Puzzling BRM issue - car turns off after 3 min.

[aNUT]

Resolution

I have a car in the shop I'm having some trouble with. It's an '06 Jetta TDI with manual transmission, engine code BRM.

The customer reported that it died on the highway driving home after work. After waiting a few minutes, he restarted it. It drove absolutely fine for 3 minutes, then died again. He repeated this process until he got home. After a cursory inspection for obvious problems he brought it to me. He determined that the battery connections are tight, the battery has good surface voltage, and that the alternator charges the battery at ~14V.

I immediately duplicated the customer's complaint, however, it appears that the car shuts down, rather than dies. The EGR throttle/ASV closes as the car cuts fueling just like if the ignition were turned off. After waiting ~5 minutes, the car starts, runs, and drives normally for another 3 minutes.

After the car turns off, the MFD stays lit, but when I cycle the key to restart, the engine does not crank, and the MFD stays blank. The only activity on the instrument cluster when the car is in No-Crank-No-Start (NCNS) mode is that the DRL light is lit when the parking brake is off, and the backlighting on the cluster is on if the headlight switch is on.

I also found that the engine need not be running for the car to enter NCNS mode. If I switch the ignition to 'on' but do not start the vehicle, after 3 minutes, the car will not crank.

I can communicate with any module in the car when it's working with VCDS, but cannot establish communication with anything when the car is in NCNS mode.

There are no fault codes stored.

I have verified that: the battery has voltage and terminals are tight. I have verified all accessible grounds in the engine compartment.

For those with manuals; I have verified that the following fuses are intact:
SB6 SB10 SB13 SB15 SB17 SB23 SB25 SB26 SB27 SB28 SB29 SB30 These are located in the box next to the battery. I have also verified fuses SC1 SC15 SC25. These are located in the box mounted into the side of the dash.

I have inspected the plenum area where the ECU is housed and found no water or evidence of rodents.

I have inspected the wiring for the cam and crank sensors, and found them to be intact.

After starting the car and waiting for it to shut down, I found no excessive temperatures on any of the relays; both under dash or on the fuse box next to the battery.

I'm a little stumped as to what to investigate next, and any help would be appreciated. Obviously, we suspect some sort of current overload, but are at a loss as to where to look next.

 

[DanG144]

I would track back to the power module. The equivalent of the 109 relay.

I will try to do it as well, with the Bentley.

 

[DanG144]

Could you post the build date? '06 saw a change in schematics.

 

[PDJetta]

Is there an X Relay, or load reduction relay, in this car? if so I wonder if its acting up, or if some main power realy is.

Wonder if it could be a bad ECU.

Things like this will keep you from sleeping at night. I am curious as to what it is too.

--Nate

 

[DanG144]

It looks to me as if it could be J317 power supply relay.
It has 458 painted on it.
It is aft of the battery in panel A on some cars, and in panel B above the driver's footwell for others.
That is what I would try first, pulling it and doing a visual and smell test. Cold solder joint, coil heating up.

 

[aNUT]

I very much appreciate the help guys.

I haven't spent more time on this today, but hopefully will have an opportunity later this afternoon.

Dan - The build date is 03/06.

This particular example has the J317 inside the car as opposed to in the engine compartment. I have another '06 here, and I'll try swapping interchanging relays between the two if they appear compatible.

 

[oilhammer]

Are there any faults in any of the other modules? Usually if the engine controller shuts down when it shouldn't, some other modules will flag a 'no communication' fault. Check the Central Electronics, ABS, M-unit (if DSG), SRS, CAN, etc.

If you cannot communicate with anything when it is acting up, I would suspect a CAN/Cluster problem.

 

[aNUT]

Again - many thanks.

I did a little more investigation yesterday with VCDS.

The car is equipped with the following modules:

00-Steering Angle Sensor -- Status: OK 0000
01-Engine -- Status: OK 0000
03-ABS Brakes -- Status: OK 0000
08-Auto HVAC -- Status: Malfunction 0010
09-Cent. Elect. -- Status: Malfunction 0010
0F-Digital Radio -- Status: Malfunction 0010
15-Airbags -- Status: OK 0000
16-Steering wheel -- Status: OK 0000
17-Instruments -- Status: OK 0000
19-CAN Gateway -- Status: OK 0000
25-Immobilizer -- Status: OK 0000
42-Door Elect, Driver -- Status: Malfunction 0010
44-Steering Assist -- Status: OK 0000
46-Central Conv. -- Status: Malfunction 0010
52-Door Elect, Pass. -- Status: Malfunction 0010
56-Radio -- Status: OK 0000
62-Door, Rear Left -- Status: OK 0000
72-Door, Rear Right -- Status: OK 0000
7D-Aux. Heat -- Status: OK 0000

The modules in bold will not communicate with VCDS when the car is in No-Crank-No-Start mode (NCNS).

