Fuel gauge works, but not in dash

-
Deemo

Deemo

Well-Known Member
Joined
Mar 27, 2017
Messages
439
Reaction score
209
Location
Tennessee
Welp, So I rewired the whooooooole car and added a spade type fuse block. Before I did all this, I was checking the gauges and as I pulled them out of the dash, the fuel gauge started working fully. Before it would go to 1/4 tank and stop. So I thought maybe bent needle. Nope. Not bent. As I push the cluster back in the dash, it shows correct. Push it almost into place. Then the gauge drops back down. Pull the gauges out an inch, goes back up. Push it back in, goes back down.

So my dilemma is that it did it before the rewiring job, and it still does it now after. I have the new clip at the tank. It's somewhere in the cluster itself. It's not touching anything it doesn't need to touch, all the wires are correct. So does anyone else have this issue?
 
It's either mechanical or electrical

1....Electrical should be fairly easy. Do some resistance tests. Carefully measure ACCURATELY the resistance of the gauge out of the cluster. Reinstall it REcheck resistance AT THE STUDS. Now move one probe from the stud to say, the connector pin on the cluster

Next move the other probe off the stud to the junction where it receives power from the IVR

Might be "something like:"

Cracked PC board trace, stud moving and somehow losing connection at the gauge

2....Mechanical........Frankly what you have describes SOUNDS more like probably mechanical.......The movement of the stud(s) causing some sort of binding in the movement, etc.
 
One of my ex bosses (Motorola, two way radio, and other stuff) "it's either a component or an adjustment"
 
I just had my stock in dash fuel gauge stop working after changing the sending unit. Thought it was a bad sender, checked wires etc... It was the connection in the back of my dash to the gauge itself, I just wiggled it a little a it sprang to life.
 
I've checked the resistance, shows right. It's got a full tank of gas as well. All the plugs are original, but I changed the wires out along with the pins and connectors in the plugs themselves. I checked and rechecked the sending unit, it's got new gas line as well along with the double c clip at the tank. Didn't know if maybe something was causing interference from another wire maybe??? Or???? I'll pull it back out and tear the gauge apart and rebuild it/clean it as best I can. Never seen this before. If it works 2 inches away from where it's mounted it should work installed lol. Or at least that's been my thoughts for 20 years
 
No interference, if you are thinking of RFI/ EMI. These are simple, "heated wire" gauges, low impedance. It surely has to be something mechanical or similar
 
Make sure your posts are not grounding out to case. If they are not centered, they can touch the case.
 
Last edited:
I don't know what kind of car you are working on. But 67-69 Barracudas, and Dusters with the Rally dash, have the instrument voltage regulator built into the fuel gauge. Others have a separate IVR, so you should check the IVR connection as well as the gauge connections.
 
Like stated above, until we know what vehicle/panel...
 
65 Valiant signet 100 /6 170 3spd. Sorry. Don't know what I was thinking. So I've narrowed it down to the circuit board. When it's loose it doesn't do it. But then my dash lights flicker with every bump. So maybe the old board is bad??
 
On my 72 Demon upon CLOSE inspection a couple of solder joints on the pins to the circuit board were loose. I soldered them and also those which appeared okay. It solved a lot of my problems. You may also add a separate ground wire. Another cant hurt.
Yote
 
Deemo said:
65 Valiant signet 100 /6 170 3spd. Sorry. Don't know what I was thinking. So I've narrowed it down to the circuit board. When it's loose it doesn't do it. But then my dash lights flicker with every bump. So maybe the old board is bad??
Click to expand...


You are makin progress. You can fix them. There's an old thread on various technicues

Here's one now

Printed circuit pins repair

Doesn't yours (or did?) have the IVR in the fuel gauge? Does it have three terminals?
 
