From: Michael Nudelman (email@example.com)
Date: Tue Nov 07 2000 - 06:29:02 PST
The way I used to do it:
1. Pick the most sensitive point on the analog portion (say, amplifier input).
Connect the AGND to CGND in this point. (whether through shells or directly).
2. Connect DGND to the same point of CGND.
3. If the device is to be grounded, connect protection GND to the same point of
4. If your cables are co-axials, the shell (especially for digital cables)
should be connected to GNDs at both ends; this will ensure the return current
path through the shell; otherwise the return current will go all over your DGND
and then through (God forbid) your AGND.
Avoid common paths for both DGND and AGND (say, a common wire, connecting both
GNDs to CGND). THis introduces GND noise to AGND.
"Pessotto, Michael" wrote:
> In a PCB with audio and video outputs referenced to AGND and digital outputs
> referenced to DGND what is the best way to connect the board grounds to the
> chassis? In our current design we have chassis connected to DGND through the
> digital output connector shells and through standoffs between board and
> chassis. The analog output connector shells are tied to AGND and are
> insulated from chassis. When we went to FCC pre-scan we saw the cables
> connected to AGND radiate with peaks corresponding to the clock frequencies
> used in the digital section. Shorting the analog connector shells to the
> chassis eliminated these peaks but introduced the noise from the digital
> plane onto our analog outputs. I thank you for your help in advance.
> **** To unsubscribe from si-list or si-list-digest: send e-mail to
> firstname.lastname@example.org. In the BODY of message put: UNSUBSCRIBE
> si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
> si-list archives are accessible at http://www.qsl.net/wb6tpu
email@example.com. In the BODY of message put: UNSUBSCRIBE
si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
This archive was generated by hypermail 2b29 : Tue May 08 2001 - 14:30:00 PDT