converting a subaru motor for aviation, some ?'s.

[soslow]

I don't know if this is the right place for this but here it is...
We are considering using a subaru ej257 (2.5 turbo) in a cozy MK 4 canard airplane. We are looking 250hp or so with absolute superior reliability. a good broad torque band from idle on up and economy of operaration are also important. Does anybody have any suggestions for turbo selection or any other advice? We are considering a crawford block, possibly with billet crank for the motor. Is the block and main bearings strong enouth to support mounting the propeller directly onto the crankshaft in this motor? the motor needs to be able to run at WOT for at least 10min safely.
sorry this is kind of scattered but I would appreciate any help or suggestions.

* Registered users of the site do not see these ads.

[ImprezaRSX]

I have been planning to do the same thing.
TWO major issues (thrust and power):
Thrust bearing is not strong enough for the prop to be directly bolted to crank. I have a possible fix. The phase I blocks have a center thrust, phase II have rear thrust bearings. The phase two crank is machined for both. Only the phase two blocks aren't. I believe a quality machine shop can machine the block for the center thrust to be used IN ADDITION to the rear thrust. This would double the thrust capacity of the engine. However, most props only turn a max of 2500 ish rpm. Making 250HP on a 2.5 L at that rpm will be difficult. This would require 500 ftlb of torque at 2500rpm... aint going to happen.
To solve this many people are running a redrive unit. Marcotti makes a real high quality redrive. With that unit, you can easily get the power out of the engine and not have to worry about the thrust bearing issue. I believe it's a 2:1 ratio. 250hp at 5000 rpm is practically no work for a nice 2.5L.

Another issue that I just remembered is oiling. The subaru engine won't take negative G's very well. Once that oil pump draws air it's over. An external sump could be made to work so that a few neg g's could be sustained. But I haven't worked on that one too much yet.

[soslow]

thank you thank you thank you, this is what we needed to know- still need some help with general biuld info so if any body has some ideas.

[ImprezaRSX]

most aviation aircraft run a low compression ratio. For this engine to create that kind of power at that low rpm you would have to run higher compression and 100LL fuel. For that I would recommend a very high quench TWE high compression piston.

Turbo's are load dependant. So You would also have to run a turbo that is a little smaller than that of an automobile. You need enough drag on the prop to help the turbo spool so it would have to be a pretty high pitch or a pretty small turbo. if all you want is 250hp, i would shoot for the smallest turbo that can flow that amount of air. A simple VF39 would proabaly suit you well. That still may even be too large an exhaust housing. To save some money I would recommend the factory WRX turbo to start off with. It's can flow 250 and has a small enough exhaust housing to help with any spool problems you may experience.

[ImprezaRSX]

Engine management is another thing I have been looking at. Subarus don't have distributors and they have electronic fuel injection when many people want a reliable carb on their plane. Fueling can be done with anything, and that isn't an issue. The issue lies in the ignition. No distributors means you need a crank trigger with some sort of control to tell your coils when to fire... Luckily, you can use the wasted spark design to your advantage and not need an extravagent distributor setup. I have been trying to make the system as simple as possible. I like the Simple Digital Systems setup if I stayed fuel injected. I would couple that with an adjustable fuel pressure regulator that I could adjust remotely from in the cab so I can have some more mixture control. Aeromotive has a regulator that I could easily engineer a remote adjuster for. A cabin mounted fuel gauge could help with this too. Closed loop is great with o2 sensors and stuff, but altitude changes in a plane happen faster than in a car so some short term adjustment would be nice. changing fuel pressure is the easiest way I can think of changing your AFRs on a digital controlled system.

[soslow]

I think we'll go with the reduction unit...
I would think engine mangement would be fairly simple.... accesport? And we intend to run 100ll fuel with a map for 91-93 octane as well

[ImprezaRSX]

it just depends on whether or not you want a back up failsafe option.

[Jeff Johnston]

Have you looked at http://www.eggenfellneraircraft.com/ ?

I don't think you need to be too concerned with negative Gs - not in a Cozy.

[Bundyboyz]

Originally Posted by Jeff Johnston

I don't think you need to be too concerned with negative Gs - not in a Cozy.

Exactly why would he ever want the plane to land? Or Dive?

[Jeff Johnston]

I don't recall ever experiencing negative Gs making a landing. As for diving, even a vigorous pushover doesn't often produce negative Gs - only Gs less than 1. The vast (vast) majority of general aviation airplanes don't have inverted fuel and oil systems, and they work just fine. I've personally done loops, aileron rolls, barrel rolls, spins, wingovers, etc., in more than one type of airplane without inverted systems, and never experienced a loss of oil pressure.

[Bundyboyz]

Every plane will experience negative G's...

I wasn't speaking in terms of the effects on the motor, I'm uncertain of them. But you have to be joking to say a plane won't see any kind of negatives.

[Jeff Johnston]

I never said it wouldn't. It probably will, but, again, lack of an inverted oil system is unlikely to be a problem. The Cozy is a cruiser, so any -Gs it sees will be of the momentary type, not sustained - and I have a hard time believeing that Subaru's oil pump can't do better than the 1940s tractor technology you find in general aviation engines. And I still don't see where you get -Gs in a landing - unless you're that guy with the Pitts that has the landing gear sticking out the top.

[Bundyboyz]

ok fair enough.

[Jeff Johnston]

Oh, and here's a picture of that guy.

[f4phantomii]

Originally Posted by Bundyboyz

Every plane will experience negative G's...

I wasn't speaking in terms of the effects on the motor, I'm uncertain of them. But you have to be joking to say a plane won't see any kind of negatives.

I would agree that it is rare to see actual negative G's in a light aircraft.

Any that you do see are so momentary that neither the aircraft structure nor your body even react to them.

But as he mentioned, you can do some vigorous pushovers and see less than 1G. You feel weightless, but anything that dips less than zero G is really brief.

I've done outside loops in a Pitts bi-plane before and only gotten about -1.5 G's and it was extremely uncomfortable....more so than a good +4G pull up.

And he also mentioned that most light aircraft don't have inverted flight systems on them. I've had the engine stop briefly doing light pushovers when the ball in the carb floats off the seat!

-Michael

[soslow]

the negative G's will not be an issue in the airplane....
we are more interested in biuld information, particularly turbo selection and engine managment

[ImprezaRSX]

I am very curious as to why eggenfellner went with a 2.56:1 reduction. That seems way too drastic to me. Knowing my subaru engines (there have been many) it would seem that a cruising rpm around 3600-4000 would be ideal. They are so darn efficient there and make so much torque.

This is a dyno graph of an engine I built last month for my roommate:


notice all the usable torque in that range. Not just that, but it would seem to me that hte longevity of that engine would be better at 3600 vs 4800 (like the eggewhatever).

This engine has alcohol injection, BUT, with the 100LL (vice 93) and cooler atmosphere (vice hot workshop with one small fan) power should be pretty close.

[CatfaceType-R]

ummm yah, what he said...lol


my friend and i wanted to do a small aircraft powered by a subaru motor but...no money