Club 80-90 Forums

Right, I've worked my way through the cooling system (flush system, new thermostat and bleed system) and am happy that the engine is definitely not overheating. So back to the drawing board again with the electrical side of things. Cutting a long story short, I've fitted a 100 ohm resistor across the plug at the sensor, across the multipin connector in the engine bay, and across the correct connections with module 43 removed. In all positions the needle sat in the centre over the LED (which is working correctly). I've fitted a new temperature sensor and also tried another thermal transistor, all with no success. The problem is the temperature guage rises with the engine temp but continues to rise, sometimes almost to the max and other times into the max with the LED flashing. Before fitting the new sensor I plugged it in and suspended it in a kettle as it was boiling, and again the needle went into the max. The fuel guage is working correctly. Any ideas? Cheers.

new sender is a wrong un

The new sensor cross referenced with the old one (.....501D) but it was a fairly cheap one from my local motor factor (intermotor). I've tried both the new one and original one in boiling kettles and they produce the same results (temp guage into the red) and also checked the resistance of both at various temperatures which produced very similar results. Could it be a problem with the 43 module? Is it possible to take it out and link out terminals in order to run the guage without the module, just to see what happens?

I seem to remember reading somewhere that, if the low water level sensor is triggered, the circuit moves the needle to the right so the driver can clearly see the LED flashing.
I wonder if that is owt to do with it ? (I take it you know "where and how" re the low water sensor ?)

I'll disconnect it tomorrow and see what happens. Cheers.

Might be no help at all, but I had a 6 month hunt for a dodgy temp gauge on my 323f.

The gauge would just go up to the top, stay there for a while, then come back down. After changing pretty much everything I discovered one of the engine earth cables was damaged. Replaced that and it was all fixed! Turns out the engine was trying to find the shortest route to earth via the temp gauge.
Might not be relevant to you, but just in case!

Rob

TeamSid wrote:I'll disconnect it tomorrow and see what happens. Cheers.

Actually disconnecting it will only do the same thing as disconnecting it will simulate a low water situation triggering the movement upwards and the flashing of the LED

Cheers Rob, I've given the earths a clean up but as I'm running out of options I might have to retrace my steps again. Are you a Volksfling goer?
Cheers Ghost', I take it that if I link out the LED at the plug it will flash constantly? Things have warmed up a little outside now so I'll nip out and have a poke around and baffle myself a little more.

TeamSid wrote: Cheers Ghost', I take it that if I link out the LED at the plug it will flash constantly?

I would do the following =

Engine cold, make sure the left hand plastic header tank is full to the brim, but it MUST have water AND antifreeze in roughly the right proportions (the sensor doesn't work correctly with just clean water).

Turn the ignition on, don't start engine, wait for the usual few flashes of the temp gauge LED to cease flashing.
At this point the gauge should of course read cold (ie not move).

Then pull off the plug on top of the sensor on the left hand header tank.

The LED should then start to flash. The needle might move over at this point (if I am correct, I might not be)

If it does not work as above then summat is wrong with the sensor or the control box on the relay panel.

If the above procedure has the desired effect on the LED but the gauge moves over BEFORE you pulled off the plug on the plastic header tank, then, well then we must think again re what is up. (You sure the fuel gauge is reading correctly ?)

Note, after "triggering" the low water sensor by unplugging it's lead, you have to turn the ignition off, then back on, to re-set it (this is a function of that control box on the relay panel, number 43 if memory serves correct, but check that detail).

Cheers Ghost', I'll have a bash at that in the morning. I retraced my steps today and didn't find anything else that appears to be suspect. I did however take module 43 off and carefully prise the casing off it, hoping to find some furred up circuitry or something. I was surprised to find a small circuit board containing some resistors and a microchip....and all clean at that. I might have a hunt for a replacement one and see if that makes any difference. By the way, does anyone know if my van should have a 'buzzer of doom'? I noticed that a wire (yellow/red) tails off the back of position 3 on the fuse board (for module 43) to position 2 where it meets another wire (blue/green). There's nothing fitted to position 2 but I notice someone has marked it with an X. What do you reckon? 1988 Caravelle 2.1dj.

Yes it will have the BOD system.

Don't know wiring off by heart so cannot comment there.

Cheers Rob, I've given the earths a clean up but as I'm running out of options I might have to retrace my steps again. Are you a Volksfling goer?

Sadly not, it's the same weekend as the 10 Under The Ben mountain bike race at Fort william

If you suspect Module 43 then remove it and re-test temperature gauge readings. Module 43 controls the function of the low coolant warning LED - if faulty it will load the temp gauge and make it read higher than normal. Once Module 43 is triggered and the coolant LED starts to flash, it loads the Temp Gauge and causes the needle to rise. The important thing to note here is how the module 43 functions - it works on the resistance of fluid. Unfortunately, poor earths will trigger module 43 to trigger just as much as loss of coolant will. If the temp gauge seems stable with 43 removed then you really need to check the earths for Module 43 before replacing it. There are two:-

a) one at the crown cluster near the door pillar (this is the earth for the micro chip in the module 43)

b) one in the engine comprtment close to the coil. This is the earth connection for the two pronged sensor.

c) the centre nut of the temp gauge (this also has micro chip in it) must not be loose, this rarely triggers the module - but worth a quick check though.

if the temp gauge is still erratic, or reading excessively high with module 43 removed, then you have a problem with the gauge temperature sender, the wiring, coolant efficiency or have changed the wrong sender or reversed them - one is a switch and the one facing the front cab is the thermal sender that connects to the gauge.

Temp gauge/Coolant LED problems should be diagnosed with module 43 removed after it is verified that the coolant system is not the problem but caused by an electrical system fault. Removing the module 43 allows you to isolate which circuit is causing the problem - the coolant level circuit or the temp sender circuit.

Right, I've just carried out a few checks, thanks to Ghost' and AngeloEvs:
LED flashes when connection at tank is removed,
Checked earths which are all OK.
Ran an earth directly to the guage. No difference.
Checked the 22ohm resistor. OK
Checked voltage across temperature guage. 10volts between earth (centre nut) and one nut, 6volts between the earth and the other nut.
I also ran the engine up to running temperature, removed the clocks and measured the resistance across the earth terminal and the yellow/red wire for the guage on the multipin connector. Result were between 80 to 100ohms.
I then measured the resistance across the temperature sensor which gave the same results.
I'm now at that point where I'm running out of things to check, assuming the results above are within correct tolerances. In an ideal world I could do with a van sitting next to mine and compare the two!

I would test the temp gauge calibration on a bench but you could use the Module 43 Socket as described in the Wiki. (Module 43 removed for access and connections at temp sender disconnected)

A 100 ohm 1W resistor across the output of the Temp Gauge should result in the needle sitting approx half way. There is a VW tolerance for this test and its one needle width either side of the LED.

A 50 ohm Resistor (47 ohm is the most common value available) but you can use two 100 ohms 1W in parallel. This should have the needle sitting far right into or close to the white band and triggering the LED.

In your case you need to be using values of around 80 ohms so try a couple of 150 1W in parallel . I woulld expect to see the gauge sit somewhere betwwen Mid and the reading you had using the 50 ohm value but not triggering the Coolant LED.

I have a spare POD so can do some comparison tests on a bench if it helps.

My gut feeling is that the coolant temp in the thermostat housing may be higher than normal as you are getting a reading of 80 ohms or the system isn't under pressure , sender is off or the thermostat is opening at too high a temperature (I use a 82 deg stat all year round and the needle never goes beyond the LED)

Did you measure the senders resistance in a pan of boiling water - should be around 100 ohms.