Why are some engines geared to reduce prop RPM?

[weilke]

Developing any engine is very expensive for a lot of reasons.

Detroit Diesel spent 1.5bil to develop and bring the DD15 engine to market in 2009 .

That's almost eclipse scale development money.

 

[Capt. Geoffrey Thorpe]

A simple statement was made (paraphrasing) "1940s aircraft engines are far more efficient than today's car engines". And that simply isn't true from the data I've seen.

Well.

The basic engine you find in today's cars is pretty much the same crankshaft, connecting rod, piston, sparkplug, etc. technology that you found in the '40s.

What has changed in your car:

Details.
Increased compression.
A lot of changes to reduce engine out emissions such as reducing crevice volumes
Lighter weight materials in some cases
More efficient structural design - reduced weight
More efficient cooling systems
Some friction improvements
Improved combustion chamber design (swirl, tumble, squish, etc.)
Better optimized ports / valves / manifolds for more power
Piston coatings
Ignition energy

Control systems - air, fuel, spark, exhaust gas recirculation, cam timing, charge motion control, etc. etc. etc.

Some of those may help an aircraft engine, others won't.

These changes were driven, for the most part by the ever changing regulations for emissions and fuel economy. And, of course, cost reductions. Your average program manager would sell his/her own grandmother for a 25 cent cost reduction...

The traditional aircraft engine was designed for a particular application. Car engines are designed to different objectives.

I would suggest that one could compare Ford engines vs. Chevy. Engine designers at Ford get graded on power per cubic inch displacement - most Ford engines have dual overhead cams to get better power at the expense of packaging space. Chevy, on the other hand, builds the Corvette with a low hood line. So the emphasis has historically been on horsepower per packaging space - not displacement - so most Chevy engines use pushrods.

Different objectives. Different designs.

Which is better?

 

[Ted]

I suspect that must be due to limited lean tolerance?

Well, this includes on engines leaned to AFRs in the 16:1 range.

I do wonder if it might have to do with the rather basic combustion chamber design that you mentioned. I've always thought that aircraft would work very well with the May head design, which was used in the "High Efficiency" Jaguar V12s. While in its stock form, that engine got something on the order of 18 mpg (which was a big improvement from the 12-13 mpg the original cylinder head design got), my '82 XJ-S V12 with the H.E. engine got 30 mpg on the highway by replacing the GM Turbo 400 with a Tremec TKO (taking highway RPM from 3000 down to 2000), removing some drag from the engine, and (yes) advancing timing. The timing by itself was worth a decent amount with that head design.

 

[Capt. Geoffrey Thorpe]

Well, this includes on engines leaned to AFRs in the 16:1 range.

I do wonder if it might have to do with the rather basic combustion chamber design that you mentioned.

Ah. Not very lean. Lean burn engines I worked on were out in the mid 20's. The problem with that for aircraft applications is that you can't maintain 75% power at altitude due to the lack of air... Plus, once you are at wide open throttle at altitude, there isn't any improvement in pumping work to be realized.

Yes, the open chamber on the typical aircraft engine does not help.

(Just to be clear, I do control system design, not engine / combustion chamber design)

 

[Ted]

Ah. Not very lean. Lean burn engines I worked on were out in the mid 20's. The problem with that for aircraft applications is that you can't maintain 75% power at altitude due to the lack of air... Plus, once you are at wide open throttle at altitude, there isn't any improvement in pumping work to be realized.

Yeah. On conventional spark-ignition engines (i.e. not lean-burn), that's typically about as good as you can get. Lean burns really do best with turbos. One of the issues with a lot of diesels in aircraft applications is that the turbo runs out, necessitating a sequential twin-turbo setup.

Yes, the open chamber on the typical aircraft engine does not help.

(Just to be clear, I do control system design, not engine / combustion chamber design)

My experience is in control system design and calibration as well, but I have a certain amount of experience with combustion chamber designs. At the very least, I've played around with a bunch of them in the real world.

 

[jhausch]

Dan has summarized the issues with the current diesel offerings well. I haven't deal with the SMA design. I like it on paper, but figured there had to be some issues with its practicality since they aren't being gobbled up left and right. The Thielert engine is an interesting concept, but the reality of it has spoken for itself. There are a number of things that I'd do differently if I were building it.

Developing any engine is very expensive for a lot of reasons. Diesels, especially so. I remember talking to Cummins engineers who explained to me that the lines for the common-rail direct injection system in one particular engine (I forget which - I want to say the 6BT in the Dodge Ram) required some incredible tweaking, to the point of a little bend here and there just to make the pressure pulses in the system (operating at some ludicrous PSI - I want to say 40,000, but forget) function properly. Just having fuel at that high of a pressure is a lot to contain by itself. Then you add in trying to get it to operate on a fuel that's not design to run that kind of engine, but does kinda sorta work...

Then you have to certify it. Which is a lot harder when you can't certify it "by similarity," as most of the engines in our aircraft have been.

To your point about diesel injector pumps/piping/systems:
I think a neat combo would be the DD cam-driven style mechanical unit injectors (which can be somewhat "timed" with variable camshaft technology)

http://transportation.centennialcollege.ca/oduffy/fuels/level 2 fuels/DDC.pdf (warning 3.5M PDF, but it has a nice explanation on this old technology)

Installed in

a 2-stroke dual-piston-per-cylinder engine like the Junkers or Gemini.
http://www.jetwhine.com/2008/08/gemini-diesel-engine-attracts-industrys-eye/

On a ported 2-stroke design like this, I've wondered if the "exhaust-side" piston and ports "survive" compared to the "intake-side".