From: Doug McKean (email@example.com)
Date: Fri Mar 24 2000 - 16:38:30 PST
Lum Wee Mei wrote:
> I have a very small design that has a servo amplifier (module) with some
> digital and analog sections. As space is a constraints, the big servo
> amplifier will be on one side while the rest of the components will be
> on the other. I was suggesting, at most 6 layers with L1-S, L2-S,
> L3-DGND, L4-VCC, L5-AGND and L6-S. In the worst case, a very small
> portion of the AGND will need to be used as DGND, if I need to place one
> or two digital components with the servo amp.
> The EMC engineer suggested otherwise - 8 layers. The extra 2 layers, are
> for routing a crystal clock signal and its gnd plane. Another
> recommendation by the EMC engineer is that this servo amplifier has to
> be enclosed in a shielded box to prevent EMI. If the PWM provides at
> most 16MHz, do we still require this shielded box as space is a real
> I was told in the end, all the grounds are to be shorted at one point.
> If that is the case, which would be the best method to do so, at the
> mating connector or within the board itself?
> Any responds will be deeply appreciated.
> Regards - Wee Mei
I've worked on servo designs.
Yes, the freqs are low.
I'm only guessing, but what your EMI engineer
*may* be suggesting, at least I would, is
prevention of ground bounce by heavy currents
in the analog section. This can translate to
common mode currents to any interface or motor
cables. It also translates to what can get
into your product as well.
Remember that the other side of EMI work is
immunity. Easy egress for energy can necessarily
mean easy entrance for energy. If the ground of
the board is susceptible to it's own bounce of
say 5, 12, or 24 volts, guess what the chances
of being susceptible to transients and surges
in the hundreds of volts or ESD testing at 8 kV?
One way, merely a suggestion, to prevent that
and increase robustness is to increase the number
of ground planes for a higher conductive ground
Regards, Doug McKean
**** 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
This archive was generated by hypermail 2b29 : Thu Apr 20 2000 - 11:35:54 PDT