Car died in traffic.., And it started right back up after havin it towed home.. **FIXED-ECU Options

[kevin311]

Thought I had this issue fixed after I replaced the igniter and dizzy last year to fix a code 12. Apparently not. This has happened twice now. Once before replacement parts, and this is the first time since new parts were installed. Left work yesterday and only got about 5 miles and the car sputtered a little and then just knocked off while I was sitting at a stop light. Had it towed home 35 miles and was able to crank it back up and drive it off the rollback.. Temp never got hot, has been staying just under midpoint of temp gauge (is that normal op temp btw?) CEL didn't come on this time either like it had before. It's a '94 GT 5sp. Not sure where to start. I've read it could be coolant temp sensor giving false reading to PCM and killing the engine bc of high engine temp. Does the 5s really have this fail safe?! Thought it was odd. Also read it could be clogged cat converter, but not throwing any codes. Considering I've replaced just about all the ignition components since owning the car I believe it may be fuel related. Any input is appreciated. Thanks in advance guys.

This post has been edited by kevin311: Oct 6, 2014 - 10:13 PM

[VavAlephVav]

Sounds like what I just went through. code 12 either means the dizzy is not sending the signal, Or the plug connector/wires are broken, Or in my case after I repaired both the ECU was burnt up.
Also be sure you have a ground wire from the intake manifold to the firewall
http://www.celicatech.com/forums/showthrea...13#post30240013

Also jump E1 to Te1 and be sure it doesn't have any codes stored.

This post has been edited by VavAlephVav: Sep 26, 2014 - 10:48 AM

[kevin311]

Yeah I suspect it could be the plug for the dizzy, but not certain. Can the ECU be disassembled or at least have a cover plate I can remove and take a peak? Few weeks ago the code 12 popped back up and after unplugging and replugging connectors it hasn't came back on since. Car has ran great the last 2 weeks of DD'ing until yesterday.

[VavAlephVav]

Yes, peel back the carpet on both sides (under center dash) and remove the 10mm nuts holding the frame, and unplug the connectors. There is also a small relay mounted to it, carefully unplug that. Then it just slides right out and yo can take the covers off it and inspect the board with a flashlight. If you've got issues you probably can see something burnt or melted somewhere.
If you see burnt or melted parts you can be sure the ECU is the problem.

This post has been edited by VavAlephVav: Sep 26, 2014 - 11:53 AM

[jordisonjr]

Is 12 the one that says no RPM signal from dizzy? Or something along those lines?
If so, that's the one I had and I just ended up replacing the dizzy with a used one and it fixed it.
Although, my symptoms were not the same as yours

[kevin311]

Ok cool. Yeah I'm familiar with the ECU location but never actually pulled it out and looked at it. It's definitely worth looking at. Far as the ground on the intake, it's good, I know that's a common ground that gets overlooked bc it's hidden.

Yeah code 12 refers to no RPM signal. First time the car had the code when I bought it (3/2012) and took the diz cap off and the pickup coil was mangled. A salvaged diz fixed the code, until last fall. I replaced the dizzy this time with a new one, code still came on intermittently. Then replaced the igniter as well. The igniter was getting so hot I couldn't touch it. The first time it knocked off (last year) I was cruising at 60mph, CEL came on and almost instantly died. Popped the hood and felt the igniter and I about burned myself lol. Hadn't had a problem since changing the igniter till now. This time it died at a stop, and no CEL thrown before stalling. Weird


Seeing as in I've replaced both dizzy and igniter in almost less than a year, I'm thinking the diz plug is too gapped, or ECU is jacked up. OOORRRRR I've got a bad wire in the harness somewhere.. Hopefully I'll get some time this weekend to tinker with it. For now I'll be DD'ing the MX-5. Thanks for the input so far

[VavAlephVav]

well the signal wires for the dizzy run straight from the ECU plug to the dizzy. more likely problems are in the connectors. more likely the one out on the dizzy because it gets so hot.
Like I show in my thread, you can get a new plug from a Camry in the salvage yard, but the colors of the wires are not in the same place. So you gotta be careful about that.
its possible you may have some other problem. but when something works one min, then doesn't work the next, then starts working again I suspect Electronic gadgets. take a look at the pics of what my ecu board looked like in that link. there were some completely melted components, but also the faces of all the big computer chips looked like they had been hot and the surface coating had peeled away.
if you see shet like that you can be pretty sure its the source of your problem.

I would probably go though the ecu pin test with a volt meter and ohm meter to check the ground continuity from the ecu plugs too.
http://bgbonline.celicatech.com/94_6gmanua...odes%202.2L.pdf
http://bgbonline.celicatech.com/94_6gmanua...out%20codes.pdf
http://bgbonline.celicatech.com/94_6gmanua...ge%20charts.pdf

[kevin311]

Awesome, thanks for that. Yeah the plug is quite brittle and the locking tab is broke so it doesn't "snap" on anymore but still stays pretty snug. Definitely does get hot too. I'll bust out the multimeter soon as I can catch a break.

[jordisonjr]

That's interesting, my CEL was on for a good period of time before any driving issues. Then it suddenly started driving rough, jerking forward and back every so often, and only died once, but started up right after and drove rough again.
I bought a used Dizzy off a member here, swapped it over, and the issue has never come up since, and this was about 2 years ago now.

