I'm assuming you are just starting with programming, if so then a common
mistake is to forget to set the watch dog timer fuse to OFF.  That will
cause problems if you don't periodically reset the WDT in your code - and
most of the common LED flash programs don't do that.

Be sure you set the watch dog timer off when you program your chip. The
setting will be in the configuration/fuses menu in your programming
software.

If you are familiar with this problem and have allowed for it, please
disregard all that preceded 8)


Dan

-----Original Message-----
From: John Bes <jbes@E-TRADE.NL>
To: PICLIST@MITVMA.MIT.EDU <PICLIST@MITVMA.MIT.EDU>
Date: Tuesday, August 24, 1999 10:33 PM
Subject: Please Help - Osc.Problem


>Hi all,
>I think I'm having a real serious problem now ;)
>
>I build a programmer and a separate test board for the 16c84.
>Programming went well, but testing doesn't. The program in the Pic is the
>famous flashing led.
>
>The problem seems to be the crystal. It doesn't oscillate very well.
>
>I have been working on this problem for 2 weeks now, and I get kind of
>desperate ;(
>Please let me now if you have a probable solution!
>
>Here are the facts:
>
>- my scope shows a mildly distorted sine wave between osc1 and osc2 (pin
>15/16).
>- when I connect the scope -> it works
>The scope has R=1M and C=33pF -> when I replace the scope with a R (1m)
>between pin 15 and 16 and parallel to that, a C (33pF)
>no result.
>- when I touch the GND side of the caps with my fingers, sometimes the led
>flashes up.
>- The R (10 Ohm) is implemented to reduce the current through the crystal.
>replacing it by a wire doesn't make a lot of diffenrence, it seems.
>- I tried several crystals (4 MHz) -> No go.
>
>- connection:
>
>        GND
>       __|__
>     _|      |_
>     _       _ 2x C=47pF
>      |       |
>      |_|X|_|
>      |      |_
>      |     |  | R=10 Ohm
>      |     |_|
>      |16   |15
>   _ |____|______
>   | Pic 16c84    |
>   ---------------------
>
>Thanx,
>John