get a slight reduction in piston speed while over square (large bore) engines get an increase in the rating.
V-12 engines, in general, Spread the load out over more bearings than a radial but had problems of their own. They are however, much more capable of higher rpm operation than a radial. They are also heavier than a radial of equel displacement.

It's rather interesting point of veiw. So - you are sure, that the problem called "floating of valves" is fabricated, aren't You? And tuning of input-exhaust cycle (I'm not sure that use a proper term, but I hope you guess what I mean) is matter of unimportance, isn't it?

All the Bristol poppet valve radial engines used 4 valves per cylinder. The Jupiter was licensed to around 17 countries? All WW II Mercury and Pegasus engines use 4 valves per cylinder.

Some overhead cam engines used truely terriable intake and exhaust ports and passages, see the Hispano V-12s

It also has a piston speed of 3,142fpm which is rather high for an aircraft engine. Hercules engines from the 1950s were pushed to around 2100hp but they kept the same rpm even though the the crankcase was redesigned with larger roller bearings to handle the increased loads.
A P&W R-2800 has a piston speed of 2800fps at 2800rpm.
the R-3350 has a piston speed of 3050.8 at 2900rpm.

THe old Pegasus had one of the highest at 3250fps at 2600rpm but then few other engines used a 7.5in (190mm) stroke. One that did was the Russian AM-38 engine used in the IL-2 but then they didn't rev that engine very high did they, inspite of the overhead cam and 4 valves per cylinder.

Small radials that reved high were the Bristol Taurus and the Gnome-Rhone "M" series. The Taurus hit 3100rpm, in part due to it's short stroke (143mm) which held the piston speed to 2,906. The Gnome-Rhone 14M hit 3030rpm but again it's short stroke held piston speed to a mere 2258fpm.

The Napier Sabre engine, a real Champ when it comes to RPM at 3850 kept it's piston speed down to 3,048fps due the 120mm stroke.

I will conclude this by noting that the Merlin had a piston speed of 3000fps and the Griffon had a piston speed of 3025fps.

It's rahter interesting but as most of examles you took designs of one company - Bristol which are not tipical at all in design (if we talk about radials). And engineers from Bristol took part in development of Napier Sabre, as you know - they took care of improving its valvetrain
Moreover to use a term "piston speed" I suppose for sleeve-valve system is not completelly correct. Maybe "speed of opening\closing valve apertures?

If to speak about classical radials we should take R-2800 or R-3350, I suppose. With their one intake and one exaust valves.

 

[Shortround6]

It's true just for an engine improved up to the limits of its design. But almost all the late-war miltary aircraft engines (especially for fighters) didn't reach their limits (at least to the end of war, some of them were upgraded during after-war years), IMHO - the era of turbo-jets started earlier of that event.
Look at a couple of examples. First - BMW-801. BMW-801D has higher boost and higher compression, and has a higher power output and lower fuel consamption on the power regime than BMW-801C. But it's not a pure example because it consumed C3 fuel instead of B4 (for BMW-801C).
But we have a pure example - Homare 23.
Both Homare 23 mod.12 and mod.21 consumed fuel with octine number "92" (with ADI). But for mod.12 compression ratio was 7.0, boost - 45.7", max. power output - 1840 HP; for mod.21 compression ratio - 8.0, boost - 49.6", power ouput max. - 2050 HP (according to TAIC). Figures of fuel consumption is unknown to me.

I am not sure about the German and Japanese engines, there are things that can change an engines ability to cool itself. For instance the later P&W R-2800s used an alumimium "muff" of cooling fins over a steel barrel vrs the machined steel fins of the earlier models and they also used a forged cylinder head with different finning than the early models. These modifications ment a lower cylinder temperature at a given level of power and so allowed a little more boost to be used.

ADI complicates things. It does two things. One is that it evaporates in the supercharger/intake passages and absorbs heat in doing so. this lowers the intake charge temperature and allows more boost (or compression ) to be used before the same critical temperature is reached. THe other is that it actually helps cool the inside of the cylinder and so agian helps lower the temperature of any "hot spots" that might cause detonation. The complication comes in because we don't know how much ADI was being used. THE More ADI per minute the greater the effect (up to a point) but can the engine stand it, I mean ther greater power output, and how long will the ADI last.
Each engine design acts a bit differently and just becasue engine "A" can reach a certain BMEP with "X" compression ratio and "Y" boost and "Z" amount of ADI does npt mean that engine "B" can do the same thing.

It's rather interesting point of veiw. So - you are sure, that the problem called "floating of valves" is fabricated, aren't You? And tuning of input-exhaust cycle (I'm not sure that use a proper term, but I hope you guess what I mean) is matter of unimportance, isn't it?.

You can float valves, it usually happens in car or motorcycle engines operating at much higher rpms than aircraft engines. While the Aircraft valves and push rods are much larger and heavier than car/cycle valves and pushrods we can go back to the speed thing. Increase the speed of the valve or push rod opening or closing by 22.4% and you have increase the energy in the valve or pushrod by 50%. Changing a car or cycle engien to a race engine can certainly bring on valve problems. Aircraft engines usually had much more modest increases in RPM. And with proper testing they could be controlled with the proper strength valve spring. Stronger valve springs to create more friction on the camshaft or cam ring and this friction does affect fuel economy a tiny bit.
Changing valve springs (double or even triple springs were used) seems like a smaller problem than redesigning crankshafts, big end bearings, and even crankcases.

Restrictive intake and exhaust porting, not enough valve area, restrictive manifolds and even restrictive intake ducts can all hurt an engines performance. Some of these problems can be gotten around by just using a little more boost from the supercharger. at least somewhat.
Some push rod heads used 2 splayed valves of generous size serviced by large well curved ports in a Hemi type (or close to it) combustion chamber. Some overhead cam engines used two parralel valves in a head with more resrictiive passages and in the case of the Hispanos and the Russian "V" series engines they used siamesed ports for some but not all the cylinders.
Given the variations in each type of engine it is not really safe to say that one type or the other allways had a certain advantage

It's rahter interesting but as most of examles you took designs of one company - Bristol which are not tipical at all in design (if we talk about radials). And engineers from Bristol took part in development of Napier Sabre, as you know - they took care of improving its valvetrain
Moreover to use a term "piston speed" I suppose for sleeve-valve system is not completelly correct. Maybe "speed of opening\closing valve apertures?

If to speak about classical radials we should take R-2800 or R-3350, I suppose. With their one intake and one exaust valves.

I thought the Bristol peaple just helped with manufacture of the Napair sleeves?

"Moreover to use a term "piston speed" I suppose for sleeve-valve system is not completelly correct. Maybe "speed of opening\closing valve apertures?"

I don't believe so, as I am trying to say, the Piston speed relates to the strain on the crankpins, crankshaft and crankcase. It has got nothing to with valve operation or the uncovering and covering of ports in the cylinder wall.

I did give you the piston speeds of the R-2800 and the R-3350.

Edit> for a 4000rpm aircraft engine see:http://en.wikipedia.org/wiki/Napier_Rapier

With pushrods............in the early 1930s

Although it did have very small valves and push rods and was followed by the Dagger with overhead cams for it's 4000rpm.