nRF51822 showing short in power after it is first powered on a custom board

[deepu]

hello,

We have developed a custom board corresponding to mbed pca10024 evaluation board.We haven't mounted the rf section and not even the crystal.The only thing that the board contains as hardware is nrf51822 IC,capacitors at AVDD,VDD.Then we tried programming the board using JLink-edu in keil and programmed it with "blinky example" provided by nordic.The programming was successful with led connected at a gpio pin started to blink for few seconds,and suddenly it stopped .When we checked the board it shows short in power.This happens for every board for which board detection and programming is successful for the very first time it is tried and then shows short in vcc and gnd

please help us sort out this issue

[deepu]

yes before testing no short but after that it looks shorted(in short detect mode)

[deepu]

Hi Guys, Thanks for your precious time and insight. Let me try to explain what we did,

After assembling the board, we do a bare board test where we check all impedance's including those across VDD, AVDD and capacitors provided for the Chips internal voltage regulators.

1. In all instance across four boards we have tested(bare boards before power up) , impedance were above 100K.
2. Then we Power up the board and test voltages. The board is powered up using a coin cell battery or a 3V supply. Measurement of voltage shows 3V. The board is disconnected and impedance checked, results when checked using a multi-meter resistance mode shows above 100K. 3)Board is powered up again, connected to a Jlink programmer and is flashed with blinky LED program. It works fine for few tries.
3. Then randomly after few tries, its no longer able to program the chip and when you check power supply impedance it shows typically less than 10 ohms(Not short but very low impedance).
4. The chip can no longer be programmed after this stage.

Analyzing more about the issue, our custom board is designed to use internal LDO. Only those components shown inside circles are assembled alongside nRF chip. Our PCB is a professionally manufactured 4 layer board with Impedance control.

The PCB board seems to be OK, it is the chip set that is getting damaged. No pop, no visible shorts. Once nRF chipset is removed the Impedance comes back upto 100K range.

Any thoughts?

[RK]

I believe we have given you all the insight we have. It sounds like you have blown something each time you test. Of course it's also possible that you have put code onto the chip and when you 'test' it later with your multimeter you implicitly power it up and the chip is now running code which sinks current and gives you an entirely bogus 'impedence' measurement. Perhaps you blew the LED.

[Alex]

Without looking at your schematic we can only guess. If you are ready to scrap at least one board you can either try removing all components one by one or apply power from bench PSU increasing current limit and checking temperature on the board either by thermometer with display or touching the board.

[Locky]

I suspect one of two things:

1. You have margin footprint for the chip. Perhaps a small solder bridge to ground
2. Check your LED GPIO circuit. As you said, the issue only appears after programming. Also double check to make sure you have set all your GPIO's properly in the code to match the board.