[reflex_b5.5]

I drove about 200 miles after fixing my melted AC switch (details here) which was caused by squished wires from a thermostat and timing belt job after the car overheated. Everything worked fine.

Then the EPC light came there was a loss of power. Took a bit and was able to re-start and get home. Then a few days later I tried again and EPC came on again.

I scanned the codes and the only stored code is P0118. This seems like it means the ECT sensor is not working.

For the AC switch repair I disconnected and reconnected a bunch of wiring harnesses including T14k, T15k, and all the center stack. Is it possible that something is loose from that and causing issues?

 

[FrescoGreen01.5]

ECT sensor is at the rear of the engine near the faux-firewall. The sensor is a known failure point in our cars.

I believe the wires run directly from the ECT connector to the ECM connector. (With engine and year Bentley schematics would tell for sure.)

First task: check the ECT. It is two thermistors in one housing. Four-pin connector, each thermistor uses two pins. With harness unplugged from ECT, you should measure a few K-ohms between two pairs of pins on the ECT. (One thermistor serves the ECM, the other the gauge on the cluster, completely separate circuits.)

(I didn't look up the pin numbers since they're hard to see anyway.)

Note to other readers: if you have VCDS, you can see the live values of both thermistors in different measuring blocks, one in ECM, the other in cluster.

 

[ylwagon]

Note to other readers: if you have VCDS, you can see the live values of both thermistors in different measuring blocks, one in ECM, the other in cluster.

Thanks for that info about looking for gauge data in Instruments; next time I'm checking the car, I'll try that.

 

[reflex_b5.5]

I took another shot at this today. The ECT sensor is cheap enough and relatively easy to get at. So I changed it out. While it was out I also checked that pin 3 on the wire harness was grounded and pin 4 had about 4.6V on it when the ignition was on without starting the car. I later realized the battery voltage was only about 9.6V.
I also noticed a hose clamp right about it was loose so I replaced that. And the coolant was low so I added some to the reservoir.
Then I jumped it to start. Everything ran fine for a few minutes. Then it shutdown by itself with the EPC light on again. I did not touch any controls other than starting the car. It was just idling.
I'm not sure where my code reader is so I have not check codes yet. Any ideas on what to look at next?

After spending a few hundred dollars to replace the thermostat and timing belt less than a year ago, I'd really like to get it working. If it's a potential I'll probably end up getting rid of it though.

 

[PZ]

Without decent voltage, nothing will work correctly. Charge and test the battery, then when started, check voltage. It's likely the low voltage caused the engine shutdown.

 

[VAGguy]

Without decent voltage, nothing will work correctly. Charge and test the battery, then when started, check voltage. It's likely the low voltage caused the engine shutdown.

Just thinking out loud. If the engine is running and producing 13 to14-ish volts via the alternator, wouldn't that keep everything going at desired voltage with the battery just being a line filter at that point? Unless the battery has a bad cell or something....

 

[sct]

My questions would be is the alternator producing 13 to 14 volts at idle? And is the system trying to charge the battery and keep the car running at idle?
I have always gone by the theory that after jump starting a car that you want to keep the revs up to give the system enough power to charge the battery and keep the engine running.

 

[reflex_b5.5]

Today, I let the battery charge a bunch before starting the car. It was at ~13.8V with the cables connected. After starting the car it was 14V+ with the cables disconnected. It dropped to 13.8V when it sounded like some fan was running. I did not see the second cooling fan going though. Then it returned to about 14V.

It ran about 15 minutes before stopping with EPC. This time it was long enough that the temp gauge inside the car showed 190 unlike yesterday where it was not warmed up all the way. The ambient today was upper 70s vs upper 90s yesterday. The battery read about 12.8V with the engine stopped but ignition on. I was able to restart without jumping. With the engine running it was over 13V again. It ran less than 5 minutes before shutting down again.

Am I possibly damaging something by running the engine? I don't hear any odd noises and it seems to be running smoothly.

 

[PZ]

It sounds like you have it fixed, but keep an eye out for possible battery drain. Check the alternator output with everything on, to make sure it's putting out enough amps to keep the battery charged.
There would be no damage from running it for a short time with a bad sensor or low voltage. Long term it could damage the cats and rings.

 

[reflex_b5.5]

Definitely not fixed as it still self shuts down by itself and shows the EPC light.
I found my OBDII bluetooth adapter. This time I cleared the codes and graphed a bunch of values. It ran for about 5 minutes before shutting itself down and showing EPC. The coolant temp values seemed reasonable so the P0118 may have been left over from the previous problems.

I've attached the screenshots of the Torque Lite graphs taken right after it died.

There's no logged codes but there is a pending P0600. This seems to indicate a communication fault. Not so surprising given that I messed with the wiring to fix the AC problem. I re-checked the connectors I disconnected and reconnected during that work. Unfortunately nothing obvious.

I've also seen posts saying P0600 could be from low battery voltage.

I also noticed the "press brake to shift" light blinked in and out after the engine died while I was not touching the brake pedal or the transmission gear selector (it's an automatic transmission).

Is this the end of DIY and I should bring it into a shop? Should I try changing out the battery?

 

[PZ]

I would replace the engine speed sensor. It can cause stalling without throwing a code. I would also double check the wiring for burnt leads at the grounds and positive leads.

 

[reflex_b5.5]

I think I've fixed it. A long time ago I had the ECU socketed and had a Wettauer chip. That chip stopped working at some point and Wett had gone out of business (temporarily?). They later sent me a replacement chip along with another chip programmed with the original programming. I had the replacement original programming chip installed. Replacing that with the actual socketed original chip seems to fixed the issue.
So it seems the replacement chip was going bad.