Hello,
Will the internal oscillator block of MSP430F4250 get damaged if the XIN and XOUT pins of the uC gets shorted?
Explanation:
I faced oscillator startup issues in a circuit consisting of MSP430F4250. It used a 32 kHz crystal with ~12pF load capacitance and the crystal did not oscillate on power-up from time to time.
In an attempt to get it running with a higher frequency crystal, I tried using a 4MHz crystal with external caps C1=C2=12 pF and internal caps disabled [effective load capacitance = (12+2)/2 = 7 pF]. The crystal did not oscillate.
The effective load cap being lower than required by the crystal (~18pF), I enabled the internal caps also (XCAP18PF) and powered up the uC [eff load cap= (18+12+2)/2= 16 pF]. The crystal started oscillating, but unreliably, as was the case with 32 kHz crystal. I probed the XOUT pin of the uC on an oscilloscope and powered up the uC. It showed a 4MHz sine wave with an initially large amplitude but decreasing with time. The sine wave disappeared in a matter of 2-3 seconds after power-up.
Code
FLL_CTL0 |= DCOPLUS + XTS_FLL + XCAP18PF; // fDCO not divided, high freq crystal SCFI0 |= FN_4; // 2.8-26.6 MHz SCFQCTL = 0x01; // N=1. fdcoclock = D*(N+1)*fcrystal = 1*(1+1)*4MHz = 8 MHz
And then I accidentally shorted the leads of the crystal with the oscilloscope probe. I cycled the power 50-60 times to see the sine wave, but to no avail. To know whether I damaged the uC, I programmed it to use the internal 12kHz VLO and it worked.. which made me wonder: can the internal oscillator block of the uC get damaged while the rest of the internals remain healthy?
Edit: I'm aware of the issues associated with probing the crystal pins directly (probe cap + input impedance). The uC was not generating ACLK output and hence I chose to directly probe the crystal pins.