[Thomachaan]

matteblue3er Thank you for the info!! I'll check out my battery draw over 45mins and see if it doesn't settle to 40mA or less. Just today I got an eBay used DME delivered, so checking these readings after install may make more sense (currently no BSD communication with IBS).

Am currently looking for someone within 100 miles of North NJ to clone my DME, otherwise gotta go with a cloning service off eBay.

EvaBmw I'm very much suspecting the same thing based on my codes and likely faults. After reading your post I watched a bunch of Youtube videos of the 6L45 and 6L50 GM transmissions. I have confidence to drop my trans pan now, something I avoided since it was serviced not too long ago at a dealership. Will remove the valve body, check my manual valve since that thing likes to break, and completely disassemble the gear position sensor (which might not be a hall sensor like some posts say). If it's a metal contact switch, I should see damage/dirt. Will use Valvoline MaxLife ATF, it's Dex VI rated and I already use it on a bunch of cars, so I can buy in bulk and do plenty of drain/fills.

It'd be weird to see something physically damaged in there because of an electrical fault. Unless the electrical fault caused a physical haywire reaction with solenoids and high pressure.

MightyMouseTech I changed that ground strap before this catastrophic fault. Used a Walmart 2awg battery wire, I'm not worried about it. Will double check with jumper wires, but the contact points are solid and should perform electrically better than the OEM braided wire.

[Thomachaan]

Finally used my oscilloscope! Tapped into the BSD line and saw what I was expecting based on bimmerlabs articles.

First Pic:
BSD signal is showing a constant 0.8v with nothing connected (waterpump, alternator, oil condition sensor, and IBS unplugged). I backprobed the DME violet wire (3 maybe?) which should be showing a square wave type signal bouncing between 0 and 12v. This constant 0.8v tells me that the chip in the DME responsible for sending and receiving communication on this line has shorted, but there's a diode between the signal and ground in the chip that keeps it from showing 0.

Second Pic:
Tapped into the PT-CAN high and low signals from the DSC. Unplugging the DSC and transmission didn't change the look of these waves. What the pic shows from top to bottom: examples of high speed CAN signals (just for reference, irrelevant), the red line is the math of adding my PT-CAN high and low together (Should be flat, cancelling each other out. Since it looks like a signal, my H & L are not perfect mirrors of each other), and the blue and yellow signals under that are the actual PT-CAN H & L measured.

The signals are there but they don't look quite right. Both high and low should be more square-like with no pointed peaks.

Third Pic:
I measured the resistance between PT-CAN H & L from the red and blue/red wires in the ECU box. Reads about 60ohms, which is correct because there should be 120ohm resistors in parallel between the lines at both the DSC and EKP (tells me that the initial electric arc didn't burn out those resistors or otherwise short the wires together somewhere).

This would have been my first post for this thread if I had an oscilloscope and knew how to use it back then. These readings are what allowed me to pull the trigger on buying a used DME to clone.

[Thomachaan]

Thank you everyone who helped with suggestions, direction, and information. OP is solved now, BSD line is working on my car thanks to a cloned DME.

I found a guy through eBay, Ivan, to do my cloning. He was local enough for me to drive to in CT. I'm in North Jersey, I would easily recommend him for cloning services, DME, CAS, whatever. PM for details.

[rjahl]

Originally Posted by Thomachaan

Thank you everyone who helped with suggestions, direction, and information. OP is solved now, BSD line is working on my car thanks to a cloned DME.

I found a guy through eBay, Ivan, to do my cloning. He was local enough for me to drive to in CT. I'm in North Jersey, I would easily recommend him for cloning services, DME, CAS, whatever. PM for details.

Thank you for reporting your success.

Too many post problems and never report their solutions.

[D733]

Hi Thomachaan, have the exact same problem but with n54. Have you tried updating the DME software before cloning your DME?

[Thomachaan]

Originally Posted by D733

Hi Thomachaan, have the exact same problem but with n54. Have you tried updating the DME software before cloning your DME?

I haven't tried DME software updates yet, but definitely gonna try at some point for that 330i power. What's the same exact problem you're facing, no BSD communication?

