I have to debug an old-style fuel gauge problem this winter, it never moves. I will first check that the gauge is wired correctly, and has power. As I see it, the gauge is really an ammeter, the current depending on the resistance of the potentiometer in the sender unit. Before tackling the tank sender unit, what sort of current should I be able to measure passing through the gauge? If it is about right, then the gauge would be indicated to be faulty. But if the current flow is high or none, then any of the gauge, the sender, or the wiring, could be at fault. But knowing what sort of current should flow would be a help to start with. Any suggestions welcome.
They are volt meters, which measure the (tiny) voltage drop across a shunt resistance, and are calibrated to read that as amps.
So the tank unit is basically a variable resistor which alters the voltage to the gauge.
The gauge usually has 12v to it and is in series with the tank unit which is grounded at one end. If this is the case, ground the other end and the gauge should read full. It might well move slowly to maximum if a hot wire type.
It's far more likely to be a faulty tank unit than the gauge or the wiring.
So the potentiometer of the sender acts as a voltage divider, the gauge then reading the proportion of the available max. of 12 volts? That makes more sense than having a continuous variable current drain, which is what I thought, from looking at the schematic, although that may still be the case, but at a lower level than what I was thinking of.
So if grounding the 'leaving' end of the gauge produces no effect, the gauge is defective. If it gives a reading, then the fault is after the gauge.
Agreed. But I can easily get at the gauge, the sender requires more work! Thanks.
I have had a look at some old circuit diagrams. Many of them have two connections on the fuel gauge, one side at battery voltage via the ignition switch and the other side connected to the tank unit which goes to the chassis earth. This suggests that the tank unit is a variable resistor and perhaps a fixed resistor which together vary the current available to the meter. Unfortunately, none of the diagrams show the internal wiring of the "Tank Unit" sketch.
They are not all like that though. One circuit had a live feed an earth feed and a third connection going to the tank unit, so this looks to be a different arrangement electrically. Another circuit showed the live side of the fuel gauge fed via the ignition warning light which appears to adopt the role of dropper resistor.
It would help if you could tell us which car the gauge and tank unit are in. There might be someone reading this with personal knowledge of that vehicle.
I will go along with Dave - it is almost certainly the tank unit at fault. I can't remember ever seeing a sound tank unit and faulty gauge. Also don't forget that the tank unit is a glorified ball valve, and if there is a rust hole in the ball it will sink to the bottom regardless of the fuel level, so it might be electrically sound but mechanically faulty, as opposed to the probable electrical fault.
I would also advise against putting full battery voltage across the gauge. The tank unit typically gives some resistance regardless of whether it is supposed to indicate full or empty, and you could ruin the gauge by giving it full battery voltage.
My first port of call would be to put a resistance meter across the tank unit, and if it shows zero ohms it is suspect.
I have had a look at some old circuit diagrams. Many of them have two connections on the fuel gauge, one side at battery voltage via the ignition switch and the other side connected to the tank unit which goes to the chassis earth.
---------------------
There is actually a third connection on old gauges via the case gong to earth. These gauges were used in the days before voltage regulators were fitted to the instrument wiring. An unregulated voltage would cause the fuel gauge to rise and fall as the current through the circuit changed. To overcome this a "control" coil is connected from battery to earth (via the case) and the combined effect of the magnetic fields of the control coil and the coil connected to the tank sender controls the needle position.
As Jim says we need to know what age and type of car you are having the problem with.
It is an HRG 1500, which probably doesn't help much. I have the circuit diagram, which shows the wiring as described earlier, ie hot to one connection on the gauge, then from the other connection to the sender, and from there to earth. I am trying to do what I can with access only to the dashboard, for now. I will be removing the cover over the tank, and therefore getting to the sender, at some later stage anyway, but I am trying to check as much as possible before I get that far. If I identify that the gauge is faulty, then I can send it away for repair. If it is not faulty, then I won't need to do that. I can ask the Association what a working sender unit resistance range is likely to be, I'm attempting now to get a picture of what's what. I have never had to fix a faulty fuel gauge.
