[rjahl]

Powerclass does not normally cause a "no start" issue but I'm not so sure about the Variant coding. Different parameter.

Powerclass would be the 325 vs 330
Variant code being 325 vs Z4 vs X3
Country code, Different Cats and tunes for different regions.

These are values withing the calibration file that the DME checks against the values stored in the CAS and EEProm. Of course they are not the only values referenced.


If you have successfully flashed the program section, next step would be NCSexpert after that, you are reflashing the EEprom directly. I don't know of any options.

[stevethresh]

What do you suggest I do with NCSEXPERT?

Also, the 9cce FRM error is from a battery deep discharge. The battery has been replaced but I don’t know how to clear the error. INPA tries but the error remains.

ERROR MEMORY
--------------------------------------
Date: Feb 17, 2020 5:25:45 pm
ECU: FRM_87
JobStatus: OK
Variant: FRM_87
-------------------------------------------------- ----------
RESULT: 1 error in error memory!
-------------------------------------------------- ----------
9CCE battery deep discharged
Error counter: 1
Mileage 148744 km
Terminal 15 0.00 0/1
Terminal R 0.00 0/1
Parking light 0.00 0/1
Parking light left 0.00 0/1
Parking light right 0.00 0/1
Hazard warning light 0.00 0/1
Follow me home 0.00 0/1
Signal or value below threshold Test conditions fulfilled Error currently present and already saved Error would not cause a warning lamp to light up

Error code: 9C CE 62 01 48 A1 00
================================================== ===========

[rjahl]

Essentially you "CODE" the DME by reading the VO from the CAS and then writing the data to the DME.

There are many DYI's available on the forums. I'm not going to rewrite them.

[stevethresh]

Thankyou. I have no problem with searching for guides, I just need to know what I’m looking for. I’ll see what I find.

[dpaul]

Originally Posted by stevethresh

Thankyou. I have no problem with searching for guides, I just need to know what I’m looking for. I’ll see what I find.

I've got this on speed dial so:

Since you've not used NCSExpert and you've had problems using WinKFP for reasons that are not clear, unless you are very confident about your setup I'd start by wiping your installation of BMW Standard Tools (BST).

Download BST from www.bimmergeeks.net. A very reliable installation set. A download of current datens is also necessary to use WinKFP but not the other BST components. I would do this in your case. You should use "BMW coding Tool" to install the datens, don’t try to copy them to appropriate directories manually.

Be clear about the difference between "coding" (what NCSExpert does) which is choosing between a set of pre-determined options allowed by the factory and "programming" (what WinKFP does) which is to completely replace the program and calibration (map) data for a given module. There are very few coding options for the DME.

If you already have some familiarity with these issues, and apologize for any redundancy.

Run NCSExpert

From file menu, select the profile "Factory Coding". Often called "Expert mode" in other installations - "Factory coding" uses a factory-set parameter list matching the VO (vehicle order) stored in CAS and FRM/NFRM. This sets up what is sometimes called "default" coding.

The other profile option is "Manipulation", which obtains the options from a text file that you edit. This allows manipulation of individual parameters. You don't need the latter now.

F1 "Start"
F3 "Chassis", choose "E89" on first popup, choose "CAS" on second popup

The above will read your VIN (FG) and your VO (FA), all NCSExpert needs to know.

F6 "Back"
F4 "Select ECU", choose "6BMOT” (NCSExpert name for the DME – it will be something close to that)
Make sure the displayed default job (“JobName”) is “SG_Codieren”, German for code the select module. It should be but if not, press F2 “Change Job” and select “SG_Codieren”

F3 “Execute Job”

JUST TO BE CLEAR, DO NOT EXPECT THIS TO ALLOW YOUR CAR TO START. I would be very, very surprised if it does. It should get rid of the 2FA3 code. If it fails, that speaks to an underlying problem with the DME or the cloning process.

[rjahl]

Thanks for the write up, I'm just too busy to do much of anything these days. I've written a new tune for my Z4 weeks ago and it's still sitting on my thumb drive.

There are a few sub routines that cause elements of the EEProm to update and I would not guarantee coding will get this done. It's just the last suggestion I have. Can't hurt and it's cheep to try.

[stevethresh]

Thankyou very much for the write up. I'll give it a go in the morning.

