MG-Cars.info

Welcome to our Site for MG, Triumph and Austin-Healey Car Information.

Recommendations

Parts

MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.

MG MGB Technical - Low Fuel Warning LED

As I understand things, the fuel gauge works by passing current, from the 10volt stabilised source, through the gauge then onto the sender unit. The resistance of the sender unit varies depending on the position of the float, moving an arm across a resistance wire. The resistance varies from about 35 ohms when the tank is full to about 300 ohms when the tank is empty. Hence the current varies from about 0.3 amps (full) to about 0.03 amps (empty). The gauge displays the current, which we see as fuel contents. (My apologies to Paul Hunt for badly paraphrasing his description of the system in the pages of Bee and Vee.)

I would like to be able to fit a flashing LED into the fuel gauge to give a low fuel, say 1 gallon/4.5 litres, warning. This would function by the voltage on the slider of the tank sender unit being compared with a zener diode reference and then operating a relay to switch the flashing LED ‘On’. Can anyone tell me how the voltage on the slider of the tank sender unit varies with fuel contents? I think it should be possible to calculate this but one would need to know the internal resistance of the fuel gauge.

In the dim and distant past, I remember learning that ammeters had very low internal resistance, but that applied to moving iron or moving coil instruments not bi-metallic strip instruments, where the deflection of the needle is caused by heating the bi-metallic strip. Anybody know what the internal resistance of an MG fuel gauge might be? And does it vary with the current passing through it?

I think the concept is simple but there seem to be a lot of variables and I don’t know what causes them to change. Any help would be very welcome.

Peter
P L Hills

Peter,

very interesting idea and i think this could be ralized too, but isn't it also possible to use a reed switch inside the sender unit as it would be more easy to modify the sender and have relyable results even when there is something wrong with the resitor inside it. I had this fault a few times on my B's and would vote for an independant curcit for a fuel warning light.

Just an idea...

Ralph
Ralph

Hey PL,
Using your own figures, compute from E=IR
You get 1/4=67 ohms, 1/2=134 ohms, 3/4=200 ohms full=265 ohms

One gal computing from a 13 gal tank = 20.384 ohms.
Using these I get current = 0.0528 per galon

You can compute the voltage for one gallon from above:

Zener Voltage = current (0.0528) times the resistance (20.384 = 1.076275 or 1.08 volts per gallon

These are rough, but probably close enough. I have no idea where to get a 1 volt zener that would conduct at this range, therefore lighting a light in the circuit with a relay or ??????

Good problem, and the problem could be solved, but I just look at the gauge where I have loaded 1 gal of fuel when empty.
James Huggins

Peter - The easiest thing to do is to just measure the resistance of the fuel gauge. Just disconnect one lead from the gauge so you are not reading the resistance of the gauge in parallel with everything else. Cheers - Dave
David DuBois

Peter, try this

http://www.allaboutcircuits.com/vol_6/chpt_6/2.html

there are plenty of ccts available by searching for 'voltage comparator'. You don't need to know the relevant voltage levels as it is adjustable to your needs. Just whack a gallon in the tank and adjust the pot until the light comes on.

Also I think that the resistor in the sender unit is not linear in nature so it is probably not possible to use a simple system such as suggested by James.

Tony
Tony Oliver

Remember that the supply to the gauges pulses on and off, unless you have replaced the regulator with a solid state one. Not a big problem, just derive your reference voltage from the gauge supply.

Personally, I find this idea of a "fuel gauge" rather a novelty*. After 25 years of motorbikes, I just zero the trip when I fill it up, then after 200 miles I fill it up again. Obviously, if the gauge suddenly read less than 1/4, I would fill it up to account for any leaks/theft.


*the gauge on mine is rather a novelty, never going higher than 3/4, even with fuel in the rubber filler pipe.
Martin Layton

The resistance of the gauge must be added to the resistance of the sender at the appropriate fuel level, then you can calculate the voltage across, say, the sender. This voltage will go up as the fuel level drops, whereas the voltage across the gauge reduces, as does the overall current in the circuit. You can then design a Schmidt trigger circuit that will flip over at the appropriate voltage.

But in my experience even with Ohm's Law when it comes down to designing circuits there is nothing like the empirical approach - i.e. first measure the voltage in the real world, then build a circuit to trigger aat about that voltage, but with a trimmer pot for fine adjustment.

However with the factory stabiliser switching on and off about once per second in order to maintain an *average* voltage of about 10v to the gauge, the instantaneous voltage in the circuit is either going to be off, or varying between 12.8v and 14.5v according to the load on the alternator. The former feature is going to cause the Schmidt trigger to switch on and off with the stabiliser unless you include a timer circuit that, once triggered, keeps it on for longer than the stabiliser off period, although maybe the switching on and off can be classed as an additional 'feature'. But the circuit will still be affected by alternator load and come on sooner when there is little load and later when there is more load. A solid-state regulator would overcome both these problems, but it seems an awful lot of trouble to go to for a gauge that goes down relatively slow and controlled way anyway. Methinks you would add more value by designing audible and visual alarms for ignition warning light on with engine running, high coolant temperature and low coolant level (BT, DT with these), and low oil pressure and level.
Paul Hunt 2

Peter,
Ralph hit on the same solution that was used on our 1980 Toyota. Foolproof and worked for the life of the car (135K / 15 years).

Tom
Tom Sotomayor

This thread was discussed between 09/08/2007 and 10/08/2007

MG MGB Technical index

This thread is from the archives. Join the live MG MGB Technical BBS now