|
|
|
|
Thread Tools | Display Modes |
03-21-2021, 04:54 PM | #1 |
Scooby Newbie
Member#: 520461
Join Date: Nov 2020
Location: Toronto
Vehicle:2004 WRX Silver |
Defective Ecu (or Icm)?
Hey guys I could really use some help with this so please dont get scared off by the amount of writing, i'll keep it very simple.
Engine: Jdm Ej205 single avcs ECU: 2001 Jdm Sti Chassis: 2004 Usdm Wrx (wired to work with avcs and a tgv delete) Ive been having a "crank no start" problem with my wrx and have been able to narrow the problem down to no spark on cylinders 3 and 4 (the back two cylinders) All 4 coils are getting constant 12v when "key on " and grounds on all 4 are perfect. I was able to trace the ignition coil signal wires back to the Ecu. I swapped the signal wires for cylinders 3 and 4 for cylinders 1 and 2 (JUST AS A TEST I WILL NOT LEAVE THE WIRES IN THIS CONFIGURATION). Now the back two cylinders have a spark and the front two dont. Im starting to think that the ICM inside the ECU is defective since it is only sending a signal out to 2 of the cylinders and the issue seems to follow the ECU pins and not the coil pack wiring. If anyone can weigh-in on this problem id appreciate that, I really cant think of another explination for this
* Registered users of the site do not see these ads.
|
03-21-2021, 09:33 PM | #2 |
Scooby Guru
Member#: 53443
Join Date: Jan 2004
Location: Anchorage, AK
Vehicle:2001 LL Bean Outback Winestone |
Have you verified that the wiring between the ECU and the is okay? If they are good then I would say the timing signal from the crank needs to be checked. You could also try back probing both of the coil wire connections on the ECU to see if there is a difference in the resistance reading. If there is a significant difference then it seems the ECU may be the problem.
|
03-21-2021, 09:55 PM | #3 |
Scooby Newbie
Member#: 520461
Join Date: Nov 2020
Location: Toronto
Vehicle:2004 WRX Silver |
I resistance and Continutity check each of the 4 coil pack wires from the ecu to the engineand all of them are working fine. Ho would I probe the coil wires on the ecu? Does that mean unplugging the ecu from the harness and checking the resistance of the internal ecu pins to the ecu shell as a ground?
Also timing was checked to be in time and crank sensor was replaced recently Last edited by Bluebieru; 03-21-2021 at 10:37 PM. |
03-21-2021, 10:46 PM | #4 |
Scooby Guru
Member#: 53443
Join Date: Jan 2004
Location: Anchorage, AK
Vehicle:2001 LL Bean Outback Winestone |
Check the resistance of the ECU contact that ties to the coil and ground. Compare the resistance readings of both sides. They should be close to the same value. You might need to use the diode mode for the check.
The crank sensor may be just fine but if one of the reluctor teeth is broken or damaged that will cause the issue. |
03-23-2021, 03:42 PM | #5 |
Scooby Newbie
Member#: 520461
Join Date: Nov 2020
Location: Toronto
Vehicle:2004 WRX Silver |
So I checked the resistance of the ecu pins between coil trigger and the ground. Resistance is identical for the firing and non-firing coils, so I’m guessing that’s pointing to a processor error. Also I just cleaned the crank gear reluctor teeth and checked timing marks again, it all looks good. Seems like my problem really is a faulty ecu.
|
03-24-2021, 08:19 PM | #6 |
Scooby Guru
Member#: 53443
Join Date: Jan 2004
Location: Anchorage, AK
Vehicle:2001 LL Bean Outback Winestone |
I suppose you could verify that the crank signal waveform is good before replacing the ECU but it does seem that the ECU is the culprit from what you say here.
