If you could go back to WW-2 with the knowledge you have now in engine design...what would you improve? No jets...

[jetcal1]

In no particular order?
Electronic engine management
Reduced reciprocating mass
Better oils
My now relatively old 2.0 liter is producing just over 2 HP per cubic inch or just over 125 HP per liter. Assuming the technology scaled up you'd have more HP from anything over 1500 cubic inches than the airframes could handle aerodynamically for the time.

Once you hit the HP/WT markers on the engines you can start designing larger offensive payloads or smaller, lighter, less expensive aircraft.
For example, an XP-77 with 1500 HP and same weight would have been an interesting proposition (Although still probably a dead end.)

 

[XBe02Drvr]

In no particular order?
Electronic engine management
Reduced reciprocating mass
Better oils
My now relatively old 2.0 liter is producing just over 2 HP per cubic inch or just over 125 HP per liter.

Whoa! Your time machine better have a truly massive transfer chamber because now you're talking about transporting a whole plethora of 21st century technologies back to the 1940s. More expertise and exotic materials than any one engineer could carry in his head.
Without semiconductors, hall effect devices, and miniaturization, how are you going to build a FADEC small enough and light enough for a single engine single seat fighter to get off the ground? A simple four channel two way radio was almost too much size and weight for the job back then. Are you going to recreate the entire solid state technology out of whole cloth once you get there? Same goes for advanced lubricants and light weight high strength reciprocating components. Are you going to single handedly leapfrog all those technologies through the intervening evolutionary steps? This I gotta see. "A Connecticut Yankee in King Arthur's Court" all over again. You'll have ole Sam Clemens rolling over in his grave. More power to you!
Cheers,
Wes

 

[swampyankee]

Bringing back any modern engine design software would be challenging; you'd need to bring back a computer (more likely, a Beowulf cluster), a large format printer and a few tons of toner.

Anything else, like traction drives instead of gears (see, for example https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19850025207.pdf) or ceramic bearing balls would require manufacturing techniques and machinery brought back. The most portable knowledge would probably be ergonomics. One wonders how many pilots were lost because they were in uncomfortable cockpits with badly laid out controls.

 

[Shortround6]

A big problem with many aircraft engines was cooling. They knew how to make power, Heck, the british managed to get 116hp out a 750cc Austin engine in 1937 (not on gasoline). It is getting the engines to survive for more than few hours or not cook their own oil at high power levels that stopped a lot of "progress".

2 stokes are not a good answer.
The problem is how much heat you are making in each cylinder vs how fast you can get rid of it. If the oil gets too hot it looses it's ability to lubricate and even the bearings hold out for a while the pistons start scuffing the cylinder walls. If the heat is bad enough the oil turns to something like varnish and then to carbon, scored cylinder walls and piston rinigs don't seal very well. Power is lost, Combustion pressure leaks past the rings, the crankcase gets pressurized and oil vapor gets blown out the breather/s.
Every advance in engine power per cylinder on air cooled engines was allowed by an improvement in cooling, either more fins or deeper fins or better baffles or something or a combination.

ANd aircraft engines had to be able to be cooled in air that was a lot less dense than cars or motorcycles operate in. trying keeping some of these modern engines making their rated power at over 20,000ft with the thin air only providing 1/2 the cooling of sea level air.

 

[jetcal1]

Whoa! Your time machine better have a truly massive transfer chamber because now you're talking about transporting a whole plethora of 21st century technologies back to the 1940s. More expertise and exotic materials than any one engineer could carry in his head.
Without semiconductors, hall effect devices, and miniaturization, how are you going to build a FADEC small enough and light enough for a single engine single seat fighter to get off the ground? A simple four channel two way radio was almost too much size and weight for the job back then. Are you going to recreate the entire solid state technology out of whole cloth once you get there? Same goes for advanced lubricants and light weight high strength reciprocating components. Are you going to single handedly leapfrog all those technologies through the intervening evolutionary steps? This I gotta see. "A Connecticut Yankee in King Arthur's Court" all over again. You'll have ole Sam Clemens rolling over in his grave. More power to you!
Cheers,
Wes

Except for the electronics (Yes I know that's a big "except") the genesis of a lot of this stuff was already starting by 1943-'44
If I may........

