Sporadic Fault Codes….

[ewo1]

24 Valve, 8.3 Liter Engine
Fuel System Troubleshooting And Repair Electronic Engine Controls

https://www.dmcpubs.com/pdffiles/873/87367126/44131/Sections/Sect10C26.pdf

 

[ewo1]

CAPS not applicable to my engine - it has a high pressure common rail…

ok...did not know that!

 

[nikitis]

Can he not pull any DTS codes?

Also my thoughts are if he can plug in a nexiq and it connect it's not the ECU directly or the network.

If it's an ohms issue nothing on the network would work even a nexiq connection. There would be too much noise on the network for devices to be able to read anything, so if this connects at all and provides Codes, it's not a network ohms issue.

If that is the case then it's something more direct. If the ECU is good and Network is good, but engine won't start, either a fuel issues, or something that drives the fuel, OR, lines to the injectors from the ECU is bad. But if network is good he should be able to pull codes as the network is used to send codes from ECU to Nexiq device.

 

[desrtdog]

I pulled a 1117 code and a 0435 code last Friday before I shutdown and cycled the key….restarted and codes moved to inactive…started to drive again and bus stalled … same codes….started again and stalled when dropped into gear, at which time I got the “comm fail” message and ABS light on instrument and bus would no longer start - just cranks, but doesn’t catch….see pics in post #40 …

When I try to scan codes now, coming up with nothing - no active, or historical…

No nexiq available…

 

[Jolly Roger bus 223]

i like a squirt bottle with gas.
make sure to hang onto the rag i have had them get sucked from trying to play with linkages and stuff and barely hanging on to it.

 

[PorchDog]

I would think that the likelihood of a terminating resistor actually failing is low-- but the fact that you know they exist, and that the bus should read 60 ohms when everything is as it should be-- gives you some good information to trouble shoot the harness. Looking for hold down brackets that have rubbed through, or corrosion in connectors, or broken wires. When you break the canbus into two parts-- with each part/segment containing the actual terminating resistor--or the module with an internal terminating resistor---you should see 120 ohms on each segment-- so if one segment shows an open, or a short, or anything besides 120 ohms, then you know you have an issue in that segment. This has to be terribly frustrating.

 

[ewo1]

No nexiq available…

You can borrow mine...but you need the software. I only have servicemaxx..
I'm thinking the clone is the same as they did provide a CD with the Cummins software, I just couldn't figure out how to install it.

 

[nikitis]

I may have a copy of that cummins insight software on Disc. I may try to install it to see if I can get it running and report back any quirks I run into or if it fails to install.

 

[ewo1]

No nexiq available…

I found my disc with the Cummins Insite software. I'd be happy to burn you a copy if you want it.

 

[ewo1]

Looking for broken wires is a painstaking time consuming job made easier with the right tools. I don't know if your familiar with the ECT3000 but it is a great tool to have in your tool box.

$75 bucks

I found one on marketplace in Houston real cheap. You might want to consider this option...
ECT 3000 on FB

ECT3000 I POWER PROBE

The new Power Probe ECT3000 Electrical Circuit Tracer is an intelligent short and open finder that makes locating short and open circuits quicker and more accurate than ever before.

www.powerprobe.com

 

[desrtdog]

Looking for broken wires is a painstaking time consuming job made easier with the right tools. I don't know if your familiar with the ECT3000 but it is a great tool to have in your tool box.

$75 bucks

I found one on marketplace in Houston real cheap. You might want to consider this option...
ECT 3000 on FB

ECT3000 I POWER PROBE

The new Power Probe ECT3000 Electrical Circuit Tracer is an intelligent short and open finder that makes locating short and open circuits quicker and more accurate than ever before.

www.powerprobe.com

I looked at a couple of videos on how to use this, and not sure how it would work if I am trying to find a short between the J1939 wires, as neither of them is a grounded wire - correct?

