2002 V40 Coolant Temp Sensor issue

[joe_padavano]

I'm having a problem with the indicated temperature from the coolant temp sensor on my 2002 V40 (US-spec, 1.9 turbo).

Long story is that I got an error code last summer. Scan tool (OTC Genisys with Operating System 3.0 and the 2005 European software) said it was the CTS. I replaced it, cleared the codes, and the problem went away. This summer, I got another error code, again CTS. This time I checked the data stream and the CTS was indicating 303 deg F. Since the car was not overheating, the electric fans didn't come on, and the temp gauge read normally, I again replaced the CTS. The new one is still reading high (about 100 deg or so bias). So far it hasn't set a code. I don't know what the threshold is. The new CTS is indicating 265-280 F under normal driving on a 70 F morning. I even checked it before starting this morning and with ambient air at 65 F the CTS was indicating 174 F. This probably explains my cold start problems last winter, since if the CTS was indicating 140 F when outside temps were 30 F, the ECU would not have gone into cold start enrichment mode.

My first question is, does the resistance go up or down with temp? If it goes up with temp, I can envision a scenario where a dirty connector is adding resistance to the circuit, causing the bias. If it goes down with temp, I got nothing.

Second, is there a curve of resistance vs. temp, or two data points (say, oh, 32 F and 212 F) so I can verify calibration of the new CTS? The old one was from RockAuto, the new one from NAPA.

There's another, possibly related symptom, which is that the data stream from the ALDL connector to the scan tool will work for about two-three minutes, then I get a data loss. I have to reset the scan tool and remove and reconnect the cable to the ALDL to get it to communicate again. I've never had this problem using the scan tool on a GM vehicle.

Thanks for any help or thoughts.

[jrmany]

That is interesting. The way i understand it is the warmer the temp reads, the higher the resistance. My sensor went bad a couple weeks ago. i had some resistors laying around, so i threw one it and it changed the temp (i just wanted to make sure that was the problem.) Since you can stream, it would be interesting to take a potentiometer and try 5 or 6 different resistances and make a graph / take photos of where your temp gauge reads. We could then see if it is constent with others' gauges and see if yours is reading right or if there is something else going on

[regent]

See this thread for CTS cal data:
https://www.matthewsvolvosite.com/forums ... or#p160670

Note that the resistance decreases when the temperature goes up.
If your sensor behaves as expected, then you my have a wiring problem as simple as dirty or loose connector pins.

On another note, I would only consider an OEM replacement sensor if reliability matters.

[joe_padavano]

See this thread for CTS cal data:
https://www.matthewsvolvosite.com/forums ... or#p160670

Note that the resistance decreases when the temperature goes up.
If your sensor behaves as expected, then you my have a wiring problem as simple as dirty or loose connector pins.

On another note, I would only consider an OEM replacement sensor if reliability matters.

Thanks for the curves. The problem I'm having is that a dirty connector or loose pin would INCREASE resistance in the circuit, causing the CTS circuit to indicate a LOWER temp than actual, not a higher one. It can't be an open circuit because the indicated temp isn't fixed, it varies as the engine warms up, there just appears to be a fairly constant 110 degree F offset. And if I remember my electronics, a fixed additional resistance would result in a different offset over the temp range as the CTS resistance changed, since the fixed additional resistance would represent a different percentage of the total circuit resistance as the CTS went from 7000 ohms to 300 ohms.

Again, as I posted above, with the engine completely cold from sitting overnight and a 65 F outside air temp, the indicated CTS temp was 174 F. At operating temp (dashboard temp gauge needle right in the middle of the range) the indicated CTS temp was 302 F to 314 F. If you back out the 110 deg offset, that puts you right in the 192 - 204 range that you might expect with a 195 F thermostat.

In any event, I bent over and bought a factory sensor and will install it this weekend. I'm planning to compare calibrations of this sensor and the NAPA sensor before I install the new one, just for grins.

[regent]

I agree, the theory is 100% correct. However, consider this also:
In reality, leak currents between pins cause all kinds of faulty readings, such as a decrease in resistance value.
A slight contamination with coolant, oil and dust can act like an added resistor in parallel and compromise accuracy. These are very low current circuits where cleanliness is important.

[joe_padavano]

I agree, the theory is 100% correct. However, consider this also:
In reality, leak currents between pins cause all kinds of faulty readings, such as a decrease in resistance value.
A slight contamination with coolant, oil and dust can act like an added resistor in parallel and compromise accuracy. These are very low current circuits where cleanliness is important.

Fair enough, however the connector at the sensor is clean on both sides. It's an environmentally sealed connector body and the terminals look fine (and are new on the sensor side connector, obviously).

I did connect the Volvo factory sensor in place of the NAPA one and got the same 110 deg F offset (190 deg indicated temp at about 80 deg ambient temp). I cleared the codes and the CEL is off for now, but will come back on (and set a CTS code) periodically. I'm assuming this is because the indicated temp from the CTS exceeds some threshold that the ECU takes to mean a failed sensor.

[regent]

Do you have the electrical schematics; there could be a wiring issue in line from the connector you've been looking at up to the ECU.