[dlshady]

Hi Guys,
I'm in need of a little help with an EEC-IV problem. I had a customer drop off his 1990 GT last week with a boatload of codes in the computer and running terribly rich. I've done the pin checks and studied schematics until I'm about to puke :barf: and I'm still not seeing the problem. Maybe one of you guys will see what it is that I'm missing.

Here's a quick rundown of what I've got:
1990 GT with an A9L processor
Completely stock motor except for an A-Trim, 30# injectors and matching C&L

A little history:
This guy bought the car with a race motor in it (running just fine) and he also bought a lightly wrecked stock 1991 LX (wrecked so light that it ran and drove fine). He took the race drivetrain out of the GT and put the entire drivetrain out of the LX in it. In the process of getting it all running, he managed to cross up the battery cables for a split second. I don't know if this is at all related to the problem, but I'm thinking its a strong possibility. Anyway, he got the car running and it ran poorly so he swapped the processor with a known good one (also swapped the alternator, just in case) but it ran just as poorly.

As the car was brought to me, it was flooding like there was no tomorrow and the Check Engine Light was on. My assumption was that it was running in "Limp Home Mode" and couldn't compensate for the 30# injectors, hence the flooding. Anyway, it initally showed the following codes: 51, 22, 53, 54, 35, 67 and four others pertaining to the missing smog equipment (which I'm disregarding). I cleared the codes, re-ran the KOEO and came up with the same codes.

After a long (and I mean LONG!) two days of pin checks and testing, it started registering a different set of codes: 21, 22, 23, 24, 34 and 67. Not sure if I pulled something or what, but now its registering a bunch of codes related to the SIG RTN. What struck me as odd is the fact that when the clutch is depressed, the car runs much better and the Check Engine Light goes off. As soon as the clutch comes up, the light comes back on and it goes to flooding again.

Here's where it gets really weird: After some testing and checking, I ended up pulling both of the plugs off of the clutch interlock switch (one plug for the starter solenoid - red/light blue wires and the other for the ECM with black/white and light blue/yellow wires). I thought maybe the switch was bad and possibly bleeding voltage from one side to the other. Then I jumpered each plug, ran the KOEO and the codes were gone. Long story short, the jumper in the starter solenoid side of the switch causes the codes to dissapear. Sounds crazy, I know, but its what I've got. I verified it by cranking the car and removing the red/light blue trigger wire from the starter solenoid. As long as the wire is attached to the solenoid and the switch on the clutch pedal is jumpered, the car runs OK w/no Check Engine Light. As soon as you pull the wire off the solenoid (or remove the jumper wire) the light comes on and the codes 21, 22, 23, 24 and 34 re-apear. I also checked resistance from pin 46 on the ECM to ground and its 2 ohms with the solenoid wire hooked up and 600+ ohms when unhooked.:curses:

What am I missing here? I can't find a relationship as far as why the SIG RTN goes away when the starter solenoid trigger wire is unhooked. I still feel like the ECM is not 100% right because I have tried both of these in another car in the shop and it runs poorly with either of them. My problem is that I hate to go putting another known-good ECM in here because there's a problem somewhere that smoked the second one. I don't want to kill a third one for no reason.....

A couple other things to note:
I have verified that there is no continuity between pin 30 and the clutch switch, which I know needs to be fixed. I don't believe this is the problem, though, because that circuit is not complete anytime that the car is in gear and the clucth is up.
2 ohms to ground on pins 20, 40 and 60
Keep in mind that the battery was momentarily hooked up backwards - may or may not have any relation to the problem

I apologize for the long read, but I'm at a loss. Its probably something so stinking simple that I'm overlooking it, but I don't know what it might be. If any of you Ford Techs read this and have other questions, I have about 10 pages of notes pertaining to voltages and resistances that I've recorded in my checking things. If you need more data, I've probably already got it so just ask.

Well, I guess that's about it for now. I look forward to see if any of you guys see something I dont in all of this!

Many thanks,
Deron

 

[Joel5.0]

WOW....:eeek: ....that one is really a good one (NPI)..