[D733]

Yes code CD9304 after valve cover job. Will send my DME for repair or cloning as well. Just wondered if you tried to update it before cloning it to restore BSD comm

[Thomachaan]

Originally Posted by D733

Yes code CD9304 after valve cover job. Will send my DME for repair or cloning as well. Just wondered if you tried to update it before cloning it to restore BSD comm

I've never seen an example of a software update fucking with BSD line communication. If all BSD components are unplugged (waterpump, alternator, oil level sensor, IBS) and there's still no square wave coming from the DME, the problem is physical (burnt out chips). Repair is possible, but the experts on our DME's charge too much. Easier to replace with a clone.

Good luck!

[fe7565]

Originally Posted by Thomachaan

Thank you everyone who helped with suggestions, direction, and information. OP is solved now, BSD line is working on my car thanks to a cloned DME.

I found a guy through eBay, Ivan, to do my cloning. He was local enough for me to drive to in CT. I'm in North Jersey, I would easily recommend him for cloning services, DME, CAS, whatever. PM for details.

I am glad to come across your thread. I am in a similar situation, but in a much smaller scale on my F30. Would help if you could give me some tips on diagnosing the PT-CAN (FACAN) bus. I do have an oscope and some decent electronics knowledge, but weak on bus signal comms.

I have two major error codes in ISTA+ that are very likely related. One says no communication with EKPS (fuel control ECU) and the other says no communication with GWS (gear shift switch). Both of these ECUs are on the same PT-CAN2 bus systems. At the same time it looks like GWS ECU is also on PT-CAN1[/B] and PT-CAN2. These two buses provide redundancy for the GWS, that's probably why gears work, but the ECU does not show up.

For the EKPS function, there is an emergency mode, whereas if EKPS does not communicate with the DME(?) or with FEM(?) that other ECU runs the low pressure fuel pump in emergency mode, thus bypassing the normal EKPS PWM input to the fuel pump in the tank. Which explains why the car runs perfectly in spite of the GWS and EKPS not present errors.

I suspect that the PT-CAN2 bus is faulty. I suspect that there is a bus error (open, short, bad component/wiring/connector) that affects GWS and EKPS.

The only reason I knew there was a fault, because I had a CID screen message about: "Fuel-System error: continued driving possible" that pops up at random intervals for a few seconds and the goes away (doing it for about two weeks).

I ran the diagnostics in ISTA for all the ECUs on the bus tree where DME, EPS, EKPS, GWS are. GWS and EKPS did not respond, neither in ISTA nor Esys.

Initially I thought the EKPS was bad and that's what ISTA pointed to too. But bought another (used) EKPS unit and that one also acts the same exact way. Does not even show up and could not communicate with it either. The other not responding ECU is the GWS (Gear switch). For this one I checked all the LEDs for P, D, R, M, S, etc and all work and all gears switch properly. This ECU seems to be on two separate PT-CAN busses (CAN1 and CAN2) and I think the data flows ok on PT-CAN1 bus that connects to the EPS and DME ECUs, but not on the PT-CAN2 bus that runs between the GWS with the EKPS.

Based on the wiring diagrams, the GWS does not have a terminator resistor. The DME does not show one either, but the FEM does and so does the EKPS between CAN-H and CAN-Low. When I measured the CAN-H and CAN-L pins of the isolated (all wires disconnected from the ECU) EKPS module, it was reading around 51 Kilo-Ohms not 120 ohms or 60 ohms. I did not check the replacement EKPS ECU yet for Ohms between the CAN-H and CAN-L pins before swapping out the old ECUs.

But, when I disconnected the new EKPS module from the CAN bus system, I measured 120 ohms between the CAN-H and CAN-L on the two can bus wires running to the EKPS (with battery power on => I know should have had it off). Isn't it supposed to be 60 ohms? I also followed the ISTA+ guide on how to check with an oscope the bus signals, but did not get smarter on what's going on (see posted). I did not get nice square signals, even as I set to 10us/div and 1V-DC ranges as directed.

[Thomachaan]

Originally Posted by fe7565

I am glad to come across your thread. I am in a similar situation, but in a much smaller scale on my F30. Would help if you could give me some tips on diagnosing the PT-CAN (FACAN) bus. I do have an oscope and some decent electronics knowledge, but weak on bus signal comms.

I have two major error codes in ISTA+ that are very likely related. One says no communication with EKPS (fuel control ECU) and the other says no communication with GWS (gear shift switch). Both of these ECUs are on the same PT-CAN2 bus systems. At the same time it looks like GWS ECU is also on PT-CAN1[/B] and PT-CAN2. These two buses provide redundancy for the GWS, that's probably why gears work, but the ECU does not show up.

