I have a very difficult fuel gauge problem. I have had the float level out three times and have checked it out thoroughly. Thus how can I narrow down the problem? What happens is that when the fuel gets to approximately 1/2 full, then the indication seems to wander and will sometime land on 1/2 full, but then while driving it will go down to 1/4 full or even lower. It has me stumped!!!
I wonder if someone can give me the specifications of sending unit resistance.
Need full, half, and 1/4 ohm resistance specs if possible. Want to substitute a known resistance for the sending unit to determine once and for all of where the problem lies. Lee
Hi Lee,
A very plausible cause would be that the float is interfering with the tank walls, or one of the tubes inside, and gets momentarily stuck, then when driving the vibrations set it free.
Other possibilities are a faulty intermittent sender, bad connections at the selector switch, gauge or sender.
Take the sender out, raise and lower the floater very slowly and check if the gauge responds accurately from empty to full.
You can measure the resistance values at the sender of the other side tank, I assume it works ok?
2 Likes
As Aristides imply. Lee - is it the same for both tanks? His points are well taken - and with only one tank involved; the solution must be found there…
If the problem is discontinuity in the tank unit’s resistor; it can only be checked by removing the tank unit and test as per Aristides - spec resistance values are immaterial.
You can separately check the gauge by connecting a 300 ohm variable resistor between gauge wire and ground. And measure the relevant resistance with gauge readings by then measuring the resistor’s resistance. The gauge should of course read (over)full with the wire grounded…
Frank
xj6 85 Sov Europe (UK/NZ)
1 Like
Yes I can do as you suggest, but was wondering what Jaguar specs were, if any, for the resistance of the sending unit. The reason I say this is I am looking at any and every aspect because of the weirdness of the problem.
One other aspect or question. The existing fuel gauge is mounted such as the original would be, however it looks different than the other gauges!! Wondering if this is the original gauge or not? And if not, someone else must have had
a problem with the fuel gauge and replaced it. If a different type of gauge and movement, could this be a possible problem? You see, I am trying to hone in on every possibility. In troubleshooting, it takes some time to fill the tank
and use up half of it etc and monitor the fuel gauge. Thus taking longer than usual to troubleshoot. As stated one would immediately suspect the sending unit. But I have had it out more than once and have checked it out every way
possible and gave it a full okay each time. Seems to operate fairly well until the gauge gets to about the 1/2 level and then it starts to malfunction. Hair raising. Wish I had another sending unit. Would like to try that even though the
existing one, checks out perfect.
Lee
p.s. I have a resistance decade box and plan on substituting it for the sending unit. Will adjust it for 1/2 & 3/8 tank level and see if the gauge is stable at that point. If not, I have a bad wire somewhere in between.
You have two tanks and senders. Test if it is limited to one sender. If so, compare senders, compare wiring, find the issue.
That’s another approach of course. But I hate to get more other stuff involved. Thus will concentrate on the one side at a time.
Lee
You’re suspecting the gauge when you have two senders to compare, so just saying you might start with identifying what is playing up 
I can suspect things at this point, however I do not want to get yet another side involved. I have enough trouble analyzing what is wrong with the one side. Hope you understand. I may have another issue entirely with the other side and do not want to get involved at this point. The auto has been sitting, not being driven for a couple of years, thus could have more than one problem.
Lee
I understand. Chances are that the other sender works fine, or at least doesn’t have the same failure - usually the floats fill or they seize.
Simulating the resistance at the sender is a good approach as well so good luck.
If you don’t have the same problem on both tanks then logic dictates that the gauge itself is not the issue.
Same logic dictates that if the problem is at a certain fuel level the sender is suspect.
The gauge is magnetic, no contact/friction parts involved inside so usually it either works or not at all.
How did you test, Lee…?
Connecting the gauge to ground through the tank unit for testing; a well working unit would gradually increase gauge reading - from ‘empty’ to ‘full’. I would regard this as proof the unit (and gauge) worked perfectly - and any failure of the gauge to ‘follow’ the tank unit perfectly would be a tank unit fault…?
To eliminate, or confirm, a gauge fault; your use of assorted ‘fixed’ resistors (instead of one variable) is fair enough - and will also indicate the gauge’s requirements. This may be relevant; your ‘different’ gage may not comply with the original specs. But if you can get the gauge to read any level by gradually ‘stepping’ resistance meaningfully - the gauge is OK, but may not match the resistances given by the tank unit?
A complicating moment may be that the tank unit resistance may not be linear - to reflect that, due to tank shape, fuel level does not in itself give fuel volume. (Though I ‘think’ gauge gradation is meant to take care of that)
You have of course used an ohmmeter on the tank unit, ideally an analogue meter, to verify its perfect output - ie, smooth changes between min and max?
David has a point, but you can just test your tank unit on the other tank unit’s circuitry - only requiring removal of the other rear light cluster? Ideally, of course, both units should be tested on the other’s circuit…
The main point is that in a confusing situation; the more data the better…
Frank
xj6 85 Sov Europe (UK/NZ)
IIRC, it is not, the pitch of the winding varies.
It could be bent or dirty so that there’s more friction at one point. The sender wire is a 3d spiral, either it’s smooth throughout the range, or it isn’t.
A valid point David, but still a rare occasion.