Colan,

     Some possibles:-

     1).As others have said, temperature variation is a real problem with
     32.678kHz 'watch crystals' - max temp coefficient tends to be
     -0.04ppm x (degree C difference from 25 degree C) squared
     but this is only -9ppm extra error across 10 - 40 degree C.
     Over a larger range I would get the PIC to log the crystal temperature
     and compensate for the coefficient in S/W.

     2). Are you overdriving the crystal - some watch crystals are 1uW max
     or less. What resister have you got in-between the PIC output pin and
     the crystal connection.

     3). Noise from digital tracks in the area can also upset the RTC - it
     tends to run fast if it has this problem - Always put a guard ring
     around your crystal and PIC input / output pins, a gnd plane
     underneath and no other tracks within the guard ring.

     4). What is the load capacitance of the crystal you are using? I
     usually use 12pf crystals, then have the cap on PIC input side = 15pf
     and 27pf on the PIC output side - rest of the capacitance comes from
     the PCB and PIC leads, (not easily controlled, which is why for long
     term, I use the Epson serial RTCs with built-in crystals, which can be
     less than 2 dollars in production qtys)

     Hope these comments help,

     Steve Lawther


______________________________ Reply Separator
_________________________________
Subject: Real Time Clock Accuracy
Author:  "Colan Tredrea" [SMTP:tredrea@AGENG.UNP.AC.ZA] at Exchange-UK
Date:    23/09/98 09:21


Thank you to those who responded to my request for help.

I need to add the following points:

1. I am unable to make substantial changes to the design.  I need to
improve the accuracy of what I already have, ie the simple crystal
oscillator using PIC16C63 Timer1.

2.  The errors I have mentioned have occured over a small ambient
temperature range of approximately 25 deg C +- 10 deg C, at the very
most.  I would like to maintain an accuracy of inside 1 second per day
over the range 10 to 40 deg C.

Thank you.

Colan Tredrea
University of Natal
South Africa
tredrea@ageng.unmp.ac.za

     <snipped>