F103RB hardware trouble

john2 · ‎2009-10-06

Posted on October 06, 2009 at 13:49

john2 · ‎2011-05-17

Posted on May 17, 2011 at 13:25

Hi,

I'm using the F103RB on a small product. Some boards are working fine but I have a couple of boards that have died or never worked. One was working, I programmed the flash but at some point after that the JTAG stopped working (ULINK2 says 'no JTAG devices found'). It now runs hot, although the code is still running. The other board runs hot and the JTAG doesn't work so I can't do anything with it. The crystal oscillator is not running on this one either. They both take a high current, some 300 mA, probably more if the psu could supply it.

Holding NRST low doesn't change the current, so I think it's not software. Seems something in the chip has died, but what - and how to find out?

I have the ADC supply decoupled by a ferrite bead, 100nF, 10nF and a 10 uF capacitor. Vbat is connected to Vdd. Each Vdd pin has 100nF and 10nF.

By lifting power pins, it seems the fault is on the Vdd side (not Vbat or Vdda).

I'd appreciate any ideas! Is there any known failure mode with these symptoms?

Cheers

John

tomas23 · ‎2011-05-17

Posted on May 17, 2011 at 13:25

Hello John,

this problem occurs, in my experience, only if you used supply with higher than allowed voltage, even short spike can do it.

Pls check other peripherals around the MCU, but my advice is to exchange the chips with new ones and make sure that your voltage regulator doesn't generate any spikes (even at zero or low current).

john2 · ‎2011-05-17

Posted on May 17, 2011 at 13:25

Thanks Edison, I'll look into that.

It's powered by a 78L33 linear regulator, static levels are fine but it could be spiking at startup.

Cheers