|
03-25-2021, 09:34 PM | #7 |
Scooby Newbie
Member#: 520461
Join Date: Nov 2020
Location: Toronto
Vehicle:2004 WRX Silver |
Ok so I tried a friends ecu that's known to be good still no start
I used romraider to read the cam and cranks signals, and both seem ok (?). The cam signal oscillates in a square wave form and the crank signal stays solid at 1 or "on" (im assuming that's because romraider logs to slowly to see each spike and dip). Romraider was also reading 114 rpms at cranking so I'm fairly sure the crank signal is good. I'm going to check all my grounds again tomorrow for the hundredth time but other than that I'm not sure what else there is to try. Can anyone confirm 100% as to what exact inputs the ecu needs to fire a spark? I know cam and crank but anything else like maf, avcs, fuel relay etc? Really anything else I can check at this point? *EDIT* would a stock security system cause the problem? *EDIT #2* I did a bit mor logging on romraider. It says my battery voltage is ~12.25v but the drops to 7.5/8.5v while cranking. Can that cause problems with the ecu? Last edited by Bluebieru; 03-27-2021 at 04:02 PM. |
04-04-2021, 02:58 PM | #8 |
Scooby Guru
Member#: 53443
Join Date: Jan 2004
Location: Anchorage, AK
Vehicle:2001 LL Bean Outback Winestone |
I would like to know what the voltage is on the controlled ground connections from the ECU to coils while the ignition is ON. All four leads should have voltage on them near battery voltage. If that is correct then that should prove out the coil circuit though you seem to have done that pretty well already.
I don't know where you are connecting to power on romraider but seeing voltage as low as 7.5 volts isn't real good. Battery voltage normally drops to around 10.5 volts while cranking and I don't think the starter would work if it was below 9 volts. The source of power to the romraider may have a connection problem or the grounding to the starter may be causing the problem. To eliminate the main ground as a problem you could use a jumper cable to bypass the normal ground lead from the battery too the starter to see if that helps. If that doesn't change things then the ground should be okay and you may need to investigate the power source side for a resistive connection issue. You can also check for grounding problems by using your voltmeter, to check for voltage while cranking the engine and using a known good ground point, like the battery post, across to the suspected bad ground point. A good ground connection will have little to no voltage on it when referenced to a good ground point. You may need to get your hands on an oscilloscope to verify the crank sensor signal. Since one coil works but the other doesn't means the ECU should be working normally. As far as I know the only signal required to fire the plugs is the crank signal. I think the CAM sensor activates the injectors but perhaps the ECU requires both signals to fire the plugs, I'm not certain of that. |
05-09-2021, 03:13 AM | #9 |
Scooby Newbie
Member#: 493479
Join Date: Nov 2018
|
I am currently having this same problem. Any updates? On mine I have checked the entire harness for breaks or any damage to the wires, tested everything I can think of, replaced crank sensor and the crank sprocket, and I’m still showing no pulse of 5v to the 3-4 side of the coil from the plug.
2005 Outback 2.5 SOHC NA MT |
05-09-2021, 07:09 AM | #10 |
Scooby Guru
Member#: 53443
Join Date: Jan 2004
Location: Anchorage, AK
Vehicle:2001 LL Bean Outback Winestone |
Where are you measuring the 5 volt signal at? Are you checking pins 1 and 4 of the ignition coil?
|
05-09-2021, 01:28 PM | #11 |
Scooby Newbie
Member#: 493479
Join Date: Nov 2018
|
Yes. I checked pin 1 while someone cranked it, 5v pulse, pin 2 and 3 key on is pos and neg, pin 4 while crank no pulse no volts.
|
05-09-2021, 01:29 PM | #12 |
Scooby Newbie
Member#: 493479
Join Date: Nov 2018
|
I measured on the plug on the harness, the coil tested fine.
|
05-10-2021, 09:11 AM | #13 |
Scooby Guru
Member#: 53443
Join Date: Jan 2004
Location: Anchorage, AK
Vehicle:2001 LL Bean Outback Winestone |
I assume you are saying pin 2 was 12 volts and pin 3 was zero volts since that is what you should see at those power connection points. Pin 4 needs to be checked at the ECU unless you have already verified that the wire connection is good between those two points. My info shows that ECU connector B135, pin 17 is a green/white wire that ties to pin 4, with a tie connection in between the two ends. Check the continuity of the wire between pin 17 and pin 4. You should have zero ohms.
|
05-10-2021, 03:02 PM | #14 |
Scooby Newbie
Member#: 493479
Join Date: Nov 2018
|
Thank you. Going to test right now.
|
Thread Tools | |
Display Modes | |
|
|