Because instruments are unaffected by low voltage, grounding is a complete shot in the dark. The answer lies elsewhere.

Just to educate me: how can an electric gauge not be affected by low voltage?

__________________
"The MB W123 is so bulletproof, you can drive them forever. Which is a good thing as it takes that long to get anywhere."
Betsie: 1984 W123 300D (hobby, 280k miles)
Myrla: 2001 Mazda Protege 2.0 ES 5spd (daily driver, 130k miles)
The Turd: 2007 Toyota Camry (wife's car, 118k miles)

i have a coolant leak, i will start a new thread

__________________
1981 NA 300D 310k miles

Every car made since the 1930's has one or more instrument voltage regulators which allow the instruments to work accurately independent of system voltage. System voltage varies from 12-14.5V in normal operation, and can be lower if there's a charging problem. So the instruments are designed to operate at a much lower voltage which can be accurately derived from the system voltage. English dashboards operate on 10V+, pretty much everyone else operates on 5V+. If the IVR wasn't present, your gauges would constantly fluctuate as equipment switches on and off.

On older cars (prior to the early 70's), the IVR's were thermal vibrators, which would be comprised of a heating coil and a bimetallic contact arm. When the IVR contacts closed, the heating coil would run hot and cause the arm to bend away from the contact. Once contact broke, the bimetallic arm cools, returning to normal shape, and the process repeats. This happens several times a minute, the rate or duty cycle dependent on the system voltage. The output voltage will be the duty cycle (always less than 100%) x system voltage. This is a mechanical form of PWM regulation. In some cases, the vibrator is incorporated directly into the sensor, in others it's a separate box in the dashboard. Modern cars use solid state devices incorporated into the gauge or the dash circuitry and achieve much more accurate regulation.

The point is, system voltage has to drop below 6V before the PWM regulator is affected. Above that threshold, the instruments will be getting a steady 5V. What good would your fuel gauge or temp gauge be if the reading was dependent on the headlights being switched off and your battery fully charged?

Good Question.

Looking at wiring diagram, it looks like the temperature gauge is in effect a voltmeter. It measures voltage to ground via a variable resistor, which is the temperature sensor. This is what I found for teh sensor characteristics:

At -20 °C 15.7 kOhm
At -10 °C 10.0 kOhm
At 0 °C 5.9 kOhm
At 10 °C 3.7 kOhm
At 20 °C 2.5 kOhm
At 30 °C 1.7 kOhm
At 40 °C 1.17 kOhm
At 50 °C 830 Ohm
At 60 °C 600 Ohm
At 70 °C 435 Ohm
At 80 °C 325 Ohm
At 90 °C 245 Ohm
At 100 °C 185 Ohm

It would appear that the higher the resistance, the lower the gauge should read.

If there was additional resistance to ground, then wouldn't the gauge read low and not high?

Similarly, if voltage on feed to gauge is low due to resistance in 12v supply, then wouldn't the gauge again read low rather than high?

I have no horse in this race - just interested in theory behind the jumper wire test.

If gauge is suspected, I would first check actual temperatures with an IR thermometer.

__________________
Graham
85 300D,72 350SL, 98 E320, Outback 2.5

A few weeks ago on my 1985 300D, I changed the radiator hoses, thermostat and because my radiator had a crack I also installed a new radiator. Prior to this work no over heating issues even in 100 degree weather. About 1 week after this work was completed my temp gauge would go to the red area within 5 minutes of starting the engine. Luckily, I have an infrared temp gun which told me the temp was really 80c. The temp gauge as bad. I had an extra temp/fuel/oil pressure gauge and that fixed the problem. I know the oil pressure side of the gauge is known for failure but I now know the temp gauge fails also. Borrow or buy an infrared temp gun to see if you really have an overheating problem.

That sounds logical. However, I went to the W123 wiring diagrams, and I don't see anything beyond the normal alternator voltage regulator. The feed to the temperature gauge appears to come right off circuit 15 which it seems is just the normal regulated system voltage. (page 120 of the diagram I have). Can you locate the regulator that controls and drops instrument voltage?

Just trying to learn a little more about these old cars!

__________________
Graham
85 300D,72 350SL, 98 E320, Outback 2.5

By the time the 123's came off the line, it would be a solid state circuit either on the circuit board or internal to the instrument, not a separate part. I may have a 124 dash in the cellar that I can play with, but no 123 parts these days. I'll have a look tomorrow.

I have been having the exact same issue for the last year or so. I have replaced the entire cooling system, and my ground strap is clean and tight. For a while, I thought that maybe the coolant flow is not adequate at idle because if I raise the idle even a little bit you can see the improvement almost immediately. When I installed a new thermostat it seemed to help a little. I know I don't have air in the system because I use a vacuum lift to fill the system. So... I am curious if it has to do with a failing temperature gauge.

Based on this statement, I'd say it's an electrical problem, because the coolant temperature cannot change that fast.

__________________
85 300D turbo pristine w 157k when purchased 161K now
83 300 D turbo 297K runs great. SOLD!
83 240D 4 spd manual- parted out then junked

I've looked at the W123 wiring diagrams and do not remember seeing a voltage regulator in the instr cluster either. I think the W123 instr cluster gauges is a bridge circuit and that is how they achieve their accuracy w/o an on board voltage regulator.

__________________
85 300D turbo pristine w 157k when purchased 161K now
83 300 D turbo 297K runs great. SOLD!
83 240D 4 spd manual- parted out then junked

Electrical problem as posted earlier in thread , because as mentioned the needle drops in seconds.

So lets factor in reduced alternator output at tickover and problem cures when revs raised.... And it only occurs at tick over.

So take a reading from alternator at tick over with meter on windshield where you can see it as you slowly apply throttle and see if the gauge drops as the volts rise .

I was thinking along those lines until Mxfrank said the gauges were compensated for varying voltage. But a failing voltage regulator could possibly be a factor?

The voltmeter test would be useful. But so would using an IR thermometer to check if the temperature does actually go up. Or even temperature indicating tape (Like this, but who sells it?)

__________________
Graham
85 300D,72 350SL, 98 E320, Outback 2.5

I think the problem with loose ground strap is all the other electric components lights, alternator and radio/ blower pulling power through the only ground path... that would be the transmission Bowden cable etc... it's not just a resistance issue... likely drops the miniscule instrument draw to sub one volt...

__________________
John HAUL AWAY, OR CRUSHED CARS!!! HELP ME keep the cars out of the crusher! A/C Thread
"as I ride with my a/c on... I have fond memories of sweaty oily saturdays and spewing R12 into the air. THANKS for all you do!

My drivers:
1987 190D 2.5Turbo
1987 190D 2.5Turbo
1987 190D 2.5-5SPEED!!!

1987 300TD
1987 300TD
1994GMC 2500 6.5Turbo truck... I had to put the ladder somewhere!

Not happening. If it was, he'd be complaining about the lights, radio, blower, gas gauge, etc. And if voltage was that low, both the gas and temp gauges would be reading low, or more likely, not reading at all. In addition, his battery would be drained in short order. And bear in mind that the instruments ground through the sensors.