I know this sounds too simple but, since the battery was installed backwards,... the built in diode of the starter solenoid may be kaput, causing problems on the starting circuit & ground.....a voltage noise test to the ground circuit of the starter solenoid/NGS/CES circuits should give the clue...

You don't mention it but, I would also verify the condition of the firewall-to-engine ground strap...if you haven't yet. Hope this helps....and please....LUK what you find out....this is a good one.

PS: Great TS Work!...BTW

 

[dlshady]

Thanks Joel,
Unfortunately, I'd already tried the solenoid and it made no diffierence. I saw the sticker on this one that said "diode protected" and it got me to thinking so I swapped it. This morning I did a little more checking and noticed that the SIG RTN continuity to ground goes away even without the battery hooked up, so it doesn't have anything to do with something being powered up. I checked from the SIG RTN terminal in the diagnostic connector to the negative battery cable (unhooked from the battery) and it is 2 ohms with the solenoid trigger wire hooked up and jumps to around 900 ohms unhooked. Just for kicks, I put a third ECM (known good) in the car since I figured it couldn't let the magic blue smoke loose as long as the battery was disconnected. It checks the same as the other two, so I think its safe to assume that the loss of SIG RTN is not happening inside the ECM. Does this sound feasible to you?
My main question at the moment is exactly where does the SIG RTN circuit go to ground? I had assumed that it happens inside the ECM but now I'm not so sure. Does anyone know the answer to this?
Well, I'm back to work now......Stay tuned for more bone-chilling excitement!:curses:

Thanks again,
Deron

 

[Joel5.0]

Deron;

This one definetely needs to be broken down. (Feel free to cut, paste, correct, change.....anything you deem necessary)

Problem Definition:
SIGRTN is lost/contaminated w/start circuit open, battery connected or disconnected.

Circuit Configuration:
Ignition Switch - Clutch Start Sw - Starter Solenoid

Shared Circuits/Components:
TFI Module Start Voltage Input
Pin 30 NDS input to ECM
Ignition Switch
Starter Solenoid/Relay

Some checks I would do:

1. Disconnect Ign Switch connector @steering column and TFI module and check SIGRTN continuity to ground, starting wire to starter relay connected/disconnected to see if there’s a difference.
2. Check resistance to ground of pin 49 (HEGO ground).
3. Do 1. with ground connection of pin 49 disconnected.
4. Do 1. w/HEGO’s disconnected

SIGRTN ground signal is indeed generated/produced by the ECM (like a filtered ground reference for the variable resistor sensors/switchs), the hooking of the battery backwards did have an impact either on the wiring or a common electrical component, not necessarily EEC-IV related but, it shares either a ground or a VPWR supply source (ie. once had a case that a shorted AODE solenoid valve assy caused havoc on the ECM and the power supply to the injectors (VPWR is shared between the injectors, ECM and the solenoid valve assy.) with no codes been generated)
I would re-verify the starter solenoid to be on the safe side. The absence of continuity of pin 30 to the clutch switch could also be key, it could be shorted to power or a non-EEC related circuit within the wiring bundles it goes through. LUK what you find out,….this one is definetely for the books. GL

PS: The HEGO's should be fried due to the battery mishap, I would keeep them disconnected during the TS and have them replaced when this problem is figured out.

 

[dlshady]

Thanks once again Joel,
The car runs pretty rich, even when the solenoid wire is connected, so I'd say you're right about the HEGOs being toast. I hadn't even thought about the possibility of them being part of the problem yet, though. Yesterday I got to thinking about my problems and I decided to chase the open circuit on the clutch switch since its the only thing so far that I've determined to be bad without any doubt. Its been my experience in the past that when you get several really bizarre problems going on that seem unrelated, often times you'll find the cause of them all in the same area (wires shorted together, etc..). Anyway, I started at the ECM since its easier to get to and went backwards from there. I didn't get too far with that before I realized that I have no idea how the wire gets from the ECM to the clutch switch (the physical routing of the wire). All of my schematics show pin 30 going straight to the clutch switch and then on down to the in gear switch but the harness doesn't run that way. I also have three different schematics and they aren't the same... They show different things concerning the differences between AOD cars and T5 cars.