For the EKPS function, there is an emergency mode, whereas if EKPS does not communicate with the DME(?) or with FEM(?) that other ECU runs the low pressure fuel pump in emergency mode, thus bypassing the normal EKPS PWM input to the fuel pump in the tank. Which explains why the car runs perfectly in spite of the GWS and EKPS not present errors.

I suspect that the PT-CAN2 bus is faulty. I suspect that there is a bus error (open, short, bad component/wiring/connector) that affects GWS and EKPS.

The only reason I knew there was a fault, because I had a CID screen message about: "Fuel-System error: continued driving possible" that pops up at random intervals for a few seconds and the goes away (doing it for about two weeks).

I ran the diagnostics in ISTA for all the ECUs on the bus tree where DME, EPS, EKPS, GWS are. GWS and EKPS did not respond, neither in ISTA nor Esys.

Initially I thought the EKPS was bad and that's what ISTA pointed to too. But bought another (used) EKPS unit and that one also acts the same exact way. Does not even show up and could not communicate with it either. The other not responding ECU is the GWS (Gear switch). For this one I checked all the LEDs for P, D, R, M, S, etc and all work and all gears switch properly. This ECU seems to be on two separate PT-CAN busses (CAN1 and CAN2) and I think the data flows ok on PT-CAN1 bus that connects to the EPS and DME ECUs, but not on the PT-CAN2 bus that runs between the GWS with the EKPS.

Based on the wiring diagrams, the GWS does not have a terminator resistor. The DME does not show one either, but the FEM does and so does the EKPS between CAN-H and CAN-Low. When I measured the CAN-H and CAN-L pins of the isolated (all wires disconnected from the ECU) EKPS module, it was reading around 51 Kilo-Ohms not 120 ohms or 60 ohms. I did not check the replacement EKPS ECU yet for Ohms between the CAN-H and CAN-L pins before swapping out the old ECUs.

But, when I disconnected the new EKPS module from the CAN bus system, I measured 120 ohms between the CAN-H and CAN-L on the two can bus wires running to the EKPS (with battery power on => I know should have had it off). Isn't it supposed to be 60 ohms? I also followed the ISTA+ guide on how to check with an oscope the bus signals, but did not get smarter on what's going on (see posted). I did not get nice square signals, even as I set to 10us/div and 1V-DC ranges as directed.

Hey so I think CAN-H & CAN-L are each 60ohms to ground, so 120ohm between the two.

The signals in your image look okay to me, if I'm looking at it right. I saw weird distortions in mine (spikes, slants, fuzz) which mostly went away after replacing the DME.

Something about your description makes me think you could have an issue with your gear selector switch in the trans (I'm assuming yours is an automatic). I also had weird fuel pump codes showing up. Never a bad idea to redo the thermal paste in the EKPS with Artic Silver or something, you've got two to play with now 😁

Ensure your battery & connections are solid, a lot of times a battery that's supposed to be good somehow fails, and a weak battery makes communication issues. Monitor the battery voltage while the car is running, if it goes higher than 14.2v, even intermittently, you might wanna replace the voltage regulate in the alternator.

Another suspect would be your JBE gateway module, or whatever it's called on the F series. I get communication error codes in my car that are never currently present, so they just intermittently microtrip.

Direct me to your thread on your issue, I wanna see how things turn out.

[fe7565]

Thank you for the quick reply and the advice.

The scope readings/guidance above I got from the ISTA guide and those are samples. I took an actual photo of my own module the oscope with CAN-H and CAN-LOW each to GND. But it looked so irregular that I thought I mucked soemthing up as far as scoping it correctly.





I need to go back and do a better job on the oscope today.


Will post updates here: https://f30.bimmerpost.com/forums/sh....php?t=1991628


I just watched a video that happens to test if there is an open in the EKPS ECU area, and it says that reading 120 Ohms indicates an open circuit. Each ECU has a 120 Ohm between CAN-H and CAN-L. I guess if you have 120 Ohm in one module and 120 Ohm in the opposite end of the network, in a closed network with no open wire the two 120 ohms in parallel read 60 ohms. If there is an open somewhere you read 120 ohm. And in my case I read 120 Ohm between CAN-H and CAN-L while EKPS was removed from the network (although with power still on).

