Temp Gauge Problem 2.1 DJ

[Richardpiper]

Hi All,

I've had what I think is a temp gauge problem for a while. The gauge rises as the van warms up. It goes right to the top and the red light starts flashing. I don't think there is any overheating- the thermostat is a new genuine replacement. Coolant changed two years ago, expansion tank full etc. You can touch and hold the main hoses around the thermostat housing when the gauge is right up, although you wouldn't want to hold them for long. Certainly not near boiling.

This is what I've done:

1. Read everything relevant I can find on this fourm.
2. Tested the gauge calibration as described by AngeloEvs- no problem- at 100 ohms the needle is over the red light, mid-scale.
3. Replaced the module 43.
4. Measured the resistance of the sender unit- this is around 95-105 ohms when the gauge is right up at the top but the engine just seems nicely warm.
5. Measured the voltage at the voltage stabiliser unit behind the dash- spot on at 10V, ignition on, engine running or not running.

I have not tested the voltage at the gauge terminals yet, they are very difficult to reach without disconnecting a load of stuff, the less disturbance the better.

This is driving me crazy, I can't think of anything else to do, and I don't want to drive the van in this state.

Any suggestions? Thanks!

Richard

[AngeloEvs]

Your gauge is reading slightly high by a couple of needle widths at 100 ohms calibration test (unless you tested the gauge calibration with the coolant sender still connected). Higher tham normal readings could be simply due to out of tolerance gauge or the Voltage stabiliser output not being regulated at 10V but something is clearly wrong as the LED will normally flash when the load resistance across the gauge falls just below 50 ohms.

Try disconnecting the connectors on the temp sender just to make sure that there is no other load on the gauge that could cause the needle to deflect. Your 95 ohm reading of the sender is about right for normal temp. The only other component is a electrolytic capacitor that is connected across the output of the gauge to Earth (located on the membrane pcb) but whilst these can become 'leaky' I haven't heard of one causing this problem. Best of luck with this one......most problems that cause higher than normal temperature readings are due to coolant issues but the Module 43. Voltage Stabiliser, coolant temp sender and wiring shortcircuits are the next suspects.

[Richardpiper]

Your gauge is reading slightly high by a couple of needle widths at 100 ohms calibration test (unless you tested the gauge calibration with the coolant sender still connected).

I disconnected the sender at the white junction block in the engine bay..... unless... thinking while write, maybe I split the wrong block... there are two in that black box!

I measured the resistance of the sender by splitting the correct block to save getting under the van, so I know I've got the right one now.
I'll look at it again tomorrow.

Best of luck with this one......

Thanks... I think I need it!
Cheers
Richard

[Richardpiper]

I took another look at everything. At 104 ohms test resistance, (using a potentiometer as a resistor) the needle is directly over the red light on the temp scale.
When the engine feels warm, not hot, the sensor has a resistance of around 98-100 ohms but the needle is up near the top of the scale. That makes no sense at all to me.
When I disconnected the white multiconnector to measure the sensor resistance, the tickover rises slightly then falls back when I reconnect. After this, the red LED flashes until the engine is switched off and restarted.
When the thermostat has just opened, the sensor resistance is 80 ohms, the needle is virtually at the top, but the red LED is not flashing.

I have worked on cars as an amateur for 30-odd years , but this has got me beat. Nothing makes sense. Anything like a bad earth would increase the reistance in the circuit, so reduce the gauge reading.
?????????

Richard

[Aidan]

i'd say there's a fault on the front to back wiring and the sender wire is earthing out before the multiplug at the dash as if you earth the sender wire it goes to full and lamp flashes that's the quick test for the guage
are you on the right sender ? black for guage, blue is ECU temp 2 signal, if you disconnect temp 2 revs will rise
I'd continuity test the wiring front to back to see if there's a short to earth, red and yellow wire should be open circuit with the multiplug off
flexi circuit can be damaged esp on the edge connections and could be something making a short there ?

I had this once on a diesel van I was working on and couldn't find the fault, but the van had a major water leak around the windscreen and the fusebox was damp and earths were pants; it had been stood a long time, the engine was fine though, in the end fitted an aux temp guage for peace of mind and disconnected the red/yellow - this was an early van though, the circuit on the late van with bod is slightly different .

