Fuel Gauge Inaccurate - SOLVED

[oldchev]

Greetings. I have a 2004 Yukon Denali @260K+ miles.
My issue today is fuel gauge malfunction. According to mileage and previous economy(?), its off 1/4 to 1/3 depending on actual. Up until last week it was or seemed accurate, but Saturday was reading under 1/4 tank and I filled it with 13.2 gallons and that was squeezing it in. The reason I kept trying to fill more was I knew (previously) it would take much more at that reading.
So today It changed from just over 3/4 to under 1/2 only driving 25 miles. Once arriving I shut it off and turned the key to on to see the reading, I watched it move a little then slowly creep up above 1/2 about midway to 3/4. Later I did the same thing and it was under 1/2 and stayed there. After driving 25 miles home it was now at 1/4. From full on dead full to 1/4 in 106 miles total.
I understand it could be one or many things. I did replace fuel pump/sending unit with factory (was it bosch? don't remember) whatever the original was... at about 180K (4 years ago) and never had an issue with fuel level ever until now.
I am thinking stepper motor for fuel gauge?
Anyone had similar issue with stepper motors for these gauges? Testing process?
No stranger to fixing stuff on this baby just want to hear from others on this. No TECH2 available but most everything else. Don't want to drop the tank (again) unless that's it. Not a fun job. All the best, thanks.

 

[MassHoe04]

You might be able to check what the fuel level sending unit is telling the gauge with a Volt-Ohm multi-meter to verify nothing wonky is happening in the tank.

Here is a link to what values are.

Fuel Gauge And Tank Capacity

I have a 2005 Tahoe. A quick google search says it has a 26 gallon tank. Haven’t paid much attention to it before but noticed today that my gauge is right at empty and the truck says I’ve used 18 gallons. Is it actually a 26 gallon tank and my gauge is just off or could it be a smaller tank?

www.tahoeyukonforum.com

 

[oldchev]

It looks like the access to the ohm wire is at the instrument cluster 24 pin connector. Using the Service Manual it says that wire would be B7-yellow wire. I'll try my Ohm reader on that wire and see what the values are. If I had a signal generator I could make the gauge move and see but so far I do not.
Anybody else please chime in. Whatever I do, I'll report back so we know how this was resolved.

 

[OR VietVet]

If was a Bosch pump, it is not OE. The two things it could be have already been said, the sender or the stepper motor. Post test results please.

 

[MassHoe04]

Bosch is certainly not junk (usually). If AC Delco is not available, for whatever reason, Bosch would be my next choice.

You are looking to confirm either the sending unit is wrong and the gauge is OK; or the sending unit is OK and the gauge/stepper motor in your cluster is bad.

I think you can measure voltage at the cluster end of the wire, but for measuring Ohms resistance... I think that has to happen right on the socket at the sending unit itself.
If your tank is full, you should expect to see either 250 ohms resistance at the sending unit terminals or a full 2.5V on the wire at the cluster.

Let us know what you find.

 

[oldchev]

Similar to what MassHoe04 posted (in the link) the values for everybody's enjoyment (attached fuel level specs).
I have all the factory manuals to follow and refer to but sometimes this forum has experienced shortcuts.
Would a Kent Moore J33431C signal generator be any help?
Looks like I'll do the easier test on the cluster Thursday.
Regarding the fuel pump assembly ... I really don't remember what it was but it was the same one going in as I took out. It was an assembly with integral sending unit.

 

[rockola1971]

Bosch is certainly not junk (usually). If AC Delco is not available, for whatever reason, Bosch would be my next choice.

You are looking to confirm either the sending unit is wrong and the gauge is OK; or the sending unit is OK and the gauge/stepper motor in your cluster is bad.

I think you can measure voltage at the cluster end of the wire, but for measuring Ohms resistance... I think that has to happen right on the socket at the sending unit itself.
If your tank is full, you should expect to see either 250 ohms resistance at the sending unit terminals or a full 2.5V on the wire at the cluster.

Let us know what you find.

Using the 2.5v at full level I would just take a potentiometer and 9v battery and turn the pot up until 2.5v is reached and mark that as max on the pot and a value not to be gone higher than. Disconnect the fuel pump and stick a negative wire off the pot and the positive on the SIGNAL wire to the cluster. See if the fuel gauge is showing the correct level. Lower pot to roughly 1.875v which should be 3/4 tank level and see if the gauge follows. Keep stepping down .625 for 1/2 tank, 1/4 tank and empty at 0.

*Make sure you are on the signal wire and not the 12v supply (source) power wire.* *Adjust voltages below the 2.5v with a voltmeter reading and pot DISCONNECTED from truck wiring harness in case you go the wrong way so you dont send 3-9v to the cluster*

DO NOT Hook up a signal generator to the fuel pump circuit. A signal generator produces AC voltage and an variable frequency sound wave. Fuel tank was never designed to have an AC signal/voltage introduced into it. It could cause a spark and an explosion.

 

[MassHoe04]

I'll be the first to admit, I am less experienced when it comes to electrical diagnostic stuff.

@rockola1971 gave some good documentation that even I might have a chance of understanding! - Thanks for that write-up!