So, looks like that there are at least 2 more ECUs with terminator resistors in that CAN bus network besides the EKPS ECU that I removed (which supposed to have a 120 Ohm as well).

I guess I need to trace all ECUs on that CAN bus and disconnect them one by one. In my case, there are only the DME (yellow), FEM (green) (any of these two would suck to go bad!) and EGS (yellow) and GWS (red), EKPS (red) and ACSM (green).




Will post updates here: https://f30.bimmerpost.com/forums/sh....php?t=1991628

[Thomachaan]

Originally Posted by fe7565

Thank you for the quick reply and the advice.

The scope readings/guidance above I got from the ISTA guide and those are samples. I took an actual photo of my own module the oscope with CAN-H and CAN-LOW each to GND. But it looked so irregular that I thought I mucked soemthing up as far as scoping it correctly.

That looks irregular for sure. Try unplugging components one by one and testing the PT-CAN lines, when it clears up you may have found the culprit. If not, the culprit could be something integral, like the DME or gateway.

From my basic electrical knowledge, my old DME not only had a failed BSD communication chip, but the voltage regulator(s) onboard the DME itself had fried. My car back then would only attempt to start if the battery was charged all the way up.

idk what the GWS is like for your car, but I found a rebuilt one for my E90 off eBay for $50 (still had a Delphi connector). Easy access with trans pan removed. Hope you can find a similarl solution, they're like hall effect sensors that blow out internally from electrical spikes.

[fe7565]

Originally Posted by Thomachaan

That looks irregular for sure. Try unplugging components one by one and testing the PT-CAN lines, when it clears up you may have found the culprit. If not, the culprit could be something integral, like the DME or gateway.

From my basic electrical knowledge, my old DME not only had a failed BSD communication chip, but the voltage regulator(s) onboard the DME itself had fried. My car back then would only attempt to start if the battery was charged all the way up.

idk what the GWS is like for your car, but I found a rebuilt one for my E90 off eBay for $50 (still had a Delphi connector). Easy access with trans pan removed. Hope you can find a similarl solution, they're like hall effect sensors that blow out internally from electrical spikes.

I will try again to take readings, armed with some more info on the F30 wiring and better oscope probes.

The GWS in the F30 seems to be just the shifter (R/D/M/S) but has both PT-CAN and PT-CAN2 connectors. It does have hall sensors. Luckily, its only about $25 on Ebay.

The DME is also about $80-120 on Ebay (old cars command much less price for parts), but I rather not think that is bad because of the re-programming etc pain.

The gateway, I think is the ZGW, and it's built into the FEM in the F30. THat would be the ultimate pain to replace because it would have to be sent off for reflashing to my car.

[Thomachaan]

Cloning/reprogramming services cost about as much as the used modules, I got lucky and found someone great off eBay nearby enough to not have to ship parts to him.

It's exciting isn't it, learning this level of detail about your car's platform? I really don't know about the F series abbreviations and modules lol. My E90 shifter has electrical connections too, but there's a separate sensor within the trans itself checking the physical position of gear selection. Your car may be different but BMW might have just renamed similar components. There are soooo many ways for these cars to electrically fail but the fix is usually something simple, just need more info. I read up for 3 months before reviving my car from death.

On this journey you'll become more of an expert than BMW techs - Invaluable experience which will hopefully help others in the same bind.

Originally Posted by fe7565

I will try again to take readings, armed with some more info on the F30 wiring and better oscope probes.

The GWS in the F30 seems to be just the shifter (R/D/M/S) but has both PT-CAN and PT-CAN2 connectors. It does have hall sensors. Luckily, its only about $25 on Ebay.

The DME is also about $80-120 on Ebay (old cars command much less price for parts), but I rather not think that is bad because of the re-programming etc pain.

The gateway, I think is the ZGW, and it's built into the FEM in the F30. THat would be the ultimate pain to replace because it would have to be sent off for reflashing to my car.

[Thomachaan]

fe7565 I'm no expert with an oscilloscope either, but it helps to have automotive backprobing needles, cheap online. I sneak in behind the pins on the connectors. I keep long lengths of wire, alligator clips, and solder to make whatever ends I need to diagnose far apart modules.

I use a relatively cheap 8 channel oscilloscope, with which I haven't used more than 2ch at a time, but could catch all 6 ignition signals w/channels to spare when needed. The investment in these tools is chump change when you consider the level of work being done, & subsequent money saved.