I also found that the key need not be switched off for the car to 'reset' after entering NCNS mode. In other words, if I switch the ignition on, after ~3 minutes, there is a click from the dash, and the car will not start. With no action on my part, after ~5 minutes, there is another click, the MFD will light up again, and the car will start. The click is the same click that normally coincides with switching the ignition on.

After clearing all fault codes; the following codes remain: It should be noted that I currently have the headlight switch and interior dimmer switch removed, so related fault codes should be disregarded.

Address 08: Auto HVAC Labels: 1K0-820-047.lbl
Part No: 1K0 820 047 EC
Component: Climatic PQ35 090 0505
Shop #: WSC 00000 000 00000
VCID: 69D2559C3905
1 Fault Found:
00457 - Control Module for Network (J519)
013 - Check DTC Memory
-------------------------------------------------------------------------------
Address 09: Cent. Elect. Labels: 3C0-937-049-23-M.lbl
Part No SW: 3C0 937 049 D HW: 3C0 937 049 D
Component: Bordnetz-SG H37 1002
Revision: 00H37000 Serial number: 00000000276476
Coding: 14018E234004150000140000001400000008730B5C
Shop #: WSC 20111 444 01184
VCID: 2A50EA90FC83
Part No: 1K1 955 119 C
Component: Wischer VW350 013 0402
Coding: 00063445
Shop #: WSC 20111
3 Faults Found:
02394 - Parking Light Left Front (M1)
012 - Electrical Fault in Circuit
Freeze Frame:
Fault Status: 01101100
Fault Priority: 2
Fault Frequency: 1
Reset counter: 99
Mileage: 134895 km
Time Indication: 0
Freeze Frame:
ON
Voltage: 11.75 V
ON
ON
OFF
OFF
OFF
01800 - Light Switch (E1)
008 - Implausible Signal
Freeze Frame:
Fault Status: 01101000
Fault Priority: 3
Fault Frequency: 1
Reset counter: 99
Mileage: 134895 km
Time Indication: 0
Freeze Frame:
ON
Voltage: 11.75 V
ON
ON
OFF
OFF
OFF
02195 - Dimmer Switch Instrument Panel & Switch Lighting (E20)
009 - Open or Short to Ground
Freeze Frame:
Fault Status: 01101001
Fault Priority: 5
Fault Frequency: 1
Reset counter: 99
Mileage: 134895 km
Time Indication: 0
Freeze Frame:
ON
Voltage: 11.75 V
ON
ON
OFF
OFF
OFF
-------------------------------------------------------------------------------
Address 0F: Digital Radio Labels: 8E0-035-593-XM.lbl
Part No SW: 8E0 035 593 E HW: 8E0 035 593 E
Component: SDAR XM H03 0080
Revision: 00000000 Serial number: AUZ4Z7F4042821
Shop #: WSC 00000 000 00000
VCID: 2B5EEF94F379
1 Fault Found:
00003 - Control Module
014 - Defective
Freeze Frame:
Fault Status: 01101110
Fault Priority: 5
Fault Frequency: 1
Reset counter: 40
Mileage: 10191 km
Time Indication: 0
Date: 2000.00.00
Time: 10:58:15
-------------------------------------------------------------------------------
Address 46: Central Conv. Labels: 1K0-959-433-MIN.clb
Part No: 1K0 959 433 AF
Component: KSG 0401
Coding: 137803281B09058F880F048051A0
Shop #: WSC 20111 444 72591
VCID: 74EC34E86EC7
Component: Sounder n.mounted
Component: NGS n.mounted
Component: IRUE n.mounted
2 Faults Found:
00457 - Control Module for Network (J519)
013 - Check DTC Memory
01312 - Powertrain Data Bus
013 - Check DTC Memory
-------------------------------------------------------------------------------

I will try to make some sense of this later this afternoon after the stream of people and cars dies down...

