legacycentral bbs

I think I have this problem figured out, but I'm writing it up for those with similar setups and problems to help diagnose it. When I find an actual solution for sure I"ll update this thread.

Ok, so I've got a rob tuned ecu, and 550 yellow top sti injectors for reference.

Symptoms:

Starting the car sometimes the engine searches for the right idle and drop quite low almost stalling. If I blip the throttle (like when reversing), and then let go it can and often will die. (May partially be unrelated - was thinking it was IAC valve as others have had the problem)

Once it got colder I noticed the car would run rough while it idled low (can't recall if that was a symptom earlier when it was warmer or not). Engine is very turbulent, car dies often. While in this mode the car won't accelerate well at all. I was leaving my girlfriend's house the other week and the car was running rough, seemed to be sputtering and wouldn't accelerate. Suddenly it was idling fine and drove like nothing was wrong. This is intermittent and one start of the engine and the car will run flawless with a solid idle. Turn it off, and back on again and it runs extremely rough.


What I've done:

At first I was thinking maybe it was a sticky fuel injector, but I haven't tested them or had them benchtested yet.

The bad lack of acceleration and bad idle had me questioning if it was a bad MAF sensor or failing fuel pump, although in my exprience, when I've seen a fuel pump die, it was either working or not.

I bought some MAF cleaner and cleaned up the MAF, after reinstalling it, the idle search seemed to drop dramatically and car ran GREAT. I took it to the store and when I came out the car died rolling out of hte parking lot, ran rough, and proceeded to die 8 more times driving home the 3 blocks I had to drive to get home.

With the car running well I pulled the maf connector and there was no change. Plugging it back in there was a definate change in the car's idle briefly (this may be normal and I don't know if that process is a potentially hazardous one). When running rough, I pulled the MAF connector and there was no change to the car's idle. It was still very rough. I changed to my stock MAF (one that can not work with the Rob Tune ecu because it doesn't account for the same levels of air flow, but I figured idle should be ok). Car idled fine. Tried a few starts with it, and got a start with the same rough idle. So it didn't seem to be MAF dependant. I wrapped on the top of the maf with a plastic end of a screwdriver a few times, and it made no difference. I have yet to test the MAF with a multimeter but I'm leaning towards it not being the MAF at this point.

I also pulled the check plate in the trunk and wrapped on the gas tank access lid that holds the fuel pump, and that made no difference (its a trick to help get the pump to act when it is cold and faulty that a mechanic told me about). The pump sounded like it was working fine, too.

Finally, while it was running rough, and with the MAF unplugged, I looked at the check engine light on the dash. The light was flickering a bit. That could be because the idle was so low and rough, but I had a thought that this could be electrical. I also remembered hearing a little click sometimes when the idle would go from rough to smooth. My car had no problems before I changed it over to the rob tune, and I was seriously hoping it wasn't the ECU. I don't think it is at this point, it is probably related to a bad splice, so I started poking around the wiring under the dash.

I actually started playing with the little 4 to 2 connector that Rob made for the spark conversion differences of the US and oversees ECUs and playing with those wires and suddenly the idle smoothed out. I couldn't make the idle turn rough again by playing with the wires, but I played with the gas pedal some to prove that it was running fine, and shut the car off. I then started it back up and it was rough again. I then went directly back to the little converter and moved it around with it and pulled on the two wires coming out of the bottom (the two side of the 4-2 conversion), and the idle suddenly went smooth again. I did this a third time with the same results. I'm not 100 percent sure this is the problem, but I'm becoming more confident it is here.

I've sent out an email to Rob and hopefully will hear back from him soon. In the mean time I might see if I can pull the cover off the little guy and see if anything looks amiss. Maybe it will just require a simple solder patch, or maybe it isn't the problem at all.

I'll update when I get more information.

-Ryan

Just a couple comments/questions for you.

Does this only happen when the engine is "cold" or can it happen after the engine is completely warmed up?

You mention pulling the MAF connector, are you talking about the electrical connector that plugs into the MAF sensor? If so, the engine should die when you unplug the MAF electrical connector. The ECU will run, albeit poorly without a MAF sensor. Unless Rob has done something different to these ECU's, I would check the MAF & MAF wiring. It also sounds like the ignition converter might need to be looked at as well.

You may also want to check this out too.

http://bbs.legacycentral.org/viewtopic. ... 30#p395930

Ok so an update. I've done a thorough visual inspection of the converter. I've pulled out all the pins in the ecu wiring harness and they seem to be just fine. There were a couple wires with tears to the shielding and one that looked like it had a lot of the wires broken. I replaced and re-soldered those wires back onto the little converter module. The problem is still there.

So I searched the web for some diagnostic documents for the ej20G which I assumed would be pretty close to the ej22t. I finally found them and found a pinout so that I could figure out what pins were what.

