Improve That Design: How Aircraft Could Have Been Made Better

[Shortround6]

I am taking a wild guess but manifold pressure would be measured at the manifold (after the supercharger and/or intercooler on the two stage Merlins) and before the intake port on the cylinder head.

 

[Zipper730]

I was wondering what fundamental problems existed with the following engines, from a technical standpoint

• I-1430
• O-1230
• X-1800/H-2600
• H-3130/3730

While I know that the H-2600 and H-3730 were cancelled because of a desire to focus on radial engines, I'm curious if either had technical flaws that were difficult to work-out?

 

[GrauGeist]

Lycoming's O-1230 may have been rated at 1,200 hp, but it was prone to failure under load and that caused aircraft manufacturers who were interested in using it, and the Army, to back away from the project.

 

[wuzak]

I was wondering what fundamental problems existed with the following engines, from a technical standpoint

• I-1430
• O-1230
• X-1800/H-2600
• H-3130/3730

While I know that the H-2600 and H-3730 were cancelled because of a desire to focus on radial engines, I'm curious if either had technical flaws that were difficult to work-out?

I-1430: Separate cylinder design, 2 valves per cylinder, smallish capacity and general crappiness.

O-1230: Separate cylinder design, 2 valves per cylinder, smallish capacity and general crappiness.

X-1800/H-2600/H-3130/3730: Liquid cooled in an air-cooled company, sleeve valves. The man driving the project became ill, so went on the back-burner.

 

[P-39 Expert]

What was an "O" configuration for the O-1230?

 

[wuzak]

What was an "O" configuration for the O-1230?

Opposed.

I = Inverted, or IV = Inverted Vee (for the I-1430/IV-1430).

 

[Shortround6]

What was an "O" configuration for the O-1230?

Opposed, that is flat.

 

[Zipper730]

Lycoming's O-1230 may have been rated at 1,200 hp, but it was prone to failure under load and that caused aircraft manufacturers who were interested in using it, and the Army, to back away from the project.

Why was it prone to failure under load?

I-1430: Separate cylinder design, 2 valves per cylinder, smallish capacity and general crappiness.

O-1230: Separate cylinder design, 2 valves per cylinder, smallish capacity and general crappiness.

What caused the general crappiness? As for the separate cylinder design, I thought that contributed to bulk more than anything else -- I didn't know it undermined performance. As for the number of valves, I'm curious as to how many the V-1710 had...

X-1800/H-2600/H-3130/3730: Liquid cooled in an air-cooled company

Pratt & Whitney never made a liquid cooled engine? Regardless, that seems more like preference over functionality.

sleeve valves.

What problems occur with sleeve valves? I do remember that the British used them in the Centaurus and I think it made a lot of smoke or something...

The man driving the project became ill, so went on the back-burner.

Was he driving both the H-2600 and 3730? Or just one of the two designs?

I also meant to ask this, but what aircraft were the H-3730 designed to power? It seems like a rather massive powerplant...

 

[Shortround6]

Why was it prone to failure under load?

For one thing it was trying to make about the same power as an Allison (or Merlin) from an engine about 3/4s the displacement, there is only so much higher rpm can do but........

What caused the general crappiness? As for the separate cylinder design, I thought that contributed to bulk more than anything else -- I didn't know it undermined performance. As for the number of valves, I'm curious as to how many the V-1710 had...

Separate cylinders mean the crankshaft and crankcase are longer, yes it adds bulk, but it also adds weight and in causes problems with crankshaft flex and torsional vibration.
Think about rear cylinders firing and trying to twist the crankshaft along it's length. The longer the crankshaft the easier it is to twist it a bit.
Allison and Merlin used 4 valves per cylinder, you want to know how many the DB 600s and Jumos used, google it.

Pratt & Whitney never made a liquid cooled engine? Regardless, that seems more like preference over functionality.

It was preference but remember that P& W made very good air cooled engines. Functionality would mean learning all the ins and outs of the vibration patterns of the liquid cooled engine layouts. P & W spent hundreds of engine test hours figuring (thousands of engineering hours?) out the vibration problems of the R-2800. Making good liquid cooled engines was not as easy as it seems, how much liquid moving how fast is needed to cool the different parts of the engine?

What problems occur with sleeve valves? I do remember that the British used them in the Centaurus and I think it made a lot of smoke or something...

British used Perseus, Taurus and Hercules engines more than they used the Centaurus and it was far from beer and skittles for quite some time.

Was he driving both the H-2600 and 3730? Or just one of the two designs?

He was basically incharge of everything and so could direct the companies efforts. When he fell ill his replacement decided to stick with what P & W knew best, air cooled radials.

They estimated in 1940 that they might have the R-4360 ready in time for the war (of course they didn't know how long the war was going to take) . the Liquid cooled engines were judged to take even longer.

