This just doesn't make sense to me. The NA BC/BF cars don't run a hot wire (yellow) to the ignitor. The JDM turbo cars don't either. Why did they do this on the USDM turbos?

On all the other cars, the ignition switch and the fuel pump and main relay are all tied together directly to the coil. On the USDM turbo this same circuit includes the ignitor, and excludes the coil. The coil gets it's power on a different circuit that also has it running through an additional fuse.

What's going to happen if I take the pair of circuits on the USDM turbo car and join them? I think it will work just fine but want to be sure. This is relevant because I am converting it to coil on plug for an Ej20G ECU.

The only other time I did this, I did it on an NA car and this didn't come up. It's confusing me a bit. Your thoughts?

Why don't you get rid of old style ignitor and plug coils, and use the WRX coils?

You can use the two wires from the ECU which fire cylinders 1 & 2 and 3 & 4 and run the new plugs as wasted spark also, when using USDM Leggy ECU?

Or am I making a mistake here? I've toyed with the idea on my n/a car as I'm running a much newer JDM EJ204 DOHC which also runs plug coils.

It's not my car or my budget. It is using early WRX 4 channel ignitor and coils. That is not open to debate.

I had planned on using the hot wire that runs to the wasted spark coil because that is what I did before. But then I wired in the ignitor and found that there was an additional hot wire to the ignitor as well. What happens if I just cap that wire? It's the one that runs to the FPR, Main Relay and pin D12 (ignition switch). It really doesn't make much sense how they did it or why this one vehicle is totally different than all other Subies in how the switched 12v runs through the car...

Unfortunately, I can't help you then. I've only got experience with EUDM SOHC n/a wiring and some EUDM EJ20G wiring, but the wires you describe... Sorry...

no don't "join" the hot circuits ..

from my read you have 4 ecu inputs to the ignitor 4 outputs to the coils and a ground

the coils get a constant? 12V pre main relay and ignitor signal

http://www.alien-speed.com/share/techdocs/94_wrx_ecu.pdf

capping sounds like a good idea?

matt do you have the ej22t diagrams in pdf?

marnix the usdm ej22t wiring is wacky different than any thing

thinking outloud now
you would want to isolate the coil as much as possible thats why on the jdm its pre relay and on a seperate circuit with the ej22t

Jaxx,
I do have the PDF, but it's a pain to scroll through. It's the download from Leggy Central with no indexing for most of the 800 pages. I printed hardcopies of all three cars that I am looking at:

USDM NA Ej22 Legacy
USDM turbo Ej22 Legacy
EUDM WRX that I got from you.

If you look at the WRX diagram you posted, the yellow wire from the coils on page 20 that feeds over to "E" runs through the Ej22T identically, EXCEPT that is the wire that runs to the ignitor on that car!!! And it does NOT go to the diode. That's what's on the other circuit.

I am thinking I want to use that wire to power the coils. And then cap off the wire on the EJ22T harness that previously powered the coil? Or do I have it backwards?

My concern is that on the Ej22T the original wire to the coil is the one that has the diode with the LG wire that feeds to pin D5 in addtion to pin2 on the main relay.

In short, both circuits are connected to the ECU and I don't want to disrupt that...

New idea. What if I snip the Ej22T chassis harness where the wire feeds to ALT2 at the fuse box and then replicate how it runs through the other cars? I can just take that snipped wire and splice it into hot wire that used to run to the ignitor.

Then I can use the wire that use to power the coil to power the new coils?

if it is identical and the yellow wire is before the diode (i believe the diode is there to protect the relay and ecu from the em created by the coil) that should be good to power the coil assuming it is the same gauge too

are the 22t coils and igniter some how integrated


its been a long time since ee100-300

Not integrated. The Ej22T predates the built in ignitor wasted spark coils. There is no cross over of the power sources to each of them. I still don't understand why the ignitor is even getting power.

Answer me this, before I post my next thought, since I always get mixed up. When you say BEFORE the diode, which way is the power flowing.

Does power come out of the ECU at b12 on a yellow wire, go to the diode, and come back out on the Lg wire, ending at d5? Or does power flow the other direction? From the ECU's point of view, which is sending and which is recieving?

power comes out of the fuse box fb-4 (which must be switched?) splits off to the coil and the diode(one way following the arrow) ..after the diode it then hits the ecu and the main relay

or does the diode function as an "or" gate hmmmm

what exactly is fb-4

I don't believe a diode can function as an "or" gate. It's purpose is to only permit current flow in one direction and one direction only. I don't understand how the wire is integrated into the main relay and ecu on that side of the diode. I guess I need to look at a diagram.

Well,
I don't know off the top of my head, but I am off to find out. I have to hand over the manifold ready to put in the car, along with some instructions, in less than 3 hours. good thing I have the complete JDM harness (minus the ignitor wiring) laid out on the floor of the garage.

I am more worried about where the heck ALT-2 is on the USDM harness. That's what I think I have to hack into and cut on the car. But that can wait until this weekend. We have some time to sort that out. All I know for now is that the hot wire that went to the USDM ignitor is going to the JDM coils...

"hot wire that went to the usdm ignitor is going to the jdm coils"

Looking over old jdm wire diagrams and not not having the usdm stuff, sounds like a good place to start. Just add a sticky note to the dash that reminds you what wires you spliced when it doesn't start.

Kevin,
That's the thing. The JDM wiring is so completely differen than the USDM Ej22T. After thinking about it a bit, I decide that was the wrong approach.I made the decision last night to leave the power wire to the ignitor cut and capped.

I tapped into the power wire for the Ej22t coil as the power source for the individual coils. I am thinking that this is going to work. If it doesn't I will cut the wire of power coming out of fuse ALT2 and splice it into the other circuit.