Jump to content
SAU Community

Recommended Posts

Hi everyone,

the car is a near stock (cat back exhaust and POD) a31 Cefiro, factory manual RB20DET. I've just changed the motor for another from a wrecker because the old one cooked and had very bad compression. I purchased an automatic spec a31 rb20det (but with clutch/flywheel on back), but switched over the loom, ecu, sensors, injectors, coils, plugs, ecu, flywheel, box, etc. Previously the old engine with these parts ran fine apart from being gutless and overheating.

Now the engine seems to be running on 5 cylinders. It sounds like a wrx and unplugging the coil for cylinder 1 has no effect. I switched coils and plugs for cylinders 1 and 2 and the problem remained for cylinder 1 only, leading me to believe the coils and plugs are ok.

I then unplugged number 1 injector plug, which made no difference to the running of the engine. Multimeter at the loom plug for this reads 0.7volts while running so I figured I had a busted injector.

Switched it for a spare today and it made no difference. This fresh injector clicked with 12v applied so it shouldn't be seized.

I ran the eccs diagnostics and got code 54: Signal Circuit From A/T Control Unit To ECM which puzzles me. The car is factory manual with the same electronics as before, so why would it be telling me it wants an auto signal? Only thing I can think of is that I've missed a plug somewhere switching the looms, if so where might it be?

Anything else I can try?

I hope people don't just dismiss this as another coilpack fault without reading because I'm sure it's not from my tests. Thanks in advance for any assistance, this has me stumped :)

Edited by l6a6w6s6
Link to comment
https://www.sau.com.au/forums/topic/94410-running-on-5-cylinders-not-coils/
Share on other sites

I wouldn't worry too much about the ECU error code right now.

It would appear that one of the wires to #1 injector has a break in it somewhere. Use your multimeter to check continuity of the wiring. 1 side will connect to the ECU (possibly pin 101). The other side is +12V when the IGN is ON.

Ok, did some tests. Both injector wires measure 12v when ignition is ON and no engine running. Tested continuity for pin 101 (white with silver dash) and it matches one pin of injector plug, pin 109 (white with black stripe and silver dash) matches the second injector plug.

I've also tried two different ecus, one an a31 manual and the other an a31 auto and the car runs exactly the same with each, and also they both give diagnostic code 54.

Any more help is appreciated.

Have you actually tested spark output from Coil/Cyl 1?

If you don't have spark it will be a faulty ignitor module (on top of cam cover at rear of motor).

The ignitor module has a transistor for each cylinder that switches the coil on and off for spark! So if the transistor is stuffed for Cyl 1 it will never spark.

It could also be a faulty/broken wire that goes into the module as well for cylinder 1.

If you measure +12 on both sides of the injector plug i would say the injector drivers are fine. (To confirm injector operation, its easiest to measure the duty cycle (with capable multimeter) of the injector whilst running, to confirm its opening)

I've run the engine with the coil disconnected from the spark plug hanging out to view, with number 2 along side it to compare. Both sparked so I assume that meant they were ok and the ignitor channels as well.

I just did another test, plugging in spare injectors into the loom whilst the engine was running (injector just loose/not in fuel rail) and could get any to make a sound.

Could the ecu be purposely cutting cylinder 1 because of the code 54?

Edited by l6a6w6s6

Hmm, you have injection and spark, but only running on 5.

Maybe there is a big hole in the piston and no compression? :(

Sounds stupid, but are the plugs from the wiring harness connected to the correct number injector and number coil? Usually there is enough wire length for cylinders that are side by side to be put on back the front.

Edited by RB30-POWER

Ok so all of a sudden it just starts working. I knew the plug was getting pulse

but wasn't sure if the injector was firing or not (rtoo much vibratio and noise t hear or feel the click. So I took the cas out of the head but still plugged in, and rotated it with the ignition on. The number 1 injector definately clicked then. Even so I decided to change it back to the original one which I tested with some direct 12v, clicked on command as well. Switched them over and then it just fired up properly like nothing was ever wrong. Unbelievable but true, this is with the same initial components with what the problem started.

Car running well now except for a slight miss as it hits decent boost, but that's not to worry I'll put it on the dyno this week just to check the mixtures and what not. Thanks to everyone who posted with ideas your help is much appreciated. Special thanks to BHdave for some solid time going through everything with me on ICQ.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


  • Similar Content

  • Latest Posts

    • My thinking is that if the O2 sensor is shot then your entire above described experience is pure placebo.
    • Here is the mess that I made. That filler there was successful in filling dents in that area. But in the middle area. I can feel dents. And I've gone ocer it multiple times with filler. And the filler is no longer there because i accidently sanded it away. I've chased my tail on this job but this is something else lol. So I'm gonna attempt filler one more time and if it doesn't work I'll just high fill primer the door and see where the issues are because guidecoat is of no use atm.
    • Ok, so I think I sort of figured out where I went wrong. So I definitely overthinked it, and I over sanded, which is probably a large part of the problem. to fix it, I ended up tapping some spots that were likely to be high, made them low, filled them in, and I tackled small sections at a time, and it feels a lot better.    I think what confused me as well is you have the bare metal, and some spots darker and some are lighter, and when I run my finger across it, it' would feel like it's a low spot, but I think it's just a transition in different texture from metal to body filler.    When your finger's sliding on the body filler, and crosses over to the bare metal, going back and forth, it feels like it's a low spot. So I kept putting filler there and sanding, but I think it was just a transition in texture, nothing to do with the low or high spot. But the panel's feels a lot better, and I'm just going to end up priming it, and then I'll block it after with guide coat.   Ended up wasting just about all of my filler on this damn door lol  
    • -10 is plenty for running to an oil cooler. When you look at oil feeds, like power steering feeds, they're much smaller, and then just a larger hose size to move volume in less pressure. No need for -12. Even on the race cars, like Duncans, and endurance cars, most of them are all running -10 and everything works perfectly fine, temps are under control, and there's no restrictions.
    • Update: O2 sensor in my downpipe turned out to be faulty when I plugged in to the Haltech software. Was getting a "open circuit" warning. Tons of carbon buildup on it, probably from when I was running rich for a while before getting it corrected. Replaced with new unit and test drove again. The shuffle still happens, albeit far less now. I am not able to replicate it as reliably and it no longer happens at the same RPM levels as before. The only time I was able to hear it was in 5th going uphill and another time in 5th where there was no noticeable incline but applying more throttle first sped it up and then cleared it. Then once in 4th when I slightly lifted the throttle going over a bump but cleared right after. My understanding is that with the O2 sensor out, the ECU relies entirely on the MAP tune and isn't able to make its small adjustments based on the sensors reading. All in all, a big improvement, though not the silver bullet. Will try validating the actuators are set up correctly, and potentially setting up shop time to tune the boost controller on closed loop rather than the open loop it is set to now. Think if it's set up on closed loop to take the O2 reading, that should deal with these last bits. Will try to update again as I go. 
×
×
  • Create New...