D6N ECM fault no start

[jrlit]

I have a D6N dozer serial number CAT00D6NJMG500913.
Occasionally when I start it, it will crank but not fire. Acts like no fuel/dead ECM. So far if I cycle the key a few times it has always eventually started. It will flash a few codes while this is happening, but once it starts the codes go to inactive and it works completely fine once running. The research I've done suggest a poor connection somewhere in the wiring but I don't understand why it would never act up while running if that was the problem. I'll attach a picture of the active faults last time it did it

 

[Nige]

The 590-9 indicates that the Machine ECM is not receiving the desired status updates from the Engine ECM via the CAN Datalink. In the strictest sense the troubleshooting will only work if the Dagnostic Code is Active. My opinion is that somewhere in a main harness connector between the two ECMs, or even in one of the ECM connectors themselves you will find condensation/corrosion/green crusties or a combination of all three.

 

[jrlit]

Ok, It does sound like my only option at this point is to start tracing wiring. It still seems a bit perplexing that once running these codes never come back up. It would seem like if it's a faulty connection they would pop up occasionally?

 

[Nige]

It does sound like my only option at this point is to start tracing wiring.

Not so much tracing wiring - looking at harness connectors.

Vibration can have a lot to do with it. Damp also.

Take the schematic and locate every major harness connector between the two ECMs, then open said connectors one at a time, clean with aerosol contact cleaner, dry and reassemble. BTW the schematic has hyperlinks so that if you click on the blue title (e.g. CONN 32) it will jump you straight to the location of that specific connector on the machine.

 

[Nige]

Did you ever get to the bottom of this.?

 

[Simon C]

Be nice to know what was found or still has problem.
Simon C

 

[jrlit]

Good afternoon, just jumped on this thread again as I have the dozer in the shop and have been digging into this some. I have not got very deep into it.And have not come to a conclusion on it. It looks a bit overwhelming trying to track this down with the amount of space You have to work with and the amount of wiring that is in these machines. The main connections on the engine ECM seemed fine. No dirt, moisture or corrosion. I did find wires going to the "operator presence sensor" in the seat that were nearly cut and frayed and likely occasionally shorting out, is there any chance something like that could have messed with the ECM? The issue is so sporadic that if I do find something out of ordinary I'll have to just run for a while to see if it does it again.

 

[Nige]

I did find wires going to the "operator presence sensor" in the seat that were nearly cut and frayed and likely occasionally shorting out, is there any chance something like that could have messed with the ECM?

Not necessarily messed with the ECM but AFAIK the lack of the correct input from the operator presence sensor will certainly apply the park brake and flip the transmission to neutral. This will happen if the operator lifts his butt off the seat while working. The operator presence sensor is an analogue input to the Powertrain ECM, not the Engine ECM.

I wasn't aware that a signal from that sensor would stop the engine on top of what I described above but it MIGHT inhibit the engine from starting if it was stopped.

Based on the Codes that appear in your first post I would say that you have an intermittent connection issue on the Datalink network that connects all the ECMs on the machine.

 

[Simon C]

A person can trace all the places the Can wires go to, but won't be easy by any means as it will require you to isolate one ECM from another. There is a guy on YOU Tube that Nige sent me to go watch that has a lot of that exact testing, that what we think is the problem and what it was in the end can be worlds apart. His channel is called Pine Hollow Auto Diagnostics and he finds all kinds of parasitic draws and other electrical problems.
I have watched about 50 of his videos and he does data driven diagnostics .

You are working on a dozer but most of all the principals are the same.

Checking all the plug connections that you find and blowing all out clean is a good start, as you may find some corrosion in one and problem solved. Have fun.
Simon C

 

[jrlit]

Well I've been crawling all over this thing, I think I've disconnected and inspected all of the main ECM plugs. The bolt on one of the connections was only finger tight but I still think it was tight enough to be making the proper connection. I've been testing the ohm reading on all of the CAN wires that I can find, so far they've all been right around 60, nothing lower than 58.3 and nothing higher than 61. There was one connector that I got a reading of 120. I believe it was CONN 35 in the schematic that Nige sent earlier in this thread. Should all the ohm readings in the entire CAN system always read around 60 or would there be a chance that one side of a connector could read 120 without there being anything wrong?
I'm sorry if I'm totally off course here, I don't claim to know what I'm doing, I'm over my head but trying to figure it out. I feel like testing the ohms would say if we have a bad connection/short somewhere, but once again if it's not currently acting up maybe that's useless?

Also, it has acted up a few more times since I first stated this thread and I've learned what to watch for. Whether they help narrow the problem down or not I have no idea. So when I turn the key to start it I can tell immediately (even before cranking) that's it won't fire because the glow plug light does not illuminate and the instrument panel does not show the engine hours on it.
Thank you guys for being willing to help on this, it's very much appreciated.

 

[Simon C]

Well I've been crawling all over this thing, I think I've disconnected and inspected all of the main ECM plugs. The bolt on one of the connections was only finger tight but I still think it was tight enough to be making the proper connection. I've been testing the ohm reading on all of the CAN wires that I can find, so far they've all been right around 60, nothing lower than 58.3 and nothing higher than 61. There was one connector that I got a reading of 120. I believe it was CONN 35 in the schematic that Nige sent earlier in this thread. Should all the ohm readings in the entire CAN system always read around 60 or would there be a chance that one side of a connector could read 120 without there being anything wrong?
I'm sorry if I'm totally off course here, I don't claim to know what I'm doing, I'm over my head but trying to figure it out. I feel like testing the ohms would say if we have a bad connection/short somewhere, but once again if it's not currently acting up maybe that's useless?

