Mass erase of factory new KL chip with SEGGER JLINK problem

[ky3orr]

Hello,

I have populated brand new KL05Z32 chip on my custom board.

Its internal 0xFF memory state prevents it from operation and causes constant reset which is visible when measuring reset signal with the scope:

It causes problems with communication to the device. Actually it is not possible to communicate (over SWD) at all:

I am using SEGGER JLink debugger. It works fine for other board (the same project) where KL05 was already mass erased.

I've been reading a lot on the web regarding tries of unsecuring and/or mass erasing, but nothing works for me now.

I've put 10k pull-up resistor on RESET line and 10k pull-down resistors on CLK and DIO lines.

It didn't changed a thing.

I have also tried a jlink script for unsecuring Kinetis found here:

https://community.freescale.com/thread/301600

It also changed nothing.

I've also tried connection from SEGGER's J-Link Commander and it tried to do mass erase also - without any luck (timeout).

Now my question is: what is actually a procedure for handling mass erase of the brand new chips?

I have the same problem on 3 out of 4 boards. I have no clue why it worked for the first time (without any additional resistors).

Thank you for your comments.

p.s. - after RESET signal release there is only 5[ms] before next reset signal mark.

[Amit_Kumar1]

Hi Grzegorz

You can check this thread   PE multilink JTAG/BDM Kinetis programming issue   Here I used PE multilink instead of SEGGER But if you do through Codewarrior/KDS process should be the same. For custom board you need to follow some hardware tips which is mentioned in the thread provided. I have programmed KL05, K20, K60 using this method on the custom boards. and many other people also got benefited from this. I hope it helps you too. Good luck with it.

Regards

Amit

[ky3orr]

Hello Amit,

I have already read your post regarding unlocking and it worked for one of my 2 new resetting devices.

The remaining one is the one I put oscilloscope drawing above.

There is no way to unlock it following any available description now.

I guess 5[ms] is not enough for segger to communicate.

Is there anything I could do instead of once again resoldering board with new device?

[Amit_Kumar1]

Hi Grzegorz

I think you must have tried all the possible ways till now. Few last thing I would like to suggest you before desoldering the uc.

1) Do the continuity test for all the GND and VCC pins. (I lost my K10 uc as I powered it on with one GND left floating so it went to permanent resetting mode and didn't recover).

2) For custom made board check if tracks has hair cracks which obstructs the flow of current.

Hope u get the solution soon.

Regards

Amit