| Electronics > Repair |
| Dead Sonance DSP 2-150 MKII Amplifier after failed update |
| (1/3) > >> |
| JimKnopf:
Hello, i got a Sonance DSP 2-150 MK II Amplifier here for repair. The owner atempted to update the device but powered off the device during update process. Now the device is dead and will not start and cannot be updated again because update works via TI LM Flash Tool on the network interface. But since the device is dead, ethernet doesn't work, no ip, no access via LM Flash Tool. The device has an TM4C1294NCPDT Corex M4. There are two eeproms, an 25L1606E which is empty and a 24LC64-I which as only few bytes in it. I desoldered both and read it via external programmer. The pcb has 2 connectors exposed, one UART which is deactivated and a JTAG connection. I couldn't get the LM Flash Tool to work with any of my jtag devices because it supports only few listed devices (NIC, ICDI eval board, FTDI Red Probe or serial UART). I'm able to connect and work with a segger j-link. I read out the 1MB flash area and compared it with the firmware update file. Maybe some code is missing or placed on the wrong sector during update. Can someone help me figure out on what address i have to place the firmware data to get the device to work? The firmware file is from here: https://danainnovationsupport.zendesk.com/hc/en-us/article_attachments/360057316992/Sonance2-150MKII_v2.2.2150.bin Datasheet for the Cortex M4 is here: https://www.ti.com/lit/gpn/tm4c1294ncpdt The memory dump is attached. |
| tv84:
--- Quote from: JimKnopf on November 04, 2021, 03:52:00 pm ---I'm able to connect and work with a segger j-link. I read out the 1MB flash area and compared it with the firmware update file. --- End quote --- What were your conclusions? Can you provide the flash dump? What about the boot log via UART? |
| JimKnopf:
@tv84 The flash dump was attached to my post. UART seems to be disabled. |
| tv84:
Jim, From a static point I cant help you very much. The FW .bin that you link is all correctly loaded in your dump starting at 0x4000. So the interruption didnt interrupt the loading of the FW file. The problem are the actions after loading. I dont recognize the initial 0x2000 bytes of your dump as a valid ARM boot block... So can't help there. It looks like something destroyed you boot zone... (? ?) The best (only? ?) solution i would say is get a dump from a similar device and overwrite all your flash. |
| JimKnopf:
I fixed it. I bought a damaged one that was sold for parts from ebay. I saw on the images that the LED were on. That means basic function works. I hooked up my Segger J-link and read the flashmemory. The startcode looks different from the device i already have here for repair. I put this new memory dump on the device i have here for repair and it started. All works well. Work is done. |
| Navigation |
| Message Index |
| Next page |