After some head scratching, I ended up taking out a spare harness that I have around the shop and my checks show the pin 30 wire going only to the round 8-wire connector underneath the MAF (where the main ECM harness connects to the O2 harness). The problem with that is that the pin 30 wire terminates in this connector because there isn't a mating pin in the o2 harness connector. My guess is that the wire terminates here because the automatic cars have a different O2 harness and would pick up this wire and carry it on down to the neutral safety switch...??? Does it split off somewhere back upstream? If so, how does this wire make its way back over to the brake pedal area?

The other thing I noticed yesterday (and this is gonna sound dumb....) What is the function of the switch on the passenger side of the clutch pedal shaft? There is the one switch on the driver side, white nylon with two sets of wires into it: one set of r/lb (starter solenoid interrupt) and one set b/w and lb/y (which I've been assuming were the NDS). The switch over by the quadrant has only two terminals, both black wires coming out in about a 5" pigtail that connect to some red/g wires. What is the purpose of the second switch.

Last bit of confuson for the day.... Right beside the connector for the switch in question above is a 2 conductor plug containing both b/w and a lb/y wires (same colors as the NDS at the pedal and there is continuity back to the pedal) but nothing plugged into it. I'm assuming that this is stubbed out here for some kind of shifter switch on an AOD car...???

Well, I've spent my morning working on other projects but I'm about to get back on the Mustang. I'll let you know what I find!

Thanks again,
Deron

 

[Joel5.0]

Regarding the other clutch pedal switch...it's not a dumb question (I had to look it up myself after getting the big ? myself), it would be the switch used for the cruise control system.

Following the diagrams I have access to (Mitchell, Motor, Veryuseful.com, Autozone.com, Fordfuelinjection.com, Products Research inc.) none mention the details of their routing or connectors. However, they all show pin 30 getting a direct feed from the ST pin @ignition switch (splice near its connector, presume its routed to the ECM connector under dash) and the other wire going directly to the clutch switch/NGS circuit.

At fordfuelinjection.com it shows that there are 2 8-pin connectors through which NGS/NDS and SIGRTN (manual or auto config) are routed. Look at the pic:

Anyway....from your observations of the clutch pedal/quadrant switches one question comes to mind.....are the right switches connected to the proper circuits?...the clutch switch used for the cruise control is connected to the stop light sw(Red-Lt Green wire) and to the Cruise Control Amplifier Module (Lt Green wire), which I assume is a NC type switch. LUK what you find out.

 

[dlshady]

I found It!:king:
Hi Joel,
Well, its done and running! Now its Miller Time.... I got it fixed about 5:00 this evening and then spent a couple hours cleaning up the shop so I don't have to go to work tomorrow. I had intended to post this earlier but I got tied up eBaying and well, you know how that goes....

The short story: My customer left the 1991 O2 harness on the motor when he installed it into the 1990 model car. Seems simple enough but for one reason or another, Ford decided to move one wire in that harness to a different position, therefore the open in the NDS circuit.

The long story: As I mentioned in my earlier post the wire from pin 30 at the ECM went to the 8-pin connector underneath the MAF (labeled "Black Connector #40 in your diagram above) and stopped there. It actually terminated in the hole marked "Manual" in the diagram above, rather than the one marked "Pin 30" and there was no corresponding wire exiting from the other side of the connector. After hitting a dead end there, I went to the pedal switch and started working my way back to the ECM. Before long I found that both wires from the NGS (the SIG RTN wire and the Pin 30 wire) go through the firewall by the brake booster and then through the round 8-pin connector into the ECM harness (brown connector #31 in the diagram above). From there the pin 30 wire goes across the firewall to the same 8-pin connector as the lead from the ECM, just under the MAF. It terminated in the hole marked "Auto" in the diagram above. For some reason Ford decided to do it this way and put a jumper wire in the O2 harness to make the connection between the two. I'll have to assume that there must have been some model or variation that needed to be able to interrupt the "pin 30" signal somewhere down under the motor, so Ford designed the harness to be able to accomodate it...??? In the AOD and manual trans V8 cars, though, they just use a jumper in the O2 harness to connect the two. Anyway, the 1991 O2 harness jumpers the two terminals marked "auto" and "NDS pin 30" together whereas the 1990 O2 harness jumpers the "auto" and "manual" pins together. (The diagram isn't 100% correct in the fact that Ford seemed to like to change the colors around a bit from year to year. The spare harness I used for testing was a 1991 and had different colors than the 1990 car, but they're similar.) What was happening in my situation was that the 1991 O2 harness was jumpering the ground from the NDS under the dash to a circuit that wasn't supposed to have a ground, and in turn leaving the pin 30 from the ECM open when it should have had a ground. There are eight wires going into the connector from the ECM side but only seven coming out in the O2 side. For some reason they changed the unused hole from 1990 to 1991.