I went yesterday evening and again this morning to poke around. Yesterday, verified that the ABS errors associated wheel sensors did in fact clear, but still have the 11/2 error (indicating loss of comms with J1939 Retarder…which I guess is maybe an internal ECM thing??). Also tried to read codes and live data (w/ ignition on) via J1587 link, as well as live data from J1939 - nothing on either and the live data message came back “no supported stream”. So started wondering if maybe the ECM is bad or in need of a reflash…

This morning, I disconnected the batteries and waited a few minutes, then checked the C&D pin resistance on the J1939, and…zero ohms…as in the wires are shorted somewhere. Since the only thing I did from the other day was pull the TCM, I did that, checked the pin 7/8 resistance … it was low (forgot to write it down, but double digit, not the 110-ish of the other day)…checked the C/D pins again after reinstalling the TCM and it was 45 ohms….changed batteries in the multimeter just in case, but similar results, with some readings starting at 0 then building …. But still not close to 60.

So, do I maybe have a slight short in the J1939 wires somewhere, and if so, how does the ECT3000 find it if neither is a ground?…do I ground one of them at the diagnostic plug? Safe to do so?

 

[ewo1]

These videos will explain better than i can...

 

[ewo1]

Mind you, the FB link is for the 3000b...which I believe is the same thing but without all the adapters.

 

[ewo1]

I looked at a couple of videos on how to use this, and not sure how it would work if I am trying to find a short between the J1939 wires, as neither of them is a grounded wire - correct? The wires do not need to be grounded for testing The ECT3000 checks for both OPENS and SHORTS.

I went yesterday evening and again this morning to poke around. Yesterday, verified that the ABS errors associated wheel sensors did in fact clear, (This is good news )but still have the 11/2 error (indicating loss of comms with J1939 Retarder…which I guess is maybe an internal ECM thing??). Also tried to read codes and live data (w/ ignition on) via J1587 link, as well as live data from J1939 - nothing on either and the live data message came back “no supported stream”. So started wondering if maybe the ECM is bad or in need of a reflash… I was curiously wondering the same thing but, you doing all the prelim troubleshooting will not only verify this concern but also save you tons of troubleshooting money!

This morning, I disconnected the batteries and waited a few minutes, then checked the C&D pin resistance on the J1939, and…zero ohms…as in the wires are shorted somewhere. Yes the wires could be shorted together Since the only thing I did from the other day was pull the TCM, I did that, checked the pin 7/8 resistance … it was low (forgot to write it down, but double digit, not the 110-ish of the other day)…checked the C/D pins again after reinstalling the TCM and it was 45 ohms….changed batteries in the multimeter just in case, but similar results, with some readings starting at 0 then building …. But still not close to 60. If the resistance reading start low and slowly climbs, there is an issue! Resistors do not respond in this manner. There could be several reasons why this is happening... water/moisture in a connector, wires shorted either to ground or shorted across another unassociated circuit and you could be reading resistance from that other circuit/short as well as cross voltage on shorted wires.... best would be to disconnect the data cable from all three places it terminates, all ends disco and try again. also, remember I spoke earlier about doing loop back testing on the wires... This is how I tested my cable for the tranny pushpad when mine went dark.

So, do I maybe have a slight short in the J1939 wires somewhere, and if so, how does the ECT3000 find it if neither is a ground?…do I ground one of them at the diagnostic plug? Safe to do so? You do not necessarily have to "ground" the wire...you just want to send an audible tone down the wire and follow the cable till you find the spot in crisis.

I worked as an underground telco cable splicer for Bellsouth back in early 2000. I was trained to work on "pulp" cable. Each individual 24 gauge wire was wrapped in plain brown paper, no color codes, no colored vinyl coatings on the wires...
I worked on 1200+ pair cables. When you open up the splice case it all look the same color.
Point I am trying to make is that we used "Tone generators" to identify each individual wire within the case. You'll be doing something real similar with the ECT3000. Put a tone/signal on the wire and follow the wire till you get to the location that produces questions/concerns.

it is a real easy and simple process to do. You'll figure it out in 10 minutes or less!

 

[ewo1]

Loopback testing...
Look at diagram I uploaded...

