"20 ohms" to hood hinge and "50 ohms" to battery post don't make sense.

Assuming your ohmeter is accurate, those values are too high. And there is too much difference between them.

What do you get on the meter when you touch the probes to each other?
What do you get on the meter when you measure hood hinge on one probe and battery post on the other?

All the ohm readings should be taken with the key off....

 

I get 12 volts at VREF pins 37 and 57 at the EEC, with volt meter to ground on battery. Also 59 ans 58 injectors pins i get 12 volts at each pin.

 

Man. That's what I call an informative post. I'll mark this one for later use.

 

When i put the two probes together i get 0 ohms. I just recheck my readings and this time I got 0 ohms from battery ground with the neg. from meter, positive of meter to hood hinge. Then from hood hinge with neg. from meter, and poss. of meter to SIG RTN on test plug and I get 20 ohms. Also tried three different places that had a good ground, I still get 20 ohms each time. Maybe ill check different SIG RTN's tomorrow.

 

That sounds right to me...the computer will ground the injector wires to fire the injectors, so they should have 12V. Test continuity between VREF and SIG RTN.

 

The exercise with the ohmmeter was an effort to determine whether the ECM is doing its job of grounding SIGRTN. With a 20 ohm resistance, it appears that it is not -- on my truck SIGRTN to ground is less than one ohm (mine is a 91 302 V8, so they might be different, but I doubt it). If you had a junkyard ECM, you could measure pin 46 to one of the many ground connections, then compare it to yours.

The 20 ohms you are measuring may be a "sneak path" through whatever is providing the 12 volts that you measure. But in any event, the resistance is not low enough.

With this, we can suspect the the ECM is internally damaged. We also still have the matter of where the 12 volts is coming from. This needs to be figured out before another ECM is installed or we risk frying that one.

My suspicion is that the ECM itself may providing the 12 volts. Perhaps some of the internal sense circuitry is shunting enough current into SIGRTN, now that SIGRTN is not connected to ground internally. But it would be good to prove that before trying another ECM.

There is a way to find out, but I hesitate to recommend it. You will have to cut and resplice your wiring harness and the future reliability of the truck may be compromised if you are not good at making durable splices.

Basically the method is to divide SIGRTN into sections and find out where the 12 volts is coming from. If you cut the chain in half, one half should stay at 12 volts and the other half should drop to something less, or to an open circuit, depending on what is plugged in, but it shouldn't be 12 volts.

Take the half that is at 12 volts, and split that in half, and so on. Eventually you will have one short section of wire with an inappropriate 12 volts on it, and you can find the short circuit.

 

I tested the MAP sensor, VREF and SIG RTN. When i put black probe to sig rtn and poss probe to VREF i get a neg voltage, because the Sig. RTN has 12 volts. I thought i was reading it wrong so i switched the probes which gives me 7 volts, as i stated on post 12. Post 11 was correct i get negitive voltage when i check the MAP sensor. So i bet all the other sensors have negitive voltage to. Can the sig rtn at the ecc be moved to other pin, that has ground?

 

It's negative voltage because the voltage on SIG RTN is higher than the signal voltage from the sensor. You must find the short to power in the SIG RTN wire. I assume it is possible for this to be a problem internal to the computer, but I wouldn't bank on that just yet. Unplug everything and check continuity between SIG RTN and every 12V power source you can find until you find where it is shorting from.

 

There are two problems:

1) Sigrtn has 12 volts on it. It shouldn't

2) Sigrtn is not grounded, it should be.

These may both be caused by internal PCM damage, or the lack of grounding may be a consequence of sigrtn shorted to 12 volts somewhere in the harness.

The reason for separating sigrtn from frame ground is to avoid picking up too much electrical noise on fairly small level signals. The analog receivers in the PCM compare the input signals from the various sensors to SIGRTN to determine a level. If they tried to be referenced to PCM ground or the firewall or whatever, there would be power supply noise layered on top of the intended input.

You could try grounding sigrtn externally. There are a couple reasons not to:
1) You haven't yet identified the source of +12 volts. You may cause fireworks and additional damage attempting to force the signal back to ground, if the source of +12 volts is high energy enough.

2) Forcing SIGRTN to ground externally is not a good long term fix, because the sensor voltages will be very noisy. If you do manage to get it to start and idle, every time you flip the lights on, turn on the AC, toot the horn, etc, noise will be injected into the 7 sensors using sigrtn and the computer will get confused.

Instead of forcing SIGRTN to ground, try connecting STI to ground (not SIGRTN) and see if the computer is alive enough to give you KOEO codes.

 

I was thinking of that. I unwrapped all of the harness and traced all of the sig rtn and it looks ok. 2 feet from the ecc pin 46 ( SIG RTN) runs into a 8 way splice. It is the first place it brances before any sensors, then test it for 12 volts. I'll disconnect the battery first to be safe. It is better to see if i have a short before putting in a used eec, thanks.

 

I just cut the sig rtn 2 feet from the ecc pin 46 and i get no voltage as soon as i connect the two ends of the sig rtn back together i get 12 volts. Tried it five time so i'm sure its a short now. If i unplug everything the sig rtn goes to and still get 12 volts does that mean its a plug shorting out? Now i have 10 feet of signal rtn line with all 7 plugs unpluged from the sensors how can i ohm test it to see which plug is bad? I'm leaving the cut sgn rtn line unhooked from the eec, battery poss. off and key out.

 

Ok, now you need to follow the 12 volts. After you cut the wire, one side of the cut should be "no volts", and the other side should be 12 volts. (You only reported "no volts", I am thinking that was on the ECC side).

The side that is at 12 volts is the problem. Let us assume that it is the sensor end of the cut. Let us further assume that you get 12 volts on that end regardless of whether the ECC pin 46 wire is connected or not. (Please tell me if any of these assumptions are wrong).

First thing to do is to find if it is switched or unswitched 12 volts. Turn the key off.
Measure the sensor end of the cut. Is it 12 volts or 0 volts? If zero, turn the key on.
Should be 12 volts now, right? (these measurements should be done with everything connected.)

Let's say you find it is switched 12 volts.
One more test:
Remove the battery positive cable
turn the key on
Measure the resistance from the "12 volt" end of the SIGRTN wire to the dangling end of the battery positive cable and report back. We want to know how "strong" a short we are looking at.


None of the sensor plugs are supposed to have 12 volts on them. It probably won't make any difference if the sensors are connected or not. The problem is probably not in the sensor plugs. But try it anyway.

You could try to guess where the problem is by pulling fuses out, disabling various sources of +12 volts. Or if you have access to (or can make) a time domain reflectometer you could determine the electrical distance from one end to the short.

But I don't think either of those methods is practical for you.

If the bottom line of the above tests reveals you have (say) a 5 ohm short to +12 volts somewhere in the SIGRTN harness, then at that point you have to either inspect the entirety of the harness or cut the SIGRTN line yet again, roughly in the middle. This should reduce the situation by 50% -- once side of this second cut will be 12 volts and the other side will not. You can focus your efforts on the half that has 12 volts. Pick a place to cut that you will be able to easily splice later.

If you had a later year truck I'd suggest looking closely at the HEGO sensor, which gets switched 12 volts as well as SIGRTN and might be shorting SIGRTN. But the Probst book for the 1988 4.9L shows no SIGRTN going to the HEGO sensor.

 

I think I'm getting closer. I cut the SIG RTN close to the ecc it no longer has voltage at the ecc end. So now I'm down to tracing down the short with the seven sensors. I unplugged all seven sensors. Then put the meter on the voltage end of the signal return and get 2 volts. I started cutting each black -wht wire at the eight way splice. I found the sensor that when cut out of the loop there's no more voltage. Its down to the Map, egr,tps,act or the ect sensors, that was causing the short! Once I trace the SIG RTN wire to the sensor, that has the two volts what do I do? There is no voltage with the key removed at any SIG RTN.

 

OK, we know now the following:

Something other than the ECC is putting 12 volts on sigrtn.

The 12 volts is coming from something that is switched with the key.

With the sensors unplugged you get 2 (?) volts. Was that 12 volts or 2 (two) volts?

If you cut the SIGRTN wire off the computer, then 2 volts (with the key on) on sigrtn is very possibly normal. With no ground on the ECM end, VREF current can feed through such sensors as the TPS and elevate the voltage of SIGRTN. That's not a problem: it is the 12 volts that is a problem.

You want to find the sensor (or the branch of the SIGRTN line) that is dumping 12 (twelve) volts onto the common splice. Two volts is OK. Twelve volts is not.

 

Look for 12V switched sources that are in the wiring harness for the computer. That could help you narrow down your search, although at this point, I believe it will just be a matter of tearing the harness apart until you find where the short is.