Talk Morgan Forums Morgan Sports Cars Morgan Aero 8, 2012< Plus 8, SuperSports, Coupe & Aeromax Fuel gauge wandering.

I just noticed this thread and I would like to second what Richard has said about earth connections.

When my car was new the fuel gauge was quite accurate. Over time it failed to register more than 3/4 full even when just topped up.
At a certain point I decided to change my dashboard and fit 3 separate instruments instead of the 3 in one original item. At that point I hoped that the new instrument would return to show full when the tank was full. After a lot of tribulation and discussions with Caerbont, the manufacturers of Smith's instruments, I was told that they had similar problems with London Taxis and the only solution that they had found was to run an earth wire from the sender on the tank direct to the battery.

I used an ohm meter to measure the resistance at the dash connection to the fuel gauge and filled the tank from empty (almost) 5 litres at a time whilst recording the resistance. In the end I discovered that there was no change for the last 15 litres added to the tank.

Fast forward to our last long tour away June. At a certain point, not long after filling the tank, I hit a large bump which shook the whole car. Much to my amazement I saw the fuel gauge started registering correctly and it has done so ever since.

I can only assume that the jolt of the bump moved some earth connection to the point at which the system was better earthed

I was told to be very careful not to over fill and stop at the first klick other wise fuel will enter the charcoal filter and bring on the Management light ,but it will clear itself when the fuel evaporates .

That makes perfect sense Peter. Given the fuel gauge is essentially a voltmeter working in a very low current circuit, a shared ground return will compromise voltage measured at gauge as total currents vary in the shared leg.

I was first made aware of this phenomena when fitting a smart external alternator regulator which piggy backed the internal one. The fitting instructions just said keep the two black ground wires separate but I later learnt one carried the rotor current and the other was part of the battery voltage sensing circuit carrying virtually no current.

A perhaps better explanation of this issue is that given a shared ground or indeed positive battery supply, the other circuit carrying a greater current will create a voltage change proportional to the current flowing through the shared cable as seen at the point where the two circuits split to individual. This change (lower if in positive leg, higher if in ground leg) will be very small and in the order of millivolts but the fuel gauge, as mentioned above essentially a voltmeter and a sensitive one at that, will reflect this voltage change in gauge needle movement. It follows that as the other circuit current varies so will the fuel gauge reading.

Note in such cases it's not the actual ground connection causing the issue but the shared circuit aspect.

This will demonstrate why I didn't do very well in my A-Level Physics exam all those years ago....

On a Multimeter, most have a screw adjustment to zero the Ohm scale. I've never figured how that works, but is it possible that a similar thing could be done to a fuel gauge, "backing-off" the Full reading, and making the needle swing over the full arc back to zero as the tank empties ?

2/10 see me