 

[stug3]

The theory of Survivorship Bias applied to allied aircraft.

 

[Zipper730]

For one thing it was trying to make about the same power as an Allison (or Merlin) from an engine about 3/4s the displacement, there is only so much higher rpm can do but........

So they were trying to do too much with too little?

Separate cylinders mean the crankshaft and crankcase are longer, yes it adds bulk, but it also adds weight and in causes problems with crankshaft flex and torsional vibration.

And this all started because of the requirement for the higher coolant temperatures back with the hyper engine, and they figured that individual cylinders would be needed?

Allison and Merlin used 4 valves per cylinder, you want to know how many the DB 600s and Jumos used, google it.

They DB-600's used 4.

It was preference but remember that P& W made very good air cooled engines. Functionality would mean learning all the ins and outs of the vibration patterns of the liquid cooled engine layouts.

So firstly: The devil was in the details? I'm guessing vibration had to do with the greater length of the engine?

Just to get back to the point: Did they ever design a liquid cooled engine prior to this point?

P & W spent hundreds of engine test hours figuring (thousands of engineering hours?) out the vibration problems of the R-2800. Making good liquid cooled engines was not as easy as it seems, how much liquid moving how fast is needed to cool the different parts of the engine?

So basically you want the coldest fastest moving fluid to cool the hottest parts and the slowest and warmest to cool the coolest parts?

He was basically incharge of everything and so could direct the companies efforts. When he fell ill his replacement decided to stick with what P & W knew best, air cooled radials.

So there was a general disinterest in the matter?

They estimated in 1940 that they might have the R-4360 ready in time for the war

If the H-2600 had been developed, would the H-3730 have likely continued along over the R-4360?

 

[wuzak]

So they were trying to do too much with too little?

Basically.

The IV-1430 had already been enlarged over the original design based on the "hyper" cylinder, which was 1008ci, IIRC, for an opposed 12.

And this all started because of the requirement for the higher coolant temperatures back with the hyper engine, and they figured that individual cylinders would be needed?

No.

Maybe it was convenient to put 12 of the "hyper" cylinders together rather than build a monoblock design (like the V-1710, Merlin and DB-60x).

Note that the IV-1430 had a single cylinder head per bank.

They DB-600's used 4.

So you can use google?

So firstly: The devil was in the details? I'm guessing vibration had to do with the greater length of the engine?

Yes, partly. The radial engines had 7 or 9 cylinders on each crankshaft throw, while the V-12s had only 2, which meant different vibration characteristics.

Just to get back to the point: Did they ever design a liquid cooled engine prior to this point?

Yes, but none that went into production.

The R-2060 was one, a liquid cooled radial.

So basically you want the coldest fastest moving fluid to cool the hottest parts and the slowest and warmest to cool the coolest parts?

You need to balance the coolant flow with the amount of heat that you can extract from the cylinders and reject through the radiator.

Cooling systems in aircraft were, generally, designed around normal or cruise power, not maximum WEP.

The limits of how long higher power ratings, such as WEP, could be used was due to the limitations of the coolant system.

So there was a general disinterest in the matter?

It was outside Pratt & Whitney's core competency.

If the H-2600 had been developed, would the H-3730 have likely continued along over the R-4360?

Given that the R-4360 was essentially chosen over the H-2600, the question is moot.

The R-4360 was an easier development path for Pratt & Whitney, since the starting point was the R-2800 cylinder.

 

[wuzak]

British used Perseus, Taurus and Hercules engines more than they used the Centaurus and it was far from beer and skittles for quite some time.

They even used the Sabre more than the Centaurus.

Centaurus development was delayed because Bristol was busy fixing and improving the Hercules.

 

[XBe02Drvr]

Maybe it was convenient to put 12 of the "hyper" cylinders together rather than build a monoblock design

P&W was a radial engine company. Single separate cylinders were their stock in trade. Monoblock design was foreign to them and too devilishly full of details that had taken other companies years and iterations to learn.
Cheers,
Wes

 

[wuzak]

P&W was a radial engine company. Single separate cylinders were their stock in trade. Monoblock design was foreign to them and too devilishly full of details that had taken other companies years and iterations to learn.
Cheers,
Wes

Not talking about Pratt & Whitney.

The engines based on the "hyper" cylinder were separate cylinder construction - the Continental IV-1430 and the Lycoming O-1230/H-2470.

I believe the P&W X-1800/H-2600, etc, were monoblock designs.

 

[GrauGeist]

Why was it prone to failure under load?

Because on the bench, it's performance looked good, but once it was installed in the Vultee and put through performance testing, the engine suffered from excessive detonation (until higher grade fuel was provided) as well as cracks in the crankcase and bearing failure.

 

[Zipper730]

Basically.

The IV-1430 had already been enlarged over the original design based on the "hyper" cylinder, which was 1008ci, IIRC, for an opposed 12.

