Another truck: 2006 Intl 8600, Cummins, Fuel gauge

The system controller is one of the more reliable components I have seen.
I have only ever replaced one. I still suspect a damaged harness.

I'm sure you are right. But now that I've gone through all the multi-pin connectors and deoxidized them, I'll feel better about the rest of the circuits. It still strikes me odd that this truck is such a departure from the diagrams you found. I never saw a single reference to the 36 pin connector on the firewall that evidently is a controller connection. I don't really care, but it just made this a bit more confusing. And apparently the sender wiring does not go through the cab/chassis connector near the doghouse in the cab. I got my continuity with it unplugged. Since these trucks were offered with different engines, it makes me wonder if they had completely different harnesses. Doesn't seem likely but I have no explanation.
My plan, if wiring the RH sender works as predicted, is to abandon the existing LH harness and make the RH the master and run a pair over the transmission to the LH tank. But I will still get under there and clip out the abandoned wiring. I'm sure the next guy won't need the confusion. Plus I'll be able to examine the rest of the harness in that area for any obvious damage. But as long as I get a good ground, and provided I can locate that common "zero reference" point, I'll pull those terminals and give them a good cleaning.

All based on results of the rewire of the RH side.

The owner, when I was complaining about not being able to fit and work under the truck said he'd have one of our fabricators build a couple of drive on ramps. I'm sure he won't have them before I'm done, but the truck shop guys might use them, enev though they have several air/hydraulic jacks and plenty of stands. Knowing this fabricator, they will weigh 1500# and I'll need a forklift to put them in place. I had to explain that if he did that, they couldn't be simple, flat top surfaces. They needed pockets or rails front and back to capture the tires. He hadn't thought about the possibility of a truck just rolling off the top. Then he started talking about 2 and 3 level ramps. I stopped listening then.

So much for the week. Now I get to try and catch up around the place. I think I'm the last guy on the road to mow. I like to wait until the clover and paintbrushes bloom and go to seed, but that still leaves a lot to mow. At least it isn't hot yet.

Have a great weekend all!

Tony
Part of the harness after it goes through the 9700 connector at the left rear of the engine
goes to the right side of the cab and then through the firewall and then back across the
cab behind the firewall to the left side. The fuel gauge wires may enter the cab on the
right side not through the white connectors on the left. I have always wondered the the
engineers did that.

I'll admit I haven't really crawled up on the right side of the engine compartment to see if there is another bulkhead, but that's entirely possible. I was just going by the continuity I found on that 36 pin connector on the left side. When I found it, IIRC, the chassis/cab connector was open, so that makes me think that particular connector goes directly into the controller, except from the front, right at the firewall. If they did such a zig-zag move from R to L I would also have to wonder why. I'm sure there was a reason, but it's anyone's guess. Seems a little strange though. Plus it doesn't match the diagram, which clearly shows it going through that cab/chassis connector. Of course, I may be dealing with someone's rigging and this isn't actually correct.

If I remember correctly its not a bulkhead connector its a big rubber grommet.

On the RH side of the engine bay?

If I have to replace part of this harness because of lost insulation and wire corrosion then I want to make my connections as close to a safe area as possible. Perhaps just ahead of wherever it penetrates the firewall

If you could access the gauge wire at the rear of the 9700 connector B11 connector
probably the simplest. Then use a good ground for the senders. I suspect you have
more than one wire in the harness with a problem.

Well I probed that connector, specifically at B11 and did not find continuity, iirc.

But if the 36 pin connector does actually go into the controller I can’t see how it would get to 9700. So Monday I’ll plug it back in, based on the assumption that it doesn’t go directly into the controller and see if I can pick it up on B11. That’s where it should be, I agree. If it’s there, I’ll find out if it is making it to 4 or 5 at the controller. If so, then the RH harness is good on the tan and is lacking on the grey (gnd) side only. I did measure 10.5 V to it at one time. If it were a clean wire and all good connectors I would expect closer to 12, but coming from the controller no telling.

I agree that there is likely more poor wiring or connectors in play here though so I plan to get under it yet to look.

mekanik said:

If you leave one side disconnected the gauge readings will be off or
the gauge may not work at all. The system controller is programmed to see
15 ohms empty and 16 ohms full.

Click to expand...

That should have read 16 ohms full and 120 ohms empty.

I knew, Terry. No sweat. Two resistors of equal value measure 50% of their value when in parallel, and 2x when in series. Basic electronics. I figured you just slipped a decimal or something. Unequal resistors require a bit of calculation, as in the formula I mentioned above. When I work on something with oddball resistors spec'd, or close tolerance, I'll play with combinations to get what I need. I keep a fair collection on hand and since the price isn't really that different in the relatively small inventory I do keep, I nearly always buy 1% precision resistors. One of my hobbies is restoration of antique radios and old ham gear. So I get to play with it sometimes. Not lately though, too much work getting in the way of my fun.

