3412e injector driver failure

[jon damon]

I have a 3412e in a wood grinder that has a puzzling issue.
It has one pair of cylinders laboring and misfiring when plugged in .
The motor will labor and smoke until the pair are disconnected and then the remaining 10 cylinders operate and accelerate normally.
I appear to have 105volts to all wires (at the related injectors) when the pair is connected.
We changed injectors to be certain it wasn't injector related.
The ecm is a 10r4088-03 with a 1251L053ks number below the part number .
I have yet to plug into scanner.

Any thoughts.
Thanks

 

[Cmark]

As always an engine serial number would be good.

So to be clear, by systematically disconnecting the harness to each injector in turn, you've narrowed down the trouble to two individual cylinders, correct?

Which two?

I'm pleased you've discovered that they run at high voltage using a multimeter instead of your fingers like some people do :/

 

[funwithfuel]

My thought, faced with a situation like this, would be to disconnect actuator harness from ECU and at injectors. Ohm out each circuit supply and return. Check every pin against every other pin and ground looking for unwanted ground or shorts between circuits.
While you have the valve covers off, check your overhead, make sure nothing dropped out or got wadded up.
Also measure resistance of all injector coils. If one or 2 are shorted that might be all you're up against. Comparative values are usually more reliable than book specs with any real hours on them.

I just remembered one other thing. The big 12s that power the really big tub grinders and wood mulches, there was a campaign I believe. 4 maybe 5 years back for injector problems, Vermeer got stuck with a bunch of them. Rumored. Might want to reach out to Cat

 

[jon damon]

Thanks for the response I will provide an engine serial number tomorrow.
The problem is either in the harness or the ecm I think.
The voltage is 105 on all wires when disconnected directly from the injectors. the two affected injectors share the same actuator harness .
Middle two cylinders right bank.
We had already swapped in two replacement injectors prior to recognizing the issue was not injector related.
From the schematic I had available it appears the four wires in the harness really appear to function as two individual signal wires and a common ground?
This is why I suspected an ecm / driver failure.
I currently have three of these 3412e's in operation daily and have had relatively little wrong with them other than random injector failure until now.
My only disappointment is not being able to find relevant trouble shooting, repair and support information that doesn't require a substantial investment at the dealer.
We keep multiples of spare parts on the shelf since we have several in operation.
I will follow up tomorrow with engine serial number and I may swap an ECM in off another similar unit.

 

[Cmark]

Yes, you're right. Each pair of injectors shares a common ground.
Attached schematic for those playing along at home.

 

[Nige]

It would be useful to have some history as to how the problem started. Has it been developing for a while or did it appear suddenly..?

My first thought with these symptoms is a wire harness break for cylinders 6 & 8 (middle two cylinders right side), most probably in the ground side as it's common to both.

As you haven't said whether or not you did any resistance tests on the injector harness, that would be my first suggestion.
Try ringing out the F729 Green wire for continuity right back to the relevant injector ground connection (Pin 27 on connector J-2) at the ECM. You could also try swapping the injector harness from that pair of cylinders to another pair just to rule that out. The injector harness for all 6 pairs of cylinders are identical.

One thing that jumps out and hits me is the splice for the grounds of the injector return. The ground side of each pair of injectors has a splice buried in the harness where the wires from the two separate injectors become one before that single wire continues to the ECM. I've had issues with dry joints on wire splices in engine harnesses of a similar design. You might also try doing the same continuity test while someone pushes & pulls the main harness around to see if anything changes electrically.

I'd do all the above before even contemplating trying it with a different ECM as a test. Please be aware that each ECM contains fuel setting adjustment specific to the engine on which it's installed so even though you may have a number of identical engines all using the same Part Number of ECM with (very likely) the same software installed in all of them the fuel settings for each individual engine will be different. Think of it as an electronic version of adjusting a mechanical fuel injection pump. There's nothing wrong with swapping an ECM from another engine as a test but it shouldn't stay swapped, if that makes sense.

The Part Number and Serial Number of the ECM indicate that it is a Reman component therefore it's not original for that engine.

