Dammit. I got bit again. Now those who know me here, know that this is a hot button issue for me, and I'm posting this to illustrate just how sneaky this problem is. I'm currently using an M644 with 20 Mhz external crystal, for a sensor processing app. I'd scanned the data sheet, and did not see any mention of the dreaded "CKOPT" fuse, and so ASSUMED that it was not an issue. WRONG.. The low fuses for my crystal should have been set to 0xD7, but I'd set them to 0xDF, which is one bit off, and activates the low power "vittoz" mode oscillator. I've been having bad and intermittent problems with the Jtag ice MkII, and I'd been having glitches in my serial output, which I was thinking were probably some issue with my code. Now that the fuses are properly set, I'll do some testing on Monday and verify with a single build on both modes, wether those glitches were caused by clocking problems. Only time will tell on the Ice problems, they come and go seemingly at random. I do wonder, is there anyone out there who actually is intentionally using the low power oscillator, with a properly designed and reliable crystal circuit? I'll repeat my advice on this, NEVER NEVER NEVER USE THE LOW POWER CRYSTAL OSCILLATOR. If you think you need it, and you're SURE you understand all the implications, and you're SURE that you've made all the measurements, then feel free to ignore my advice. For hobbyists, and even most commercial users, this mode is just absolutely begging for trouble. -- There is no computer problem which cannot be solved by proper application of a sufficiently large hammer.
Message
CKOPT Again..
2009-04-12 by David VanHorn
Attachments
- No local attachments were found for this message.