Pantera DeTomaso Temperature Gauges

Over the years I have heard that the temperature gauges in Pantera’s were a bit off. And, I’ve heard that adding a 10 ohm resister in series with the temperature probe would make the gauge more accurate. In fact, at least one vendor offers a nifty plug and play setup to do just this. Our car came with the 230F gauge – which shows 160F as the centered point on the scale. With a 190F thermostat – the car always looks like it is running warm. And, when the car is actually hot – as in I set the radiator fan controller to come on at around 210 and off at 220F, well, it looks like the thing is really too hot, with the needle way over to the right. I was also aware that later cars came with 260F gauges. So, I set out to find a 260 on eBay. Sure enough, one showed up and found a new home. Naturally, I was curious how accurate it was. And, well, for that matter, how accurate the original was. Early on I tested the 230F gauge to see if the Echlin TS6628 temperature sensor would work. I set up this apparatus to see what I could learn – an aluminum container filled with water that I could heat with a heat gun, temperature sensor and temperature gauge in the water, two volt-ohm meters to see what the sensor’s resistance was, and the temperature gauge left in the dash, with the engine wires extended to the temperature sensor in the aluminum can. Oh, I used a screw driver to stir the water in the can.

5 / 10

Figure 5 – Circuit Model

Note, I included the “+” and “Term” labels from the back of the meter on the diagram above to designate where wires went. See Figure 9 for a shot of the back of the meter.

I then mathematically modeled the circuit, knowing that resistors in series add and resistors in parallel add the inverse. Note – the meter is set up such that there is battery voltage across one coil while the temperature sender powers a separate coil in the meter. As such, the meter reading is proportionate to the difference between battery voltage and the reduced voltage due to the temperature sensors resistance.

Next I experimented with various resistance values for R0 and R1 until I got the resistance the meter saw at a given temperature (combination of the temperature sensor resistance (R) plus, R0 plus R1) closer to the resistance generated by the temperature sensor at the same temperature.

OK, I know this is complex – and likely confusing. Don’t get the idea I figured this out on my first try. Initially I sort of figured I’d possibly need a resistor in parallel with the coil in the meter that senses the temperature sensor, and might also need a resistor in series with the coil in the meter that senses the battery voltage. Turned out I didn’t need either.

My modeling suggested that I needed around 200 ohms in parallel with the sensor (R0). Turns out, after experimenting with resistors and the gauge, I needed 220 ohms in parallel (R0) and 1/2 ohm in series (R1). I did not need R2 or R3.

The resultant data are plotted on Figure 6.

Circuit Model