Re: [SI-LIST] : Number of GND/Power pins in a connector ?

About this list Date view Thread view Subject view Author view

From: Lum Wee Mei (lweemei@dso.org.sg)
Date: Thu Mar 23 2000 - 15:18:50 PST


Since we are in this question, I would like to add one :

Is there a preferrence to use female connector on a board? I was told that male connector is not prefer because the pins may acts as antenna loops when the board is left alone.

Can someone enlighten me on this subject?
Thanks.

Scott McMorrow wrote:

> Stuart,
>
> "It depends" is the correct answer.
>
> The best way to size the number of ground and power pins in
> a connector is to simulate the regions on either side of the
> connector. The connector model should be fully coupled for
> all pins including power and ground. You should grid the ground
> and power planes around the connector edge and within the
> grid include the decoupling capacitors. A connector swath
> should be simulated with data transmitted in both directions
> (if bidirectional) and driven by worst case driver edge rates into
> transmission lines which are referenced to the non-ideal power
> and ground grid. Multiple data patterns should be simulated
> for worst case pattern sensitivity.
>
> These sorts of simulations will answer several questions:
>
> How much data skew and jitter is caused by connector crosstalk,
> and return path effects for each ground/power pattern?
>
> What happens when a signal is referenced to a ground plane
> on one side of the connector and to a power plane on the
> other side?
>
> How much noise is induced by data switching into the ground and
> power grids on either side of the connector?
>
> What is the worst case instantaneous voltage differential
> across the ground pins on either side of the connector?
>
> What is the worst case instantaneous voltage differential
> across the power pins on either side of the connector?
>
> And for bonus points:
>
> If you have accurate modeling of the power switching currents of
> the devices on either side of the connector, you can simulate the
> worst case connector and plane noise due to power transients
> and data switching transients.
>
> Depending on the edge rate and the quality of the connector you
> may find that although you need few power and ground pins to
> facilitate clean power delivery, you may need many more power
> and ground pins to facilited clean signal delivery.
>
> regards,
>
> scott
>
> --
> Scott McMorrow
> Principal Engineer
> SiQual, Signal Quality Engineering
> 18735 SW Boones Ferry Road
> Tualatin, OR 97062-3090
> (503) 885-1231
> http://www.siqual.com
>
> Stuart Adams wrote:
>
> > How do I calculate the number of ground and power
> > pins I need in a board-to-board connector.
> >
> > The number of grounds is a function of switching
> > current and how many signals will switch at the same
> > time, correct ?
> >
> > More grounds is always better but what about power pins ??
> >
> > If one or two pins are sufficient to carry the max steady
> > state current and my board is well bypassed with bulk and
> > ceramic caps, is there a need for alot more power pins ?
> >
> > -- Stuart
> >
> > **** To unsubscribe from si-list or si-list-digest: send e-mail to majordomo@silab.eng.sun.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
> > ****
>
> **** To unsubscribe from si-list or si-list-digest: send e-mail to majordomo@silab.eng.sun.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
> ****

**** To unsubscribe from si-list or si-list-digest: send e-mail to majordomo@silab.eng.sun.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
****


About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Apr 20 2000 - 11:35:52 PDT