1. Miniaturization of electronic valves was already underway via aircraft radars, the Manhattan project, etc. (And other forms of electronic components had been hardened and miniaturized enough for proximity fuses.)
2. Mechanical fuel controls were already in production for early jet engines (I believe the first primitive EEC's were being piggy backed on engines like the J35 by the late '40's.) (Given a manual override with an auto enrichment valve tied to EGT, I believe the slow accel problems similar to those experienced by turbine fuel controls could be eliminated.)
3. The Germans had introduced the single Power Lever concept via the BMW 801
4. Synthetic oils were being developed in Germany via Fischer-Tropsch
5. ADI and NoX were current technologies
6. We have zero concerns about emissions or warranties
7. Mechanical Fuel Injection exists (See #2)
8. 115/145 fuel is available with lots of lead versus 93 Octane for the performance I cited in my example
9. We only need 400 to 800 hours between overhaul

You've read my comments about the constraints caused by a lack of production or a lack of available engineering, etc., in other posts.
Yet given the extremely relaxed standard of a 800 hour TBO, and the fact that aircraft engines rarely use full WEP.....
This is now more a question of integration of engineering disciplines than actual development of totally new sciences and technologies that are foreign to anything then in production in 1941-45.

I think it's doable.

 

[Shortround6]

Ah, things are not quite as simple as they appear.
granted they were not supercharged but did you know the First Porsche engine to exceed 1 hp per pound was the 4.5 liter 917 flat 12 engine of 1969? forged titanium connecting rods and other parts/materials not used in WW II. People keep wanting to compare modern engines and keep forgetting that the high performance aircraft engines of the day, on hp per pound basis took several decades for even race car engines to beat. And the race car engines were nowhere near as a reliable or durable as the aircraft engines. Your genesis theory needs a bit of work. The electronics in the proximity fuse consisted of a few vacuum tubes that were miniaturized. Want to try to build even a simple computer from a 1980s-90s car using vacuum tubes? reliability of said vacuum tube computer would be measured in minutes if not seconds simply due to the number of parts. BTW at the start of the proximity fuse program they would have been satisfied with a 20-25% dud rate if I remember correctly.
While looking for more information I found this image.

Magnetic core memory, patented 1951, that "Chip" stored 18 x 24 bits. It replaced vacuum tubes.

I would also note comparing lead content is a lousy way of comparing fuels. For example the post war Mil-F-5572 fuel specification allowed for 4.6cc of lead per US gallon for not only 115/145 but for 100/130 and even 91/96 fuel. You had to use better base stocks and other components for the higher grade fuel, not just more lead. The civilian ASTM D910-48T specification only allowed 4.6cc per gallon for commercial 115/145, both 100/130 and 108/135 were only allowed 3 CCs per gallon and civil 91/98 octane (yes, 2 points higher) was only allowed 2 CCs of lead.

Wright had to develop a brand new way of putting fins on cylinders for the above 1200hp R-1820s, also used on the 1900hp R-2600s and the R-3350s. They machined grooves into the cylinders that were larger at the bottom than at the top(surface) and formed sheet metal into a W configuration, very long outside arms and very short inside arms. This was "rolled" into the grooves so that the bottom of the W was swaged into the groove for mechanical interlock.
In a lot of cases you not only need the design of the new parts, you need the designs for the machines to make the new parts and in some cases you need the designs for machines that can make the machines that can make the parts you want.

 

[jetcal1]

Ah, things are not quite as simple as they appear.
granted they were not supercharged but did you know the First Porsche engine to exceed 1 hp per pound was the 4.5 liter 917 flat 12 engine of 1969? forged titanium connecting rods and other parts/materials not used in WW II.

Please remember, I was basing HP on HP/Displacement not weight:
"My now relatively old 2.0 liter is producing just over 2 HP per cubic inch or just over 125 HP per liter."
The 4.5 un-turbocharged was already over 500 HP? Which is pretty much close to my 2 HP per cubic inches.

My comment on lead was based entirely on it's ability to lubricate the components in the valve train along with it's coating properties for the rings and cylinder walls. I wasn't using it for the gains in HP. But for the extra lubricity under higher boost and temperatures.

As far as the electronics? I'm still suggesting a hydro-mechanical control with the addition of a EEC for monitoring the engine for EGT and or boost, again, well with in the limitations of electronics of the time. (I agree the sampling rate won't be so hot.) I'd agree the box will be pretty much the same size as a then contemporary auto-pilot or some of the radio components of the day. The idea is to maximize the advantages of a single power lever like the Kommandogerät.
(The VT did indeed have a low reliability rate at the beginning, but my system is not undergoing ridiculous levels of acceleration or spinning 10's of thousands of times a second, I believe my electronic EEC will be substantially less sophisticated than, say....the B-29's fire control system.)

