Newbie instrument cluster woes

[BarnfindDan]

Hi there

I am in the process of restoring a dormant 323i coupe which needs a lot of TLC. I have attacked the oily bits and now have an acceptable idle which is half the battle.

Current issue is that the instrument cluster is generally unresponsive - all dials are not doing what they should. I have dash lights and 12v to the appropriate blue and white plug pins (following the Bentley manual) but the rev, temp and petrol gauges just don’t move. Car has a new fuel tank and sender, new thermostat and I have pulled the instruments apart to tighten the ground nuts.

Previous S.I. board NiCads leaked pretty bad so I replaced the board with a second hand unit and reinstalled NiMh batteries remotely so now the S.I. lights work well. Could this be the issue as the new batteries are not up to full charge yet? The instrument PCB looks good and I have checked for continuity between the board and instrument pins and they all check out.

I am not sure what more I can do! Any suggestions are welcome. Thanks.

[Steve]

Hi and welcome to the E30 Zone.

Have you checked out our Wiki? https://www.e30zone.net/e30wiki/index.p ... n_Problems

[BarnfindDan]

Yes - checked that. That’s what lead to new sender and bits above. To be honest I have only tightened the brass nuts this evening and have not had a chance to check if that solves outer gauge issues but will give it a test tomorrow.

I’m more concerned as to why none of the gauges work while the lights do and the correct voltages are being supplied.

I did swap the cluster for a well used unit to see if the issue persisted, and again no rev counter or temp action, although the fuel gauge went straight to half full. I assume this is because the donor cluster was set for a 65l tank and not the smaller tank I have installed.

[BarnfindDan]

Thanks for quick reply Steve, really appreciate any pointers at the moment as I am contemplating re-soldering all joints which will be a total ball ache.

[Brianmoooore]

You say you've replaced the SI board and its batteries. In my experience, the type of SI board which uses 1.2 volt rechargeable batteries are now very thin on the ground. Most now are the later type, that uses 3 volt non rechargeable batteries. Are you sure you haven't fitted 1.2 volt batteries to one of these boards?

[BarnfindDan]

Yes, very sure. The batteries that came off both original and second hand boards were the red Varta type. I had them matched for modern NiMh batts 700MAh as per the forums.

[BarnfindDan]

So could the batteries be the issue? I assume not as the dials get their supply independently from the blue and white plugs no?

Worth noting that the gears all look good inside the instruments too. The car has only covered 55K kilometres.

[BarnfindDan]

Temp gauge does flicker and raise very slightly (still below 0 or blue) when ignition is turned on if that helps

[flybynite]

I had them matched for modern NiMh batts 700MAh as per the forums.

The ones in the photograph are NiCd not NiMh. Did you find any NiMh?

I haven't found NiMh AA that are 700 mAh only AAA.

[BarnfindDan]

Well spotted flybynite- I took them to the electronics quarter of town and they matched them as best they could. The batteries work the S.I. board well but not the gauges.

I remote mounted the batteries in a dedicated box under the dash in case they leak again.

If you want NiMh 700’s I can look about for you - there is a really good source of electronic bits here in Asia.

If you remote mount the AAA’s it shouldn’t make a difference what size they are.

[Brianmoooore]

Original batteries were NiCd - NiMH didn't exist back then, however NiMH is a direct replacement.
The effect the batteries can have on the cluster is that when they fail, they tend to short circuit, pulling down the voltage that the cluster electronics runs on.

[Contours]

When my s.i batteries failed some years ago, this did not affect the instruments. You might have to look at connections to the sender units, faulty stuck gauges, faulty sender units and/or maybe poor earths.

[flybynite]

Is there any reason a larger AA battery would not work, say 800 mAh or 1000 mAh ?

[Brianmoooore]

Capacity of the batteries (within reason) isn't important or relevant.
Check the voltage across each of your replacement batteries to check they are charged. They should really be charged before fitting, if they're not supplied fully charged.
Temp. gauge and tacho. are often affected by battery issues, but not the fuel gauge. Switch the ignition on and short pin 4 (brown/green wire) of the white instrument cluster plug to earth. This should make the fuel gauge go to full.
You can try the same thing with pin 26 of the blue plug for the temp gauge.

[BarnfindDan]

Is there any reason a larger AA battery would not work, say 800 mAh or 1000 mAh ?

Apparently the higher the mAh, the longer it would take to charge, if the charging apparatus of the dashboard could charge them at all. Best to keep as close to spec as possible.

[BarnfindDan]

Capacity of the batteries (within reason) isn't important or relevant.
Check the voltage across each of your replacement batteries to check they are charged. They should really be charged before fitting, if they're not supplied fully charged.
Temp. gauge and tacho. are often affected by battery issues, but not the fuel gauge. Switch the ignition on and short pin 4 (brown/green wire) of the white instrument cluster plug to earth. This should make the fuel gauge go to full.
You can try the same thing with pin 26 of the blue plug for the temp gauge.

Very handy information, will give that a shot as I simply cannot see anything else likely to cause fault.

[BarnfindDan]

So resoldered the grounding nuts on the back of the instruments and after reinserting and smacking the binnacle a few times I now have a fuel level showing - still no temp though which is worrying. I’ll look at testing the temp sensor next before checking for breaks in the wiring. This is turning into a massive ball ache.

