'05 CDI "communication fault" N14/2 Glow Output Stage 2538-4 jay-bob?
 

I have a current code 2538-4 communication fault Glow Output Stage. I also have a few stored codes that my MB II will not erase because of a communication fault. Can I assume both are caused by the same thing? There are other modules that I access and can clear codes just fine, but the engine module is not letting my MB II communicate. My STAR computer is not working now, so the MB II is all I have. How about it jay-bob, you got any ideas?

On the 648 (and 642) the Glow Plug Relay talks to the ECU on LIN bus (single wire slow speed serial) and shares that line with the voltage regulator.

This is the routing of the data wire from the ECU to the glow plug relay and alternator.
All of these are 0.75 mm (or about 18 gauge American) solid blue according to my Star Finder

Pin 66 on the ECU, goes out to a Splice Z26/1, in the harness near the EGR valve
Branches to Pin 1 on the alternator regulator connector and Pin 12 on the N14/2 glow plug relay

On the GP relay multi wire connector
pin 1 cylinder 1 2.5 mm (similar to American #14) black/blue
pin 2 cylinder 2 2.5 mm black/violet
pin 3 cylinder 3 2.5 mm black/red
pin 4 cylinder 4 2.5 mm black/yellow
pin 5 circuit 87 (protected engine management logic power) 0.5 mm (#20) red
pin 6 circuit 31 0.5 mm brown
pin 7 cylinder 5 2.5 mm black/green
pin 8 cylinder 6 2..5 mm black/white
pin 9 , 10, 11, no connection
pin 12 LIN bus 0.75 mm blue

The big connector with single wire on the GP relay is circuit 30, straight battery feed, via a 10 mm (or about 6 gauge American) via fuse 69, 125 Amps, in F32 (the 'front prefuse' box in the passenger foot well behind the shielding plate).

Thank you! I just replaced the alternator, so could that be the problem? The car did not have this issue before the alternator was changed. The car had a parasitic draw, and that was the last thing that could be the culprit after all the other systems were ruled out by millivolt test on fuses. Turns out, that was the problem.

Well I did have an issue with a NAPA replacement alternator having a dead LIN bus chip in it. Maybe the alternator regulator is sucking down the LIN bus...

I got a Genuine Bosch rebuild from my Euro parts source (S&J Euro Parts). It came in a new Bosch carton, and looked new.

About an hour ago, I removed everything so I could get to the alternator to check the plug. It all looked good, with everything connected properly. I scanned for codes again, and got the same message when checking "CDI 3 Common Rail Diesel Injection 3" I got- "Fault in communication with ECU"
I tried clearing codes in other modules, and they cleared just fine (mostly under voltage codes). I believe this is all related, but have no idea how to fix it.

EDIT: Also, the alternator is charging at 14.6 volts +-

The 'fault in communication with the ECU' error code isn't the same thing as the first error code saying the ECU couldn't communicate with the Glow Plug controller.

The 'fault in communication with the ECU' error is telling you your scan tool can't seem to interface/find the N3/9 main CDI control unit.

Was the ignition on when you scanned for codes? If not the scan-tool won't find the N3/9 module because it isn't powered up.

Yes the key was on, as I was able to clear codes in other modules. I was also able to see six faults in the CDI module, I just was not able to clear the codes. Four were stored, the other two were the glow plug controller (currant and stored). I know they are two different faults, but I can only assume they are caused by the same thing, as they both showed up after I changed the alternator. Could there be a fuse that is blown? Since the scan tool worked to clear codes in other modules, I assume at least the diagnostic circuit/fuse is good.

I'm confused.

In your first sentence above you indicated you were able to see 6 codes and erase 4 - this indicates at that time your scan tool was able to see/communicate with the N3/9 CDI main control unit. As I recall the glow-plug output stage error is from the CDI control unit?