[stevethresh]

The process was successful but as you expected the car still doesn't fire however the 2fa3 and 2fa4 errors. Now I'm left with the following:

E R R O R M E M O R Y ----------------------
Date: 19.02.2020 12:59:05
Model: E89X
Scope: all control module
JobStatus: OKAY
Quantity: 19
----------------------------------------------------------------
ADR Gen. name JobStatus
92 VIRTSG92 ERROR_ECU_SERVICE_NOT_SUPPORTED
----------------------------------------------------------------
ADR Gen. name JobStatus Number of errors Error code Kind of error
00 JBBF OKAY No Error stored
01 MRS OKAY No Error stored
12 DME/DDE OKAY No Error stored
17 EKP OKAY No Error stored
18 EGS OKAY No Error stored
1C LDM/ICM OKAY No Error stored
29 DSC OKAY No Error stored
36 TEL/MULF OKAY No Error stored

40 CAS OKAY 1 Error stored A0B4 60

41 DWA OKAY No Error stored

50 SINE OKAY 1 Error stored 9D12 11

56 FZD OKAY No Error stored
60 KOMBI OKAY No Error stored
62 MOSTGW OKAY No Error stored
63 MASK/CCC OKAY No Error stored
64 PDC OKAY No Error stored

72 FRM OKAY 1 Error stored 9CCE 62

78 KLIMA OKAY No Error stored
================================================== ===============

[dpaul]

Did you ever try CAS realignment in INPA (from MSV70 main menu, shift-F7)?

[stevethresh]

The result from either start or program says 'process in progress'

I also spotted another error has popped up since trying which is:

$SGBD: MSV70 Version: 6.00


================================================== =========================================
$DATAID: 004 9PP0L 921S 0BI1S
=== 19.02.2020 === 14:25:57

Current kilometre reading: 148746 km
__________________________________________________ _________

comment line 1:
comment line 2:
__________________________________________________ _________


$ Error location number: 12103
$ ErrorLocationText: 2F47 EWS Irreversible control unit error
57012F47520400000328000000000500000000000500000000 000500
Error: 1 (1) No: 12103 2F47 EWS irreversible control unit error frequency: 3
error type: 2 (150) control unit defective
Readines Flag: (17) Test conditions not yet met
error status: (34) Error currently present, not yet debounced
MIL status: (48) Error does not light up the warning lamp (MIL)
P-Code: F_LZ: 40
environment text / entry at 0 km(n.) 0 km(2.) 0 km(1.)
Motor speed 0.00 rpm 0.00 rpm 0.00 rpm
Spannung kl. 87 0.00 V 0.00 V 0.00 V
Engine control status ENG_LOCK ENG_LOCK ENG_LOCK
Time since start 0.00 s 0.00 s 0.00 s
__________________________________________________ _________

[dpaul]

OK, so CAS and DME are not communicating correctly.

Lets go back to the start - you had a "crank but no-start" condition with no clearly conclusive error codes that you ascribed to a "water-damaged" DME.

You replaced the DME with a used one, cloned to the original using BDM but still had a "crank but no-start" condition, again with no helpful error codes until you tried EWS aligning and now have a very clear fault.

So my question now is why did you think the DME was the problem in the first place? Perhaps the issue lies in the CAS module?

One thing you might want to rule out first (although I think there would be clearly relevant error codes) is that if the DME actually failed, that it did not destroy ignition coils by allowing excess or continuous current to be passed. I remember reports of this sort of failure (with an MSD80) although I imagine it's rare. Just pull the coils and check with multimeter.

Hopefully, posters with extensive experience like Levanime or Hassmaschine might chime in.

[stevethresh]

DME was definitely water damaged. There's another post of mine that shows the damage. The board was covered by an inch of water (including the EEPROM chip).

The 2fa3 error mentioned the EEPROM and that's what the guy who cloned the DME thought hadn't come across properly.

I can check the coils. I'm also going to see if ISTA gives any suggestions.

Thanks for trying to help. This is proving most frustrating!

[rjahl]

Try reading the following thread

https://www.e90post.com/forums/showt...1169885&page=5

Read the discussion with DRC38 regarding his fried DME and his efforts to recover it. In the end, he was successful

[stevethresh]

Thanks for the suggestion. I don’t currently have access to a BDM reader so I’d need to workout how to make the change through the OBD port.

[dpaul]

Originally Posted by stevethresh

Thanks for the suggestion. I don’t currently have access to a BDM reader so I’d need to workout how to make the change through the OBD port.

You are not going to be able to do that with BMW standard Tools (INPA etc) or likely any inexpensive aftermarket tool.

Chinese clones of aftermarket OBD tools like KessV2 are available but have limitations. Some memory areas are not accessible to them because the toolmaker did not know how to achieve a high level of authorization with the DME. For example, in MSD80 and MSS60 DMEs, KessV2 clones can read almost the entire memory but not the region where the ISN/SK is stored and a few other small areas. There is a broader range of areas to which writing is restricted, like the first entry in the AIF etc.

I have no idea if the memory locations you'd need to fill with FF are restricted in the MSV70 or not.

The advantage of BDM is that authorization is not required - all flash rom is open to read and write, at least in the MSV70. A BDM tool like a KTAG clone is cheap but of course, you'll need to remove and open your DME and make electrical contact with the BDM pad in one way or another. All described pretty well in hassmaschine's thread.

[stevethresh]

ISTA claims to have done the alignment but I've not yet been able to start the car or confirm with INPA that the error is gone. I suspect my battery is currently too low on charge to start the car anyway after looking at one of the other errors so it's being charged overnight.

