Need an EFI tech!

[ripperdog]

I need some help from someone experienced at trouble shooting EEC-IV systems. Some time ago I swopped a EEC-IV system in to my '85 Mustang GT convertible and have never been able to work-out all the bugs.

The car starts and runs but exhibits all kinds of drivability issues including a wandering idle and difficult warm engine starting. It will almost always show a check engine light after running for a short period of time.

When performing a KOEO test the following codes are indicated: 67, 85, 22, 34, 51, 53, 54

All the sensors, switches and actuators that are producing fault codes check-out O.k (with the exception of the cannister purge solenoid). The EEC-IV computer is from the Ford Cobra mass air conversion kit (includes mass air meter & 24 lb injectors). The main wire harness was sourced from a Mustang specialist junk yard. I requested a harness from a '91 model to match the engine harness, but who knows what they actually sent me?

I suspect a problem with the main wire harness but would like to solicit a second opinion before removing it. Pulling that damn dashboard back apart in order to remove the harness is my worst nightmare! Someone Please Help!!!

[Fat0eknee]

yeah i dont really know the codes off hand but if you have a stock motor those 24 lb'ders need to go

[tmoss]

If you don't have the GT40 heads and Cobra intake to match the EEC/24# injectors, you might be running rich. You know the Cobra used a 19# mass air and the EEC made the fuel corrections for the 24# injectors right? So, you need to run a 19# mass air meter with the Cobra EEC and 24# injectors.

[PKRWUD]

Was the engine fully warmed up when you ran the KOEO self test? What is your TPS voltage at idle (with the engine warmed up)? Have you tried a different BP sensor?

[Unit 5302]

Sounds like something's broke or not working right.

[ripperdog]

Yes, the engine was warm when the KOEO test was run...actually I have run the test several times just to confirm that it was throwing the same codes consistantly. It's been a while, but I believe that I set the TPS voltage at .96v. I don't have a spare MAP/BP sensor, but the one I installed was new and I know that when I tested the output, the voltage was in spec for my elevation.

Yes, the '93 Cobra computer was calibrated specifically to work with it's 70mm mass air meter and 24lb injectors. They will not run properly if the parts are changed individually but, if the parts are used together, than it should work just fine...at least according to Ford.

Thanks for all the input. Please feel free to chime in with any additional ideas or suggestions!

[tmoss]

Originally Posted by ripperdog

I don't have a spare MAP/BP sensor, but the one I installed was new and I know that when I tested the output, the voltage was in spec for my elevation.

The voltage is just a carrier, the EEC actually gets a varying frequency with changes in altitude. The frequencies are 122.4hz for 17" of mercury to 159.6hz for 30" of mercury.

[ripperdog]

tmoss,

You are 100% correct. My mistake...I've been away from this stuff for almost 4 years. The MAP/BP sensor's digital output is converted to analog by the computer and can only be read as frequency with an oscilloscope or a specialized MAP/BP tester. I have access to neither at the moment so I'll just have to wait untill I can borrow a "known good" sensor.

The thing that struck me as most odd was that, according to the KOEO codes, a combination of 7 sensors, switches and actuators are reading faulty or out of range. I can see the possibility of a couple being bad but what are the odds of 7 going bad at once? Can one bad sensors cause all the rest of them to fail? Or should I be looking for a common thread such as a faulty SIG RTN wire?

I was hoping to find someone more experianced at dealing with these kind of problems to get me pointed in the right direction. I'd rather fix it in the car than have to rip-out the whole system and disect it on the garage floor.

Thanks for the help...keep the ideas coming!

[tmoss]

Some newer multimeters have a frequency function and very high input impedance, so the frequency can be read with them too. Yes, I would look at the sig rtn wire or the regulated 5v vref that they use. I use an older Fluke 8060A for that function. I have some nice color diagrams you could use at:

www.veryuseful.com/mustang/tech/engine

[Capri306]

Sounds like bad grounds or something not pinned-out right in the harness. I'd double check the wiring second, the ground connections first, all with a schematic and test light...or something like that. Hopefully it's something relatively simple and easy to fix. Good luck!

[Jeff Chambers]

Originally Posted by Capri306

Sounds like bad grounds or something not pinned-out right in the harness. I'd double check the wiring second, the ground connections first, all with a schematic and test light...or something like that. Hopefully it's something relatively simple and easy to fix. Good luck!

Grounds are the first thing I'd check as well. I've seen multiple sensor codes like this before and found that the EEC ground (located up in the right kick panel) had broken. Fixed the ground, fixed the problem.