However your screen shots in post #5, after you disassembled things to verify that the plug to the alternator was all good, indicates your scan tool can no longer see/communicate with the N3/9 control unit. So at this time you can't see any codes until you can get your scan tool back into communicating with N3/9.

If the car starts and runs you have power to the CDI control unit. If the CAN wiring to N3/9 is good your scan tool you should be able to communicate with N3/9. You have to get the communication working before you can really continue.

Given the LIN bus is apparently shared with the Glow plug output state and the alternator, and the error appeared immediately after alternator replacement, it seems kind of likely that the new alternator doesn't understand its supposed to communicate with a LIN bus and has done something to take down the bus.

As Jay-bob said in his post he and others have replaced alternators with NAPA sourced remans and had issues. I realize your Bosch alternator should be of higher quality but I'd bet there are a couple of different versions of that alternator that share the same basic case and maybe you didn't get the version that is LIN capable?

You need to figure out why your scan tool isn't able to communicate with N3/9 - I don't think that problem has anything to do with the new alternator or the LIN bus. If you can get N3/9 back into communication I'd disconnect the plug to the alternator, which would isolate that LIN bus connection, and see if that brought the LIN bus back on-line which should erase the Glow-plug output state error in the CDI control unit.

Thank you Tim for the great information!

"In your first sentence above you indicated you were able to see 6 codes and erase 4 - this indicates at that time your scan tool was able to see/communicate with the N3/9 CDI main control unit. As I recall the glow-plug output stage error is from the CDI control unit?"

I was not able to erase any codes in the CDI control unit, I only get the "fault Communication with ECU". Yes the glow-plug output stage error is from the CDI control unit. It had 2 codes that were stored and current. The other 4 were stored. My scan tool is able to communicate with all the modules, and will read and clear all codes except for the CDI module, it will not clear codes. That is the part that is baffling me. The car starts and runs, battery charging at 14.6 volts, but has a CEL and glow plug light on constantly.

Have you tried to change out the GP controller to eliminate hardware issue first?

I still don't think I'm getting my point across.... Let's try it this way.

Right now, today at this moment can you communicate with the CDI control module N3/9? Can you get a menu that shows you options to take to read codes or look at live data in the N3/9 module?

I have a couple of aftermarket scan tools that I use as a backup to my SDS system, its not at all uncommon for those tools to not communicate with some of the control modules in the car. I've also seen them communicate with a module on one day and not communicate on another day.

Remember that the path for communication to the ECU goes like this:

OBDII port -> ZGW (Central Gateway) -> Engine CAN bus -> N3/9 ECU

And from the ECU to the N14/2 Glow Plug it is a LIN bus that leaves the N3/9 ECU and goes to the N14/2 Glow Plug and the G1 Alternator regulator.

So we have 2 things that have to work.
First the scan tool has to see N3/9 through the central gateway via the Engine CAN bus.
Then assuming it can see N3/9 and communicate with it, then the LIN bus has to be operating for the N3/9 to communicate with N14/2.

You can't directly talk to N14/2 with the scan tool, you can only get get N14/2's status as N3/9 has determined.

So if the LIN bus is down then you're not going to see the N14/2.

Another thing to try: get in, close the door, put key on position 1 (just the radio on).
Navigate to the MPH display on the cluster.
Push in the dimmer knob 3 times rapidly, you should see Ub = 12.xx V
Start the car and watch the voltage. If it immediately goes to 14.x V then the regulator is not communicating and it's in default (make nominal voltage all the time) mode.
If the voltage walks up from 12.xx to 14.xx over the span of about 15 seconds then it is getting orders from N3/9 to walk the load in. Meaning the LIN bus is working.

The regulator part should be
Mercedes PN: 0031549706
Bosch PN: F00M145225
Others cause lin faults

You guys are awesome!
 

I'll say it again, you guys (jay_bob, TimFreeh,dieselbenz1) are awesome! Thanks for the great information!

