temperature measuring problem

[rainagain]

Years ago when I converted a Renault Clio to run off veg oil I used cheap ebay temperature probes to measure things like fuel pump temperature to make sure it was warm enough to switch from my starting fuel tank containing diesel to my main tank containing veg oil. I simply extended the wires between the LCD display and the probe, I had no problems with this and everything worked ok.

I’ve tried to do the same thing on my current car to measure the inlet temperature, as it’s a diesel as well and I’m interested if the inlet temperature is staying approximately the same as the external temperature, as at the moment it draws air in from behind the front left wing (fender) with no direct cold air feed.

I very carefully ran the wiring from the top of the dash across to the passenger (left hand side) of the car and out through the bulkhead with the rest of the factory cables for a neat job. This meant extending the cabling from the probe to the LCD display by ~2-3m. The probe itself resistance is about 10k at room temp so I thought that a couple of extra ohms in a longer cable wouldn’t make any difference to it. However it now over reads by ~10 degrees C!

This is the bit I can’t get my head around is as the probe has a negative temperature coefficient, that is to say when it heats up it does the opposite of metal and its resistance drops, so if anything tiny bit of the extra resistance in the longer wiring should cause it to under read.

I’m thinking about putting a resistor in series with the probe to ‘calibrate’ but I’m worried this will make it non linear, has anyone else ever experience anything like this themselves?

One thing I should mention is I actually have three temperature probes, so three pairs of wires, what I did to make things neater was I connected all the black (gnd) wire for each probe together as this allowed me to run a single four core wire from the cabin to the engine bay (one shared ground and three signal wires). I did check this wouldn’t affect the readings by checking this setup on the bench with short wiring before I installed them as I wanted to make sure this had no effect on them, which on the bench it didn’t.

[msc]

What are you using for meters and how are they powered?

I think your problem is in trying to share a common return.

Mike

[JRMichler]

It's possible that you're picking up some high frequency noise that's faking out the electronics. If that's the case, you can try twisting the wires about one turn every two or three inches. If that doesn't do it, then you need shielded wire. And shared common returns are always bad.