w123 flickering gauges

[Hibbo]

Hi y'all, I have posted a couple of times on this forum but am still pretty much brand new to it, so I'd like to say hi; Hi. I have had a good read through and people obviously know what they're talking about, so please allow me to share my woes.
After buying a 190d as a cheap diesel runabout earlier in the year and being well impressed with it, I have this morining purchased my very own 'Cyprus taxi' - a 1983 300d. The bodywork is excellent and apart from the usual steering box play and a few squeaks here and there I'm made up with it. However - The fuel and temp gauges like to flicker and dance about, above their settled positions. Not knowing how much fuel you have is a bit of an inconvienience, and the temp gauge bouncing to the top is not reasuring.
The gauges are fine when the engine if off, and tend to quickly settle down when the engine is idling in N. As soon as you put it into gear they're off again.
I have made the following assumptions, and am hoping someone can tell me if I'm on the right track or talking pish.
I assume these gauges are fed +12v and then earthed through their respective sensor? If this is the case a breakdown in insulation due to chafing on bare metal could cause extra earthing thus flicking the pointers up? I am thinking that the movement of the engine when drive is taken up is pushing the wires against some bare metal or an earth wire somewhere.

Does the above make sense? If so can anyone point me to where these particular wires pass or tell me to get a grip!

Sorry to waffle on so much!

On a different note the sunroof doesn't work, I can hear the motor running, but the mechanism is knacked in some way. Is this quite difficult to get at? Does the headlining have to come off?

If you've made it this far, thanks very much.

Hibbo.

PS. What temp should this engine run at? The 190d (602 engine) sat at a shade over 80c, but this one is over 90c.

 

[TimN]

It may be just a lose or dirty connection on the back of the instrument cluster. Remove the large connector and refit. Also check the fuses to ensure electrical continuity.

The circuit board in the instrument cluster can suffer from corrosion. Take it out and inspect it. solder bypass wires if necessary.

If your up to it I can send you my notes on refurbishment of the steering box. Just send your email address to my private message box.

 

[television]

Loose connection, corroded connection, I would do as Tim says clean all the fuse holders. If they are all jumping around then the fault will be on the earthy end, that is the common connection in the return path, though saying this I see that the fault is OK at idle. I cant remember off the top of my head if the 190 has a OVR if so it is the most likely cause. Any earth cables on the engine should be checked,
This is where a moving coil volt meter comes in very useful, you can see the fluctuations on the 12v lines, a digital meter has a gate, but the gate times are too slow to see spikes or the like.

Malcolm

 

[Hibbo]

A story of pointless electrical tail chasing!

Thanks for the replies, I have noticed that the gauge do their flickering when I bounce the front of the car, but not when I rock it (from the side). I think the common earth thing may be a red herring, and both senders are knacked. This does sound unlikely, but I'm pretty sure the temp sensor is goosed as I connected a variable resistor between the temp sensor cable and earth, adjusted it to give a mid scale reading, bounced the car and the gauge is steady (as far as my Dad was telling me). Not absolutely sure, as it was difficult to hold bits of wire on an old volume control, on to neg terminal and on to the sensor wire, then bounce the car. I've ordered a new temp sensor so will see what that does.

I did have an AVO on the 12v system, and it is steady. The reason I had the meter out in the first place was due to my own buffoonery, and I really did end up chasing my tail; allow me to explain (standby for extreme boring waffling):
Whilst trying to free the adjustment bolt on the steering box, the grolly bar I was using hit the back of the alternator and proceeded to attempt to weld itself to it (the engine wasn't running but the battery was connected - yes, i know!).
It was only a quick short but I wanted to check that the alternator and electrical system wasn't damaged, so I started the engine, disconnected the battery and checked the system voltage: 6v. Reconnect the battery and it went back up to 13ish.
Revving the engine (batt disconnected) made no difference, then after a couple of minutes it would return to 13.8v. I repeated this a few times and it was always the same. I was now thinking maybe the rectifier pack was knackered, or the alternator was taking ages to fully energise.
I then noticed a chattering relay sound when the batt was disconnected, but stopped when the batt was in again. I also stopped when the voltage (alternator only) returned to 13v. I was now really confused. More checking revealed that it was the glowplug relay that was chattering, and the glowplugs were staying on after the engine had started.
Then I finally realised what was going on; (For whatever reason - I've since read about afterglow mods, or maybe just a loose wire - I'll check) The glowplugs were staying on with the engine started, with the batt disconnected the poor wee alternator couldn't privide enough current to power them, and thus the glowplugs pulled the system voltage right down. With the battery back in the system the voltage returned as the current demand could now be met. The relay was chattering as 6v obviously isn't enough to energise its coil.
So the conclusion is, my worrying about any damage I'd done with the grolly bar has caused me to overload my alternator and do god knows what to the glowplug relay contacts! There's a moral there somewhere! Everything seems fine though, so I think I'll just forget about it. (I'll not mention pulling my hair out and hoping beyon hope that there was a main fuse somewhere after all the electrics died completely. I then spotted that the earth strap had came off its chassis clamp!)

