Ls swap fuel gauge ??

[Bluebell]

Have 97 c1500. Put a 2004 5.3 in it.
So far only test it idling but it idled fine and gauges did what they should. But of all things that possibly could go wrong, noticed the fuel gauge is past full, 3 o'clock position and stuck there. I've had truck for few years and no issue before.
I'm guessing it's ground issue but not sure where to even begin to look.
Seem straight forward until I realized the wiring going into the passenger side firewall, is some sort fuel gauge buffer?
Tried following a test procedure but nothing changed and the wording used became more and more unclear as to what they were saying.
I believe the 4 pin square connector under the brake booster is the fuel gauge wiring.
I disconnected it and the fuel pump still worked, so either I was probing the wrong connection or the fuel pump uses different connector.
The 4 pin square connector. I tried jumping wires per instruction and no change. Gray wire was showing 10-12v but what I read said it shoukd be more like 4-5v.
The 4 pin connector had gray wire. Purple/white, black/white and green.
I did ground the blsvk/white wire and the gauge moved like 1/8' once but that's it. Never moved again trying replicate the addition of verses without.
I do understand I could trace wires from the tank but I really feel like it has be related to the obs engine harness. Just not certain where to begin. Trying avoid popping the loom open, spent alot time making the wiring nice and wrapping the loom with fabric tape. Lol
I'm thinking there's a ground issue with the buffer module but not certain how to approach that.
Any input appreciated. I'll keep plugging away as I get a chance to but this has me stumped.

 

[Shwa Kid]

What did you change as far as your fuel system goes? Are you still using the stock sending unit in the tank? If it worked before your swap, focus on what changed.

I think you're on the right track with the ground, check the connector back at the sender, not the one up at the firewall. My gauge went to full without it. Check that none of the pins pushed out of the connectors or anything. Here's a diagram from an 89, I don't have a newer one but it should be similar. Basically the pink wire has some voltage from the gauge cluster, the fuel level part of the sending unit acts as a resistor, which then connects to ground via the blk-wht wire. I think it connects to the frame near the sender.

You must be registered for see images attach

 

[Bluebell]

I didn't touch the fuel side, my year pump is supposedly fine for ls swap. I disconnected the lines at the firewall and bought pfte adapter to go to ls quick connects.
Was able test the buffer module behind the glove box and I was getting voltage on the purple wires black/white low reference. Which is not right.

 

[Bluebell]

As of right now I'm stumped.
Tested purple/white wire from fuel pump and found 12v. Should be ohm resistance.
Then I test negative cable disconnected to negative post and found 5v.
Then switch continuity test. Removed the 2 positive cables that are on my battery. The positive to starter shown no continuity. But the positive cable going fuse box, shown continuity to ground, and thats when positive and negative are removed from battery.
Pulled every fuse and relay, underhood and inside truck, no change in continuity, from disconnected positive cable to ground.
The only way to eliminate continuity, from positive cable disconnected to ground, was disconnect the c100 connector from the firewall.
Even with the pcm connections removed and all sensors and starter disconnected. I'd get continuity unless I removed the c100 connector. Which suggests to me there's a positive wire. Maybe orange. That's going to ground somehow.
Figured a sensor might been pinned incorrect but would think unplugging would rule that out.
I don't even understand how it was running and all sensors showing up in live data on torque app.
Thoroughly lost at this point, hopefully going back over the wiring and pinouts reveals the secret.

 

[Bluebell]

So this is what I've found so far.
Ohm meter across purple and ground, at square 4 pin connector at firewall, does vary. It was .15 ohms when tank was emptyish and near 100 ohms when I filled tank up. Supposed be 0-90 so close enough.
I put ground wire to the purple/white output on the buffer module and the gauge immediately went to E.
I did figure out that the gauge wiring goes from firewall, into passenger firewall behind glove box to buffer module, then across dash to gauges.
So all that is as it should be.
But somethings not making any sense as gauge still reads past full.

The buffer module has pink wire shows 12v.
Then gray wore showing 5 volts
Then there's a purple/white that comes from the square 4 pin connector.
Then there's another purple/white for out put.
Then a black/white wire, this wire goes to ground but it also shows 1.5v. I disconnected it from buffer module and it shown 1.5v coming out of the module. Which seems odd, and the other end I clipped, goes ground.
So gauge works, sender wires from tank do vary resistance but can't seem figure out how to get a and b to connect, so the gauge works.

 

[Bluebell]

So after trying bypass the fuel buffer module, it finally sunk in that the buffer converts the varying resistance/ground signal from fuel sender to a voltage signal, that goes to the gauge.
So next step is remove cluster and figure out why the fuel gauge isn't grounded.