NXP LPC1857 goes into unprogrammable state

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

NXP LPC1857 goes into unprogrammable state

575 Views
jean-riegardtva
Contributor I

Hi

We as a company have developed a LCD display board for our project based around the NXP LPC1857. We have been at it for roughly two years now and we have come across a somewhat disturbing phenomenon. While debugging some of the boards with the J-Link Ultra+, they sometimes tend to go into a state where the can not be programmed again. When you hit download and debug in IAR, the code is built and attempted to be programmed to the LPC1857. The J-Link lights up like a christmas tree flashing a number of different colors across all three LED's and then the message pops up that states "Failed to get CPU status after 4 retries. Retry?" At this point we can not salvage the LPC1857 chip. It still runs the last version of code that was flashed to it, but it will not be picked up by IAR, repeating the procedure above. This is quite disturbing as we have lost a number of processors this way and it is quite sporadic. We have investigated the possibility of it being a DAP error but it is not that. We were able to recover other LPC chips from DAP errors and following the same procedure does not work.We have tried using J Link Commander but it also is not able to recover the damaged chips. The only way to recover our boards is to have the NXP 1857 chip replaced and that costs a lot of time and money. Do you have any advice for us?

Kind Regards

Labels (1)
0 Kudos
1 Reply

427 Views
bernhardfink
NXP Employee
NXP Employee

We have seen sporadically electrical damages on the JTAG/SWD port pins by connecting JTAG debugger boxes.

If you have supplied the board and the JTAG debugger by different switch power supply domains, there might be a huge difference in the electrical potential.

With the physical outline of a JTAG connector it is not guaranteed, that the GND pins are connected first, so it could well be that 2 signal pins get connected first which have a difference in potential of several 100V. For a very short time there is a peak current which could destroy a JTAG/SWD input pin. Nothing you could really avoid, only something which you can protect against.

For boards getting connected to debuggers regularily, we recommended to add protection diodes to the JTAG/SWD input pins. Or you alway connect the debugger to the board before you connect it to power.

If you stiill can access with the ISP mode using the UART interface, but can't connect with JTAG/SWD anymore, then most likely you deal with this issue.

Regards,

Bernhard.

0 Kudos