[Fixed!] 1998 Nissan Frontier with KA24DE -- Temp Gauge

I tried tapping on the gauge... no change. I also checked the grounds. I have done this in the past (before posting here) and they are clean. I think at this point I may remove the cluster and take it to a place in town that does this type of electronics after I confirm that they work on this type of thing unless anyone has any better ideas. One interesting thing is that Thursday I started the truck and the temp gauge immediately went to normal! It was 85 degrees outside, but I would still expect the gauge to read cold until the thermostat opens.

Things I've learned... 1. I never thought about the voltage on the wire from the cluster to the sensor. Makes sense, just never thought about it. 2. I didn't know the sensor varied so much in resistance. I still don't really understand that. Seems to me that if the engine sensor reads 181 degrees that the gauge sensor would stay pretty steady at around 130 ohms or so. 3. I still don't really understand how the temp gauage takes the resistance from the sensor through the circuit board and converts it to a reading, other than lower resistance makes a higher reading.

JimT wrote: I still don't really understand how the temp gauage takes the resistance from the sensor through the circuit board and converts it to a reading, other than lower resistance makes a higher reading.

Pull up the link at the bottom for a voltage divider calculator. V1 is what the cluster uses to supply the sensor circuit. I don't know what that is but 12V will work for this example. R1is the resistance inside the cluster. 200 ohms will work for that. Vout is where the cluster takes measurements it then displays on the gauge.

R2 is your temperature sensor. Put in 400 then 200 then 100 then 50 ohms into that field to simulate a cold start then warming the engine up. You'll see Vout incrementally change from 8V cold to 2.4V hot. That's your normal operating range.

Now you can simulate some fault conditions which could cause the gauge to peg hot to see what they do to Vout. First change R2 to 0 which is a short to ground on the sensor wiring.

Then put R2 back to 100 ohms and change R1 to 20,000 ohms which simulates an internal fault in the cluster.

Now for "Bonus material" :silly: leave R1 at 20,000 and change R2 to 10,000,000 to simulate a DMM voltage check with the sensor unplugged.

www.digikey.com/en/resources/conversion-...ator-voltage-divider

Matt T wrote:

JimT wrote: I still don't really understand how the temp gauage takes the resistance from the sensor through the circuit board and converts it to a reading, other than lower resistance makes a higher reading.

Pull up the link at the bottom for a voltage divider calculator. V1 is what the cluster uses to supply the sensor circuit. I don't know what that is but 12V will work for this example. R1is the resistance inside the cluster. 200 ohms will work for that. Vout is where the cluster takes measurements it then displays on the gauge.

R2 is your temperature sensor. Put in 400 then 200 then 100 then 50 ohms into that field to simulate a cold start then warming the engine up. You'll see Vout incrementally change from 8V cold to 2.4V hot. That's your normal operating range.

Now you can simulate some fault conditions which could cause the gauge to peg hot to see what they do to Vout. First change R2 to 0 which is a short to ground on the sensor wiring.

Then put R2 back to 100 ohms and change R1 to 20,000 ohms which simulates an internal fault in the cluster.

Now for "Bonus material" :silly: leave R1 at 20,000 and change R2 to 10,000,000 to simulate a DMM voltage check with the sensor unplugged.

www.digikey.com/en/resources/conversion-...ator-voltage-divider

If I understand correctly, then the gauge actually reads Vout and reacts to that. Aha. So both R1 and R2 affect Vout. Therefore, if R1 has deteriorated over the years and varies resistance, that is one thing that can cause Vout to plummet and indicate a problem. Furthermore, if I hooked up a voltmeter at the same spots where the circuit measures Vout, I could watch the voltage vary.

JimT wrote: If I understand correctly, then the gauge actually reads Vout and reacts to that. Aha. So both R1 and R2 affect Vout. Therefore, if R1 has deteriorated over the years and varies resistance, that is one thing that can cause Vout to plummet and indicate a problem. Furthermore, if I hooked up a voltmeter at the same spots where the circuit measures Vout, I could watch the voltage vary.

I think you've pretty much got it. Though there may be electronics between Vout and the gauge itself.

If you backprobe the temp sensor you will be able to watch Vout changing. Problem is that when you see Vout plummet you have no way of knowing which resistor has changed. And when you unplug the sensor to troubleshoot further the problem can self correct very quickly which makes this difficult to diag with 100% certainty.

I think you've pretty much got it. Though there may be electronics between Vout and the gauge itself.

If you backprobe the temp sensor you will be able to watch Vout changing. Problem is that when you see Vout plummet you have no way of knowing which resistor has changed. And when you unplug the sensor to troubleshoot further the problem can self correct very quickly which makes this difficult to diag with 100% certainty.[/quote]

Very good, thanks. I'm going to consult with the electronics place I'm thinking about and see what they tell me. Thanks everyone.

Update: I got the cluster removed Sunday, dropped it off to Southern Electronics Monday, got it back Tuesday. Reinstalled Tuesday. I'm hesitant to say it because of all the false starts, but I think it's fixed! I've made three trips in it and no gauge misbehavior at all. Sweet! Thanks everyone! Where do I find that fixed icon?