 

[thepumpguysc]

Not knowing anything about this partic. mdl engine.. I was thinking along those same lines as Nige..
Doesn't the injector have to be "entered" into the computer?? So simply swapping inj's wouldn't be advisable w/o entering the specific info into the ecm..
& the wiring harness that connects the inj.'s would be suspect..

 

[Nige]

That's why I was suggesting eliminating all the possibilities that it could be a harness (either the engine harness or the harness specific to that pair of injectors) before proceeding further.
No E-Trim Codes on the injectors on HEUI engines like this one. Injectors can be swapped ad nauseum without having to do any reprogramming. Fortunately we've long since gone past the days of Prime & Non-Prime injectors for HEUI engines, there should only be the one type (Non-Prime) out there.

 

[jon damon]

Thanks for you input here's some more info
ESN 4cr01048 Arrangement #131-4016
I checked for harness issues continuity in the wrong places etc. found nothing .
Problem according to staff that has been operating machine stated all was fine until start up at beginning of the day .
No obvious degradation of sorts recognized during prior days use.
I have now swapped a spare ECM and the engine runs fine on all cylinders.
It appears the ECM is having an internal problem.
Nige the cylinders effected are 5-7 opposite bank from ECM mounting.
Since this is a reman computer and the machine is 19 years old are there additional thing I should do or look for externally other than the obvious harness failures that would could this failure ?
Would a shorted injector coil cause the ECM driver to fail?
I had read that Cat had protection circuitry in the ecm that would detect such a failure and deactivate the energy to the affected injector.
I want to minimize and potential reoccurrence of this sort of thing $$$$ is there an ohm range for the injector coils and relevant sensors that could cause failure.
Are there any legitimate resources for ECM repair other than CAT that know there stuff?

 

[jon damon]

I apologize for typos above

 

[Nige]

The ECM could have an internally fried circuit that feeds those two cylinders, or the possibility exists that it didn't shut down correctly the last time it was de-powered. It may be possible to open up the ECM and replace the faulty component. I realize that the ECM is "sealed" but if you put your nose next to the square plugs where the machine harness connects can you detect a burnt smell..? I don't know of any legitimate sources to repair an ECM but maybe one of our US members could suggest something.

ECMs do sometimes fail. I wouldn't have thought that a failed injector or a harness failure would cause an ECM to fail internally. Certainly I've never seen that personally as far as I can recall in 20 years around HEUI engines.

Personally my next move would be to get Cat ET hooked up to the failed ECM and see whether it communicates via the Data Link. I've seen weird things happen to ECMs between an engine being shut down and fired up again. You have to remember that the eCM is a dedicated computer but if it loses part of the operating system when it shuts down it has no self-repair facility or start up in Safe Mode like you have on your PC. It either works or it doesn't.

Back to what I posted above, and a question. Where did the replacement ECM come from that is now installed..? Because I can guarantee that it will have different Full Load & Full Torque fuel settings in it to the one that was removed. FYI the settings for engine 4CR01048 are -4 & -14. The different settings in the replacement ECM won't stop the engine running on all 12 but I wouldn't suggest running it like that for long.

 

[jon damon]

The ECM is off a similar application but at a significantly lower horsepower setting.
The engine serial number for the donor ecm is 4cr01105. AR128-2326.
The original unit is a 990 hp platform and the donor is a 750 horsepower platform.
I wasn't planning to try and utilize the donor for anything but troubleshooting.
I wasn't concerned going to the lower HP ECM since cooling etc is set up for significantly higher power level.
If I were going in the opposite direction 990 ECM on 750 platform I would have had greater concern.
There appear to be several rebuilders in the US most only appear to do truck related ECMs but I will inquire as to if Industrial poses a difficulty.

I really do appreciate the assistance. Thank You

 

[jon damon]

Any other insight would be appreciated.

If there are any sources of CAT ET's available for sale or a suitable option for these industrial engines I am interested in purchasing one.
I have been borrowing one randomly and I hate doing that but we have several of these units now and I think it is time to purchase.
Thanks

 

[funwithfuel]

You might wanna do a search on that. I think quite a few guys are using Texa software on a loaded laptop while other guys are using CAT ET from unsupported third party.

