Bump....

This problem showed up again after being in hiding for several months.

I called my local Chevy dealership about the problem and they told me they would be "glad to look at it" and informed me that it would take at most 4 hours to fix @ $90.00/ hour.

Any new ideas?

 

New ideas? You did not even entertain the suggestions I had 9 months ago....

 

I checked every suggestion you had that was practical. 2 out of the 3.

Refrence post #5.

The one I cannot seem to check is the voltage and ground of the wiring during the failure. Why? Becuase it doesn't stay in a failed condition long enough to pop the hood and stick a meter on it.

I am sure you know this sensor is located right next to the exhaust manafold. Trying to grab that in a hurry will leave you burnt.

I am looking for more suggestions, or practical methods of finding this problem.

 

There are just not too many directions to go here. With regards to the circuit, it does not get any more simple. One sensor, two wires, connected directly to the PCM. The PCM provides a 5 volt signal, and the sensor changes the resistance to ground(provided by the PCM) with temperature. Other than that, you have the PCM converting the analog signal, and communicating the output to the IPC....via the class two serial buss. A pair of mechanic gloves carried in your car, will ensure you don't burn your hands, and having a meter at the ready, will help get this isolated. Checking the resistance of each wire from the ECT connector to the PCM. Checking the continuity of each wire with a meter(set on tone) while physically shaking down the harness.

How far you go, depends on how badly you want this resolved.

Good luck.

 

Thanks for the suggestions, I will dig into it some more when I have the time.

I may build some type of piggy back harness that gives me some convienint taps for my multi meter so that I can read an actual resistance value real time, from inside the car.

One question:

The sensor is using a 2 wire setup as you mentioned, how should I test each wire? I understand the continuity test perfectly.

What resistane should I see between the two wires on the wire harness side?

Should I see 5v measured between the two wires on the w/h side?

Also I assume I should see very little resistance between the wire harness conector pins and the PCM connector, and I should not see continuity to ground from either wire in the harness, if I do this would indicate a short right?

Just making sure I have figured out all the failure modes and a way to test them.

Thanks again.

 

Great idea.

A continuity test is great for checking opens/shorts, but the resistance needs to be evaluated for each wire. Test each wire from the connector at the PCM & ECT.


With both connectors off, you should have infinite resistance between the wires.


I don't know what "w/h" means.....

You should measure 5 volts between the yellow wire(5 volt output from PCM) and brown wire (ground from PCM).....at the ECT connector.


[/QUOTE]
Correct.

 

Any updates on this? I'm having the same exact problem, just changed my sensor and the problem persists. Although not exactly the same drive cycle, I've had the car sit for a week and when I start it, I'll get the check gauges light and 250+ temps after 10 seconds... At an idle!

 

I haven't been able to troubleshoot this problem yet. I just got done with a brake job on my Daily that turned into a ball joint/ half-shaft extravaganza. After that my wife's A/C stopped working and I have been elbow deep in a VW trying to make sure she keeps her cool.

I hope to take a good look at it soon. Perhaps this weekend. Feel free to update this thread with your progress as well. The more the merrier!

 

If it's gonna be $360 tops (4 hours @ $90/hr) I might just go to the dealership and have them look at it. I spend that much on tickets and I haven't gotten one in awhile so it would be worth my while to check it out... I set aside $100 each week for tickets anyway lol so it's meant to be used.

Also, has your issue ever given you a CEL? Every time the temp goes crazy it feels like a misfire, the check gauges light will come on and the car will hesitate for a split second. My A/C also turns off when it says it's overtemp too. There are also times that the temp has gone to a point that it said "xxx F" on the DIC. Hopefully our problem is the same and when one is resolved the other can do the same.

 

Wow...what a rebel. Will you sign my yearbook?

 

Mmm, sarcasm.. Tasty. I set it aside in case I get a ticket, I don't go out trying to get them, so when I don't get one for a while I can reward myself. It's more of a 'car fund' so any of that money is used for the car

 

I will be taking a deep look at this problem this weekend. I hope to report back finding the root cause.

 

Checked all the electrical connections today, all seems good. 5v is present with key on. Contunity between PCM connecter and ECT connecter.

