Crazy fuel gauge

Technical question, vehicle is 1996 D90 300Tdi. The signal from the fuel gauge sender is obviously variable with the fuel height, and goes up and down according to how the fuel sloshes about in the tank. The fuel gauge, however, should show a steady reading. The signal is somehow damped. Is this in the sender or in the gauge? Asking because I have recently fitted a new fuel sender which worked perfectly for 3 months. Yesterday I let the tank get very low (right on the lower edge of the red sector) and then brimmed it. As I drove away from the petrol station, the needle was waving about like a drunken sailor between about half and full. Today it is the same. It will stay steady on F for a while (done under 20 miles on full tank) and then drop to 1/2 and start waving around again. Fault in sender or gauge? Possible link with letting tank go empty and then filling to brim? Any help appreciated.

Mine had a problem where if it was over half full it showed empty with warning light on. As soon as it got back to half it worked again. Bought a new pickup pipe/sender and it fixed the problem. I think the 'clever' bit is at the tank end.

Thanks for that. Today, it is behaving completely normally. Yesterday, it was flickering about like the old Morris Minor ones used to, and generally waving between 1/2 and 3/4. Today, it is where I expect it, on the F mark and moving only slightly depending on whether I am going up or down hill. That's exactly what I would expect. I'll keep an eye on it, but it might just be 'one of those things'.

BlackDogZulu said:

The signal is somehow damped. Is this in the sender or in the gauge?

Click to expand...

The LR gauges, and indeed almost all of them (before ECU's got involved) were always a bimetallic strip which worked with heat - basically when the sender float is at the top the resistance is very low, lots of current flows and the strip heats up, and bends moving the gauge to full - as the fuel drops the resistance increases thus less current therefore less heating and the strip bends less and the needle position reflects this. That is why when you have a full tank of fuel and you switch on the ignition the needle doesn't just fly to the full mark instantly instead slowly sweeping across. This is what create the hysteresis of the gauge, i.e. what stops the needle flying about. So in answer to your question it is the gauge that "dampens" the needle as a direct function of how it works.

If your's is moving about then that is an indication of a bad earth or wiring fault which is messing up the current flow giving muddled readings or the gauge has a bad connection to the the strip and or the strip has finally started to yield making and breaking the circuit.

Out of interest, mechanical voltmeters use magnets and a wire coil which is why a volt meter given 12V will fly up to 12V immediately as the coil creates magnetism and moves the needle. It would not be much good in most instances to have a voltmeter that lags behind.

Huddy84 said:

Mine had a problem where if it was over half full it showed empty with warning light on. As soon as it got back to half it worked again. Bought a new pickup pipe/sender and it fixed the problem. I think the 'clever' bit is at the tank end.

Click to expand...

Nothing clever in the tank at all, the sender is is a rheostat that is actuated by a float.

Your issue will have been worn away contacts in the "top" section which would have shown very high resistance or open circuit, i.e. made the gauge think there was no fuel as electrically high resistance or open circuit would send no or very little power to the gauge which would show low fuel and set the warning light on.

discomania said:

The LR gauges, and indeed almost all of them (before ECU's got involved) were always a bimetallic strip

Click to expand...

discomania said:

Nothing clever in the tank at all, the sender is is a rheostat that is actuated by a float.

Click to expand...

This makes sense to me. When I fitted the new sender, before I put it in the tank I experimented by moving it through its range to see if I had cured the fault (gauge wouldn't read below 1/2, meaning I ran out of diesel on my first trip). I noted that the gauge worked over the full range and responded to movements of the sender rheostat, but not rapidly. I had forgotten about the bimetallic strips (last seen in 5th Form Physics). So the 'damping' if you can call it that is in the gauge, in the nature of how it works. Thanks for that.

It was very flickery for a while after filling up but now, for the second day, it is behaving normally. Perhaps it was the Land Rover Spider working out behind the instruments.

So, I have filled it up twice since the last post. The needle flickers drunkenly for the first 20-30 miles and then seems to settle down. From 3/4 tank downwards, it's as steady as a rock. What can it be about a full tank that make the needle waver about? I am mystified.

BlackDogZulu said:

So, I have filled it up twice since the last post. The needle flickers drunkenly for the first 20-30 miles and then seems to settle down. From 3/4 tank downwards, it's as steady as a rock. What can it be about a full tank that make the needle waver about? I am mystified.

Click to expand...

How fast does the needle move when it is flickering?

There could be an issue with the contacts in the "full" position and it may be that it is not making good contact, so when the circuit is broken the needle will start to fall back towards empty until it makes contact again. As I said in my post above it works with the heating of a bimetallic strip, so if it looses power it will fall until it gets power again, it should cope with the normal sloshing about of a fuel tank as the hysteresis effect of the bimetallic is pretty good at giving the impression all is dead level and stable so if it cannot manage at full tank I suggest there is problem with the sender at full.

Other things to consider:
Gauge fault - at full it is in the highest current - highest heat so an electrical fault here is possible.
Wiring issue, but almost certainly not or else it would show at any fuel level unless it is a dry connection that is struggling with the slightly higher current of "full tank".

discomania said:

How fast does the needle move when it is flickering?

Click to expand...

Thank you for the reply. The flickering is quite fast - not quite 1950 Morris Minor fast, but much faster than the normal damped movement. When I tested the movement before I installed the sender the movement was quite stately, perhaps 3-4 seconds from full to empty, no matter how quickly I moved the float and lever. This is a lot faster than that, which makes me think of a loose connection at the gauge end. I have just dug up an old fuel gauge from my original Ninety, so I might substitute that and see how it goes. Cheers.

BlackDogZulu said:

Thank you for the reply. The flickering is quite fast - not quite 1950 Morris Minor fast, but much faster than the normal damped movement. When I tested the movement before I installed the sender the movement was quite stately, perhaps 3-4 seconds from full to empty, no matter how quickly I moved the float and lever. This is a lot faster than that, which makes me think of a loose connection at the gauge end. I have just dug up an old fuel gauge from my original Ninety, so I might substitute that and see how it goes. Cheers.

Click to expand...

If it is fast then I would suggest it does sound like the gauge, failure of the bimetallic strip or thin feed wire could cause an electrical issue that would show as a flicker or odd reading.

Bypass the fuel sender, link the two wires out or deck the gauge feed wire (green and something I think off the top of my head) to the body and the gauge should go to full, see if the gauge flickers about now.

I'm suspecting something loose in the gauge, and a substitution might sort it out. However, I brimmed it yesterday and the gauge is behaving totally normally! Sods Law.

Cheers