Hi all. The context is: for a customer, I'm trying to repair the MIDI scanning electronics of a piano keyboard that was on the market for only a short while in the early 2000's. From contacting the original developer and doing some troubleshooting, the diagnosis became that the MC9S12UF32 was probably faulty.
I finally was able to obtain a couple of these MCUs (albeit -PB instead of -PBE, which I don't think should matter) and after desoldering and desoldering and connecting the board to a USBDM off of Amazon (this one), I found these observations:
After seeing this post I tried flashing with version 4.12.1.140, but that did not work either.
After seeing this post I've tried powering the target from a PSU and also doublechecked my connections - it seems like they are good; also because at first, the chip is detected properly.
Any advice well appreciated, thanks in advance.
Solved! Go to Solution.
Hi,
I was able to program the images you provided to a UF32.
It required updating the description of the UF32 memory regions in the USBDM device data.
The version uploaded to Scourceforge (USBDM_4_12_1_290_Win) has these changes (along with a number of others related to the UF32):
https://sourceforge.net/projects/usbdm/files/Version%204.12.1/Software/
bye
Hi,
I was able to program the images you provided to a UF32.
It required updating the description of the UF32 memory regions in the USBDM device data.
The version uploaded to Scourceforge (USBDM_4_12_1_290_Win) has these changes (along with a number of others related to the UF32):
https://sourceforge.net/projects/usbdm/files/Version%204.12.1/Software/
bye
Alright I've just tried again with:
Programming of the target flash failed!
Reason: Program data not within target Flash Memory
bit cautious to try anything else now without first consulting you, @pgo
Hi,
I don't blame you for being cautious.
The programmer expects memory images to be in paged format. It is possible the image is linear.
Anyway, if you can provide the flash image I will check the format and advise.
This assumes the the image is not confidential.
Best to do this by Private Message.
bye
PS.
You can check for problems by doing a verify (with Security=Unsecure). This should report the same error rather than reporting that it is secured!
You can do a rough check if it is a linear image by selecting Linear Image before loading the file and doing a verify. It should report failure to verify rather than complaining about being out of range.
I rather hate the UF32.
bye
Hi ,
I have uploaded a new version of USBDM that copes with UF32 better:
https://sourceforge.net/projects/usbdm/files/Version%204.12.1/Software/
Tested with a UF32 programming and mass erase.
(I now have only 1 UF32 that is not bricked. They are a pain!)
bye
Hi,
I'll see if I can find a UF32 to test with.
Note that this particular chip has a errata that may be affecting the outcome:
https://www.nxp.com/docs/en/errata/MSE9S12UF32_1L47S.htm#xxMUCts01498
I suggest you check the mask version and if the above applies.
This chip should always be programmed without security.
Please also supply:
The last two are reported by the programmer. Screen grab the Interface page of the programmer.
bye
PS. I have spent some time refreshing my memory on the UF32 and how the programmer handles it.
The first error your received Unsupported operation or options for this chip was because you were trying to program the chip with security enabled. Because of the chip errata this option is not supported as the chip may not be programmable again unless special steps are taken beforehand (see errata). This is a check built into the programmer early on.
Unfortunately this check is absent from the mass erase button which was added later. Mass erasing a UF32 renders it unusable. Note that mass erasing is not unsecuring the device - it simply erase the entire device. For a non-UF32 device this would leave the chip partially unsecured and able to be programmed. I will correct this in the next version but this is of little comfort for you!
I will do some more testing if I can find a UF32.
Hi pgo,
The mask version is indeed 1L47S, so it applies.
The OS is Windows 7, 64-bit edition, Service Pack 1.
The USBDM hardware is reported as: USBDM-JS16-SWD-0001
BDM firmware is 4.12.1
screenshot of the Interface page:
So... seen as Mass Erase did work one time for me, this chip is bricked? I have a third and last one to solder on... perhaps there's something I could try to see if it's revivable...
Yes - the mass erase using the previous version of the software would have triggered the bug in the chip and bricked it.
I know of no way of recovering a bricked chip.
The current version uses a different strategy for Mass Erase Now:
https://sourceforge.net/projects/usbdm/files/Version%204.12.1/Software/
I believe the programming works in both versions provided unsecure is selected as the security option. Other choices are rejected. However I have not tested the previous version.
I recommend EraseMass as the Programming Erase Mode.
The current version has been tested with a UF32.
bye
Alright, I see. I will solder on the third lucky candidate and use unsecure & EraseMass while programming with the new USBDM version. Fingers securely crossed...
Please use the current version as it has been recently tested with UF32.
Yes, forgot to add that just now.