Once I found this, I called the owner to verify my suspicions. He verified that he did use the 1991 O2 harness so he brought me the original one and sure enough, they were wired differently. Then I installed the correct O2 harness and checked to see if I had a good SIG RTN (which I didn't before). It measured 2000 ohms to ground, but pulling off the starter solenoid wire now made no difference...so I figure I've made some progress. Then I installed a third known-good ECM and low and behold, I got 2 ohms to ground on the SIG RTN circuit. :joy:

Once I finally got everything put back together that I'd disassembled for testing, I started it up, backed it out of the shop and let it run for a while. As luck would have it, the damn thing ran out of gas and died just about the time it went into closed-loop. Yep, my heart skipped a beat and it took me a few minutes to figure out what had happened....:shakehead: Once I got some gas in it though, it ran like a champ.

I guess the moral of this story is that "just because it plugs together doesn't mean its meant to go together" or "don't go mixing and matching harnesses", whichever you choose . Or maybe it should be "Pay attention to which battery cable is red..."

I still don't know what cooked the computer but I'm pretty certain that it was the mis-matched harness because it killed two processors, not just the one that was in the car when the battery got crossed up. Just a thought...

It's 2:30 now and I'm tired so I'm off to bed. Thanks so much for all your help!

Deron

 

[Joel5.0]

GREAT!....glad it's finally id'd and fixed...thanks for the info on the wrap up...details in wiring anh?,..... assumptions that all wiring are the same for specific years, proves again are wrong....details, details....enjoy the Miller's, I'm going the Medalla (local brand) route. Here ....you really deserve one...after doing a great job fixing that problem. GL

 

[302man]

Thats great you found it and got it working. But, did some checking on what you said about the oxygen sensors. I looked on Autozone's site for O2 sensors and they all came back with the same part number. 87, 90, 91 and 93. All the same Bosch 13942 sensor. I am not calling you a lier, so don't feel I am attacking you in any form. I don't get into the wiring stuff too much, but its really weird to say the least. I work on mustangs too for fun and I started reading this thread and I had to look for my own knowledge. From what I knew, 87-93 O2's were all the same. Like I said, don't take this as an attack. Did you find any places that offered a different part number for a 91 VS 90 oxygen sensor ?

 

[dlshady]

Accidentally hit the send button. See my reply below!

 

[dlshady]

Hi 302man,

You're right (as far as I know) in the fact that the sensors are the same for all the years but the difference is in the short harness that goes to them. The main ECM harness runs from the computer, up and out the firewall (into the engine bay) and Ys off, one leg going across the firewall to the brake booster area and the other going forward across the passenger shock tower to the air meter area. Just underneath the air meter, this harness terminates in a round 8-pin connector. There is a seperate short harness that goes from that connector, down to the front corner of the oil pan and then back down each side of the motor to the O2 sensors and the low oil sending unit. The differences in the two years is how the wires are configured in this harness, not the O2's themselves. And as far as I can tell, there's only one wire that's differentin a different location. Truth be known, I could have probably disassembled the connector and moved the wire from the hole that it was in to the empty hole next to it and been fine, but I'm not a big fan of mixing and matching parts, and I feel like doing that might have caused more headaches for someone down the road. Since I had access to the original O2 harness that came out of this car I decided to use it. Otherwise I would have just made do what I had.

This isn't the first time I've run across this type of thing. Ford seems to be big on using the same connector in a particular location but moving the wires around into different holes. I built a homemade 2003 Cobra for a customer not too long ago out of a 2000 V6 car and you'd be surprised at how the 03 Cobra engine harness plugged right into the V6 car's dash harness, but if I'd hadn't paid attention we would have had stuff going on like the fuel pump running when you turn on the left blinker..... The plugs were all the same but that's about it.

