Looks like my last message was silently discarded. If it shows up again, then sorry for double-posting.
----- Ursprüngliche Nachricht ----- Von: Mark Rages Gesendet am: 15 Jun 2009 21:59:40 > 2.5% is over temperature (0..85C). I made a mistake when I wrote > 'frequency' above. It also varies with voltage a little bit -- looks > like it's calibrated at 3V. Still, stable enough for async comms. > I would guess that the oscillator is similarly stable at other > frequencies, but the datasheet doesn't guarantee it. The calibration > is just a value factory-stored in segment A memory that you can read > and program into BCSCTL1 and DCOCTL. Okay, then when using the calibration values, it might be sufficient for UART operation on this device. On the 1611 or 1232 (the two we are uisng) the temp drift is specified at 0.1%/°C, so it makes 8.5% over standard temp range. And this is only for 0.75MHz. (usually they pick the best case for datasheets, so it is likely that the drift is significantly higher at 4MHz). That's not good enough for UART, along with the fact that you have to do the absolute value calibration yourself too. Since in the original post the MSP was not specified... JMGross
