nRF51 DK reboots when trying to connect to custom board with needle adapter

[DerPMO]

I'm having major troubles with my debugging setup using a nRF51 DK (PCA10028 V1.2.0).

When I try to connect to the board I want to debug/program (which has an BlueMod +S = nRF51822), the nRF 51 DK seems to be rebooting (LED5 turns of for a second then comes back on, on the PC the JLINK Drive dis- and then reconnects).

Recovering device. This operation might take 30s.
ERROR: Cannot connect to any nRF device. Please make sure a device is
ERROR: connected to the debugger and supplied.

I've had this problem for a while which popped up once in a while at what seemed to be random, I figured it was because the onboard battery of my custom PCB was empty. And indeed when I recharged it, most of the times it would connect on the first try again.

After a while this started to happen more often, which seemed to correlate with me soldering stuff on the PCB. Turns out that wasn't the problem, apparently somehow my JLINK Needle Adapter wire became defect, no connection on Pin 4 TCK. (it is a weird setup with some adapters J-Link 9-pin Cortex-M (Debug Out) -> 20-pin 0.1'' JTAG -> 10-pin 0.1'' JTAG -> J-Link 10-pin Needle Adapter because apparently there is no easier way to get from 9-pin Cortex-M of Debug Out to JLINK Needle).

After replacing it, I could get to work again.

Long Story short: Now I have a PCB that I cannot recover anymore, every time I try to connect the above described behavior can be seen. Other PCBs still work, so this time I can rule out a cable defect.

Any suggestions on why this problem persists, how I managed to brick my nRF51822 or maybe even how to recover it?

Thx, Philip

[FormerMember]

One suggestion is to check if there is any short-circuits that aren't so easy to see on the board.

[DerPMO]

For the case that someone stumbles upon this question: It is possible to get the CPU into an unrecoverable state just by having some errors in your code (although I'm still not sure what the cause is). My problem actually was that the battery is soldered to the board and I didn't have any way to do a hard reset (power cycling). After I disconnected and reconnected the battery, the "bricked" PCB worked again. I guess having a button to trigger the chip reset line also would do the trick.