[Brianmoooore]

No connection to the earth nut makes the gauges read slightly high rather than not read at all, so I don't think they are the cause of your problem.
There's a soldered joint near the top of your pic. above (just below the lamp holder position) that looks 'dry'. Reflow that with a little new solder, and any others that look dodgy. Pay special attention to the soldered connection for the blue and white plugs. They're not a weak point, but sometimes they get rough treatment from someone who can't work out how to release the plugs.

[BarnfindDan]

Thanks for that B’mooore, I think I’ll have to reflow the board.

Just shorted out the brown plug which had no effect on temp gauge and now I have located the C101 plug (it’s the square plug round the side of the fuse box) I’m going to try grounding pin 4. I’ll post results when done.

[Ian325i]

Did you ever get the gauges working.......! let us have the outcome.

[BarnfindDan]

Hi Ian - sorry for being quiet for so long, I had been waiting for a secondhand cluster to arrive from Denmark, it took a month to get here to Kuala Lumpur!

Popped the shipped cluster into the car and the fuel gauge fired up directly which is good. Temp gauge still flatlining - so I turned my attention to the sensor. Having a M20B23, my temp sensor has two pins which I had previously shorted to help diagnose the issue (previous shorting had no effect leading me to believe it was the cluster at fault).

Out of desperation I chopped the two connectors from the harness and replaced with shiny new ones and guess what - the sodding temp gauge flicked to full hot. This means that the cluster is ok but the connectors were corroded.

But still, when the new connectors were attached to the temp sensor there was still no discernible reading when the engine was hot. I pulled the sensor and stared at a short, stubby sensor which has no resemblance to the 3-4cm long sensor usually situated at the brown plug. My car has a strange, ineffective temp sensor which is the actual culprit.

Now currently looking for a two pin temp sensor for the car. RealOEM is showing part numbers 62110788115 and 62111268355, both of which have only one pin. I actually have one of those here with me (an OEM Cambiare VE718002) which is no use to me. Current thinking is that I look for a sensor similar to the E21 6 cylinder.

If anyone here can throw the appropriate part number at me I would be eternally grateful - without this sensor I am risking cracking the cylinder head as the engine feels rather hot when running but I can’t be sure.

[Brianmoooore]

AFAIK, the one pin sensor is a direct replacement for the two pin. The one pin version earths through the engine metalwork rather than a wire.

[BarnfindDan]

I thought as much, but I really wanted to keep the car as factory spec as possible. As I said before it has only 55k on the clock (although I now have three clocks so it could read anywhere up to 256k depending on which I put in...) and sports the original interior and carpets. Last owner even left a shiny new air slide valve in the footwell which I understand are rarer than rocking horse excrement. Good job too at the valve in the engine was corroded beyond salvation.

Is there no OEM part that will work?

[Brianmoooore]

IMHO, the single pin sensor is still ' factory spec.' Parts design evolves over the years, and this is simply an example of that. Brake pad and tyres are two examples. The original brake pads would contain asbestos, but the ones on your car now don't, and you'd find the handling and grip considerably poorer on the tyres it was supplied with.

[BarnfindDan]

Fair ‘nuff. I’ll bolt the one I have in and see if that solves the problem.

[BarnfindDan]

Yep, new sensor with one pin works, although still not convinced the reading is right.

Ran the engine (stationary) for 10 minutes and needle shifted slowly to just short of half way. While this is encouraging, the top rad pipe was fairly sizzling and I could see some feint smoke coming from the front of the block. This could be burning surface oil/grime but still it is concerning.

I’ll re-bleed the coolant system tomorrow and have another go, but before I do that just thought I’d check to see if the now redundant cable on the loom (the one that is supposed to plug into the now deleted second pin of the temperature sensor) needs to be grounded? Or should I just leave it flapping about and insulated?

[BarnfindDan]

It’s all getting a little silly now...

Original VDO pre-facelift with modified S.I. board and non functioning petrol gauge, a post-facelift VDO board with US style fuel gauge and broken odometer gears and a pre-facelift Moto-Meter for a 4 cylinder with a clock instead of a rev counter.

[Brianmoooore]

Yep, new sensor with one pin works, although still not convinced the reading is right.

Ran the engine (stationary) for 10 minutes and needle shifted slowly to just short of half way. While this is encouraging, the top rad pipe was fairly sizzling and I could see some feint smoke coming from the front of the block. This could be burning surface oil/grime but still it is concerning.

Doesn't sound too far out to me. If the engine's been running for a while with the car not moving, the 'stat will probably be wide open, and the temperature now being controlled by the bi-metal strip in the middle of the viscous coupling for the fan. Drive it, and it'll probably come down a bit.

I’ll re-bleed the coolant system tomorrow and have another go, but before I do that just thought I’d check to see if the now redundant cable on the loom (the one that is supposed to plug into the now deleted second pin of the temperature sensor) needs to be grounded? Or should I just leave it flapping about and insulated?

The spare wire already is a 'ground', and can be left connected to the plug. The wire is still present (but totally unused) on all M20s, right up to the end of production, since it seems, no one told whoever made the injector looms for BMW that the sensor had been changed to single pin.
The wire was deleted from the rest of the engine loom, but was still fitted to the body loom of some E30s, and went to pin 20 of the C101 engine loom socket. Unfortunately, from about 1989 onwards, pin 20 was reassigned to feed 12 volts power to the ABS ECU, so if a late engine loom was plugged into an E30 with the ground wire still in place in the body loom, the result is clouds of acrid smoke and ruined looms, as some on here have found to their cost.