 

[Fless]

Note that the signal wire sender voltage is inversely proportional to the indicated fuel level. Less voltage = more fuel level.

 

[oldchev]

I have a conflict. A chart copied from this forum (first one) shows just the exact opposite volts and ohms readings for the entire range most notably Full and Empty. The second chart is from my OEM shop service manuals for 2004 (Yes, I have all three). Is it possible the first one is from an earlier manual? Can anyone explain the variation?

 

[oldchev]

I see when I posted the charts, they flipped. So the one I stated as first is the dirty looking one (GMC Fuel level#2), whereas mine is the clear white one that says Fuel Level Specs.

 

[Fless]

Ok, so I'm very curious about this. Seems like less resistance (nearer 0 Ohms) would allow a higher sender voltage on the circuit. However, what I don't know is if the BCM (if that's what controls the gauge) says higher voltage = lower level on the gauge display.

 

[oldchev]

Here is what I have discovered so far. Today 2/18, I checked the signal circuit at connector 152 (under the fuse box, eng comp).
I pinned into the signal circuit (purple) and low reference circuit (black). Using my DMM, read Ohms. I have two meters so checked both on each test to verify. To clarify, The fuel level meter is the gas gauge.

Test #1: Fuel level meter read one line under 1/4 @ 198 Ohms.
I added 6 gallons gas.
Test #2: Fuel level meter read one line above 1/4 @ 170 Ohms,
I added 5.84 gallons.
Test #3: Fuel level meter read one line below 1/2 @ 120 Ohms,
I added 6.05 gallons, apparently topping it off.
Test #4: Fuel level meter read FULL @ 46 Ohms.

I know what some have said about the specifications however my 2004 GM service manual has the chart (attached below) with 40-66 Ohms = 88-100% or basically FULL.
My Fuel Level Meter seems to drop from Full to about 1/2 when using about 6 gallons and from 1/2 to about 1/4 using 6 more gallons the rest is anybody's guess.
So with all that said, not having a signal generator to test the actual stepper motor Fuel Level, I am leaning towards the stepper motor being defective since the ohms read outs better define the actual fuel level.
Can anyone think of any further tests I can do without purchasing a TECH2 scanner? I have an Innova 3130c scanner that does a lot but no signal generator properties.

 

[rockola1971]

Check power at fuel tank plug. Is it there and within spec? If it is then sounds like your stepper motor is bad OR your needle is loose on the stepper motor output shaft (Like the inertia belt slipping on the outer ring of a harmonic balancer)....you'll never get a good reading.

 

[MassHoe04]

Check power at fuel tank plug. Is it there and within spec? If it is then sounds like your stepper motor is bad OR your needle is loose on the stepper motor output shaft (Like the inertia belt slipping on the outer ring of a harmonic balancer)....you'll never get a good reading.

Mine reads about 1/8 more than actual. It reads 1/8 on gas and 6 MPH on the speedo, before I even turn the ignition on. But, that is because PO put some cluster decals on the faces of the gauges and did not align the needle when they put it back together.

With two gauges being off and the PRNDL display contacts needing re-soldering... I'll probably just send out for a rebuilt cluster. At some point. No rush there. I got things like control arms and tires to get first.

In the meantime, I subtract 6 MPH from what I see on the speedo and I use the digital information center to tell me how many estimated miles are left in the tank.

 

[oldchev]

Update:
After two full fill ups, the gas gauge reading is close if not at the standard reading. I have nothing to attribute this to. I have monitored the distance and gallons used, then on fill up (all the way) it corresponds with what actually was used and the gauge needle seems to be back to normal and is relative to actual fuel level. Ain't that a peach!
So, watching it, but so far so good.

 

[OR VietVet]

When you say: fill up (all the way), do you mean fill till pump nozzle kicks off or put as much gas as you can in there? You will damage your evap system if you put as much as will hold in there.

 

[BadOlHoe]

Did a quick search and this post is relevant to almost the same issues I have had in the last few months. A lot of the info was over my head seeing as I have next to zero electrical knowledge... other than don't stick a fork in a toaster. With that said I am having similar issues. Is there a simple fix to a fuel gauge all of the sudden reading 3/4 at full fill up and then once it hits 3/8 it's like you can watch it go down to 1/8?

Apologies for hijacking this post

 

[oldchev]

Further update 4/07/23:
From the original post, Fuel Guage was displaying low reading faster than use. It "corrected" for a couple months then back to, or worse. My gallons consumed on the trip meter was accurate as each fill up corrosponded to the gallons used. So I went with that so as to not get caught short. Per the previous posts, I did a Digital Multimeter check of ohms at connector 152 (under the fuse box, eng comp). This led me to believe the sending unit was ok as the ohms vs gallons used was very close.
The correction process I used was to send my instrument panel to Dr. Speedometer for all new stepper motors, plus he also replaces all light bulbs. Now the guage displays accurate and corrosponds to fuel used.
With a little homework, I was able to narrow down the problem, which is now solved.

 

[Fless]

Thanks for coming back and explaining the solution. Gauge stepper motor issues are common in the 2003-06 clusters. Easy enough to replace if you have basic mechanical and soldering skills.