I've a feeling there were quite a few ways of doing it before hotwire gauges with voltage regulators arrived. Moving coil meters could have a second winding to compensate for varying battery voltage etc.
I'd agree about not putting full battery volts across the gauge for a long period. Just flicking it to ground would show if a moving coil type was ok. It's the later hotwire type which moves slowly.
The HRG 1500 had the Singer 12 mechanicals so it probably has the similar gauges etc. The Singer wiring diagram shows 2 connectors on the gauge, one to the ignition and one to the sender.
The obvious first test is to put a volt meter between the "battery" terminal on the gauge and the chassis, and you should see battery voltage indicated when the ignition is switched on. If not you have a wiring or fuse problem.
If you disconnect the wire from the sender from the gauge and connect an ohm meter to that ire with the other side of the meter to the chassis, you should see some resistance, not an open circuit and not a dead short.
If you have fuel in the tank, then rocking the car with the meter still connected should show a variation in the resistance.
If those tests succeed then there is probably a fault in the gauge. If you don't get the expeted readings from the sender then you have a wiring problem or a sender problem.
When you can get to the sender, try the ohm meter test again with the meter connected directly to the sender terminal. If it looks OK then the wire from the sender is faulty.
The HRG 1500 had the Singer 12 mechanicals so it probably has the similar gauges etc. The Singer wiring diagram shows 2 connectors on the gauge, one to the ignition and one to the sender.
The obvious first test is to put a volt meter between the "battery" terminal on the gauge and the chassis, and you should see battery voltage indicated when the ignition is switched on. If not you have a wiring or fuse problem.
If you disconnect the wire from the sender from the gauge and connect an ohm meter to that ire with the other side of the meter to the chassis, you should see some resistance, not an open circuit and not a dead short.
If you have fuel in the tank, then rocking the car with the meter still connected should show a variation in the resistance.
If those tests succeed then there is probably a fault in the gauge. If you don't get the expeted readings from the sender then you have a wiring problem or a sender problem.
When you can get to the sender, try the ohm meter test again with the meter connected directly to the sender terminal. If it looks OK then the wire from the sender is faulty.
Jim
Beware of one thing, somewhat counter intuitively, if you remove the wire that goes from the gauge to the tank the gauge should go to full.
There is all you want to know about these gauges at
formatting link
and following pages. It is actually about the MGA ones but I am pretty sure yours will be similar if not the same.
Yes, I have that pretty much sorted out. I was just hoping to find out what sort of resistance to expect from the sender unit, and identify whether the gauge was faulty or not before getting that deep.
By the time I get down to the sender, I will have no problem removing it if needed, then I will be able to exercise it at will. Having first covered the hole in the tank, to prevent unwanted fume ignition!
I just read the MG information, and then compared that with my Lucas wiring diagram. The MG notes state clearly that the gauge should have an earth connection, but my wiring diagram does not show that at all.
You really can't go by one maker's wiring diagrams and expect it to be correct for anothers. You may get more of a guide by the wiring colours. But on a car that old, I'm not sure how standard they were.
If both terminals on your gauge show resistance to chassis along the lines of the figures quoted for the MG, you might reasonably assume you have something similar.
Don't forget you don't always have to have an earth wire. The mounting clamp for the gauge might be doubling as a conductor.
which happens to be what I have for my non-functioning ammeter. It all looks good, the needle moves freely, it passes current, but there is no response. Weird.
The fuel gauge system is now operable, after buying a replacement sender, as the old one was dead, and supplying an earth connection to the gauge body. I am in the process of transferring the float and rod from the old unit to the new unit, not a straightforward process.
Perhaps I should have explained. An ammeter is simply a volt meter which measures the voltage drop across a series resistor - commonly known as a shunt.
MotorsForum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.