I was a little confused as to how to get the resistance readings out of the MAF, so I just did the voltage diagnostics.

The voltage for the power line with the ignition on was like 12V which was right in the range. I also checked the voltage on the signal wire and the ground wire (pin 9 and 10) with the ignition turned to hte on position and the car idling.

I also double checked all my wiring changes when converting to the ej20g ecu. All of it seemed pretty sound visually and physically, although I didn't test it all with a multimeter or anything.

I tested the car while it was running rough and running fine. I think the values were pretty close if not the same, but I failed to write them down and don't remember them. So here are the values for when the car idled fine.

Ignition on:
pin 9 - 0.26
pin 10 - 0.16 (that seems a bit high to me. It should be a ground wire so it should be 0 right?)

Car idling:
pin 9 - 1.26
pin 10 - 0.06


The signal wire seemed to be within the range, although the ground wire seemed to be a bit off.

With the car running, I pulled the MAF sensor connector off of the MAF. The rpms dipped slightly but otherwise no noticable change in the car's idle. When I put the connector back onto the MAF with the car idling, it had a hickup. It had a slight sudden stutter then ran normally. This is NOT how it should behave correct?

Again the car seems to consistently run rough at this point when I first start it. While it is running rough, sometimes all it takes is a slight tug on the ignition converter to get it to run fine. It isn't as easy to get it to go back to running rough, which if it was an electrical fault you'd think it should be just as easy to get it to run rough or run smooth by just jiggling some wires around. Maybe it is all coincidental, but it seems to respond consistently.

It does seem my MAF may be faulty, yeah? I've been suspecting the converter this whole time, and have tried to troubleshoot it but I may have to send it back to Rob for a more thorough test.

Any thoughts?

Ok so no one has replied to my most recent post yet, so I'll just add some more info.

I borrowed another MAF from a friend and hooked that up to the car. Oddly enough, this one has the same part number but only has 4 prongs instead of 5. Mine has 5 prongs. I don't know if that makes any difference, I think there was an additional ground or something in there on some of them?

I was able got the car running with the new MAF and when I disconnected the harness connector from the MAF sensor, the car died. I tried this a few times. I then plugged in my MAF into the car, and did the same procedure. Well guess what? This time the car died each time as well. Tonight is the first time my car died when doing that with my MAF.


I poked around and did some voltage readings again with the new MAF in the car. Again, with the ignition turned on, but not running, the signal wire was within its designated range, but the ground wire was at 0.16, and at idle it dropped to 0.06.

I checked the voltage going to the ignition converter from the ecu as well. The power was 11.4 volts, and the ground was at 0.5 volts while the ignition was turned to the on position. The volts read 13.7v for the power and 0.4v for the ground while the car was idling.

What should I be seeing for a ground voltage? Ideally it would be 0v, and I've heard that anything below 0.09 was fine somewhere, but 0.4 seems high.


The only other thing that I could think of that I could check would be the two ignition wires that go from the converter into my wiring harness. What I did was wrap the yellow and green wire around the pins that had been in the stock ecu and dropping some solder on them to hold them in place. Maybe I could benefit from snipping and connecting those wires directly either by soldering them or putting in some connectors.

Ok Update number 3:

I rewired the two wires coming out of the ignition converter to see if my "wrap around the pin and solder" method was not working well.

I started the car and ran fine. After playing with the wires a bit and the converter box I managed to get it running rough again. I went and started pulling spark plugs to see which cylinder wasn't firing. Turns out the front two cylinders were both not fireing. So that's what, cylinder 1 and 2 I think? I'll have to double check. So if both of them are not firing, then it isn't likely that I have two bad pins in the ecu, its more likely there is some sort of wiring short going from the 2 yellow wires to 1 yellow wire conversion.

I'll keep troubleshooting but I'm getting closer.

-Ryan

Ryan,

Thanks for the updates. It sounds like you may be on the right track with wiring. Keep us posted on what you find.

I tried rewiring both the yellow wires going from the harness to the converter, and in the process the blue jumper wire came off the board. I tried my best to get it back on but after putting the convert back in the car it still wouldn't run right and this time, no amount of "futsing" with it would get cylinders 1 & 2 to fire. I got a hold of Rob and he told me just to send it back to him and he'd get me another one hopefully in the next couple weeks, maybe after his vacation.

So for now, I'll give an update once I get the new one put back in the car.

That sucks, but sounds like sending it back to Rob is the best option.

Got word back from Rob today. My converter is in the mail. One of the surface mount resistors had a bad solder job. It looked good visually but on the bench the problem I had reported showed up to him as well.

As always Rob was very helpful and responsive. Took a little time due to being on vacation but I am excited to get the car back on the road.

Sent from my Galaxy Nexus using Tapatalk 2