[ripperdog]

Well, I found the problem. There is a burnt trace on the ECU circuit board leading to pin #46 (sig rtn). This explains why all the sensors on the sig rtn circuit were reading out of range and why the engine was running so badly. The cause of the problem, however, was a bit more difficult to find.

I didn't know at the time but, when I bought the harnesses I was incorrectly supplied with an O2 sensor harness for an automatic transmission car instead of manual transmission. Automatic cars supply 12volts through the neutral switch circuit where as manual transmission cars work with the 5volt vref. The cross-over is done in the O2 harness. The first time I stepped on the clutch it dumped 12 volts down the sig rtn wire and cooked the computer.

The fix was simple...just move one wire in the O2 sensor harness plug. Unfortunately, now I need a new computer. Anyone have one for sale?

Anyone planning an automatic to 5spd tranny-swop should make sure that they get a new O2 sensor harness if they are planning to change to a 5spd computer.

Thanks again for all the help! With any luck I should have this thing back-up and running sometime soon.

[tmoss]

Are you sure you can't scrape both sides if there is good trace and solder a wire to bridge the bad trace?

[Capri306]

Originally Posted by ripperdog

The fix was simple...just move one wire in the O2 sensor harness plug.

HA! I'm not surprised in the least amount by this news, but still, ain't it a hoot? Electronics and electrical problems in general are usually like this. Guess that's why I'm studying to be an electrical engineer: if you know what to look for, you're 90% there.

I'm very happy that you found out what it was so quickly, though. Props to you, but sorry I can't help you with a new EEC-IV PCM. I'd probably look for one in a boneyard out of a REALLY wrecked Mustang, because unless it was absolutely FUBAR in the passenger side, the computer should be fine.

tmoss: I think he'd be best off finding a whole new PCM because if there was a voltage surge up to the PCB, not to mention through the signal circuits, it's cooked and may present problems later down the road, even if the gap is bridged. There is a gap there for a reason, and bridging it would be like sticking a piece of aluminum foil into a fuse socket " 'cuz the damn thing keeps blowing." hehehe, i have to laugh a little because I've done that before to get home...but, I digress.

[tmoss]

The signal return is a ground circuit, sothere is a chance the opening race did not damage anything above the ground plane. I'd give it a try anyway.

[88fivepointoh]

Originally Posted by ripperdog

Yes, the engine was warm when the KOEO test was run...actually I have run the test several times just to confirm that it was throwing the same codes consistantly. It's been a while, but I believe that I set the TPS voltage at .96v. I don't have a spare MAP/BP sensor, but the one I installed was new and I know that when I tested the output, the voltage was in spec for my elevation.

Yes, the '93 Cobra computer was calibrated specifically to work with it's 70mm mass air meter and 24lb injectors. They will not run properly if the parts are changed individually but, if the parts are used together, than it should work just fine...at least according to Ford.

Thanks for all the input. Please feel free to chime in with any additional ideas or suggestions!

Absolutly, if they are all matching parts they will be just fine. #24s on a motor that is stock will run fine. I ran them just fine for quite a while. #19's are to lean in my opinion, 24's are good.

[ripperdog]

I've been seriously considering bridging the burnt area of trace with a piece of lite gauge wire but I'm just not sure that I have faith in either this ECU or my soldering skills. However, if I change to a computer from something other than a Cobra, I will have to "pony-up" for a new air meter as well. I guess I should find a cheaper hobby!

[PKRWUD]

I've been warning people about that for years, but from a slightly different angle; never use a manual tranny ECM in an automatic car without clipping the wire going to pin 30, or else the ECM will fry.

[Capri306]

My apologies, tmoss, I didn't comprehend the fact he's using a Cobra PCM. In this case, I'd go for it, too, and try to jumper the severed connection. Be sure to use sandpaper or at least scratch the PCB until it's clean enough to solder!

ripperdog: if you don't think you have the experience or ability (nothing against you at all) to solder the gap, I'd suggest you at least try to find someone who does it professionally, even if it costs a couple of bucks. You might want to find somebody in an electrical or electronics trade, or somebody that has a lot of TVs sitting in their yard. Well...mebbe not that last recommendation, but you get the idea. Hell, grab a (good) tech at Best Buy and ask him to take a look at it 'after hours'! Might be worth it.

PKRWUD: do you remember a loooong while back when you asked me about why carbureted drag race cars use 'longer runners', and street cars used 'short' ones? Well...I have the answer for ya. PM me if you wanna hear my rant/babbling about it. ...and yes, that harness pinout is a b!tch, ain't it? Makes you wonder what they were smoking that day in the lab...