The 36 pin connector to the controller (there are two, actually...on behind the kickplate and one on the firewall). The tan and grey wires I never could find are actually on the engine compartment side. I traced them back to the connectors at the sending units and the RH side was good, very low ohms. The LH, not so good. So, figuring I'm running new wire at least that far anyway, I snipped them and soldered on a 150 ohm resistor to each tan wire and chassis grounded the greys through the resistors.

Unfortunately, there is absolutely no change in the behavior of the fuel gauge. The only thing different from the schematic is a tan wire A36A that comes from J36A and goes off to "Telematics GenI Module (2J) CH3_36 that has a note on the schematic that simply says "W/ Aware". That I have not looked for so obviously not tied it to the tan from the senders.

Part of what I find confusing is the way they show 3 tanks, 3 senders, but only 5 pins at the controller for the tans. The greys are common of course. The tank specifically labeled for the 8600 has it's tan N36A running through a connector associated with the transfer pump. There it changes to J36A and joins J36B on pin 4 at the controller. So that tan never goes through connector 9700 at all. Might note that the Telematic reference only applies to the two parallel tanks on N36E and N36D until they pass 9700 and change to J36A to pin 5 and J36B on pin 4. So N36A/J36A/N36E/J36B and all the tan wires are common, with only the Telematics leaving that schematic on A36A, assuming that module is used. Looks like a lot of wasted wire to me, and since they are all essentially a loop, I don't see how they get a separate read from the tanks.

In any case, with my resistors in place and the greys grounded....I expected some changes due to eliminating anything untoward under the chassis. Unless there is a 3rd tank hidden out somewhere, I'm beginning to question the controller.

What am I missing?

The controller processes the signal from the tank senders and transmits it to the instrument panel on the data link.
It could be the instrument panel causing the problem. The attached diagram is from the manual for your truck.
It does not show three tanks you might be looking at the wrong diagram.
I would suspect the instrument panel before the controller.

That may change my testing setup. I had already figured it was the cluster, and the owner was pushing so I bought a used one and it was very clean, likely out of a wreck, and I had to transfer a couple of the gauges from the original to get the full set back that the truck was equipped with, but there was no change in the behavior. It was cheap, and a calculated gamble. Cost less than a single trip to the dealer to scan it, so that's where he wanted to spend the money. Knowing as much as I do know, with the schematics and information you've provided, I probably would not have done that. But it's done. That's one more reason I am beginning to suspect the controller. I have another 8600 in the yard, but it's a 2012, so entirely different setup. If I had an identical model I might be tempted to swap in it's controller just to turn the key on and let the gauges self test and see what happens. Not actually sure I don't, so I might look around. I've eliminated all the wiring between the senders and the 36 pin connector and subbed in known good wire, so this is getting puzzling and looking less like a simple wiring issue to me.
We have a Diesel Laptop system, but I never use it and don't know if it would show anything. I figure if it would, the truck guys would have found and fixed it by now. It's been sitting on my list for at least a year.

I'll take that schematic and see what changes there are and adjust to suit and see what happens. I'll be back!


Quick glance says ground the grey (pin 26) and one lead of a 75 ohm resistor (half a tank for two tanks) and connect the other lead to pin 5.

Another thought that may be a clue. When you key on, it POSTS, then everything reads fine. Then a second set of tests apparently happens and the air pressure is analyzed and if recognized as low, the light comes on and the piezo beeper starts....it is at that point that the fuel gauge does it's thing....red light, gauge around to 6:00.

You have pretty much covered all the bases. The controller is programmed for each individual truck.
The fuel gauge would be a programmable parameter as the senders for dual or single tanks are the same.
I have seen controllers "Forget" part of their programming and something start to not work. In all cases it
was a light somewhere on the truck that quit working and started working properly after the controller was reflashed.
It might be worth a trip to the dealer to get them to reflash it and see if it solves the problem.

That may be what it comes down to. I'll make my new cable runs permanent and anchored properly so they won't be tempted to start picking my work apart. Then, the way I see it, it only leaves the controller. If a fresh flash won't take, I'm sure they will be willing to sell me a new one. But this truck is only a $75K truck, and I have two machines down right now worth a combined $1.2-1.3mm, so it will just sit until I get the more important work done. Beside, I believe the intent is to sell or maybe rent this truck, but the machines are integral with the operation of one segment of the business that happens to be lagging behind production and these machines really need to be working, not waiting on a water truck.

I appreciate all your help Terry, and when I get back to it, I'll come back to this thread and pick it back up.