Thanks Art.

Your explanation seems to be right. I don't understand why it's not
published in a datasheet. I am going to do some measurement with RC
oscillator tomorrow. I was certain I was able to wake up the PIC during an
added zero byte of at the beginning of a transmition running at 19200. There
isn't any problem to change a timing, but I love clean solutions.

Igor


-----Original Message-----
From: pic microcontroller discussion list [mailto:PICLIST@MITVMA.MIT.EDU]On
Behalf Of Art
Sent: Monday, June 09, 2003 11:20 PM
To: PICLIST@MITVMA.MIT.EDU
Subject: Re: [PIC:] WAke-up from sleep


We have experience with crystal oscillators. Believe me, they do take that
long to start, especially if you want it to be stable. A premature
activation of the chip with an ambiguous oscillator condition could lead to
problems.

I'm sure this is an arbitrary delay imposed by the chip maker, but it is
probably based on reality and done for a good reason. They can't predict
what type of crystal you will use either. There are different types too, no
doubt this delay takes into account the 'worst' case regarding this issue.

Just be happy you aren't using an LC network based oscillator! They make
crystal oscillator start up times look like a jackrabbit compared to a
snail.

Good luck to all.

Art

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu