Okay I know this has been asked probably 1000 times, but I've read through just about all the threads I could find, and I'm still really struggling...
So my '92 940 Turbo will not start when cold without starting fluid. Once warmed up, it runs and starts fine.
Check engine light is showing 1-2-3, so I started by replacing the ECT. No change.
Next, I looked into wiring by testing continuity at the connector to the ECU.
Testing resistance between pins 5 and 13, here's what I found:
ECT unplugged: infinite resistance
Jumped both pins in the ECT's connector to ground: 0 ohms
Jumped both pins in the ECT's connector to eachother (but not to ground): 0 ohms
Plugged the ECT back in: 4800 ohms (was ~45F so this seems about right)
So based on the information I can find, there is nothing wrong with my ECT or the wiring between it and the ECU. I bought a new (used) ECU and it still has the same problem. Now I'm just stuck and cant think of what else could be the issue... is there anything else that could cause the 1-2-3 code and cause the cold start issue?

 

indeed, per what you say, that sounds like the ECT is good, and yes, that 1-2-3 code is for a bad ECT. Did you clear the codes, and does that 1-2-3 code come back ?

does your 1992 turbo have the 5th 'cold start' injector? that's ONLY used to start the engine when its very cold, later cars did away with it and just increased the amount of regular injection for the same net result.

 

Thanks for the reply,
Yes, I cleared the code, and it came back (1-2-3).
My car does not have the cold start injector

 

The OBD1 is far from perfect, it's better than nothing but you can't rely on it much... Is the 1-2-3 code gone for good?
If the ECT is fine, let's assume the replacement is OK, then look at fuel pump, may be weak when cold, battery of course should be up to snuff, and all tune up items should be OK. It is a fuel issue for sure if the starting fluid gets it going...

 

So the car had the issue when I bought it a few months ago... Previous owner didn't know what the problem was and just threw parts at it. He put both new fuel pumps in it, filters, etc. I've tested fuel pressure at the rail (don't remember what it was, but it was in line with what it's supposed to be). He also did many ignition components such as plugs, wires, distributer cap, coil, crank sensor, MAF and probably some other stuff that I don't remember. New battery as well.

 

One thing that I'm suspicious of is that the previous owner said he put a new head on the motor. Maybe he didn't reconnect everything properly... there is one wire/connector under the intake manifold that isn't connected to anything and looks exactly like the injector connectors. Looks like its in the general location of where the cold start injector would be, so MAYBE the car is supposed to have a cold start injector and the previous owner put a new head AND intake on that doesn't have one ??? am I thinking too far into this?

EDIT: upon further investigation, this random wire/connector is not quite long enough to reach up to the bottom of the intake... But nonetheless should still probably be hooked up to something. Nothing else looks out of place though.

 

One big question with cars that have had "everything" replaced is what kind of parts were used. These days the aftermarket stuff is often subpar. So, hard to say what issues the car had before all the work and new parts?
Since you are still getting that code, may be address the ECT again with an OEM unit.
I don't think this car ever had a cold start injector, and don't worry about the loose connector.
Your problem is still FUEL, so I'd concentrate in that area...

 

I totally agree that it's hard to know what new parts are any good.
Can the new ECT still be the issue even if it reads correct values at a variety of temperatures?
I've tested the resistance at 40F, at operating temp, and as it was cooling down, and all values seemed to be correct.

 

Well, if it reads OK at testing then it should be OK... But it could be something else, there is where the OBD often fail to pinpoint correctly, even later OBD2 systems are far from 100% correct.

 

Okay so I still haven't made any progress. Does the ECT system have it's own ground somewhere? I know that theoretically its just a measure of resistance through the sensor, so it shouldn't need a ground outside of the ECU. But I'm confident that my sensor is good, wiring is good, and ECU is good... so the only other thing I can think of is that it'll be a ground somewhere but I don't even know how to begin looking into that.

 

the ECT grounds through the engine block... the engine has two different ground points where it ties into the wiring harness, at either end of the fuel rail mounting bar.. one of those grounds is for low level signals, and the other for stuff that uses more power. and there's a main ground wire from the alternator to the car chassis

 

Just checked... all those grounds are there/secure but the two from the fuel rail mounts into the harness have been cut and spiced back together. Is there any possibility that this could cause the ECU to not read the signal from the ECT? (i don't see how it could, b/c I am getting the correct values at the pins into the ECU, so maybe ECU ground or something?) I don't want to go poking around in the harness any more than I have to.