 

[oilhammer]

I'd be curious to find out where that 'click' is coming from. It might be the load relay that kills power to things during cranking (although you'd think that would not let the engine stop running once it had been started).

 

[jcrews]

Make sure terminal 15 isn't dropping at the ign switch first. Backprobe or tone it out.

Can you see terminal 15 dropping in the central electronic measuring blocks?

 

[VLS_GUY]

The fact that you have a loss of communications to several controllers at once, have had a fault code set for a CAN BUS failure (1312) and have a 0457 Electrical Control System Control Module failure code at at least two separate controllers points to either a CAN BUS wiring problem or the Electrical Control System Control Module(J519) is toast. The J519 is under the instrument cluster right on the relay carrier.

 

[40X40]

Cool (interesting)problem and I am stumped.

I'll suggest you try what the dealer mechanics do when they are stumped and throw parts at it..... starting with the ignition switch.

Good luck and for heavens sake keep us informed!!!!

Bill

EDIT: Load shedding is triggered by low voltage, could the car be sensing a low voltage situation that is somehow self correcting/resetting?? Swap in a known good battery and check and clean all grounds. (you probably already did those things....)

 

[oilhammer]

The fact that you have a loss of communications to several controllers at once, have had a fault code set for a CAN BUS failure (1312) and have a 0457 Electrical Control System Control Module failure code at at least two separate controllers points to either a CAN BUS wiring problem or the Electrical Control System Control Module(J519) is toast. The J519 is under the instrument cluster right on the relay carrier.

I am leaning towards the Electrical Control Module as well. That part has been superceded/updated multiple times since that car left the factory. But I gotta warn you, the soft coding will not transfer over to the new module... you may want to actually go through the long coding helper in VCDS and jot down a few key things, because recoding the new module can be a pain since all the boxes/options are moved around and have different values needed. Been there, done that...

 

[DanG144]

The driver's door wiring is a common place to have CAN bus wiring issues on these cars. The original loom is a few mm too short.

Replacement looms are reasonable in cost (in my opinion) and easier than using multiple repair wires.

In the few I have done (or watched someone do) I was amazed at how many broken wires and how much insulation was missing. It surprised me that there were not more codes and issues.

 

[oilhammer]

Good point Dan, but it seems that only causes a CAN-bus failure to go to sleep, which usually just causes a dead battery.

 

[DanG144]

Brian,
I am a bit ignorant in these matters. The battery drain is indeed what usually brings me to the door wiring.

But looking at the wiring for the can bus, it looked as though if the wires were shorted, instead of broken, that they could interfere with the communication with other devices. As I indicated, I am not very well versed in communication bus issues, so I could be off the mark.

 

[oilhammer]

I am not sure a shorted CAN bus would cause this, more it would cause the CAN system to revert to 1-wire (backup) mode. But usually, the door harness wires that break are the large B+ and GRD wires, not the CAN bus wires. However, since that module cannot power up properly, it never sends an "all clear" to the central module... at least that is what I think is happening.

Still, in any event, aNut can easily pull the boot back and take a peak... if nothing else to rule that out.

 

[runonbeer]

subscribed.

 

[VLS_GUY]

The type of failure I am thinking about here is the "babbling idiot" problem that affects low level serial interfaces of all types. In this failure mode a controller like the J519 in this problem continuously transmits high priority messages and monopolizes the network. This does not allow other nodes to establish communications with each other and communication error codes are set as a result.
Every CAN BUS has to deal with this problem. In fact if the the codes are transitory and have no functional impact fault code 00457 can be ignored according to VW. Good CAN Bus network designers just try and make sure they design networks with extremely low probability of the "babbling idiot" problem on normally functioning hardware. For more on the problem a good paper: http://paws.kettering.edu/~jpimente/flexcan/EFTA05_FlexCAN-paper.pdf
I would like to connect a CAN BUS analyzer to this car to see what is actually going on. I think something like this would be handy to have on modern cars:http://www.microchip.com/stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=1406&dDocName=en546534 Note the tools for designers that come with this tool.

 

[aNUT]

Resolution -

Faulty J329 power supply relay. This relay is located (on this particular example) on the bottom left corner of the J519 Vehicle Electrical Control Module. (It would be the hardest one to get to...)