1) disconnect data cable at the 4 location on left side of page. and also at the ABS location.
You want these circuits open.
Key off
2) test resistors (ohms) at diag port C&D...Good???

set DMM for continuity test,
3) test continuity at diag port, F&G, there should be no continuity here as the ABS side is open. If you do get a resistance reading. ....problem!

YOU NEED TO FIND CONNECTOR C303 AND DISCO THE RESISTORS FROM THE CIRCUIT.
The reason you want to remove it is so there are NO set values on the wires when you test for continuity.
If you can't find there is a work around...

4) Now that the circuit is open check for continuity on C&D (without resistors in circuit). there should be no continuity. If you do get some sort of reading, problem! At this point change the setting to read DC volts...is there any voltage here?

There should be NONE.

If you do find voltage that is a clear indicator that "A" wire is shorting/crossed with a power wire probably inside the loom somewhere.

If there is zero volts but you still get a resistance reading then "A" wire my be shorting to either ground or to another wire that does not have voltage on it but see a "Load" and that's why you get a resistance reading. Again...Problem!

Oh and BTW, I realized a misspoken mistake.
Earlier I said I did not see where the datalink was connected to your multiplexer, well it is...
Could the Multiplexer be the problem..??? IDK !!! But we can test for that!

Once you disco those components and test the datalink, reconnect them back one at a time and retest the datalink each time to see if there are any changes which could identify a problem spot.

loopback testing is gonna be tricky because the datalink wires are in parallel to each other so if we short out/jumper together say the "Instrument panel" data wires we will see the short on the meter but if the damage is say on the "Multiplexer" side, it could be deceiving and confusing. In the end though you will have identified for sure the datalink cable needs help.

Now if you disco the 5 connected locations and all is clear and you still have no communication then it either a bad ECU or bad power/ground to the ECU.

 

[desrtdog]

Where did you find that version of the wiring diagram? I’m not seeing it on the vantage portal under my bus number (F27077. Below is the version I have - it is for air brakes and does not include C303 and the resistor.

 

[ewo1]

Where did you find that version of the wiring diagram? I’m not seeing it on the vantage portal under my bus number (F27077. Below is the version I have - it is for air brakes and does not include C303 and the resistor.

Interesting, yours is dated 11/05/2008 and the one I found is dated 03/21/2010.
The datalink on your drawing seems to be more complex...
Yours has two circuits, the J1939 and J1587.
the drawing I uploaded only has the J1939.

really interesting... I do not have an answer...yet.

I will look into it. -->...edit -->looked into it...Ok, so I clicked on the wrong drawing from my files folder...my mistake. gonna work from the service manual directly now...

The trouble shooting steps I mentioned earlier, still the same.. disconnect the "Field end" end and test from the diag port but this time retesting again as you reconnect back each connector on the field side in order to observe any changes as you reconnect.

 

[ewo1]

So the difference between both drawings, both are accurate by the way, is that the one I posted is for Hydraulic Brakes and your is Air.
I should have picked up on that.... it my rocket scientist training....majored in Lunch in high school... damn '70's....the list goes on.....

 

[desrtdog]

So here is the corresponding hydraulic brake schematic…it also has the J1587, and shows the resistors…but it is dated earlier than the one you sent…
Wondering if there’s a version for air brakes that matches your file?

 

[ewo1]

I am gonna try to do this page by page, best if you print them out as I go along.

I will reference each page by the PDF page number and not the actual page number on the doc. Makes it easier for finding it when you open the PDF
I will not list "Pg or PAGE", just the number followed by the sheet number. The sheet number is on the print and tells us where to go next.

369/air drawing
bottom right-J1939 diag interface 17

362 Instrument cluster
"A" connector - GREEN
take a look at pins 15& 16, 17 & 18.

17& 18 - J1587 circuit
16 & 16 - J1939 circuit

Introduction to SAE J1587

I believe one of my earlier posts/links to another thread mentions instrument cluster "comm fail".
It would be worth looking at the connections behind the cluster. Would be real silly if you found a loose connection there...