Temperature gauge accuracy

I'm trying to get an accurate reading for my stock temperature gauge.

I replaced my mechanical regulator with an electronic regulator from RTE, and calibrated my temp gauge to 10, 23, 73 which is standard.

Digging around on the web and making some phone calls, the consensus seems to be that Mopar gauges should read 190 in the middle.  However, my 69 A108 FSM shows a resistance wire going from the sender to the gauge, and I confirmed that mine reads 5.8 ohms, which is what jkr also reported here:
https://vintage-vans.forumotion.com/t32426-a100-temp-gauge-resistance-cable#222866

Now here's where it gets interesting.  I bought a new sender from Napa (Echlin), and the spec sheet says 15-19 ohms at 220 degrees.  So with the resistance wire on my van, I interpret this to mean the gauge should see 20.8-24.8 ohms at 220 degrees.  If that's the case, how on earth could the temperature be 190 in the middle of the gauge (23 ohms)?  If I removed the resistance wire, it sounds like it *would* be around 190 in the middle.  I pulled my sender off the engine, and while still running through the resistance wire to the gauge, put it in a pot on a burner, and my temp gauge reads ~207 degrees.

Does everyone also think we should be at 190 in the middle to the gauge?  Do all Mopars have this resistance wire, or could it be calibrated based on where the sender sits in the engine?  17 degrees is a hell of a ways off, which is why I'm concerned.

I have not seen that resistance-wire in a passenger car Mopar - my 69 Charger definitely does not have one. The sending unit has the same parts no (IIRC) and I installed a new one from rockauto.com lately to my A100. At approx. 175°F it reads somewhere leftish inside the two marks of the gauge.

Problem is, that the gauge is most likely somewhat inaccurate. The original mechanical voltage regulator obviously gives you somewhat higher readings on the gauge compared to electronically regulated 5V.

I do it the "easy way": drive the truck and measure the upper cooling hose with an infrared thermometer when hot and compare to the gauge reading. I test it every now and then (normal driving, stuck in traffic etc.). At least, this is an easy option with our engines right next to the driver seat

Why do you say the mechanical regulator would obviously read higher than the electronic regulator?  I’m using the IVR from RTE, which is supposed to match the original style (turns on and off rather than generating a steady output).

I wonder if they just assume the engines in the vans will get hotter, and use the extra resistance in the wire as a way to not concern the driver?

Of course it turns on and off - but even their website says, that it outputs "steady 5V". That's on/off on steroids (multiple thousand times per sec). Effectively, it is a different output to 5-6 on/off cycles (12V, 0V) as the mechanical ones.

Further, our A100 has the limiter build into the fuel gauge (at least 68 and 69) and not as a dash-mounted limiter (my charger has such a limiter).

And I can compare, as I have replaced both, Charger & Truck, with linear limiters (5V steady, of course taking care of heat). Both read lower now than before.

Not on the IVR. I threw a voltmeter and scope on it, and it pulses below 10Hz. They literally tried to replicate the mechanical regulators. Totally different than a linear which looks like steady 5V on voltmeter and scope.

In any case, it’s certainly more precise than the mechanical - which is gonna just be in the ballpark.

Do you know if there’s a diff in output between the internal in the fuel gauge (like the vans have), versus the external? I wonder if the internal regulator could have anything to do with the resistor, because if the voltage output on the internal regulator is lower or higher or average, it would certainly need to be compensated for.

Thanks for the insight. As far as I remember, the fuel-unit works exactly the same as the external one. IIRC, even some passenger Mopars use the fuel-unit.

But if it pulses with 10Hz, their website is wrong - not only regarding the "steady 5v" but also the application list of cars....

Anyway, using a IR thermometer to check the readings would be my way, because my feeling is that you'll never get the gauges to read perfect

I don’t recall the exact hertz of the IVR, but it’s definitey slow (and watchable on a voltmeter).

Like I said before, perfectly calibrated gauges show me reading 207 at the center of the gauge with an industrial thermometer (1% F degrees accurate) which is much more accurate than an IR gauge. I’ll try the IR gauge next - but that’ll probably just leave me with more questions.

Well I answered my own question today.

I re-read the FSM about testing the gauge, and it says:
Disconnect the molded terminal and wire from the temperature sending unit on the engine.  Connect one test lead of Tester C-3826 (10, 23, 73 ohms) to the molded terminal and the other test lead to a good ground.

So this means that later model Dodge A van gauges are actually calibrated including the 5.8 ohms of series resistance built into the sending unit wire.  Said another way, as opposed to the gauges in passenger cars, the van gauges should be calibrated to 15.8, 28.8, and 78.8 ohms!

I do not feel that they are actually "calibrated" For me, its more "assume, the temp might be in the range of". It reasonably accurate in terms of "yesterday it showed X so it'll be somewhat the same today". And if it leaves the two marks on the right hand side - be worried

the resistance wire I refereed to does not really add to the gauges final reading from what I know. it DOES however slow the radical changes of the system that reflects as wild temp swings of the gauge. if you cut the wire off and just run a copper stranded piece in it's place the swings are very noticeable. ask me how I know.... I had to retrieve mine from the trash can and it is still all coiled up around the sending unit

jkr wrote:the resistance wire I refereed to does not really add to the gauges final reading from what I know.

It sure does make a difference on the gauge.  If I pull the sender from the engine and put it in a pot of boiling water, and compare that to jumping the wire directly from the gauge to the sender (without going through the resistance wire), the needle is clearly in a different location.

So maybe the resistance wire is to avoid the jumpiness like you're suggesting, which for some reason only occurs on the vans, and not the cars?

That's something odd which I don't understand either.

I re-calibrated my gauge today, so for posterity:

If you calibrate your gauge at 10, 23, 73, and exclude the sender resistance wire, your gauge will read:

Slightly below C for 73 -- close enough where it would be in spec and/or you wouldn't care.
Between the T and E of TEMP for 23.
The second to last hash mark (between the P in TEMP and H) for 10.

So the middle and high temp values would be misleading.

I run both the stock gauge and an aftermarket autometer mechanical gauge in my truck. normally it is between the E and M of the stock dash gauge and about 180 - 190 on the other. the 360 intake I use had a provision for a vacuum ported fitting that I removed and installed the probe from the mechanical gauge. a hot day and slow moving city traffic I have seen 210-230 but that doesn't worry me much.

Actually a great idea running a mechanical gauge. I wonder how I could fit that in and run both -- given I don't have a 360 intake.

Brandon2 wrote:Actually a great idea running a mechanical gauge.  I wonder how I could fit that in and run both -- given I don't have a 360 intake.

You could use something like this.