USBDM stopped detecting Kinetis target

[gustavsl]

I was using a FRDM-KL25Z as a debugger for a custom board using MKL26Z64. Debugging and programming all went well using CodeWarrior 10.6 until, suddenly, when I tried to program the board, I got an error:

ARM GDI Protocol Adapter : Can't connect. BDM has no connection to target

The only thing I did before the error happened was to update USBDM to 4.11.1.20, but I managed to program the board once with the updated USBDM and drivers. After that, I just got the mentioned error.

I tried using a standalone USBDM programmer (USBDM-JS16-SWD-SER) with the custom board, but the error persists, meaning it must be something with the target processor.

I tested the setup with Windows XP SP3 (using CodeWarrior) and with Linux (using KDS 3.0.0).

Any tips?

Thanks.

[pgo]

Hi Gustavo,

I have just re-tested a FRDM-KL25Z board with the latest version on sourceforge and I was unable to find any problems with the stand-alone programmer.

I also tested with Codewarrior.

• Could you provide a copy of the binary file that you programmed to the KL26 in case it is somehow related to it?
• You should also select the mass-erase option as it may be more reliable.
• Finally, have you programmed a bootloop firmware into the onboard KL25 chip?

bye

[gustavsl]

Hi, pgo.

This is the binary file I programmed:

https://dl.dropboxusercontent.com/u/10218454/Baba.elf

Yes, I have programmed a bootloop into the KL25Z.

I double checked the SWD connection between the boards. I see the reset pulse on the oscilloscope, but still nothing.

Thanks

[pgo]

Hi Gustavo,

I would expect activity on RST, SWCLK and SWD when connecting.

I had no problems programming the image you provided - but this was using a FRDM-KL26 board and the on-board interface.

There could be some issue with an external cable that I am not aware of.

This is the last part of the waveform triggered on falling edge of reset when doing an auto-detect.

This is a close up of part of the waveform - the rounding is BW limited 10MHz on the DSO

[gustavsl]

Hi, pgo. Thanks for testing.

When I try to "detect chip" using ARM Flash Programmer, I see the reset pulse and immediately get the "can't connect" error. SWD and SWCLK are high and show no activity. I checked continuity between connections and it seems OK. Any other things I may look for?

Thanks

[pgo]

Hi Gustavo,

I can't think of a reason for having activity only on the reset line.

Could you try without any target connected - You should see something like:

(Results depend on the oscilloscope - the activity on SWD & SWCLK line will be more visible if the communication speed is set to 250kHz.)

The initial pulse is switching the interface to SWD mode from JTAG.  Close up it should look like:

If you don't see this I would seriously suspect the cable.

Random list of suggestions :smileyhappy::

• Confirm operation with the on-board target chip (KL25) (J11 closed)
• Confirm that J11 is open on the FRDM-KL25 board.
• Confirm cable connection continuity to FRDM board
• Confirm that there are no shorts between pins.
• Try a USB2 port
• Try high and low communication speeds
• Try manually selecting target device and use Mass Erase Now to erase the device.
• Use Detect Chip and please provide a screen grab of the actual error pop-up you get.  I would expect :

That's about all I can suggest!

bye

[gustavsl]

Hi, pgo.

I've managed to get a SEGGER J-Link to test with my custom board. The debugger is working fine, but also can't detect target. It shows:

WARNING: RESET (pin 15) high, but should be low. Please check target hardware.

I'm thinking it may be something with the target processor, hence it does not connect either with USBDM or J-Link.

I'll try replacing the microcontroller and see what happens.

Thank you.