# How do I get my EG33 to run on a air frame?

Discussion in 'Subaru' started by jake2465, Sep 1, 2017.

1. Sep 1, 2017

### jake2465

#### Member

Joined:
Sep 1, 2017
Messages:
9
1
Location:
Huber Heights, Ohio / USA
Hello,

I just finished rebuilding my EG33 and I am now to the point where I need to figure out how I will get the engine to run on its own without being in the car. I know people have done this already, but I cannot find any web pages where someone has explained just how to go about it. I would like to use the OEM harness and computer if possible. One man I spoke to said that my 1996 SVX would likely run OBD2 and that would pose some big problems for me and that I should consider getting a older OBD1 computer and harness.

Is there a modified schematic that could show me how to trim and mod the wiring to accommodate the computer and engine so this can work?

2. Sep 1, 2017

### Chris In Marshfield

#### Well-Known MemberHBA Supporter

Joined:
Jan 29, 2008
Messages:
1,200
985
Location:
Germantown, WI USA
BoKu likes this.
3. Sep 1, 2017

### pictsidhe

#### Well-Known Member

Joined:
Jul 15, 2014
Messages:
5,894
1,501
Location:
North Carolina
if you have the donor car, start with the whole harness. Many engines will run with just the essentials plugged in, but will just throw codes. I've heard that new, virgin, ECUs won't whine about the lack of, eg, the rear washer fluid level sensor if it had never encontered one. Another option is disabling unwanted stuff with a tuning setup like HPtuners. I'd just try it first, though. It's worked for me with 3 EFI into dinocar transplants. Ross will no doubt have a better answer.

4. Sep 1, 2017

### jake2465

#### Member

Joined:
Sep 1, 2017
Messages:
9
1
Location:
Huber Heights, Ohio / USA
The big thing I was wondering about was if the computer would end up getting picky and just throwing the engine into limp mode without the car transmission being connected and such... I have no idea. I understand the basic theory of these MPFI systems, but they have so many accessory systems that communicate with the ECU that it makes me think just yanking out the harness with the ECU and pugging the engine sensors to it on some test stand and expecting it to run smoothly would be wishful thinking on my part.

BoKu likes this.
5. Sep 1, 2017

### rv6ejguy

#### Well-Known Member

Joined:
Jun 26, 2012
Messages:
3,550
2,574
Location:
Russell Sherwood is the only guy I know who has successfully flown the EG33 for a long time with the OEM ECU. Almost everyone else I know has switched to SDS. The factory harness is a nightmare and you're likely to get a ton of codes and possibly the dreaded limp home mode unless you take care of all these things. Many a Subaru powered airplane has ended up in a field when people didn't get all the gotchas handled. There are a number of 4 cylinder EJ engines and some EZ30s using the OEM ECU successfully but the same things apply. Some of these have had issues as well.

Russell is the only guy you should listen to if you are set on using the OEM ECU. He's not on here but does post on the Yahoo FlySoob group. Russell is a smart engineer and usually willing to share what he did with other folks.

Watch the speed sensor input, bad COP connections and be sure to run it a lot on the ground under high power before you take to the air.

Last edited: Sep 1, 2017
wsimpso1 likes this.
6. Sep 1, 2017

### jake2465

#### Member

Joined:
Sep 1, 2017
Messages:
9
1
Location:
Huber Heights, Ohio / USA
It would depend on the pricing of the SDS system. Up until now, I have never heard of it. Recently I have had some expensive expenditures such as rotor blades and a sprag unit. I did not want to also sink another few thousand in a stand alone fuel management system just to get the engine to fire off when I really won't be ready to get off the ground for a few years to come. I don't want to price myself right out of my project before I really get started. I still need to get my tubing so I can weld together the main frame and start hanging components off of it.

7. Sep 1, 2017

### jake2465

#### Member

Joined:
Sep 1, 2017
Messages:
9
1
Location:
Huber Heights, Ohio / USA
My idea was to have the engine runnable with that OEM system for a while so I can do some ground testing and things like that. I don't know how comfortable I would really be with a 22 year old computer and harness being my lifeline between me and the ground.