 

[Nige]

Regarding your last question HEF is littered with threads posted by people who have bought copies of ET through non-official channels and can't get them to work correctly. I'm not saying there aren't options out there that will work, just be careful.

Back to your two engines. 4CR01048 is specced for 990BHP @ 2100 RPM with 2970 ft.lbs of torque @ 1400, software Part # 154-4648. The corresponding figures for 4CR01105 are 750BHP @ 2100 and 2250 ft.lbs. @ 1400, software P/N 130-5809. going through the Parts Manuals for the two engines a thought struck me. Is the failed ECM the old type with the rectangular 4-bolt cover directly underneath the two plugs for the wiring harness where the software module actually plugs into a socket on the motherboard..?

 

[Nige]

This is the cover I'm talking about ....... If you have this type of ECM under the cover will be a plug-in module approx 1.5" long x 1" wide loaded with the software.

 

[jon damon]

The failed ecm is the reman and it does not have the 4 bolt cover. The 750 hp ecm may have the cover it is the original ecm from new I think.
I will check for certain in the morning.
I think I know where you were going though possibly just swapping the software eprom?
The failed reman unit appears to be two piece that just unscrews the two halves.

 

[jon damon]

I will say the timing and fuel curve program differences are amazing though.
The difference in the motors tone and response on the 990 hp engine vs the 750 hp is immediately noticeable.
The 990 has always been crisp and rapid to accelerate compared to the 750 and 860 hp versions on my other machines. Although the 860 is somewhere in the middle..
While I still have a couple mechanical 3412s which have been some of the most reliable engines I have ever had. One at 650 hp and 800 hp on the other.
Most of the issues we had from any of these engines is from operators neglect.
We bought all of the ones we have used and some of these machines appeared to have not had proper care for quite sometime.

The mechanicals never seemed to respond anything like the 990 3412e though.

One thing I have noticed which is particularly noticeable on one of the 750 hp machines is that when it is running on the cold start side of it's programming it hunts almost like it is experiencing a correlation problem.
Is this commonplace or is it possibly pointing to something else?

 

[Nige]

The failed ecm is the reman and it does not have the 4 bolt cover. The 750 hp ecm may have the cover it is the original ecm from new I think.
I will check for certain in the morning.
I think I know where you were going though possibly just swapping the software eprom?
The failed reman unit appears to be two piece that just unscrews the two halves.

Swapping the eprom if possible between the two ECMs would have been quite an elegant solution. It's what we used to do all the time back in the day. It drove Cat nuts ...... and is probably part of the reason for the new version of the ECM.
The failed ECM is the newer externally-programmable version. The 10R-4088-03 Part Number indicates that it was manufactured some time after Nov 2010. If you were to remove those screws and open the case it would no longer qualify as a Reman Core and that is going to cost you $. IIRC the price of a Reman is about half that of a brand-new ECM. Currently your non-working ECM would be good as a Reman Core for a replacement ECM. The replacement would be blank so a dealer would have to flash the software and load the setting into it corresponding to the engine on which it was going to be installed. The latest version or 10R-4088 is -04, a new ECM would be a 156-7172.
Back to my question regarding smell. Does it smell burnt when you put your nose next to the 2 x 40-pin plug sockets..?
Bescuase if it doesn't I go back to what I said earlier about hooking ET up to it to see whether or not it communicates. I've had success before re-flashing (reloading) the original software when it got corrupted or lost data during a shutdown. Note that this will NOT work for an ECM that has a burned-up driver (back to that smell busniess again). When it comes to flashing software only Cat Dealer versions of ET will do this, the Customer version will not, so it's something you'd have to take it to a dealer to have done.

 

[Nige]

One thing I have noticed which is particularly noticeable on one of the 750 hp machines is that when it is running on the cold start side of it's programming it hunts almost like it is experiencing a correlation problem.
Is this commonplace or is it possibly pointing to something else?

It could possibly be nothing more than someone has activated Cold Mode in the engine software configuration. Until the coolant is up to a certain temperature (65degC from memory) the engine will only hit on as many cylinders as it needs to maintain RPM under no load. As soons as it warms up it should hit on all 12. Only a poke around with ET would determine if there is a problem or not.