I have read online recently that  and I quote "The ECU between Automatics and Manuals are exactly the same. Subaru has a different computer to manage their automatic transmissions." 

I have a 2002 Impreza Outback Sport and while trying a mod to regulate the signal to the ECU from the bank 1 number 2 O2 sensor to trick the ECU into believing the Cat. is efficient, I may have burn up the ECU. 

So... Does the above statement have any merit?? I have found and obtained an ECU from an donor car that was spot in identical except that it was an automatic instead of a manual. 

I am reluctant to swap it and turn the key just to find out if it will work properly. I found out the hard way what can happen playing with the electronics already and don't need to cause any more problems than I already have. 

My understanding is that you're safe in swapping the ECUs, as they're the same in both AT and MT cases, and there's an input (via the connector/harness, of course) that informs it which type of transmission is in place.  The (automatic) transmission is managed by a separate controller, the TCU, which communicates with the ECU.

I assume you've restored your O2 sensor "mod" to its original state.

That's what I was hoping to hear. Thank you 

And indeed I have removed and un did all modifications and everything is back to factory specs. Also have a new O2 sensor (downstream} going in as well in case I fried that too. 

 

I'm going to do the swap and I'll report my findings here. Ever since my failed mod it's throwing code 0037 (bank 1 sensor 2 heater circut low voltage or something to that effect). Which was not present until I shorted that exact heater circuit trying to mod it. 

Okay so absolutely no change to anything after the swap. 

Ever since the failed mod the voltage from the downstream O2 sensor has been between .32v and .33v . I'm fairly certain it should be between .01 and 1.0v. hence the DTC and the 0037. 

It hasn't actually thrown a code yet but scoping the sensor while running is showing me the same results as the stock ECU was. 

 

Is there a fuse or relay on the O2 sensor circuit ?? I believe both O2 sensors are powered together from the same circuit, so I should have lost power to both. 

Which would lead me to believe either the sensor is a brand new defect or I have a wiring issue. 

As mentioned before I fixed all the wires I had cut  or spliced. I'm fairly certain I didn't smoke any wires anywhere in the process and I never even saw a spark or anything arching or smelled any thing burning. 

Any input as to what else I could look for or test to help find the source of my problem? 

 

Oh and yes the #1 sensor is operating as it should no problems at all. 

Guess I need to follow the loss of voltage and determine the cause  and focus on the fault code instead of what I'm guessing I did to hurt it. 

Look to the exact schematic for your model and year.  My guess would be that the wires go straight from the ECU to the sensor (i.e. no fuse between), so you're probably down to a wiring fault.

for service manual and wiring diagrams, try here: http://jdmfsm.info/Auto/Japan/Subaru/

 

I’d trace the known issues specifically  the code and back track that wire if you have no voltage and think you should. I’m busy and only have this a cursory glance so I may be mis understabding

To confirm in case you’re hesitant since you still have issues - AT and MT ECUs are interchangeable.

ATs and MTs can in a few years, particularly that 00-02 range you’re in, have differing trigger points on the crank and cam sprockets which does lead to some swap and “signal” issues. This can make for some uncertainty in commentary online as people discuss this. There are issues swapping some things but  are no issues swapping just the ECU like you did. 

I've driven about 40miles with multiple stops and no CEL , but it's just a matter of time. 

I'm certain what I did , could not have caused any wires to melt. The short circuit I made intentionally was done with a resistor and capacitor and done on the 1 volt signal wire. 

So yes I will start with chasing the codes and I will forget about the the possibility that I screwed up . Well....at least for now anyway. 

So this hack was just about slowing down (damping) the response from the sensor.

A working converter will generally show a pretty flat rear O2 signal around 700 millivolts.

GD

.Yes , thats exactly what I was after. Just trying to tame the signal from the downstream sensor. Because I can't pass smog here in California when my subi keeps spitting that dreaded P0420 code !! And a California Cat is around $600 installed. 

The resistor and the cap for the mod were like 3 bucks 🤣

So I had to try it !!

.... But look were it got me. 

 

 

As I type I am testing the wiring . Just had the CEL come on , this time is was p0136 *circuit malfunction" which completely makes since. Just need to pin point the malfunction now!

 

37 minutes ago, CodemanLivs said:

.Yes , thats exactly what I was after. Just trying to tame the signal from the downstream sensor. Because I can't pass smog here in California when my subi keeps spitting that dreaded P0420 code !! And a California Cat is around $600 installed. 

The resistor and the cap for the mod were like 3 bucks 🤣

So I had to try it !!

.... But look were it got me.

There are other hacks I've read about for this pesky 0420 thing - one is to use drilled-out spark plug antifoul spacers to back the sensor out of the exhaust stream a little.  Is there anything to that approach?

14 hours ago, jonathan909 said:

There are other hacks I've read about for this pesky 0420 thing - one is to use drilled-out spark plug antifoul spacers to back the sensor out of the exhaust stream a little.  Is there anything to that approach?

Actually that does work really good from what I've heard, but it's not very incognito. I was just at the smog test shop last week and they went over and under every inch of the car with a mirror on a stick and a flashlight just before the failed me for insufficient data, system not ready. 

 

Sheeeeeeeee-it.

On 9/1/2022 at 9:56 PM, jonathan909 said:

There are other hacks I've read about for this pesky 0420 thing - one is to use drilled-out spark plug antifoul spacers to back the sensor out of the exhaust stream a little.  Is there anything to that approach?

Works on 2004 and earlier subarus before the rear sensor became more utilized by the ECU. Pre 2004 the rear sensor data is practically used as just a toggle switch “good” or “not good”.  Very good success rate and worth trying in areas that are not asinine on inspections. They’re available premade on eBay for cheap  

There were also mini catalytic converter like inserts you screw in front of the O2 sensor that had a minuscule amount of catalyst just for the sensor stream. But I haven’t seen one for a few years and I never used those. 

On 9/4/2022 at 7:03 AM, idosubaru said:

Works on 2004 and earlier subarus before the rear sensor became more utilized by the ECU. Pre 2004 the rear sensor data is practically used as just a toggle switch “good” or “not good”.  Very good success rate and worth trying in areas that are not asinine on inspections. They’re available premade on eBay for cheap  

There were also mini catalytic converter like inserts you screw in front of the O2 sensor that had a minuscule amount of catalyst just for the sensor stream. But I haven’t seen one for a few years and I never used those. 

Exactly

And

Good info

Thanks

On 9/1/2022 at 6:56 PM, jonathan909 said:

There are other hacks I've read about for this pesky 0420 thing - one is to use drilled-out spark plug antifoul spacers to back the sensor out of the exhaust stream a little.  Is there anything to that approach?

Actually that does work really good from what I've heard, but it's not very incognito. I was just at the smog test shop last week and they went over and under every inch of the car with a mirror on a stick and a flashlight just before the failed me for insufficient data, system not ready.