[Richardpiper]

i'd say there's a fault on the front to back wiring and the sender wire is earthing out before the multiplug at the dash as if you earth the sender wire it goes to full and lamp flashes that's the quick test for the guage
are you on the right sender ? black for guage, blue is ECU temp 2 signal, if you disconnect temp 2 revs will rise
I'd continuity test the wiring front to back to see if there's a short to earth, red and yellow wire should be open circuit with the multiplug off
flexi circuit can be damaged esp on the edge connections and could be something making a short there ?

I had this once on a diesel van I was working on and couldn't find the fault, but the van had a major water leak around the windscreen and the fusebox was damp and earths were pants; it had been stood a long time, the engine was fine though, in the end fitted an aux temp guage for peace of mind and disconnected the red/yellow - this was an early van though, the circuit on the late van with bod is slightly different .

Thanks Aidan, I'll get the multimeter out. I'm testing the forward-facing sensor, red/yellow and brown wires.
An aux gauge looks like a good option.
Cheers
Richard

[Aidan]

often damage from wires touching hot exhaust manifold in that area, originally they would have been cable tied out of the way but if engine has been out then may not have been quite so tidied up

[AngeloEvs]

I would measure the resistance across the gauge with everything connected as normal but with the LSCU module 43 removed and gauge reading high. You can do this at the LSCU module 43 socket but instead of inserting a dummy resistor across the terminals I referred to )in the post that you read about calibration checks (in WIKI too), insert two male spade connectors and attach the meter probes to them. Your readings are spot on and if you get the same or similar redings with the test just described I would check the guage by removing the dash. Power the dash up by connecting Pin 3 to zero volts and pin 8 to +12V. Place a 100 ohm resistor across the Temp gauge output terminal nut to the centre nut and watch the needle. It should stop left of the red led. If it continues to climb on a bench test then either the guage is faulty or the capacitor is leaky (unlikely but not impossible). Also take a resistance reading of the gauge output to pin 3 earth. This should be very high but be aware that there is a microchip in the gauge which controls the coolant LED and responds to pulses from the LSCU module 43 if coolant is lost. This chip could be the cause of the problem as it connects to the output side of the gauge as well as the 10V regulator, earth, etc, but you will be the first person to have experienced it failing and causing this problem.

[Richardpiper]

I would measure the resistance across the gauge with everything connected as normal but with the LSCU module 43 removed and gauge reading high. You can do this at the LSCU module 43 socket but instead of inserting a dummy resistor across the terminals I referred to )in the post that you read about calibration checks (in WIKI too), insert two male spade connectors and attach the meter probes to them. .

OK, after a busy week, I've found time to try that out.
The meter reads 705 ohms (!) when the thermostat opens. (Gauge up at top, red LED flashing this time). I assume I've done the check wrong in some way, don't see how- meter attached as above.

Looks like I might end up fitting an extra gauge from Just Kampers. Main argument against that (apart from cost) is that you have to cut a hose and insert a kind of tee piece to hold the sender, apparently no suitable tapped hole in the block.

Cheers
Richard

[AngeloEvs]

the picture in the thead.........https://club8090.co.uk/wiki/Co ... alfunction you measure across the same terminals as the resistor is fitted into but no resistor, just a couple of male spades pushed into the socket so that you can place the meter probes on those.

A reading of 705 ohms and a temp gauge reading that high suggests that the fault is on the dash or in the gauge, etc. I would remove the dash and bench test the guage (whilst still fitted) and confirm that this is definitely faulty.

[Richardpiper]

Well, I decided I'd had enough so I bought a VDO gauge on Ebay. Using the VW sender, on a try-out it reads 90C as the thermostat opens. That's close enough for me, so I think I'll just be disconnecting the original gauge and using the bolt-on replacement.
Thanks to all for the advice, particularly AngeloEvs.
cheers
Richard

[AngeloEvs]

If you are removing the gauge I am happy to check it for you and see if it can be repaired if faulty. You will lose the LED Low coolant warning with it disconnected/removed but you cannot leave it connected faulty as it will most likely affect the reading of the new gauge.