blinking coolant temperature LED

[runningfix]

Ok, time to press reboot…

1. the alternator charges ok, producing 14.03v
2. the alternator tested ok on the bench using an oscilloscope
3. the primary battery receives 13.45v during alt charge
4. the auxiliary battery receives 13.22v during alt charge
5. the temp LED goes through a self test at ignition on and off at start-up idle
6. the voltage at the temp gauge 10v post is 10.22v prior to alt charge
7. the voltage at the temp sensor post is 9.2v prior to alt charge (cold)
8. the voltage at the temp gauge 10v post drops suddenly to 9.29v with alt charge
9. the voltage at the sender temp gauge post drops suddenly to 8.8v with alt charge
10. the temp LED goes solid with alt charge, the alt LED goes out.

I’m going to trace the where of the voltage leak / drop when the alt charge kicks-in.

[runningfix]

Well, tracing the voltage drop forward lead me back into the cluster and the voltage regulator. This is a new regulator (eBay not Brickwerks). However, I can’t see that the drop is significant enough to warrant the temp LED going solid. At engine idle under no charge the input leg of the regulator reads 12.20v and the output leg 10.25v. As soon as the alt charge kicks in the input leg reads 13.57v and the output leg 9.8v (albeit a bit erratic). I read on another post that the 10v regulator tolerance is +/- 0.5v so this is within tolerance, which leads me back to dbeierl’s original conclusion that the gauge driver circuit and the LED must be at fault, despite a self test (of sorts) and the needle working fine.

So, do I have the confidence to replace the gauge, hmm not sure. If I can find one on eBay at a decent price perhaps I’ll have a go, until then I’ll just have to live with it, and I might order a Brickwerks regulator as a precaution.

Thanks all for guiding me through this maze.

[runningfix]

RE: solid temperature light on T25, 1983 DG 1.9 WBX

Ok, came across a used temperature gauge at a decent price from a reputable parts seller over here and mustered up the courage to tackle the PCB and replace what seemed to be a faulty gauge in the in the cluster (turned out to be much easier than I was expecting). However, no change. The new gauge acts in exactly the same manner as the old gauge – from the point of turning the key on until the drive is over the temp light remains on constant – solid not blinking – while the needle works fine just touching the 8:00am position of the LED under normal operation. Note; previously I was getting a self test of sorts, but that soon stopped and the LED just remained on constant. This new gauge act’s exactly the same.

I can’t believe that what dbeierl describes as a rare occurrence can happen twice in succession with different gauges, so have to believe the problem lies elsewhere. Given all the voltages testing fine, the voltage regulator, the alternator testing fine, what else can be left – the PCB itself (interestingly there was some slight delamination and a small tear/opening on the first track above the yellow/red input post of the temperature gauge but when examined under a backlight there was no obvious crack/break in the copper track ...and I’m not sure if that track is related to the temperature gauge).

Any suggestions on what/where to test next gratefully received.

tbh I don’t find the light that bothersome (better than blinking), but it’s now become a mission to solve the conundrum.

[runningfix]

SOLVED: solid temperature light / LED T25 1983 DG 1.9 wbx

Cause: incorrect PCB voltage regulator / stabalizer

Testing the cluster on the bench produced the same solid LED when the VR was placed in-line with the temperature gauge. 12V direct to the gauge presented a blinking light, so that pointed to the VR. Why I didn’t perform that simple test at the outset is beyond me.

Anyway, so nobody else makes the same mistake - suppliers over here were sold of out of the PCB voltage regulator for my westy (and don’t publish any tech details), but I found what I was told was the correct VR on youtube https://www.youtube.com/watch?v=OfOqbUmbCaY and ordered the same (LM2940T-10/NOPB Voltage stabiliser LDO, fixed 10V 1A THT TO220 TEXAS INSTRUMENTS). Sadly that was my mistake. I’ve since learned https://www.youtube.com/watch?v=lZuWcstQShA that this chip needs a capacitor, but more importantly is a switching chip ”… is a buck converter is so it approaches 90% efficiency by switching so there is no waste heat.” My guess is that the switching plays havoc with the temperature LED and that’s why it presented itself as a solid light i.e. always on, not blinking.