I agree we did not touch on if the engine was going to be air cooled or water cooled, or major changes to manufacturing techniques. The intent would certainly be to avoid clusters of little ladies welding our cylinders together like some WWI engine, or radiators so large that they have the drag of a house.

Again, not based on weight, but displacement, I believe my boosted engine can achieve 800 hours TBO at two HP per CID using 1940's technology.

 

[ThomasP]

Not sure if this applies to this post, but I would bring materials technology also. The alloys used in modern manufacturing would be valuable, as would the alloys/materials used for cutting tools. As mentioned up-thread some of the technology of engine design might not be realistic to incorporate, but the ability to manufacture most modern alloys of aluminum and steel should be manageable for the most part, including some alloys with much higher strength, wear, and temperature resistance. I am not sure how great an impact on engine power this would have but I think it would be significant, plus reliability and durability of engines would improve.

Also modern formulas for high speed steel and carbide cutting tools would be manageable for the most part, as long as the raw material was available (ie tungsten, molybdenum, cobalt, etc). Production rates would increase, quality control of parts and reliability of engines would improve.

 

[XBe02Drvr]

I think it's doable

If you can do it, more power to you. I'm still sceptical.
Cheers
We

 

[tomo pauk]

Some options that might work on technology and machine tools of the day. A big V12 (35+L), with 2-stage supercharging, 2- or 3- speed drive, a power-recovery turbine, direct injection, single-lever operation, water injection. Should be getting 2500+ HP on 100 oct fuel, and 3000+ on 130 grade after some development.
Shortcoming? Big airframe needed.

 

[jetcal1]

If you can do it, more power to you. I'm still sceptical.
Cheers
We

Please keep in mind, I'm not looking for modern levels of reliability, nor is this a constant output, but a peak WEP power. We're not trying to pull 5K + plus HP out of a Merlin for ten minutes. (3.3K HP, yes. Not 5K +)

 

[swampyankee]

Outside of racing and other games, power per cubic inch is really unimportant; what's much more important is power per unit installed weight, including things like mission fuel and cooling system. Note that I am aware there is a relation between displacement and weight.

Certainly, modern technology such as FADECs is completely impractical, and would remain so until the invention of the transistor (Bell Labs) and integrated circuit (TI). A mechanical, analogue fuel injection system could be produced, although whether it would be practical and sufficiently superior to the carburetor technology available is moot. Quite a lot of modern engine technology has been driven by emissions requirements -- I would be really surprised if either US or non-US engine makers would have bothered with the research in combustion and engine control that's led to street engines that can produce levels of power that a 1960s racer would envy without the need to provide performance while meeting emissions and fuel economy requirements.

Possibilities:
1) Improved alloys. These could include non-combustible magnesium alloys, aluminum-lithium alloys, titanium alloys, and various refractory superalloys. Somewhat less glamorously, improved bearing alloys.

2) Improved sealing technologies.

3) Improved manufacturing techniques, some of which would be practical with 1940s machinery (factory machinery lasts a long time: there were machines in quite a few 1960s factories that had been built to be powered by line shafts, but converted to individual motors)

4) Better understanding of engine vibration characteristics.

Mechanical superchargers are pretty much non-existent for current aircraft piston engines. Were one to bring back modern turbocharger technology, it may be possible to get a two-stage turbocharged, intercooled engine into a fighter. While this would be highly non-trivial, it could permit aircraft operations to 50,000 ft (see https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19980137599.pdf, which was planning on three-stage turbocharging for operation at over 80,000 ft).

 

[Deleted member 68059]

likewise any significant gains in structural weight reduction from accurate detailed vibration and resonance analysis.

Engineers were aware of torsional vibrations and how to calculate it even before 1900 as it was a very serious problem in ship propellor drives, it was being determined with more than sufficient accuracy by the late 1920`s to understand how to optimise the stiffness of cranks etc. French engineers had already invented the torsional dampers commonly used even now in the 20`s (Salomon for one). Ready-reckoning had been eliminated as a design method for solving vibration problems by all serious aero-engine designers by the mid-30`s; it was simply much more labour intensive to calculate than it is now.

Likewise the structure of aluminium castings was very good by 1930`s, and was being done to more refined degrees by the 30`s than automotive mass production engines were being done until very recently (mostly for cost reasons).

The main area for weight reduction is brought about by increase in crank speed, which enables engines half or a third of the size to produce the same power. To achieve that needs several things to be done all at once, which are all inter-related and could not have been done in WW2 (oil, bearings, metallurgical purity etc).