I will try this today. I just put a brand new AGM battery in the car, and tested the charging system. I get about 14.6 volts with my probes directly on the battery with the engine idling. Yes that tells me that the alt is charging, but it may be in default mode as you say. Again, VERY GOOD information!

Yes, that is the first thing I did, and it changed nothing. I now have a new glow plug control in stock! (a good thing, as we have four cars that it fits between my brother's family and I).

Yes I can, and have been able to (with my Icarsoft MB II) establish communication with the CDI control module N3/9, and can view a list of current and stored codes (six total), but get a communication fault when I scroll down and highlight clear fault memory (second photo post 5). I will try jay_bob's test as described in post 12 later today. I have not tried to view the live data, so I will look at that as well.

video link
 

It seems to me that the alternator is communicating, at least that is the way it appears in this video; https://photos.app.goo.gl/Dz5empcYjqEHCmXt9
Although it only gets to 14.1 volts, and I did start it once before making the video, the voltage does raise gradually.

Also, I tried to scan codes again, and I could not even access most of the modules without getting a communication error. However, I was able to access the diagnostic section, and got a 930C and 930E.

Based on your video and Jaybob's excellent rendition on the N3/9 architecture it looks like your new alternator is indeed in communication with N3/9 and under its control.

But since the check engine light and glow plug light on the dash are still on I'm guessing there still is some sort of issues with communications between N3/9 and the glow plug controller N14/2.

Since the error with the glow plug controller happened at the same time your new alternator was installed I'm still thinking there is some sort of incompatibility between the new alternator and the LIN bus that is causing issues with the communication to N14/2.

But from the images in your last post it appears to me that your "MBII" scan tool is having problems communicating with your car at this point in time. Both errors seem to be telling you the MBII scan tool can't initiate communications to the car. I have no idea what a 'knowledge base error' and a 'synchronization of the DTC " means - do you have any documentation or guidance with the MBII scan tool that info on 930E or 930C error codes?

I don't think those codes are MB-specific, I think they're associated with the MBII scan tool?

At any rate until you can get the MBII scan tool back into communicating with N3/9 or come up with another scan tool that can you're kind of unable to proceed.

With engine running you have communication errors?
With engine not running you can see modules but not clear codes?

I am still doing research on 930C&E. I found this so far:
https://mbworld.org/forums/e-class-w211/695674-what-caused-error-code-930c.html

It may be a problem with my scanner, so I will take an hour (or more) today, and hook up my extremely slow SDS computer, and see if it (and I) can figure it out.

This problem was only the last time I used my MB II on the car. A couple days ago to make the video, I could access and clear data in every other module save the CDI module. I was able to see the DTC's, but not clear them in only this module. I then tried it the next day, and I got the 930 codes, and could not continue.

Stay tuned, I will get back to the car this afternoon, hopefully with more good info after scanning with my (extremely slow) SDS computer.

Should your scanner operate differently ie not be able to communicate with engine running vs engine not running and scanner working this would indicate? One major difference is the alternator is spinning putting electrical noise on the Lin and causing these scanner issues.

I don't think the engine running or not should have any effect on the scanner working or not. I often use it for live data, and that requires the engine to be running.

Consulting with my parts guy, these numbers are not correct for the '05 CDI. He looked up the proper part#, and they are backordered everywhere (his suppliers). I also looked on ebay, and these numbers are not for the CDI either (for C , ML etc).

Ooops didnt mean to misslead you my notes must be wrong.

No problem, I just want to make sure I get the right part. It is not an easy job to replace the alternator in this car. However, with the new information I have, it may just come to that.