The 1008 arrangement was also a V from what I recall.

No. . . .Maybe it was convenient to put 12 of the "hyper" cylinders together rather than build a monoblock design (like the V-1710, Merlin and DB-60x). . . . Note that the IV-1430 had a single cylinder head per bank.

They really should have redrawn the contract as technology advanced to allow for monoblock construction -- actually, why didn't they?

So you can use google?

That's how I got the information, you said don't ask, use google. I used google.

Yes, partly. The radial engines had 7 or 9 cylinders on each crankshaft throw, while the V-12s had only 2, which meant different vibration characteristics.

Which they'd have to work out.

Yes, but none that went into production.

But they would have a body of knowledge on radial engines, to some extent, anyway.

The R-2060 was one, a liquid cooled radial.

I'll see what I can find online

You need to balance the coolant flow with the amount of heat that you can extract from the cylinders and reject through the radiator.

Understood.

Cooling systems in aircraft were, generally, designed around normal or cruise power, not maximum WEP.

So the key is providing proper cooling for cruise power, but being able to cool the engine enough for WEP settings?

It was outside Pratt & Whitney's core competency.

I'm actually curious why the USN was so interested in developing the design at all. That surprised me more than anything else.

Given that the R-4360 was essentially chosen over the H-2600, the question is moot. . . . The R-4360 was an easier development path for Pratt & Whitney, since the starting point was the R-2800 cylinder.

What I was curious about was basically, from an academic standpoint, is basically...

1. If the H-2600 was given the go-ahead: Would the R-2800 have likely been given the go-ahead or been cancelled?
2. If the H-2600 was given the go-ahead: Would the H-3730 likely been given the go-ahead, essentially, by extension?
3. For the R-4360 to have been given the go-ahead, would it have required the cancellation of the R-4360 radial design?

 

[Shortround6]

They really should have redrawn the contract as technology advanced to allow for monoblock construction.

Mono block construction dated to WW I, see Hispano Suiza V-8s.
You seem to still be confused as to who was designing what.
Continental and to some extent, Lycoming were building what the Army told them to build. If the Army said "design separate cylinder engines" that is what they did.
If they had designed monoblock engines the Army would have said "that is not what we ordered, we are not paying for it".
When all the Army was ordering was 2 cylinder test rigs it didn't make much difference. When it came time for the 12 cylinder versions things went to crap in a hurry.

What I was curious about was basically, from an academic standpoint, is basically...

1. If the H-2600 was given the go-ahead: Would the R-2800 have likely been given the go-ahead or been cancelled?
2. If the H-2600 was given the go-ahead: Would the H-3730 likely been given the go-ahead, essentially, by extension

If you cancel the R-2800 (which was earlier in timing than the H-2600) you give Wright market domination in the 1500 hp and market for several years (maybe 4 or 5).

According to most sources, Mead became interested in Sleeve valves at some point in 1937. Work on the R-2600 (yes R-2600) began in August 1936. This would be P & W 4th two row radial and that added to their experience with single row engines (at least 4 different ones). However upon learning that Wright was building an R-2600 in March of 1937 P&W modified their engine to 2804 cubic in to top Wright. The R-2800 had at least year and probably more head start on any of the Liquid cooled P & W engines and given P & W expertise in air cooled cylinders (and the fact that the R-2800 used the same size cylinders but not the same construction as the R-2180) that lead may have been even bigger.

In any event it is claimed that P & W Estimated the date for start of mass production for the X-1800 as some time in 1942. Jan 1942 saw the A series R-2800 being phased out of Production by P & W (none would be built in Feb) and the single stage and two stage R-2800s being phased in, Ford was also building B series R-2800s in Jan ( a few dozen) nad would stop Building A series engines by April (which would see 463 B series engines built).
Canceling the R-2800 in favor of the H-2600 would have been a catastrophe for both P & W and the US war effort even if the H-2600 had worked and even if it entered production in the 2nd half of 1942.

 

[wuzak]

According to most sources, Mead became interested in Sleeve valves at some point in 1937.

It seems to have happened during a tour of UK aero engine manufacturers. Mead was impressed by the prototype Sabre.

 

[Zipper730]

Continental and to some extent, Lycoming were building what the Army told them to build. If the Army said "design separate cylinder engines" that is what they did. . . . If they had designed monoblock engines the Army would have said "that is not what we ordered, we are not paying for it".

No, I get that. I just think the US Army should have re-written the contract when time came to go to 12-cylinders. I'm not sure how easy it was to deduce the effect of two cylinders in a test-rig versus twelve cylinders.

If you cancel the R-2800 (which was earlier in timing than the H-2600) you give Wright market domination in the 1500 hp and market for several years (maybe 4 or 5).

I asked if the H-2600 had been given the go-ahead would the R-2800 have proceeded.