I’ve since found the correct VR replacement on Samba https://www.thesamba.com/vw/forum/viewtopic.php?t=718717&highlight=replacing+voltage+stabilizer (LM7810 as a compatible replacement for the original TCA700Y). All installed and the temperature LED working as it should.

On another positive note, all the muddling about in solving this sure did fast track my knowledge on the westy's electrical circuitry, and that's a bonus.

[dbeierl]

[NorthOf50]

[runningfix]

Hi NorthOf50

I'm no expert but those numbers look as they should to me so it does suggest a fault in the temperature gauge LED internals. Are those numbers at ign on or when the engine is running. Can't recall if its this thread or one on club80-90 that has details on taking the gauge apart if you're that way inclined (worth a read as there's info on the capacitor function and a potential wire short), opting for a replacement gauge or a GW audible as you suggest.

Relay 43 is on the A pillar above the relay panel (just below/under the air vent). It's coolant related only and doesn't factor into the LED self test.

[NorthOf50]

Those readings are ign on Eng off.

If I correctly understand the description of capacitor fault in the gauge from reading the previous posts, it sounds like it causes the light to continuously flash, not cause it not to flash. Or would it just fry the circuit to the led entirely?

If replacement of the cap will get my light working again, that is something that I can do. If not and its gauge internal then I'll just opt for the GW audible as the rest of the gauges etc all work fine. I'll just have to give up the coolant low warning, but I'm not sure if I have a 42 or 43 relay, will check later today.

Thanks!
_________________
'84 1.9WBX Camper (Penny)
'88 2.1WBX Camper (Martin)

[runningfix]

If I recall correctly from what others have told me on this thread, the LED has a micro circuit embedded within its resin that causes it to flash if the voltage drops below 5V (the engine temp sensor voltage output reduces the hotter the engine gets). As I understand it the capacitor in the gauge interrupts that voltage temporarily while it’s being filled (the self test) and then again if it’s being emptied faster than it can be filled (engine overheating) both occurrences causing the LED to flash. If the capacitor isn’t working properly then it can cause the LED to flash, but if it’s not working at all then it could be that no power is getting to the LED for it to be able to flash (btw that’s my logic, not a known fact). Others on this thread know much more about this and can no doubt advise better.

Personally, I would look for a replacement gauge to test before taking your gauge apart or going for an audible alarm (which is really meant to compliment the warning light not replace it).

Fyi, Club8090 also have a good wiki on the temperature gauge system https://wiki.club8090.co.uk/index.php/Coolant_and_Heating_Flashing_red_light_on_dash

[Achilles3588]

Nursing a headache after reading thru 14 pages of posts on this thread for about the third time, so humor me and let me post up. '91 Syncro Diesel Doka

Baseline
- Vehicle has the GW Gauge Cluster PCB replacement and Coolant Alarm Kit installed
- New Coolant Level Sensor and terminals replaced at the pressure tank, all working fine until recently.
- Coolant recently replaced w/ 50-50 during engine swap
- Coolant Level at pressure bottle well above Min. line
- Fuel Gauge working fine
- Vehicle fitted with Coolant Level Control Unit P/N 191 9191 376A (small cube)
- Battery fully charged => 14.9 VDC +/- across terminals
- Replaced Temperature Sender w/new

This is what I am seeing:
- Key ON (initial): Coolant Temp Gauge LED Flashes for ~3 sec, no needle deflection
- Key ON, engine COLD, after ~10 seconds or so, LED flashes, gauge needle pegs high, alarm sounds. This is a new development.
- Have checked the ground connections at the Temp Sender and at the Coolant Level Sensor with a VOM. They look good. Measuring voltage across the temp sender terminals gives a fluctuating voltage reading between 9.6 and 10.0 VDC. But when measuring these wires individually, the voltage readings are steady. 10.0 V on the power side, 13.9 V on the ground.
- All grounds behind the relay panel appear solid and relatively corrosion free (not perfect of course).
- Connections at the 14 pin connector also appear to be making good contact.

