I have exactly the same issue.
I'll be honest, I've been lazy/procrastinating fixing it-- all I really need to know is when I'm close to out of fuel, so it isn't a big deal yet. But I did my research.
Mine will sit on 3/4 full for maybe 100-150 miles, then count down to empty from there.
The fuel sending unit is a variable resistor with a float, as the float inscribes an arc across the resistor, the resistance value changes, and the fuel gauge will read in a different position. Fancier models might have 2 floats, and in 1990 ford incorporated a module, called an anti-slosh module I think, that's there just to delay any changes (so your fuel gauge doesn't bounce around whenever you hit a bump and the fuel sloshes and the float goes up for a split second).
There are two possibilities: sending unit or fuel gauge in the instrument cluster.
Yeah, you could have a wiring issue, too, but a straightforward open circuit, or high resistance in the circuit, would not explain our symptoms. An open circuit would read either empty or full all the time (depending on how your sending unit works, if high resistance = empty tank, or full tank), and a high resistance circuit will read either emptier, or fuller, than normal, across the entire spectrum. A wiring issue between the sending unit and the gauge can not explain a tank that truncates at 3/4 full. It has to be teh sending unit, or the gauge.
With regards to the sending unit, one possibility is that the resistor is dead/open past 3/4 full. I'm not actually sure that this would make any sense.
I need to reread my electrical manual for my specific 1990 ford, but let's just make up an example, of a gauge where high resistance = full and low resistance = empty. If the resistor is open past 3/4 full, then it will read 100% full until, suddenly, it reads 3/4 full.
One can also imagine a scenario with a bad sending unit where there's a single dead spot, and the tank would read 100%, 90%, 80%, 70%, 100%, 50%, 40%, etc., where as the sending unit passes over that dead spot with infinite resistance it'll suddenly read full again.
Most common is just a sunken float, leading to an empty tank all the time. Obviously we don't have that problem either.
The only conceivable scenario is that the float, or the contact inside the variable resistor, is hanging up on some physical obstruction at 3/4 full and can't read higher.
Any problem in the sending unit would require replacing the sending unit, of course.
What I did was fill up my tank, crawl under the car and unplug my sending unit and put an ohmeter across the ends. I got a good value for full, according to my ford electrical manual, although resistance was a little high.
I then checked at 3/4 tank, when it started to go down,
and then I checked when it was approaching empty.
All values were a little high, corresponding to some higher resistance from age/corrosion/what not, but, they were about in the right spot. It wasn't like I saw 150 ohms at full and and at 3/4 tank both, I got an even spread. It can't be my sending unit.
If you DO see the same resistance value at full tank and at 3/4 tank, where you think 3/4 ought to be, then you've got a sending unit problem.
Furthermore, my gauge will seem to default to 3/4 full when turned off. I understand that this isn't right; others' go to empty when the car is turned off and the key pulled out.
By the way, my gauge, due to the higher internal resistance, always used to read 1/4 full for empty, and 5/4 full (overfull) for full. If you're wondering how I can get 150 miles on a quarter tank, I can't, that's a half a tank.
So I'm sure it's my gauge.
I've got a junkyard instrument cluster that I'm going to try, and I just haven't gotten around to installing it.
Last edited by BerniniCaCO3; 02-26-2012 at 11:37 PM.