Fundamentally it is very difficult to "downsize" aero engines, because they operate on a totally different duty-cycle to the automotive racing engine, and by the time all considerations are taken care of, you basically just end up with an aero engine again. Which was exactly what Porsche discovered in WW2 when the RLM asked them to design an aero engine.

The area where the engines could have been dramatically improved is the combustion chamber/piston crown, which could have yielded a large leap in performance with the available tech of the day.

 

[XBe02Drvr]

The main area for weight reduction is brought about by increase in crank speed, which enables engines half or a third of the size to produce the same power.

Doesn't the weight of a recip engine planetary reduction gear assembly increase dramatically as the required speed reduction increases, offsetting some of your HP gain from higher crank speeds with more weight? Unlike turbines with their smooth power that don't pound the gears like a recip does.
Cheers,
Wes

 

[Shortround6]

I would note that Porsche made two attempts to break into the aircraft engine market in the 50/60s and 80s.

In the 50/60s they tried to use the 356 4 cylinder 1.6 liter engine, it was called the 678 and came in several versions. The direct drive version offered 52hp for take-off at 3200rpm and max continuous of 50hp at 3150rpm. Several versions used reduction gears and one gave 75hp at 4,600rpm. However it weighed 247lbs with the reduction gear. For that kind of weight you could get an O-235 Lycoming that would give you 108hp for take-off at 2600rpm.
In the early 80s they tried again and used the engine out of the 911 sports car.

See Porsche PFM 3200 - Wikipedia

the important part is as as follows.
"With about 3.2 litres (195 cubic inches) displacement, the normally aspirated N-series models produced about 210 hp, while the turbocharged T-series produced about 240 hp. This was roughly twice the horsepower of a conventional lower-rpm design of the same displacement. With single-lever operation, fully aerobatic fuel and oil supplies, fuel injection with automatic altitude compensation and optional turbocharging, the PFM 3200 series were some of the most advanced engines on the market. However they were heavier and when fully accessorized and larger than competitive engines, especially in cross-section (important for aerodynamics.) Also they generated more aerodynamic cooling drag due to the fan system used for cooling"

Granted the early 80s are not today's knowledge but then Porsche 911 engines weren't economy care engines either. The conventional lower-rpm design of the same displacement engines dated back to around WW II in basic design or layout.

See; Porsche PFM
for a rather scathing criticism, maybe the author is biased, I don't know.

 

[swampyankee]

Shortround, I think a second issue that Porsche ran into is that the aviation world is somewhat fussier about after-market support, especially if there are reliability or serviceability problems, than is the automotive community. After all, if the engine in your 911 stops, the worst is that you'll get passed by a Cinquecento.

 

[Shortround6]

People keep coming up with the idea of using modern technology without appreciating the old technology. Which for high powered aircraft engines was arond the same level as race car engines only the aircraft engines had to last a lot longer.

and they keep bringing up displacement which neither the engine builders nor the airframe builders cared about.

There were a number of powerful car engines that could/should have provided inspiration.
1939 Mercedes 154.

A 3 liter engine of up to 473-480hp with two stage supercharging. And yet Daimler-Benz took years to incorporate any it's bigger features into an aircraft engine.
The state of tune (power output) and rpm limit changed according the race.
for the Germans part of the problem was that it didn't run on gasoline, another problem was the fuel.

to get todays power output you need todays materials and today's manufacturing techniques and today's methods of testing/examining the parts.
Titanium for example was used as a trace alloying element in certain weldable stainless steels, but not as a primary material for anything like engine parts or framework.

 

[Deleted member 68059]

Doesn't the weight of a recip engine planetary reduction gear assembly increase dramatically as the required speed reduction increases, offsetting some of your HP gain from higher crank speeds with more weight? Wes

Hi Wes, you`re quite right to flag that as a problem, what happens is that you keep the normal reduction gearing ratio and therefore end up with a smaller diameter prop to keep the tip speed below about 0.9 Mach or so, increasing the blade count may be necessary then to absorb the extra power. That obviously has its limits, but you never want a really extreme difference anyway because if the engine is running THAT fast (like an Indycar engine or something) it will wear itself out too quickly to be of use in an aeroplane anyway.

A good illustration of how hard it is to adapt automotive engines to planes is Burt Rutans Pond-Racer, although I`m sure eventually they could have been solved given another chunk of cash to develop for a year or two.