Any help is appreciated.
_________________
Larry V.~
'91 Syncro Doka 1.9L TDI 'AFN'

[runningfix]

Hi Achilles3588

Is the needle pegging fully to the right, or just high?

I'm not familiar with the 91 Syncro, but for a 1.9 1983 WBX if the needle is pegging fully to the right I would be looking for a grounding issue in the yellow/red wire that runs from the temp sensor forward to the 14 way PCB connector. The needle pegging is exactly what happens when you touch the yellow/red temp sensor end to ground (done as part of some tests) and could happen if the wire jacket has worn through somewhere along its run exposing the copper and coming into contact with ground.

[Achilles3588]

[runningfix]

Hmm, above my paygrade methinks. If it was me I would be thinking that 9.6v at the PCB temperature sensor terminal (yes) is lower than it should be from engine cold, but not so low as to peg the needle so high (others here can be exact on what 9.6v translates to in F/C deg coolant temperature). To me it suggests a gauge sensitivity issue or perhaps some erratic voltage behaviour (was it working correctly in the past). The coolant level side should have no impact on the needle, and with relay 42/43 removed you should still get the LED self test (do you). Since the LED and alarm go quiet with relay 42/43 removed, I would be looking at the coolant level side (have you tested the relay is working correctly) before moving onto the temperature needle side to figure out the why of the 9.6v.

A test lamp is meant to test between a negative and positive source – so an in-line test between two negative points shouldn’t generate a light, but might have a capability to test continuity with a short residue flash (?).

btw, I believe you should have relay 42 not 43 (or an older tall 43) in your year/model. What do you have?

[Achilles3588]

[runningfix]

Ok, reading between the lines I’m assuming you’ve recently replaced the original PCB with a GW board and alarm following which all was working as it should except there was no movement from the temperature needle, hence the logical step to replace the engines temperature sender. Since then you get the LED self test as normal, but now the needle climbs steadily until (after circa 30 seconds from engine cold) it stops just at the overheat zone, the LED lights up and the alarm goes off (as they both should). Yes.

I would question if the new temperature sensor is faulty or if it’s the correct one for your engine. It’s not a grounding issue in my mind otherwise the needle would climb more quickly than 30secs and keep going past where yours stops. Have you put the old sender back in to see if it all returns to the original state with no LED and alarms (and no needle movement). Temperature senders are only a few dollars so I would be inclined to check if you have the correct or a faulty sender and order another before getting into the gauge internals.

[Achilles3588]

[Achilles3588]

It was good to step away from this for a while. Head clearing exercise.

So started by checking the coolant level sensor ground - no problem there.

Then checked continuity of the power side - the blue green wire from the coolant level sensor to the junction box in the Vanagon engine bay. Aha! Found and repaired a break in the wire not 4" from the sensor.

Problem solved!
_________________
Larry V.~
'91 Syncro Doka 1.9L TDI 'AFN'

[sagebus]

Really happy you solved that mystery Larry, good for you. Simple little things like this can be so frustrating for some folks like me.
_________________
1984 Vanagon GL, Wolfsburg Westy 1.9 ltr

[Achilles3588]

[SyncroHead]

I learned somethig new this week. Don't know if this has been written about before...

Crazyvanman has described the issue where if you have the issue with the coolant light blinking while the level is full and the coolant is, well, cool - then is may very well be a deteriorated component in the coolant gauge's blinking circuit.

What I learned is that upon starting a van that has this issue, if you turn the key to the ON position without starting and let it stay in this condition for a bit (I've done about 6-8 seconds) the blinking light will go OFF when you start craning the starter and stay OFF.

I've not done any research to determine why this happens, but I've now seen two vans where this is the case. So, for now at least, I'll just pretend my Syncro is a diesel and I'm waiting for the glow plugs to warm up.

Regards,
Jim Davis
_________________
"A Vanagon? It's not a car or a van. It's a hobby!"

Check out:
www.VANAVATION.com
www.SyncroSafari.org and
www.Vanagons.org