|Hey, PICers --
The 12C671 is a marvellous little machine. The chip of the future.
(Well, the near future, anyway.) But the data sheet looks like they put it
out in too much of a hurry. There are some gaps in the coverage.
Like OSCCAL: the trimmer for the built-in RC clock. From the description,
it sounded like a neat idea, so I did a little playing around with it. With
disconcerting results. I guess I need help.
The 12Cxxx OTP chips are so cheap that I buy 'em by the stick and use 'em
like Kleenex. But for this kind of playing around -- burn one, run it once,
and burn another -- I bought four of the JW type. The Datarase II wipes 'em
clean, four at a time, including the factory trim values at 03FFh. It doesn't
ask if you remembered to write those down.
So I labelled the four JWs: A, B, C, and D; and decided to calibrate each
one, learning at the same time how to use OSCCAL. The test program reads a
trim value from program memory, installs it in OSCCAL, and lets me measure
clock frequency from CLKOUT. There's one data point for one chip. Slow work.
And the data points don't make any sense yet, because I don't know what is
supposed to happen. Page 44 of the data sheet: "Adjusting CAL3-0 from 0000 to
1111 yields a higher clock speed. Set CALFST = 1 for greater increase in
frequency or set CALSLW = 1 for greater decrease in frequency." Oh, yeah!
The only consistent behavior I've found so far is the setting CALFST = 1
turns the clock OFF.
So hey, all you PICers -- anyone care to enlighten the rest of us?
-- Mel Evans aol.com (727)595-7685 mevans1027
More... (looser matching)
- Last day of these posts
- In 1998
, 1999 only
- New search...