From: Jim Thompson
Subject: Re: Querry, Kick Starting Crystal Oscilator.
References: <3DCA905A.59FD928B@mmm.com.DELETETHIS> <3DCAE105.6CAAD7B9@mmm.com.DELETETHIS> <3DCBDC8E.2E54EBC@mmm.com.DELETETHIS> <3DCBF14F.A3B9BD76@mmm.com.DELETETHIS>
X-Newsreader: Forte Agent 1.91/32.564
Date: Fri, 08 Nov 2002 17:39:49 GMT
NNTP-Posting-Date: Fri, 08 Nov 2002 12:39:49 EST
Organization: Cox Communications
On Fri, 08 Nov 2002 11:15:59 -0600,
firstname.lastname@example.org (Roy McCammon),
In Newsgroup: sci.electronics.design,
Entitled: "Re: Querry, Kick Starting Crystal Oscilator.",
Wrote the following:
|Jim Thompson wrote:
|> On Fri, 08 Nov 2002 09:47:26 -0600,
|> email@example.com (Roy McCammon),
|> In Newsgroup: sci.electronics.design,
|> Article: <3DCBDC8E.2E54EBC@mmm.com.DELETETHIS>,
|> Entitled: "Re: Querry, Kick Starting Crystal Oscilator.",
|> Wrote the following:
|> |Jim Thompson wrote:
|> |> |Just looking at the specs on the processor.
|> |> |use this crystal in this circuit and wait this long).
|> |> |
|> |> Can you reference a processor data sheet?
|> |Jim, I cannot reveal the processor, but it is nothing
|> |special in this regard. You simply hang a 32KHz crystal
|> |on it and it oscillates after a while. The reference
|> |design holds the chip in reset for 2.1 seconds for
|> |the purpose of letting that oscillator start. The
|> |manufacturer isn't much help (they are digital guys).
|> |All they say is they don't know how fast it starts
|> |and that we better stick to the reference design.
|> How long is it taking to reach equilibrium?
|I cannot scope it out right now, and even if I
|could, it almost certainly would not be a worst case.
|Typically, I would expect it to be stable in a
|few 100 ms or so.
|Can we have a discussion about the technique
|in general without referring to a specific part?
|> ...Jim Thompson
You're being too evasive for me to discern whether it's a "Q" problem
or a bias stabilization problem.
If it's a problem with getting the bias centered I'd just momentarily
short the crystal with an analog switch. (Or use a tri-statable
inverter or buffer.)
If it's "Q" that's the problem I'm not sure what you can do... unless
you can live with less accuracy for your clock.
Or use an external low-power crystal oscillator that runs all the time
(watches do seem to run for quite a long time on a button battery ;-).
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice:(480)460-2350 | |
| Jim-T@analog_innovations.com Fax:(480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |
For proper E-mail replies SWAP "-" and "_"
Don't buy ANYTHING from CSENSORS.............Contact me for details