New information
 

After waiting an entire seven minutes for my system to boot up (seemed like an eternity), I got my SDS to work. It showed much of the same information that my MB II did, but I was at least able to clear the stored codes. I also did an alternator test, and according to the SDS, it failed (view photos). It still has the 930 codes, and was no help with how to fix it. I have 11 photos of the screen, and at least 2 issues still remain. Although I don't want to believe it, the alternator test is pretty conclusive. I really do NOT want to R&R it again. Thanks again you guys for all your knowledge and help with this, and I hope the SDS screen photos will help in finding the problem with the car.
One question for the SDS experts: The last photo shows a list of events and codes, have all these happened to this car, or are these just "possible".

930 & others
 

I am not sure how to fix the problems, as the part of the SDS system that shows diagrams and repair info shows an "Internal Error...... file not found". The 930 code is still there, and the system will not let me go further without displaying the internal error message. This is probably a fault with my software ($99 hard drive from China). Please view photos, and let me know if you have any ideas. Thanks!

The long list of error codes are the entire set of possible errors with some additional information and troubleshooting sets. They are not indicating the error occurred, you're supposed to use them for specific testing of components like you did with the alternator.

I have no idea what the alternator test is supposed to show. There are voltage lines and current traces so I'd guess N3/9 is supposed to manipulate the voltage regulator via the LIN bus and see associated changes. I don't know if the high/low traces are supposed to be upper/lower limits? Not sure how this test is supposed to work? If N3/9 commands the alternator to ramp-up the alternator to 100 amp delivery where is the associated current supposed to go?

Agree it seems like your version of SDS doesn't like the results of the test.

I've seen the 'file not found' error before on many bootleg SDS systems. I'd guess its just missing data that was not correctly copied by whoever set up the system. Not sure its really fixable, way over my capabilities.

I went on the EPC and looked up the alternator and regulator part numbers for the 211.026 series... there are 4 possible different alternators and 6 different voltage regulators that can be used depending on model year and engine serial number. Kind of a mess.

Are you still getting the unable to communicate with the glow-plug module error?

Thanks for your knowledge and willingness to help. I am thankful that the list of error codes is NOT specific to my car!

What I gather about the alternator test is that the top and bottom black lines are the max. and min. limits, and the blue line being the results of the test. Once the test started, the RPM was raised by the computer, and about half way through it said to turn on headlights, rear defroster, and A/C with fan on high. On the lower graph, the blue line dipped below, so I assume that is how it failed the test. I might try doing the test on the other CDI (we have three and a Bluetec between my bro's family and I). I also might swap the voltage regulators as well. The car I am working on belongs to my niece, and she really needs it back. If the problem travels to my car, then I know the regulator is the problem. It is not all that difficult to just replace the regulator. Only the air cleaner box, and intake hose need to be removed. The alternator can then be removed and turned with the regulator facing up to change. To remove the alternator, the coolant needs to be drained, as well as removing the cooling fan and radiator hoses. And yes, my bootleg SDS must be missing something. It also gives me an error (Java?) when accessing the EPC. I am not sure if it is a software or hardware problem ($60 ebay Dell computer). I would love to have that part of the program, as I am always looking up part numbers. I sent an email to my parts guy with the VIN so he can get the correct part number for me. Just looking it up without the VIN, the part was not available from his source, but on backorder. If I had the correct part number, I might be able to find it. Thanks again, Rich

I don't think you can swap regulators between alternators without some knowledge of the configuration of the alternator and the regulator.

In the EPC I'm seeing that some regulators have to be used with certain alternators so mixing and matching might bring unpredictable or undesirable results?

If you send me the VIN I can lookup what the MB part number of the correct alternator and regulator is supposed to be for that VIN. I don't think that's really going to help much though because you have a Bosch re-manned alternator and there isn't any cross-reference in the EPC for Bosch re-manufactured products.

I would run the alternator test on one of your other CDI's and see if it passes. Given the flakiness of the bootleg SDS systems I'm not sure I'd take your SDS test generated results as gospel.

Are you still getting the N14/2 unable to communicate with the Glow plug controller error? If I were in your shoes I'd zero in on that issue before proceeding with other diagnosis. If the N14/2 error was still there I'd disconnect the alternator LIN wire and see if the N14/2 error went away. That would confirm that the new alternator isn't correct and is taking down the LIN bus communications with the glow plug controller.