Loose pins on the circuit boards is typical to all models. they never were soldered, just crimped. We can add a little solder but... if the pin should pull out of the board in the future it will peel the copper trace away with it.
There are models that don't have printed circuit boards. Instead, every wire is attached at the final destination, much like our ALT' gauge is. All these gauges could be wired with ring terminals on their posts.
The lighting is a different hurdle. We would need different lamp sockets. Our sockets depend on printed circuit board and chassis ground through the inst' housing.
So what this all means is you could attach the blue fuel sender wire directly to the fuel gauge ( assuming that is the loose pin/fault ).
If I'm not mistaken, yours is a stamped tin inst' housing. 4 little legs in the corners serve as stops and chassis ground points. Maybe just fully tighten the panel to the dash so the lights, etc... get a good ground or attach a actual ground wire form inst' housing to somewhere. This added ground wire is recommended typically too so sometimes what vehicle/panel doesn't matter so much. Good luck with it.
 
I have gone through my whole car with brand new wiring harness I've made. I got two 66 valiants I'm fixing to be working on and one I get to keep after scavenging it to make the other run. Worst case I'll swap gauges over to the square style. But I've soldered all the pins, and still nothing. I'll try adding a ground wire to the fuel gauge itself. If that doesn't work then I guess that's another reason to go with aftermarket auto meter gauges ;). This is as far as it goes up. Pic below And the tank is full.

IMG_0929.JPG
 
I'm pretty savvy on wiring. I could maybe delete the board and directly run wires from a plug I can get at the junk yard off a newer car and that way I can still unplug the gauges and make a new backing for the original board to swap that with. Just take the new plug and run the wires directly to each gauge and each light.
 
If the gauge moves at all there is some current passing through it.
It is possible that the actual fault is in the senders current path. First the wire from gauge to sender must be good then the continuity though that sender and to chassis ground must be good. More than a few times a rusty ground jumper ( on the fuel line at the sender ) has caused gauge operation to come and go. Previous threads read "My fuel gauge works fine until it rains" or just the opposite, "Hit a large puddle and the fuel gauge suddenly started working".
I mentioned the loose circuit board pin due it this being the most common fault location. It doesn't rule out those less common fault locations. I wouldn't rework the entire inst panel for a fault in 1 wire/path.
 
Well. Instead of reworking it I pulled the blue wire out and put a connection straight to the gauge. Still does it. Ring terminal is connected to it. Nothing is touching anything that it's not supposed to touch. I'm totally at a loss. It's in the last two inches of having the gauge cluster in the dash lol.
 
Here's an idea:

You DO have the 3 terminal fuel gauge with the built in IVR, is that right? That thing needs a ground to operate "I believe."

Maybe we are on the wrong track here. Maybe when you think it stops, it is properly grounding the IVR, which then provides less power

Maybe the gauge itself is AFU

Here's what I'd do. Get yourself a resistor or two, or a spare fuel sender, or maybe you can easily pull that out.

In any case set up resistance to read 1/2 and then full tank and see what it does under that "known" condition. That is to say, you can check the sender, you can't exactly check the gauge without a known resistance

c-3826-jpg.jpg
 
To ask this another way, in your photo "almost there" what is the resistance of the sender?
 
So you have "weird one" we haven't seen before. Along with where is the limiter located? OEM mechanical or aftermarket solid state regulator? Questions like "How much fuel is in the tank ? Do you think the gauge is accurate when it does read higher?" Are both of the nuts that attach the gauge to the board snug making good connection?" Maybe a pic of the back of the panel will help.
See some of us know that some solid state regulators will pass the full 12 volts when not chassis grounded. That 12 volts would produce a higher "wrong" in a bad gauge/circuit and it can cook/ruin a good gauge when the sender path is good. Regardless how much we know about how it works, no matter how many faults we have seen before, our diagnosis of this "weird one" is dependent the details you provide.
 
Ok. I'll get the multimeter back out and pics of the back of the gauges.
 
You must log in or register to reply here.
-

Back
Top