[fe7565]

Originally Posted by Thomachaan

fe7565 I'm no expert with an oscilloscope either, but it helps to have automotive backprobing needles, cheap online. I sneak in behind the pins on the connectors. I keep long lengths of wire, alligator clips, and solder to make whatever ends I need to diagnose far apart modules.

I use a relatively cheap 8 channel oscilloscope, with which I haven't used more than 2ch at a time, but could catch all 6 ignition signals w/channels to spare when needed. The investment in these tools is chump change when you consider the level of work being done, & subsequent money saved.

I am making a couple of back probes as it is certainly a necessity. Will order some real ones as well.

My oscope has only two channels, but I only used one channel until now. I may buy one of those PC oscilloscopes for these not so fast signals. I like the idea of much easier manipulation of the readings and settings within the PC Software.

Although, I think my Sigent 100 mhz can be connected to a PC, a small pocket size oscilloscope module would be easier to use. I was looking at videos of them pushing that Picoscope oscilloscope, it’s a UK company. But I think I could find a cheaper hardware. However, I think it’s the PC oscope software that really what makes a difference not as much as the hardware.

[fe7565]

After a couple of false starts managed to find the problem. It was one of the PT-CAN wires in the engine bay between the GWS and the EGS (Transmission control). Pin 6 going from the engine bay to the transmission connector is open somewhere. I was not able to trace where exactly. What I did is to connect it with pin 8 which goes to another module. Pin 8 and Pin 6 supposed to be connected anyway in the harness somewhere on the way to te EGS.

As soon as I made that connection, the rest of he bus at GWS and EGS read the correct 60 ohms vs the 120 ohms before. The two termination resistors for my car made on 2012/3 are in the FEM and the KOMBI. The BMW wiring info is incorrect for my car, because the 2012/7+ and not the 2012/6-diagrams were applicable.

[Thomachaan]

Great news, so no need to repair the wire, just jumped to another one? That's a win!

It'd be good to eventually find out where exactly that issue is. Whether fluid leaks, accident damage, rubbing, pinching, the issue could spread to another wire.

As an update to my car, no direct issues to report. I have a P0420 code going that I think is related to my missing underside splash shield, and intermittent ghost codes for DSC/KOMBI no communication, but those may have been present before my sparkage incident.

High speed oscilloscopes are getting cheaper and cheaper, making analysis of these communication lines super simple.

Originally Posted by fe7565

After a couple of false starts managed to find the problem. It was one of the PT-CAN wires in the engine bay between the GWS and the EGS (Transmission control). Pin 6 going from the engine bay to the transmission connector is open somewhere. I was not able to trace where exactly. What I did is to connect it with pin 8 which goes to another module. Pin 8 and Pin 6 supposed to be connected anyway in the harness somewhere on the way to te EGS.

As soon as I made that connection, the rest of he bus at GWS and EGS read the correct 60 ohms vs the 120 ohms before. The two termination resistors for my car made on 2012/3 are in the FEM and the KOMBI. The BMW wiring info is incorrect for my car, because the 2012/7+ and not the 2012/6-diagrams were applicable.

[fe7565]

Originally Posted by Thomachaan

Great news, so no need to repair the wire, just jumped to another one? That's a win!

It'd be good to eventually find out where exactly that issue is. Whether fluid leaks, accident damage, rubbing, pinching, the issue could spread to another wire.

As an update to my car, no direct issues to report. I have a P0420 code going that I think is related to my missing underside splash shield, and intermittent ghost codes for DSC/KOMBI no communication, but those may have been present before my sparkage incident.

High speed oscilloscopes are getting cheaper and cheaper, making analysis of these communication lines super simple.

Yes, I need to investigate if the harness is rubbing anything or getting burnt by the exhaust. At some point the wires from pin 6 and pin 8 merge into one wire that goes into pin 6 of the round transmission connector.

But the harness runs behind the engine block down to the transmission. Hard to intercept it. And pin 8 had a good reading (50K ohm between pin 8 and pin 7) to the EPG, so the break is somewhere between pin 6 and where pin 8 and pin 6 merge.

Been looking at used EBay harnesses ($70) but cannot tell where that connection may be.
But surely... if it's caused by rubbing/burning, then other wires may be next.