I used the RC oscillator reliably on hundreds of 16F872-based boards=20
using a 4.7k resistor and 56pf resistor.  It's been a while, but IIRC=20
the frequency was around 1Mhz (but don't hold me to that).  Also never=20
understood why ballpark values weren't put in the datasheet, and=20
would've liked something more "known", but RC was great for debugging --=20
I could drop in a much larger cap and slow down the PIC to a crawl, so=20
LED's would serve as a crude logic analyzer.

Cheers,
-Neil.


On 8/22/2012 10:29 AM, Peter P. wrote:
> I made the RC osc work on 16f57. It requires a minimum of 18k as R to wor=
k at all. I chose 22k for reliability. With 18k and 47pF at 5V and 30C it r=
an at about 100kHz (at CLK2 =3D=3D CLK1/4). I would really like to not than=
k mchip for not wasting the 3 minutes it takes to put in some decent limits=
 for the R and the C. The 18k minimum R is significantly different from pre=
viously listed minimums (about 10 times higher than the one implied for 16c=
57, the last chip of this class for which I found data in its datasheet).
>
> With lead frame capacitance given as ~5pF for DIP I infer that the RC osc=
 can go about 8 times faster than with 47pF external cap, namely about 800k=
Hz maximum core clock (6.4MHz at CLK1).
>
> -- Peter
>

--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.