The printing on the relay (again, in this example) is 460; part number 1K0 951 253A.

Many, many, thanks to all who devoted time to finding a solution to this issue.

 

[40X40]

Thank you for the learning opportunity! Nice job, BTW.


Bill

 

[whitedog]

aNUT, thanks for the heads up to this one. It certainly goes in the oddball problems for no start. What I really would like to know is the path that got you there.

 

[aNUT]

aNUT, thanks for the heads up to this one. It certainly goes in the oddball problems for no start. What I really would like to know is the path that got you there.

Based on the behavior of the car, I suspected a current overload. Fault codes indicated an issue with the J519 vehicle electrical control module. Additionally, it was plausible that the other symptoms like CAN system issues could be related to a J519 issue. I systematically removed relays one at a time from the J519 to determine which ones were necessary for running the engine. When I removed the J329 power supply relay, the car behaved as though it was in NCNS mode. I substituted the J329 from another BRM car, started the problem car, and it ran until I shut it off 30 minutes later.

 

[aNUT]

aNUT, thanks for the heads up to this one. It certainly goes in the oddball problems for no start. What I really would like to know is the path that got you there.

Rather; I stuffed a few hundred dollar bills into the end of my Magic Guru Wand, shook it vigorously at the ECU, blew three kisses, and all was well.

Hopefully He-Who-Should-Not-Be-Named will not return before the cam is smoked

 

[DanG144]

Great Job.

I was trying to research from the Bentley and the wiring diagrams. They were of minimal help.
It does show the J319 on a couple of drawings, but it shows no current path through it.
The only real hint of its purpose was on figure 44 page 97-43, which does indicate that it is a (terminal 15) relay. This would indicate that it is probably a power multiplier for the ignition switch on.

I admire your troubleshooting method.

 

[oilhammer]

aNut, good job! I figured it must have been related to that module somehow, as that was the only thing that has a central control over a bunch of items. Glad you found it, and thanks for reporting back! I'll file this in my things to watch out for section in the ol' noggin'

 

[whitedog]

OK, so you knew that it was a current overload because it seemed to get hot, then stop working after a consistent 3 minutes, then after five minutes it would be cooled off enough to work again. Since you could get the same NSNC with just leaving the key on, you could eliminate the fuel system and any mechanical things on the engine.

So I see how you got through that, but then there are the Codes ( I cleaned up to just the basics here):

Address 08: Auto HVAC Labels: 1K0-820-047.lbl

1 Fault Found:
00457 - Control Module for Network (J519)
013 - Check DTC Memory
-------------------------------------------------------------------------------

-------------------------------------------------------------------------------
Address 0F: Digital Radio Labels: 8E0-035-593-XM.lbl

1 Fault Found:
00003 - Control Module
014 - Defective

-------------------------------------------------------------------------------
Address 46: Central Conv. Labels: 1K0-959-433-MIN.clb


2 Faults Found:
00457 - Control Module for Network (J519)
013 - Check DTC Memory
01312 - Powertrain Data Bus
013 - Check DTC Memory

Three of them point to control modules, two of which specifically say J519. ( I wonder if the radio code is just missing that part of it's label?)

So from there, you were pointed to the J519 and from there it came down to trial and error, and finally a substitution.

I write this to help myself remember some excellent troubleshooting, so I hope I have it right.

 

[aNUT]

That was the general process.

The Guru Wand is an inside joke Runonbeer and I have. We only wish to be so mystical when we encounter stuff like this.

I knew I'd found the problem relay as when I removed it, the cluster went blank, except for the DRL indicator. I also had the luxury of having another relay to try, so my guessing didnt cost anything.

I cleared all codes, (there were a ton given all the stuff I had unplugged) and performed another auto-scan today. All the codes cleared except for the dimmer switch code in the central electrical module. I suspect that switch is bad since they fail so commonly, but I won't be chasing that now that the car works. The dimmer works for what its worth.

I would like to tip my hat to Dan as he really called it as far as a bad relay on the J519.

 

[DanG144]

You are too kind; I had the wrong relay.

 

[whitedog]

Sometimes a sign post points in the wrong direction, but it gets you thinking about directions and can unlock the brain to looking at things differently. That is why, I have been known to ask ladies about things just to help get my mind working differently. I have also found that it helps to tell someone about what is going on, but just do it in my mind.