RE: [SI-LIST] : Power Supply decoupling for Video Accelerator

About this list Date view Thread view Subject view Author view

From: S. Weir ([email protected])
Date: Sat Apr 21 2001 - 16:13:39 PDT


If you elect to use an LC filter, be careful that the filter does not
peak. A badly behaved LC filter is often worse than no filter at all. I
recommend that you place an appropriately chosen damping resistor across
the inductor to contain peaking. Alternatively, if the analog Vcc load
current is small, you can use a series resistor to the PI network. This
sort of network appears in PowerPC's. The advantage of this approach is
that it not only damps the inductor, but also the ESL of the capacitors in
the PI network.

Regards,

Steve.

>>>one very good ground plane and one high
capacitance capacitor at the entrance of the power supply into the board
and one well selected capacitor (parasitic inductance, charge, ...) close
to the chip pins.
>>>
may not be enough, at least the part with "one well selected capacitor
(parasitic inductance, charge, ...) close
to the chip pins". Of course, it depends on your application, but generally
you want well-decoupled chip and this may mean well-decoupled planes and the
VCC/GND pins run to their respective planes with short fat traces and vias.
However bunch of different grounds etc may be an overkill, or even a
detrimental practice. What happens sometimes (and I did see exactly that)
the application people in the chip manufacturing company create a small eval
board and connect it to the power supplies in such a way that the DUT
requires some crazy de-coupling techniques.
When you have something really sensitive (PLL loop VCC, for example) you may
want to filter the VCC by a simple PI, using an inductance and an additional
cap. In case of more sensitivity you need to use something drastic. But you
have to find out how sensitive your chip is.

Mike
At 11:49 AM 4/20/01 -0400, you wrote:
>[email protected]

**** To unsubscribe from si-list or si-list-digest: send e-mail to
[email protected]. 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 Jun 21 2001 - 10:11:41 PDT