 

[XBe02Drvr]

you keep the normal reduction gearing ratio and therefore end up with a smaller diameter prop to keep the tip speed below about 0.9 Mach or so, increasing the blade count may be necessary then to absorb the extra power

And thus wasting some of your added horsepower in propeller inefficiency. Unless, of course, you have the technology and materials to build blades of the scimitar fan variety seen on upgraded Hercs and Hawkeyes. Did they have Kevlar back in 1940? Don't think so.
The Beech 1900s I was flying in the 1980s were touted as the first mass production airplane to have Kevlar propellers.
Cheers,
Wes

 

[Shortround6]

Getting back to this.

Electronic engine management

we have been over to some extent but those old engines used some pretty simple ignition systems. Duel plugs for both better ignition and redundancy. Fired by duel magnetos, one magneto fired one set of plugs (most of the time) and because there was not a wide variation in operating rpm they often (very often) used fixed ignition timing.
Yes a modern computer controlled ignition system could probably provide much better performance and allow the the engine to operate closer to the edge.
However provide redundancy you need two complete ignition systems. Which runs up weight and cost, You also in WW II would have to introduce new sensor technology. They had EGT gauges (exhaust gas temperature) and even exhaust gas analyzers (basically an oxygen or other gas sensor) but they were analog, depending on the value being measured they simply passed a different value voltage to the gauge. The gauge was little more than a voltmeter with a new scale pasted on and calibrated to the expected values. You either need a digital sensor or an analog to digital interface for your electronic engine management. Or you are operating at 1950s technology level with an analog computer.

Reduced reciprocating mass

This seems to be assuming better piston and connecting rod materials and/or better manufacturing techniques. Titanium so far seems out of the question. It wasn't really available in large pieces until the 1950s, during WW II it seems to have been used as a trace (low percentage) alloying element. The engine makers of th etime were constanly pushing the materials envelope as it was, yes in the 75 years since WW II things have gone much further but it is a question of not just changing existing designs but coming up with new, for the time, industries for industrial processes.

My now relatively old 2.0 liter is producing just over 2 HP per cubic inch or just over 125 HP per liter.

How much does your relativity old two liter engine weigh?
and can it produce 125hp per liter for 7 1/2 hours, 5 minutes at a time with 5 minute "cool down" periods (at around 60-70% power?, maybe more) between each full bore 5 minute run?
That was the US standard for War Emergency Power. The US would not approve a WEP rating unless a test engine had completed 7 1/2 hours at that rating.
Now if your 250hp engine weighs much over 250lbs then it is of little interest to the aircraft engine makers. The were looking for 1hp per pound of engine weight and could care less about cubic in or liters.

BTW a 2 liter 250hp engine at sea level becomes a 2 liter 125hp engine at just over 21,000ft unless you increase the level of supercharging by a factor of 2.

Assuming the technology scaled up you'd have more HP from anything over 1500 cubic inches than the airframes could handle aerodynamically for the time.

Unfortunately the basic engine operating conditions do not scale well. Small cylinders cool better than big ones, there is more surface area of cylinder wall, piston top and cylinder head to dissipate heat through for the volume of fuel burned than a larger cylinder.
Gasoline only burns so fast, as it was one reason for dual ignition in large aircraft engines was to get flame fronts started in different areas of the cylinder so that the flame fronts would move across the piston and pretty much complete the fuel burn by the time the crankshaft hit about 20 degrees past top dead center and the rest of the pistons travel was pushed by the expanding but already burned gases. Since you can't speed up the rate of combustion much without some really advance combustion chamber shapes or flow patterns (or triple ignition) and since the engine also has to return at least useable cruise performance there is only so much you can do to biases the engine towards the high rpm of the range.

It was these fundamental facts that had Halford (and others) trying to build 16-24 cylinder engines of similar displacement to 12 cylinder engines in order to get high rpm.

You can reduce the reciprocating weights. you can't double the speed of the flame front travel in the cylinder. You might be able to improve cooling with different materials and more coolant volume, for liquid cooled engines, larger coolant passages?, bigger coolant pumps with higher flow rates may not work as the fast moving coolant may not transfer enough heat in the shorter period of time?

Once you hit the HP/WT markers on the engines you can start designing larger offensive payloads or smaller, lighter, less expensive aircraft.
For example, an XP-77 with 1500 HP and same weight would have been an interesting proposition (Although still probably a dead end.)

Unfortunately, what with having to come up with a pretty much a new manufacturing base for most/all of these improvements the aircraft are going to be far from cheap. The smaller and lighter is debatable depending on just which improvements or how much of jump in technology you can accomplish.