I didn't manage to free the steering box adjuster, just locked it back up. I'll have another go when I get the car up here next weekend (left it at my Dad's)
I'd really appreciate a copy of your steering box gen TimN, but I can't send PMs (because I'm new to the board?), I don't mind posting my email address on here, or is that not the done thing?

 

[stever]

With regard to the steering box, I got my bolt free using a 19mm deep socket connected to a breaker bar which I then persuaded with a lump hammer. Soaked the allen bolt with WD40, then held the 19mm with a ring spanner and freed the allen bolt with a long allen key. Do you know that to tighten it you turn anti-clockwise?

 

[TimN]

Yes. What you are doing is raising the pitman shaft further into the steering box thus engaging the gears further. remember to adjust it with the steering facing forward and leave 10mm of slack at the steering wheel edge or you could split the box.

 

[Hibbo]

A partial success

:???:
Thanks for the replies.

I had another bash at the car today, the first thing I did was to replace the temp sensor, problem still exactly the the same. I then found a connection in the front n/s of the engine bay (right by the glowplug relay) where a temp sensor connector connects to another wire that then goes to the temp sensor (bound in with the glowplug harness). I'm guessing this is a throwback to when/if the temp sensor was mounted there and just measured engine bay air temp. Anyway, this wasn't the problem.

I then (finally) removed the instrument cluster (yes yes, as I was adivsed to do ages ago), this was MUCH simpler than I thought, dinna ken why I avoided it so much. When I had a look at the back of the cluster, there was a link between two pins which look dodgy, and sure enough it fell of (almost said 'came off' ) as soon as I touched it. I resoldered this.
I then took the gauges off the back, and found a copper track peeling away on the triple gauge board, so I traced it around and soldered a link from two convienient spots. I also cleaned the board up a bit, lots of corroded old flux between solder joints etc. There was a fair bit of oil splattered around, but nowt to get worried about.

I refitted it, and - spot on, the gauges where completely steady. However, the fuel gauge was completely steady on its (lower) endstop, no movement at all. I took the cluster off again and checked again inside, the fuel gauge to move ok when power is applied to it (the AVO was enough to make it move). I think the fuel sender is knackered but couldn't be arsed removing the boot lining so called it a day.
I don't think the fuel gauge was working in the first place, it was just the erratic flickering that made it appear to be giving a reading.

I took some photeys of all this pish, I might upload them if anyone particularly cares, but I know photos can help alot.

If anybody is still reading this, good god, well done! (now get a life! )

 

[turnipsock]

You have been watching to much Still Game.

The fuel gauge thing and the temp gauge are two different circuits. The low fuel light is also not connected to the fuel gauge, so even if the gauge is faulty, the light will still work as it should.

The fuel tank sender is a wiper arrangement. I had seen threads about using optimax or ultimate cleaning up the sender unit. I didn't think it would have much effect as I thought these fuels were only going to have an effect when they are burning. But...after about two months, my fuel gauge started working properly.

The temp sensor is resistance thing. You can use an avo to simuate a gauge. 12V-AVO-sensor-ground.

 

[Hibbo]

Aye, ken ken.

I'm actually English, but have lived in Dundee for 6 years so have picked up some of the lingo eh? (eh'll hay a peh, eh?)

I'm aware that the temp, fuel, and low fuel are all seperate circuits, I thought that the fault was common (as it was) as both gauges were bouncing in unison. It now transpires that there were two isolated problems; 1) bouncing gauges, and 2) fuel gauge not working. As I mentioned, the filckering made it seem as though the fuel gauge was giving a reading Now that the flickering is sorted, it is clear that the fuel gauge it knacked.
I don't think the wiper is stuck, as there is no movement of the needle whatsoever.
From what I could gather prodding aboot with a (really old and crap) avo, putting power in on pin1 (of the cluster) and grounding pin2 (might of been 3) made the fuel gauge move. I thus deduced that pin2 (or 3! - i was on the correct one!) connects to the sender. I then checked the resistance of pin2 (or 3!!) to earth on the free plug - about 40ohms, and kept an eye on it as I poured in a couple of gallons of fuel, it didn't really change, but this was with an appalling avo, so is completely fallible. I may well be barking up completely the wrong metaphorical tree with my pinouts etc, and what I was seeing/reading was due to parallel paths and such like, I just dinna ken!

It should all become clear when I get to the plug on the sender unit (the carpet is rivetted in place over it, I just couldn't be arsed doing it today). Simply be a case of grounding the gauge and low level wires and checking for FSD and illumination respectively. Then resistance checking through the sender itself.

All will be revealed!


PS. A bloke walks in to a bakers in Glasgow and says 'excuse me, is that a cake or a meringue?' to which the baker replies "aye, you're no wrang, it's a cake!"

 

[turnipsock]

I seem to remember that 40 ohms was the max resistance. It drops down to about 2 ohms at the other end.

The problem must be the sender unit.

Does the low fuel warning light come on when the fuel is low?