Temperature gauge test

[2ndwind]

I could not find any info on how to test the temp gauge for my 1982 FI (same as others years?) other than a comment from Mark in a post to attach 12V to peg the needle. Through trial and error, here is what I did:

Using a trickle charger set at 6AMPS
#1 When looking at the back of the gauge, attach positive 12V to lower left terminal (GN) and negative to lower right (V). Should peg the meter. If this doesn't happen, the meter coils are probably burnt up inside.
#2 Positive 12V on lower left terminal (GN), negative on lone UPPER terminal (BN), attach a 10K Ohm variable resistor to upper terminal (BN) (with the neg 12V too) and lower right terminal (V). Adjusting the resistor should variably deflect the gauge needle. This can also be done by attaching the resistor between GN and V with the power hooked up as in the beginning of #2 but the deflection response is much slower.

Any improvements on this testing technique by someone who knows what they are doing is most welcome.
For example, I don't know if running 12V right off a battery would give the same results.

[spider2081]

I have a wire diagram for a 1981 with wire colors. yellow/black is 12V, Green connects the sender to the gauge, white/black is the ground. The gauge consists of 2 coils with 12v connected to the tie point of the the coils. One coil connects to ground through the white/black wire. The other coil connects to ground through the Green wire through the temp sender in the head to ground. Disconnecting the sender when power and ground are connected should cause the gauge to peg full hot. I don't know what the resistance of the sender is at 190 degrees so I can't tell you how tot test the accuracy of you gauge. I tested mine by using a spare sender and putting it in boiling water.
I found the needle on my temp gauge moved aprox 10 degrees with the panel lights on. I resoldered the ground tab to the gauge and added a ground wire for the gauge backlight. this cured that problem.

[So Cal Mark]

I use a variable resistor on the wire to the sending unit. The trick is to find the specs for the gauge at the low, mid and high ranges. By using a resistor on the sending unit wire you're testing the integrity of the entire system

[baltobernie]

A cheap infrared thermometer http://www.harborfreight.com/non-contac ... 93983.html is the easy way to check your cooling system components.

[spider2081]

Good point on the hand held temp gauge. Harbor freight has a lot of inexpensive stuff that is good enough for me as a back yard mechanic. I bought a number of seldom used but nice to have tools from them. One being a hand held temp gun another being their fuel injection test kit. I questioned the accuracy of the pressure gauge in the kit and tested it against a calibrated one from our shop. Darn thing was so close it shocked me.

[So Cal Mark]

the infrared temp gun at HF is only about $26 these days and is well worth the investment. They work well for testing plugged radiators as well as other components

[spider2081]

I tested the spare gauge from my 81 spider.
Looking at the back of the gauge the top terminal has a white/black wire on it and is the ground connection for the gauge and the dial light. Below the light on the left side the letters GN are stamped next to a terminal with a yellow/black wire on it. To the right the terminal has 12v stamped near it and then the letter V next to the terminal. A green wire is on this terminal.
To get the gauge to work (deflect to the right) I had to apply 12VDC to the yellow/black wire on the left side of the gauge and the power supply ground to the ground terminal at the top of the gauge.
With nothing connected to the green wire the gauge did not peg even though the Fiat Electrical Diagnostic Manual stated it should.
I found the gauge reads; 120 degrees with a sender resistance of 950 ohms, 190 degrees with sender resistance of 175 ohms and 260 degrees with the sender resistance set to 75 ohms.
Gauge does read approx 210 degrees with the used sender in boiling water so I think the resistances are pretty close.
Hope this is of some help.