Well, that's it on my end. I hope this makes more sense now. It's been fun.....

Deron

 

[sy733]

Regarding the other clutch pedal switch...it's not a dumb question (I had to look it up myself after getting the big ? myself), it would be the switch used for the cruise control system.

Following the diagrams I have access to (Mitchell, Motor, Veryuseful.com, Autozone.com, Fordfuelinjection.com, Products Research inc.) none mention the details of their routing or connectors. However, they all show pin 30 getting a direct feed from the ST pin @ignition switch (splice near its connector, presume its routed to the ECM connector under dash) and the other wire going directly to the clutch switch/NGS circuit.

At fordfuelinjection.com it shows that there are 2 8-pin connectors through which NGS/NDS and SIGRTN (manual or auto config) are routed. Look at the pic:

Anyway....from your observations of the clutch pedal/quadrant switches one question comes to mind.....are the right switches connected to the proper circuits?...the clutch switch used for the cruise control is connected to the stop light sw(Red-Lt Green wire) and to the Cruise Control Amplifier Module (Lt Green wire), which I assume is a NC type switch. LUK what you find out.

Holy post of the year ....I checked out a 91 standard GT today and under the stock MAF when I unplugged the connector, the lb/y auto and manual pins need to be swapped on the black connector#40 to be correct....Where can we get a connector diagram that shows the #XX's ?? .....very good post dlshady and joel....thanks....


I called ford today to see about buying a correct o2 sensor harness and they said they only sell the connectors with about 1 foot of wire x 8....I'm going to try and swap pins in the factory connectors first...If that can't happen, I'll buy the ford connectors and solder something up...


Great job on the troubleshooting dlshady!!! :thumbsup: This is a great post!

 

[HYBRED]

It was the O2 sensor harness that fried the EEC. I remember helping a guy on here last year that kept on frying EEC's because he had the wrong O2 harness when swaping engines from one car to another. I always make it a point to tell persons doing a 4cyl to v8 conversion to get the correct O2 sensor harness. There is an auto harness and one for a manual. Also to note, the earlier 87-89 used different harness from the 90-93, due to the O2 sensors. A 4 wire sensor vs a 3 wire sensor.

just because it plugs in doesn't mean it is right is a good statement. sometimes Ford changed the color of the connectors but not the shape. So look for mismatched colors. Ex. the headlight harness plug is black for 87-89 and blue for 90-93..... you cant mix 'em either.

 

[92cleancoupe]

I wish i had seen this earlier..
The fast way to solve this WHOLE problem, very few people know, is to add a wire going from pin 46 off the computer to ground...

 

[302man]

I wish i had seen this earlier..
The fast way to solve this WHOLE problem, very few people know, is to add a wire going from pin 46 off the computer to ground...

How would grounding a ground wire solve this problem ???

 

[mustang30212]

Old thread...but very helpful!!

RH

 

[killacal]

TTT for lots of great info!

 

[liljoe07]

I wish i had seen this earlier..
The fast way to solve this WHOLE problem, very few people know, is to add a wire going from pin 46 off the computer to ground...

PIN 30 was sending voltage to the SIGRTN, which should have zero voltage under any conditions, adding a ground wire in the case above, doesnt stop the voltage from reaching the SIGRTN. IF the SIGRTN wire was damaged in some way then yes it would help

 

[gnvair]

I see too many people band aiding their issues and some are just hard headed and think that because it runs that all is well. As for getting the correct O2 harness, I found a source for new O2 sensor harnesses. http://www.foxresto.com/products/byCategory/91/Hardware_Wiring_Repair

 

[killacal]

Seeing as the only difference between the two harnesses is that one pin, has anyone confirmed that you can just re-pin the connector (switch the "Auto" pin to the "Manual") and vice versa? This seems like a much easier solution than going to buy a new harness.

EDIT: I found a very good post that shows the different 02 harnesses and pin configuration.
Here it is:

http://forums.corral.net/forums/showthread.php?t=1094179