I've seen in hasmachine's threads that he refers to a xprog-m device. Do you think this will this do what I need if I have to manually interface with the DME?

https://www.ebay.co.uk/itm/X-PROG-Bo...aa0e%7Ciid%3A1

[icon2015]

I suspect, as dpaul already indicated, that your issue is, at least now, at the CAS module. I would, first, check the ground cable of the engine, and second, I would get the CAS out of the car, open it and look for anything unusual on the motherboard.

[rjahl]

Originally Posted by icon2015

I suspect, as dpaul already indicated, that your issue is, at least now, at the CAS module. I would, first, check the ground cable of the engine, and second, I would get the CAS out of the car, open it and look for anything unusual on the motherboard.

While I agree that the problem could be more complex than just the DME, he is communicating with the CAS over can so it's probably OK.

What I don't understand is the relationship with the vendor that cloned this DME. Clearly the job was botched, maybe he needed an eeprom file as a base or maybe he should have just created a virgin DME. But at the least he should have supported the OP.

If the vender has just failed him, maybe he should mention his name so others do not fall victim.

[dpaul]

Originally Posted by stevethresh

ISTA claims to have done the alignment but I've not yet been able to start the car or confirm with INPA that the error is gone. I suspect my battery is currently too low on charge to start the car anyway after looking at one of the other errors so it's being charged overnight.

I've seen in hasmachine's threads that he refers to a xprog-m device. Do you think this will this do what I need if I have to manually interface with the DME?

https://www.ebay.co.uk/itm/X-PROG-Bo...aa0e%7Ciid%3A1

X-Prog is for eeprom programming - the external eeprom in the MSV70 can be programmed using X-prom without removing it from the circuit board but it still requires removing the DME, opening the case and attaching wires/clips. X-prog is not an OBD device.

Most DME program and data are stored in flash memory internal to the microprocessor. It remains unclear to me if there is anything important for starting the engine stored in the MSV70 external eeprom. In any case, a BDM device is the appropriate approach to apply the remedy suggested in rjahls link.

The most typical use of devices like X-prog is to alter data in the CAS module relating to key codes and KOMBI modules relating to VIN and mileage.

Your comment about needing to have the battery on charge raises the question about proper voltage control when using WinKFP or other tools. You can cause all sorts of problems if you let the bus voltage drop below 12.6V. Many if not most folks use high current supplies to keep the voltage well above this level when flashing modules - that would include manipulations of EWS/CAS using INPA. Personally, I connect a 13.8V/40A supply any time I am programming or coding.

[dpaul]

Originally Posted by rjahl

While I agree that the problem could be more complex than just the DME, he is communicating with the CAS over can so it's probably OK.

What I don't understand is the relationship with the vendor that cloned this DME. Clearly the job was botched, maybe he needed an eeprom file as a base or maybe he should have just created a virgin DME. But at the least he should have supported the OP.

If the vender has just failed him, maybe he should mention his name so others do not fall victim.

I agree. This vendor should be able to offer more support

[stevethresh]

The winkfp process said it was successful and the ZB number shown after for the DME matched the number it stated it was going to install.

I've checked the first three ignition coils and there is no resistance or infinite resistance:

X.X - no resistance between these two pins even if reversed (showing 0)
.X.

infinite resistance in any other combination of top and bottom connections (showing 1)

Also not getting any resistance reading from the output to any of the input pins (showing 1)

Anyone know if that's right?

[dpaul]

Originally Posted by stevethresh

The winkfp process said it was successful and the ZB number shown after for the DME matched the number it stated it was going to install.

I've checked the first three ignition coils and there is no resistance or infinite resistance:

X.X - no resistance between these two pins even if reversed (showing 0)
.X.

infinite resistance in any other combination of top and bottom connections (showing 1)

Also not getting any resistance reading from the output to any of the input pins (showing 1)

Anyone know if that's right?

Here's a wiring diagram that is at least approximately correct:
https://www.newtis.info/tisv2/a/en/e...nition/x7JApNv
WARNING: this may be the incorrect diagram because I don't have your VIN and am generally unfamiliar with European model specifics. So enter your VIN on this site and get to the correct diagram. This won't affect the pin numbers on the lgnition coils but will affect the number of the Fuse protecting the circuit (which you should check)

Your measurements probably indicate OK - resistance between pins 1 and 3 should be 0.4-0.8 ohms (but not zero!) and there should be no connection (infinite) between pin2 and either pin 1 or 3. Other than checking the fuse, that's it

All that said, likelihood is DME/CAS problem. I still have some concern about your report that WinKFP gave you 'insufficient authorization' errors the first time you tried it. WinKFP is a pretty robust piece of software. Also concerned about INPA hanging when you tried resetting EWS initial values. Have you retried that using a good stable power supply?

If it were me, I'd buy a clone KTAG and modify the binary the way drc38 did in Rjahl's link. Alternatively, there's no reason the vendor who cloned your DME could not do that very easily.