When it came up with me I was originally told bad ECU by members on here, however, I'm not sure how accurate that info was.

[kevin311]

Yeah that's what's odd. Originally even with the code 12 the car drove fine. After the salvage replacement it was fine for about a year, so I figured it gave out, def got my $35 out of it. So went new and still had the code. Replaced igniter with new and cleared the code finally. Until a couple months ago it's been intermittent. Last two weeks it's driven great with no CEL. Cranked it up yesterday just to listen to it. It's odd, it runs great until it reaches operating temp and then it stalls out. Did it twice to confirm. Looks like I'll have a chance today to open up the ECU and put a multimeter on the wires for the diz and hopefully get to the bottom of this nagging issue.

[VavAlephVav]

While you're at it take ohm readings of your coolant temp sensor. Should be a chart in the manual that shows what it will read. Its the kind of thing that can make the computer act weird if it's reading the wrong temp.

[kevin311]

Will do, when I noticed that, the post I saw on another website about the temp sensor seems feasible. I'll give them all a check today if I don't get rained on.

[Smaay]

check the grounds. the jerking and sputtering sounds just like an ECU ground wasnt solid

[kevin311]

Alright, time for an update.

Checked the coolant temp sensor prior to cranking, outside temp here today's around 65*F. Resistance was just under 3 kOhms, which by the graph is accurate I'd say. Fired her up and before it reached OP temp it tried to shutter off. Long as I fed it some gas it would stay running, like crap, but running. Ran it until normal temp then shut it off and measure resistance again. This time measured at .24 kOhms which again looked accurate. So I'm convinced te sensor is fine.

Checked resistances in pickup coils at diz, they're well within spec. So diz is OK.

Checked trouble codes, codes 12 AND 13 had been stored. 13 is new, when I checked the codes two or so weeks ago, 12 was the only one stored. Reset everything and 12 have erased but 13 still remains. Which is an interruption in RPM signal. By the BGB most likely solution is dirty or loose connectors and/or terminals. We'll see.

Checked the grounds Smaay and all are connected, clean and tight.

Pulled the ECU and took a peak inside, looks like there's a few cooked resistors. At least from what I can tell they're cooked, I'm no electrician, but they don't look right I'll try to get a pic up. How do I post pics in the forum?! Lol

This post has been edited by kevin311: Sep 29, 2014 - 4:59 PM

[brk34]

Smell the ecu if its burnt it should have a distinct smell.

[VavAlephVav]

Pulled the ECU and took a peak inside, looks like there's a few cooked resistors. At least from what I can tell they're cooked, I'm no electrician, but they don't look right I'll try to get a pic up. How do I post pics in the forum?! Lol

you upload them to a site like photobucket or whatever you use. then copy the direct link you get from the photobucket page, and when you post here use the button just above the message box that says insert image and copy the code you pasted from photobucket.

for the most part if you see anything that looks mostly brown or black like its cooked that's bad electronics.

a couple months ago when I got mine there were two dudes parting out 94's, I bought the ecu from one guy, the other guy may still have it.
http://www.6gc.net/forums/index.php?showtopic=93407

he told me $60 shipped. just be sure its a '94 5sfe manual, federal emissions. he's from MA so it should be,
the 94 and 95 should be the same but 96+ will be ODB2

[kevin311]

[kevin311]

Sweet it worked. At first glance I didn't notice anything but looking at the second pic the large gray resistors and powdery blues to the left caught my attention. Maybe they're supposed to be that way but the only resistors I've ever seen are striped for the resistance values, and typically glossy. Someone who knows what they're looking at please feel free to chime in. Thanks to all of you guys thus far

[VavAlephVav]

hmmm, I don't think there's anything wrong with those particular resisters. I don't remember why but I think those are normal, just some special type. be sure and check both sides of the board and look for things like a capacitor with goop coming out of it or burn marks, and melted pieces.
the thing I'm wondering about is the way some of those computer chips look like part of it is shiny black and the other part of it is dull gray like the protective coating has melted away or something. several of mine looked even worse.
and I would use a flash light to look carefully at the 6 cylinder shaped black and silver capacitors, especially the big one in the middle-left. Often you can see physical defects but, even if you can not that doesn't prove the electronics work as they should. especially if it runs most of the time and only acts up once in a while.

This post has been edited by VavAlephVav: Sep 29, 2014 - 9:18 PM

[VavAlephVav]

this is from the trouble shooting with codes I linked before - go all the way down to where they have a flow chart for DTC 12 and DTC 13. because you have the 13 too I'd look at the plug first

DTC 13 - RPM SIGNAL
This code indicates that a momentary interruption of the "G"
and "NE" signal from the distributor to ECM has occurred, but has
returned to normal. This malfunction is usually a loose or dirty
connector or terminal. For further diagnosis, refer to appropriate DTC
12 chart.

http://bgbonline.celicatech.com/94_6gmanua...odes%202.2L.pdf

what you do is start it up, and if you can get it to stay running for minute you go out there and kind of wiggle the plug on the dizzy and see if it makes a noticeable difference or starts acting weird when you mess with it.
you may just have a bad connection. or connector.