Also, it has acted up a few more times since I first stated this thread and I've learned what to watch for. Whether they help narrow the problem down or not I have no idea. So when I turn the key to start it I can tell immediately (even before cranking) that's it won't fire because the glow plug light does not illuminate and the instrument panel does not show the engine hours on it.
Thank you guys for being willing to help on this, it's very much appreciated.

You need to Voltage drop test the power to the Main Relay, Start Relay and some other places for voltage loss of potential. Do you know how to do or need some instruction. I will have to check schematic later this afternoon for some simple tests to rule out dumb stuff. Maybe Nige will have a few suggestions as your glow plug light, and no engine hours is something else.
Simon C

 

[Cmark]

If a module isn't awake then it won't be broadcasting on a data link. When looking for "data link" faults, it's just as likely to be a problem with a power or ground not turning a module on.

 

[jrlit]

You need to Voltage drop test the power to the Main Relay, Start Relay and some other places for voltage loss of potential. Do you know how to do or need some instruction. I will have to check schematic later this afternoon for some simple tests to rule out dumb stuff. Maybe Nige will have a few suggestions as your glow plug light, and no engine hours is something else.
Simon C

I have a vague idea how to do this but honestly I'll take any instructions you're willing to give. TIA

 

[Simon C]

If a module isn't awake then it won't be broadcasting on a data link. When looking for "data link" faults, it's just as likely to be a problem with a power or ground not turning a module on.

Cmark is correct in checking that there is clean power to the Engine ECM and Powertrain ECM. There was a post recently on a skidsteer with data link problem by code, that I commented on and followed.
In the end it was a faulty main power relay that was partially burnt out.
My schematic is RENR 8189-06 but does not cover your machine so rather than lead you down a rabbit hole I will wait for someone else to comment on the exact pin numbers. If you send me the schematic unless yours is on paper I could give some exact tests to do.
Simon C

 

[jrlit]

Here is the schematic

 

[Nige]

I think I've disconnected and inspected all of the main ECM plugs.

Without a close look at the electrical schematic, which I don’t have close handy at the moment, I would say that’s only half the story. If you trace the CAN - and + wires on the schematic I’m sure that you will find locations where they pass through “intermediate” connectors. These could be anything from 12 to 24 pins and any one of them could be loose, corroded, damp, or damaged. Don’t ask me how many of them there are but the schematic will show every one. TBH each one of them should be checked and ticked off the list. A good dousing with aerosol contact cleaner at the same time would not go amiss.

 

[Simon C]

So what Nige has posted is in need of doing but might want to test a few of these things I have posted below to rule them out.
So I looked at the schematic, and from the front description on the cover it is a little off from your serial number, but I could be wrong, Nige or Cmark might know. But thanks for the schematic, So here are a few tests you could perform to verify if you have clean power and clean ground.

#1 Test your Relay 2 Main ( Part of Fuse/Relay Base ) that is found on the schematic in area G5 for voltage loss where it outputs by Blue cable. With Key Switch Turned On and some lights and some fan running to flow current check by hooking Positive meter lead to Battery 24 Volt Positive post, and Black meter lead on Output Blue Cable connection you should have tops 0.4 volts or so on Meter.

#2Check the Relay3 Main also with same test as #2

#3 Just above on schematic test UF7 Powertrain ECM Fuse with the Black test lead to see how much loss with key on as well. Should be less than 0.4 volts on meter.

#4 Check UF 8 and UF 2 fuse as well.

There is a whole bunch more fuses that could be checked as well. I open the schematic and take pictures with my phone so that I can find them all. If your fuse panel is dirty or corroded, it might be a good idea to clean this up.

Your Powertrain ECM has Battery Positive pins also that could be checked.
Your Engine ECM has Battery Positive Pins that can be checked also.

All these positive tests are done while the key is on and the positive meter lead is on the Battery Positive 24V post, with your black meter lead testing all the fuses and relays for loss of potential.

All the grounds at the computers can be tested by backprobing the Battery - (Negative ) pins that are listed on the Powertrain Control Module such as Pin # 13,23,57,70, with your Red Meter lead while your black Meter lead is hooked up to the Frame Negative Post of the Battery. The Engine Control Module Ground connections can be tested the same way as well.

Dont go to the machine without pictures on phone or paper copy so you see evrything.

Simon C

 

[Simon C]

Without a close look at the electrical schematic, which I don’t have close handy at the moment, I would say that’s only half the story. If you trace the CAN - and + wires on the schematic I’m sure that you will find locations where they pass through “intermediate” connectors. These could be anything from 12 to 24 pins and any one of them could be loose, corroded, damp, or damaged. Don’t ask me how many of them there are but the schematic will show every one. TBH each one of them should be checked and ticked off the list. A good dousing with aerosol contact cleaner at the same time would not go amiss.

What Nige posted before I finished my typing will need to be tested if you find nothing wrong in the other testing.
Simon C

 

[Simon C]

Forgot to say that the ground connections at the ECM's that go back back to Battery Negative Post should only show a loss of 0.2 volts maximum..
Anything else you test that has a high reading, post it so we can look at schematic to tell why.

Simon C

 

[Nige]

So I looked at the schematic, and from the front description on the cover it is a little off from your serial number, but I could be wrong,

Schematic UENR5402 covers the OP's machine.