I ran a pigtail into the car to monitor the voltage/resistance at the ECT connector. I will report back when I can observe behaviour during a failure event.

 

Cool, if you don't find anything I'll take mine to GM and see if they can find something else, here's hoping our problem is the same

 

Here is what I have checked, and the results:

PCM Pins Resistance to Ground:

Pin 80 (Brown)-> Ground @ PCM Case = 4 ohm
Pin 80 (Brown)-> Ground @ Engine Cradle = 9Mohm

Pin 74 (Yellow)-> Ground = Infinity

PCM Connector to ECT Connector (unable to test resistance between connectors):

Yellow Wire = Continuity
Brown Wire = Continuity

Measured at ECT Connector:

Resistance between wires (both connectors off) = Infinity

Resistance of each pin to ground (both connectors off) = Infinity

Resistance of each pin to ground (PCM connected key off)=
Yellow Wire=4.5k ohm
Brown Wire=7 ohm

Voltage between brown and yellow wire (PCM connected, key on)=4.98V

Voltage between brown and yellow wire (PCM connected, key off)=0.00V

To me everything I measured seems ok. Remember all the measurements I made were on the car while the malfunction wasn't happening. I imagine when the problem occurs the values above may change.

With this in mind I decided to run a lead from the ECT connector into the CAB so that I could look at voltage and resistance real time as the car was driving, and hopefully watch the values that occur when I see my false temperature peaks.

I added two quick connect wires into the pigtail that goes to the ECT connector, I then zip tied it out of the way and along the top of the LH bank of cylinders.



I then re installed the fuel rail cover and ran the wire through the grommet where the hood release cable passes through the firewall.



With the wires inside the cab I soldered on 2 insulated alligator clips, and attached them to my multimeter.

Monitoring Resistance:



Monitoring Voltage:



I drove around today trying to recreate the erratic temperature readings so that I could monitor the voltage to try and see what was happening. The problem did not occur.

It is possible that in all the "messing" with the wire harness and connectors I have unknowingly solved the problem. I will report back if I have any news.

If there are any other checks I should perform please let me know

Thanks for the help!

 

Update:

Today I measured the voltage present at the sensor during a warm up. Engine started at 90F and I stopped measuring at 220 F.

Who can tell me why the large voltage shift at 120F?

 

Looks like you found your problem.

 

So what do you think the problem is?

As I see it, I found something different that what I expected, however the DIC readout just kept on trucking like this was an expected input. I measured a similar condition yesterday (with less data points) to find the same results.

So far all the measurements made were under normal operating conditions, no wild indicated temp swings.

 

Looking at the data, you know that from 90 degrees to 120 degrees that can't be right. As the temperature increases, the resistance in the circuit decreases. So the voltage should have started high (close to 5v), and steadily decreased, as the engine coolant temp increased. For whatever reason, it started reading right, just above 120 degrees.

 

So I tested this a few more times, each time the voltage changes abruptly and exactly at 120F (or when the voltage at the ECT conntector drops below 1V)

Here is why this is happeing.

From GM's point of view our engines must run in any concievable environmental condition. That means starting in sub zero conditions, or out in the desert.

If GM used one voltage range (the range that appears before 120F OR the range that appears after 120F) it would not be able to accurately determine temperature due to leaving the normmal bounds of a 0-5V sensing system.

I made a curve fit to each seperate "curve" in the graph I posted and crunched the numbers.

If the Lower voltage range were used, ECT temps above 160F would cause the PCM to read a temperature less than 0.

If the higher voltage range were used ECT temps below 60F would give a PCM read of greater that 5V.

Both of these are out of range of the standard 0-5V system, so in order to maintain good resolution and the ability to read very low and high ECT temps 2 ranges were required.

I bet if you test any other voltage at ECT connector it would have similar behavior.



Back to my problem:

Since digging through the system I have driven the car for a few hundred miles with out this problem resurfacing.

It is impossible to prove that I have fixed this type of problem since it is intermitent, it is only possible to prove that I haven't fixed the problem some time in the future.

As for now things look good, I will continue to run my DMM in the car and look for the problem to resurface for the next few weeks or so. If all is well I will remove it and call it good.