8. Sep 1, 2017

### rv6ejguy

#### Well-Known Member

Joined:
Jun 26, 2012
Messages:
3,550
2,574
Location:
I say experiment with what you have. If you get frustrated, we have a solution for you. You'd be looking at $1500-$1800 depending on coil options and ECU options.

Just be **** careful, especially if this is a rotor craft. Lots of folks have ended up bending planes with the OEM ECU which wasn't designed for aviation but can work if you cover all the bases.

9. Sep 1, 2017

### jake2465

#### Member

Joined:
Sep 1, 2017
Messages:
9
1
Location:
Huber Heights, Ohio / USA
Ross, I would like to hear more about what you have to offer. Ultimately I will end up transitioning to a stand alone system anyway. I would only want OEM for ground testing (sand bagging the helicopter to the ground and running power through it). But, actually flying that machine in that OEM configuration would make me a little uneasy. I mean, sure, I can drop the collective to the floor and shoot a auto into someone's bean field and probably make it out alive, but I would rather just avoid that scenario. Plus, it would be likely that I would ball the helicopter up in the process.

10. Sep 1, 2017

### rv6ejguy

#### Well-Known Member

Joined:
Jun 26, 2012
Messages:
3,550
2,574
Location:
If you checked out the Subaru Page you'll see a couple helicopters there, one even with an EG33. We can drive the factory coils but a few people have had problems with the coil connectors and poor contacts. A number of people are ditching the COP units and using one of our coil pack options.

We replace the factory magnetic crank sensors with a Hall Effect. Watch out for crank pulleys which spin the outer ring though, It's happened to a couple of customers. You could flywheel mount the magnets and mount the Hall sensor on that end too.

Many people ditch the factory intake manifold which is 35 pounds if I recall and fab something lighter.

Factory TPS is usable but you may ditch the 2 barrel TB for something lighter if you change the manifold. Ford 5L stuff works good here.

Watch the oil cooling, that's probably a bigger challenge on a rotary wing aircraft. Coolant too if you're pulling a lot of power at slow speeds or hover for a while.

cheapracer likes this.
11. Sep 1, 2017

### cheapracer

#### Well-Known Member

Joined:
Sep 8, 2013
Messages:
5,059
3,327
Location:
Australian
Ross has been a member here for a long time, is an aviator, helps others frequently and even recommends competitors products if it's right for them over his own product.

We are lucky to have people like him in the industry.

delta likes this.
12. Sep 1, 2017

### jake2465

#### Member

Joined:
Sep 1, 2017
Messages:
9
1
Location:
Huber Heights, Ohio / USA
Yes, that intake manifold is quite heavy. I was surprised at how much it weighed. Fabricating one would be the best thing to do, but I don't recall what the calculation is for matching internal manifold volume and max torque peak. I suppose I could just find the volume of the current manifold and fabricate something similar to it and be decently close to the numbers.

Does the SDS system use O2 sensors? Would it be a plug and play kind of system where I would only need to connect the fuel pump, ignition, starter solenoid and other accessories like that? Or, would this system require that the engine be put on a engine dyno and have a custom fuel / ignition map to make it come alive?

13. Sep 1, 2017

### rv6ejguy

#### Well-Known Member

Joined:
Jun 26, 2012
Messages:
3,550
2,574
Location:
You can see a composite manifold here: http://www.sdsefi.com/air34.htm and Russell's looks similar, I'll try to dig up that photo later today. I built one for mine here: http://www.sdsefi.com/air42.htm

We generally don't use O2 sensors feeding back to the ECU for mixture control in aircraft, only for monitoring and tuning.

Our 6F system would be easy to install. You'd have to mount an air temp sensor, drill the balancer for magnets (template included) and graft the injector and sensor connectors (and COP wires if used) onto our supplied harness which would be custom made to your length specs.

We'd have a base map entered to get it started and you'd use a wideband O2 to tune the system with the supplied programmer. The optional PC data logging would aid in that process. No dyno needed but you'd have to get the engine under load, tied down perhaps.

