Tripminder Calculations

[Lutrova]

In other news, I've completed the shop manual diagnostic test for erratic fuel level, and while I know more than when I started, I still can't say I've found a smoking gun.

For reference, this is the diagnostic test:



So in order, this is what I've found so far.

B1 - Trunk Ground: The ground in the trunk is secure, with 0 ohms between the ground connector at the level sender and sheet metal. Sender connector terminals are free of corrosion.

B2, B3 - Resistor Wire Bypass: Jumping the sender connector with 10 and 72 ohm wires both displayed empty readings at the fuel gauge, which suggests the fault is not with the level sender.

B4, B5 - Fuel Wire: From the instrument cluster connector, fuel wire reads open (>20k ohms) when disconnected at level sender and 0.4 ohms (<3 ohms) when jumped at sender connector. The fuel wire and the intermediate connection don't appear to be bad.

B6 - Gauge Voltage: Fuel gauge has 12.12v (>10v) between ignition and ground terminals, which is a pass.

Having arrived at the end of the test, the manual suggests the fuel gauge should be replaced. Or the 'fuel gauge ground circuit A7 bus bar' could need servicing, whatever that is. I guess I didn't really investigate step B0, as the level reading varies regardless of other electrical loads in the car. Just swapping out the fuel gauge is a bit of a bummer, since these things are so hard to come by and I seem to have two that look to be in great shape but are apparently bad.

[gadget73]

I think the buss bar is the big multi-tab thing under the dash near the gas pedal where a bunch of things plug in.

[Lutrova]

The vertical metal piece that a couple of grounds and the message center logic module bolt to? I think it might also be a support piece for the whole instrument panel? If it were to need service, the way to test it would be to touch a multimeter at one end and the other at some known good ground and check for no resistance?

[gadget73]

I believe thats the one yes. Only thing I would really imagine needing to do is unplug and re-plug each of those grounds. If something were crusty you'd find it with just a visual inspection of each connection. If nothing is really bad, just re-seating everything ought to be enough to sort out any questionable connections.

[Lutrova]

Yeah, nothing so far has looked crusty at any point in the system. Connectors, grounds, circuit boards, all clean as a whistle. I suppose all the leaves is some internal fault in the fuel gauge. And if the analog anti-slosh modules routinely go bad, then that's probably what happened here, except there's no simple way to bypass it.

If I were more electronically-adept, I think it'd be possible to rig up and arduino to replace the main circuit board for the gauge. The VFD looks to be self-contained on its own board, and the external connections are just five wires. I'm sure for someone who has some experience with these things, it wouldn't be a very difficult project.

One other thought, I've noticed the gauge will sometimes show one reading when first turning on the car, and then something wildly different after starting the engine. Similar behavior happens with my aftermarket radio, which will retain memory of its settings between turning off the car and setting it back to On or Accessory, but once I start the engine the radio resets itself. Is it possible there's some power surge or other funkiness that's throwing off the fuel gauge at ignition? How would you go about looking into that?

[gadget73]

cranking voltage will be low thanks to starter load, but most of those circuits should be turned off when cranking anyway. The start circuit typically powers just the starter, ignition, and EFI systems. Everything else drops out both to help unload the battery and to save it from funky voltages. If yours doesn't, maybe thats the real problem.

[Lutrova]

This cutoff is done at the starter solenoid? I assume that would be the place to test for a fault like this? Or maybe watching voltage on the terminals at the back of the fuel gauge?

My electrical know-how is slowly growing, but still ends at tests described in the shop manual.

[gadget73]

its in the ignition switch itself. Have to verify what contacts go dead when cranking. I'd probably just pop the small wire off the starter solenoid so you can hold it in crank mode as long as needed for testing. I forget what the EVTM calls them without looking but there is the group of stuff thats hot in accessory mode and in run, the group of stuff thats hot in run, and a few things that are hot in start.

a funky switch might cause this to happen, or if someone has added aftermarket crap and connected things together that ought not be connected it may do it. remote start, alarms, that sort of BS tends to tap into these circuits.

[Lutrova]

No evidence of aftermarket work in that realm. Someone butchered the radio connector and bypassed the amp, but that shouldn't come back to the ignition switch.

Something is definitely funky in the dash wiring. I believe with cruise on, I can resume the set speed by activating the turn signal. I once assumed there was some short in the steering column, but having opened it up enough now, I'm afraid the issue probably lies deep within the dash and a proper inspection may require complete disassembly, maybe even removing the harness.

I bet there's one little thing out of place in there that's causing all of these strange issues.

[gadget73]

radio does have key-on and steady-on power but if those were crossed nothing would turn off with the key ever.

without looking at the cruise wiring I don't know for sure but there might be a shared ground in there somewhere between turn signals and the horn circuit. The cruise buttons power through the horn relay and it "sees" a different resistance to ground on each button. If that ground is funky it'll do odd things. Probably worth a look in your EVTM at the first couple pages where it shows the locations of all the grounds. That may show you the common link.