Post #1

antwas

Watchlisted
Thread Owner
Joined
04.06.2026
Messages
4
Reaction score
0
Location
United States
I’m able to successfully read and write to the DME in bench mode using Xhorse Multiprog. I use a paid service to get the flash back to stock because an attempt was previously made to install BM3 on the dme the was locked and that bricked the ecu. DME has no communication with car and I attempted to update it with ISTA+ and it’s staying to replace the DME.
 
Ratings and Reviews
Total ratings: 0

Post #2

The symptoms you're describing — stock flash restored via bench but DME still has zero communication on the car — almost always point to an incomplete flash restore.

Key question: was the full flash restored or only the program/calibration area?

MG1CS003 (B48 in your F36 430i) has multiple flash regions. If the failed BM3 attempt on a locked ECU corrupted the boot sector or RSA/signature area, and your paid service only restored the application/calibration blocks, the DME may appear fine on bench but won't properly boot and communicate on the vehicle CAN bus.

Why ISTA+ says "replace DME":
That message simply means ISTA+ can't see the DME on the bus at all. It's not a hardware death sentence — it just confirms zero communication.

What to check:
  • Ask your flash service exactly which areas were written back. You need a full bench restore (all blocks including boot/RSA), not just application + calibration.
  • After the restore, read the DME back with Multiprog and compare checksums against the stock file — do they match across all areas?
  • On the car side, check CAN activity on OBD pins 6 & 14 with a scope or CAN sniffer. If the DME is transmitting nothing, it's not fully booting.

Note on ISN:
If the ISN was corrupted during the brick, the car won't start, but the DME should still communicate on the bus for diagnostics. Since you have zero communication, ISN is not your primary issue right now — flash integrity is.

Share the HW and SW numbers from the DME label (or from the Multiprog read info) and confirm which blocks your service restored, and we can narrow it down further.
 

Post #3

I pulled a few identifiers from the file. I’ve also attached the original flash file after it bricked.


Hardware (HWE):


  • Hardware_DME8R1_SP11

Hardware Application (HWA):


  • DME8.4.x_B48_V1

Bootloader (BTLD):


  • MDG1G_LK-CB_011_253.5_3.2
  • MG1CS003_LK
 

Attachments

Post #4

That file confirms the problem. The entire internal flash is 0xFF from offset 0x000 onward — the first 512 bytes (and almost certainly a large portion beyond) are completely erased. An 8 MB read full of 0xFF means the internal flash is essentially blank.

What this tells you:

The MG1CS003 on the SPC5777 has both internal flash (bootloader, RSA/security blocks, startup code) and external flash (application + calibration). Your Multiprog read of the internal flash came back wiped. The failed BM3 install on a locked DME erased the internal flash, which holds the bootloader (MDG1GLK-CB011253.53.2) and low-level startup code.

This is why you have zero communication on the car. Without a valid bootloader in internal flash, the DME never boots — it can't initialize CAN, so ISTA+ sees nothing.

What you need to do:

  • Your paid flash service — ask them specifically: did they write back internal flash or only external flash? Based on this dump, internal flash was never restored.
  • You need a donor/stock internal flash for this exact HW revision (HardwareDME8R1SP11 / MG1CS003 / B48). The identifiers you pulled (RBA1DGS1, SW 10SW002540) came from the external flash which still has data — but the internal flash is what's dead.
  • Write the stock internal flash back via Multiprog in bench mode. After that the bootloader should be functional and the DME will boot and communicate on the vehicle bus.
  • ISN sync / personalization will be a separate step after communication is restored.

Bottom line: The DME isn't bricked in the hardware sense — the internal flash just needs to be reprogrammed with a matching stock bootloader image. Focus on getting the correct internal flash content written back and you should regain communication.

Do you have access to a donor internal flash for this HW, or do you need guidance on sourcing one?
 

Post #6

A few options:

  1. Donor DME — Find the same HW revision (DME8R1_SP11 / MG1CS003 / B48) from a known-good unit and read the internal flash with Multiprog. This is the most reliable route since you know it's clean.
  2. Your paid flash service — Go back to whoever did the stock restore and ask them specifically for the internal flash image. They likely only wrote back external flash. If they have the full backup (internal + external) from before the brick, that's your file.
  3. BMW flash containers (SWE/PSdZData) — The bootloader and internal flash content can sometimes be extracted from official BMW SWE update packages via ISTA+ programming data. Your bootloader ID (MDG1GLK-CB011253.53.2) can help locate the correct package. However, writing raw SWE content to internal flash via bench requires knowledge of the exact memory layout and addressing — not straightforward.
  4. Forum / community — Post a request in the appropriate section here with your exact HW identifiers. Someone with the same DME revision may have a clean internal flash backup they can share.

Options 1 or 2 are your fastest path. A donor read from the same HW revision is ideal — it's a direct 1:1 write back via Multiprog with no conversion needed.

Important: if you go the donor route, do not copy ISN/personalization data from the donor — only the bootloader and startup code areas. Otherwise you'll have a key mismatch that prevents starting even after communication is restored. Your flash service should be able to help merge the correct sections if needed.
 
Activity
So far there's no one here