ECU not booting, ECU not grounding main relay pin 8

Sooo my 89 GGSX has been sitting for awhile. T-belt snapped shortly after I bought it, sat for about 6 months. Got around to doing the head and belts. It ran great but had a no-start when hot at that point, and tags were expired, so I parked it for about another 8 months (I have a few cars). I went to move it/wash it a couple weeks ago and it just did a no-start. I noticed the CEL wasn't coming on, I had no spark, and no fuel pressure. So I checked the main relay (aka MPFI relay), and found I had power into it on pin 10, but no ground on pin 8(from ECU). Pulled the relay and bench tested, works good. Checked ECU ground on ECU pin 104 and IGN pwr on ECU pin 110, good ground on 104, 12v on 110. Then I checked continuity between main relay pin 8 and ECU pins 63 and 66. Perfect continuity (.03 ohms). I pulled out my ecu which had been replaced by the previous owner, caps look new, board looks clean, it has a sticker from the company that rebuilt it. Hell it even smells good. However it obviously isnt grounding pin 63 to close the main relay. I pulled the cover off the main relay, and closed the contacts by hand and the CEL comes on. when I attempt to start it the second set of contacts in the relay close, and I observe spark on my tester, however it still just cranks. It HAS been sitting for awhile though, but I put fresh non-ethenol(no e-10) gas in it a couple weeks ago when I tried to start it, and it was stored nearly on empty. I'm not sure where to go from here really. /ubbthreads/images/graemlins/banghead.gif It seems like there is a fault in the ECU, but maybe I'm overlooking something. I'm going to continue poking at ecu pins to see if I can't find another fault. All testing is being done with a Fluke 88v. Any DSM gurus have any help? I'd appreciate it.
-Clancy

 

click

 

Quoting Speed_Racer:
I pulled out my ecu which had been replaced by the previous owner, caps look new, board looks clean, it has a sticker from the company that rebuilt it. Hell it even smells good.


The ECU was done once already. I have the paperwork. I pulled it and looked and didn't see any leaks. Admittedly its something I have not had to deal with yet, but I have known about this issue forever. I was just looking at Terry Posten's thread on his ecu work. I think I may just be sending my ecu. Any telltale signs I may see other than just acid, leaky nasty caps?

 

Look for IC110 on the board, one of the vertical ceramic boards. Check for continuity between pin #1 of IC110 and pin #110 on the back of the ECU. Then, check for continuity between pin #4 of IC110, and pin #63 on the back of the ECU. I'll bet you that last one is open. If so, you will need to solder a jumper wire under the board to bypass the open trace. Both those traces go underneath a surfacemount cap, where acid is trapped, and continues to eat the traces for years.
What most people keep failing to realize about these "remanufactured" ECU's is that they ended up at the recycler for a good reason, ACID DAMAGE from bad caps! /ubbthreads/images/graemlins/banghead.gif Most of the repair jobs I have seen are the bare minimum to get it running and out the door. All they care about is that it lasts the 6 months or 1 year of the warranty. So, oftentimes there is lots of acid still left underneath components, continuing to do damage, or many traces that are partially eaten that are just left that way because "it works". Very few places will do the level of repair that SteveP, Terry, or Dave at ECMTuning will do, disassembling the ECU so that all the traces in the area can be inspected and all the acid can be scrubbed out.

 

[stevep]

Quoting keydiver:
Most of the repair jobs I have seen are the bare minimum to get it running and out the door. All they care about is that it lasts the 6 months or 1 year of the warranty.


I doubt they care how long they last at all, as long as it's cheaper to throw another ECU at any returns that might show back up. It clear they do these as cheaply as possible. Often times the work does more damage than the acid did.

Then there is the fool who not only doesn't clean the boards but he drowns the area in RTV to hide the mess.

 

Yeah, thanks guys. I'm definitely sending it to SteveP. Terry recommended him to me. Even though its just a GGSX ecu, I plan on using this engine and ecu in a Suzuki Samurai at some point after I swap in a 1g dsm turbo. So its worth it to me to repair it.
I pulled it back out to send it to Steve, and decided to look closer. This is what I saw

There is a bit of corrosion on the board.

 

[Telecaster]

Any update on this? I have the same problem and i've checked the following for continuity:

pin #1 of IC110 and pin #110 ecu
pin #4 of IC110, and pin #63/66 ecu

I can also ground pin # 8 on the mfi to get power to pin 63/66 to power the ecu. But wired correctly, i'm not getting any power.
Whats the ground source for 63/66 or should i say where does it continue to out of IC110?

 

[stevep]

Quoting Telecaster:
Any update on this?


I haven't received his ECU yet.

In your case if grounding pin 8 of the MPI relay turns the power on the the car starts then perhaps IC110 pin 1 isn't seeing the 12v from the ignition switch IG1 terminal, IC110 pin 4 isn't connected to pin 63/66 anymore, or the transistor on IC110 is blown. This assumes your ECU hasn't got bigger issues.

Send it in and I'll fix it up too. /ubbthreads/images/graemlins/smile.gif

 

Just wanted to post a 'thanks' and another datapoint to this problem. I've got an '92 Talon TSi AWD ecu installed in an EDM GSX/GTI model Galant (Dynamic-4) that's been turboconverted with VR-4 pistons, Evo3 16G turbo etc.

I sourced this ecu used from ebay, and promptly replaced the capacitors which hadn't leaked much yet. I cleaned it as best as I could, and it did work fine for about a year, until now.

The car just wouldn't start after a couple of days of sitting, and I noticed that the CEL wasn't lighting up at all like it should. After verifying that the wiring harness, MFI relay, grounds and 12V inputs were fine I started examining the ECU. I found out that (obviously) the MFI relay wasn't being grounded but otherwise the ECU would work normally if given 12V supply into MFI supply input pins. I also managed to trace the +12V signal lead to IC110, after which I was getting a little despaired since it seemed like the problem was on that hybrid board. /ubbthreads/images/graemlins/smile.gif

However, thanks to keydiver's post which I found via google, I checked the connection between IC110 pin 4 and ECU pins 63/66. Sure enough, the connection was broken somewhere along the way. Doesn't seem like C38 would be the culprit since the area around it is completely clean. Could be broken somewhere else, maybe under the cooling profile under which it is hard to see.

Anyway, running a jumper wire on the back side of the board corrected the problem. /ubbthreads/images/graemlins/smile.gif

 

The problem is that doesn't really correct the problem. If the acid wasn't right out next to IC110, under C38, that means it is back under the heatsink, attacking the capacitor matrix. If you don't pull the heatsink off and clean up the acid, you will have more issues down the road, because the acid is still in there attacking other traces. /ubbthreads/images/graemlins/frown.gif

 

Yeah, that's what I was a bit worried about too. The board is actually quite clean as the caps hadn't leaked much yet when I replaced them.
However apparently some acid had traveled near the filter capacitor matrix. I will need to take that off and clean up in the near future, right now I just needed to get the car running asap.

I've been thinking of converting to a newer flashable EVO5-8 ECU actually, since these seem more reliable and much more sophisticated. They aren't really expensive either and apparently 2G ecu connectors would fit as well.