Yes I am still getting the N14/2 unable to communicate with the Glow plug controller error. This is from my parts guy, I sent him the VIN:




A 004 154 03 06
* REGULATOR SWITCH
ALTERNATOR
Code: F211/F220;
Replaces: A 003 154 71 06
Replaces: A 003 154 59 06
Replaces: A 003 154 41 06



I will disconnect the LIN connector, and see what happens. I will also run the alternator test on my other CDI.

Agreed that it might not matter using the VIN to find the regulator for a reman alternator, but I can PM the VIN to you and see if you come up with the same info.

Glow plug control NOT the problem!
 

I disconnected the LIN connector and scanned for codes, ONLY ONE came up and it was for the alternator! I plugged it back in, checked for codes, and the glow plug fault came back! :)



I really don't want to pull the alternator again, but it looks like I may have to do just that. :mad:

Well that's some progress.

I put the VIN of my CDI in and the EPC indicates the Regulator is indeed part # 004 154 03 06.

There is also a note saying that this regulator should be used in conjunction with alternators 013 154 00 02 or 014 154 07 02.

If you want to PM me your VIN I can see of yours is the same.

BTW I think you could ignore the LIN interface error with the alternator and just run it as-is.

The ramp-up-ramp down functionally of the LIN bus is mainly a fuel-saving scheme where the alternator defers charging under some conditions when the engine is decelerating.

I bet the check-engine light doesn't even come on when the LIN bus to the alternator is down?

The plug has to be on for the alternator to even work (it has two wires). If I just deleted the LIN wire only, then yes it may not turn on the CEL or give the glow plug control code.

Your VIN comes back the the same part numbers as my VIN.

VR - 004 154 03 06

Used with Alternators
014 154 07 02
013 154 00 02

It appears the original alternator is the 013 # and it has been superseded by the 014 #

FIXED!!!
 

I got home from my vacation to find the Genuine Bosch voltage regulator in the mail box. I installed it by just removing the alternator from it's mounting, and then turning it with the regulator facing down. I was able to replace the regulator much easier this way. I put everything back together, and NO CEL or glow plug DTC!!!!! The bad regulator appeared to be a cheap Chinese copy, and NOT a genuine part. Thanks again for all your expert help!

UPDATE:
 

Well, that was short lived. After replacing the alternator, regulator and installing a new AGM battery, I thought the car was fixed. But after sitting for a little more than a week waiting for a new turbo cartridge to arrive and get installed, the battery was COMPLETELY DEAD! And I don't mean just "mostly dead" (guess the reference), I mean so dead that charging it over night only brought it up to six volts. Totally ruined battery. On top of that, with another battery installed, it is only charging (or should I say not charging) at 12.3v. My guess is that either the BCM is bad, or the alternator is bad again. I will be "coding out" (deleting/removing) both the Tele Aid and alarm system with my STAR SDS computer. I believe either one or both of these systems could be responsible for the parasitic draw. The Tele Aid system is obsolete, and the alarm is useless anymore (ignored by most people). I am hoping one or both turns out to be the cause of the draw, so I can have some closure. Stay tuned.......

Finally fixed!!!
 

After removing the alternator and sending back the regulator, I installed the replacement regulator (under warranty), and the car now charges the battery. I decided to have the alternator tested (at the 'Zone) before installing it. I found out that the CD player was in a tizzy, and was the source of the draw. I was told that the radio has not worked since about the time the battery started going dead. I removed both fuses (CD & radio), and the draw went away. The car sat for almost two weeks while waiting for the replacement regulator, and still was able to start. Once I put the fuses back in, opened the CD changer door, the radio came back and everything works! I used an amp clamp to monitor the modules gradually sleeping, and confirmed that the draw was gone.