Here is a photo of Russell's engine and intake

jake2465 likes this.
14. Sep 1, 2017

### jake2465

#### Member

Joined:
Sep 1, 2017
Messages:
9
1
Location:
Huber Heights, Ohio / USA
I have a huge roll of carbon fiber that I could use to make a fashionable intake manifold. The main concern I have with doing that is knowing what epoxy I should use on it. Just a couple hundred degrees would probably be enough to over cure regular epoxy. a long time ago I used to deal with epoxy from "Hysol", but it was some pretty thick stuff and mainly used for patch jobs, not full blown components.

The SDS system sounds like a viable option for me. I have a CNC Bridgeport mill, so making brackets or anything of that sort would not be too much trouble. I already have an AEM wide band O2 and the gauge for it.

getting the engine under load may be a bit tricky, but I may be able to cobble something together to make it work out. There is a guy in town that runs a repair shop and also has a engine dyno. He just does not know a whole lot about mapping and sensor work as far as I know.

15. Sep 1, 2017

### rv6ejguy

#### Well-Known Member

Joined:
Jun 26, 2012
Messages:
3,550
2,574
Location:
With the data logging and your wideband you could probably tune the system pretty easily yourself on that guy's dyno.

I'm no composite expert but there are a few on this forum who could suggest a high temp resin. Otherwise a steel on aluminum intake could be made. We have some weld on injector bosses which could be helpful but the cost of 6 of those plus 6 new injectors would add a lot of cost. The factory injectors work fine.

16. Sep 1, 2017

### pictsidhe

#### Well-Known Member

Joined:
Jul 15, 2014
Messages:
5,894
1,501
Location:
North Carolina
If you want to make a new manifold, the lengths and diameters of the runners determine the RPM of the torque peak. For factory tune, you keep the runner length divided by the area the same. It's a Helmholtz tuned system, like a bass reflex speaker. The plenum volume is pretty much the bigger the better, though there are diminishing returns as it gets huge.
Ross, I'm curious as to what makes OEM ECUed engines stop, other than wiring problems?

17. Sep 1, 2017

### jake2465

#### Member

Joined:
Sep 1, 2017
Messages:
9
1
Location:
Huber Heights, Ohio / USA
I was thinking of holding on to the lower runners that house the stock injectors and just bolting the modified manifold on top.

If you like, send me a message and we can talk more about what options would be best for my application. I was not really planning on another purchase in the near future, but from what I have seen, this system is would be a wise choice. Much cleaner than the OEM system and adjustable as well.

18. Sep 1, 2017

### jake2465

#### Member

Joined:
Sep 1, 2017
Messages:
9
1
Location:
Huber Heights, Ohio / USA
I have the wire schematics for the Subaru SVX. It is basically a sea of wires and switches all over the place. The problem I am running into is that few people want their engines to run outside of their cars. I have spoken to some knowledgeable folks that are into tuning, but they mainly only go as far as piggy back systems on top of the OEM system.

I believe the issue is that the manufacture of the car has tried to make their ECU's as automatic as possible. Really, the ECU is a engine / transmission ecu. It runs both. So, it would be very likely that if I tried to run the engine without that computer getting some kind of feedback from the transmission, then it will flip out and proceed to make my life miserable. And finding people that actually know how to deal with this is like finding a needle in a haystack. As far as I know, it's definitely doable, just not easy.

19. Sep 1, 2017

### rv6ejguy

#### Well-Known Member

Joined:
Jun 26, 2012
Messages:
3,550
2,574
Location:
I made my tube lengths suited for about 4500 rpm cruise. The stock EG33 uses a variable resonance chamber valve which is ECU controlled. You don't need that for an aviation application.

Because there are so many chassis sensor tie-ins to the ECU, when those sensors are not present or give readings outside set parameters, you can trigger a code, limp mode or protection shutdown like fuel pump relay going off. There have been many limp home forced landings with OEM ECUs- Subaru and GM especially. Many people don't appreciate the complexity of modern ECUs. Some of these have MILLIONS of lines of code and are checking many things for OBD compliance and diagnostics. Few people understand all that's going on and what to do to safely turn off certain unneeded functions for aviation unless they wrote the software for that ECU.

20. Sep 1, 2017

Joined:
Jun 26, 2012
Messages:
3,550