From: Ray Anderson ([email protected])
Date: Sat Sep 23 2000 - 02:11:02 PDT
Don't know about the HP8591E (9 kHz to 1.8 GHz), but I've got a
similar instrument, HP8594E (9 kHz to 2.9 GHz) that allows up to 30 VDC
at the input connector. I'd suspect the 8591 has similar input voltage
capability, but you need to verify it prior to connecting it to a
live circuit. A broadband DC block at the input of an analyzer that
doesn't allow DC will provide the same capability.
A couple things to be concerned with. At frequencies above a couple
hundred MHz the location where you probe on the PCB can make a very
large difference in the levels of RF you will measure.
You might think that doing an A/B comparison between a system with
one set of decoupling caps versus the same system with a different
set of caps and measured at the same place on the PCB would be a
straight forward experiment to interpret. Not always so.....
Recently we did quite a similar experiment, i.e, measure the spectral
signature at a specific place on a PCB, change a whole bunch of decap
values and then remeasured the spectral signature. In this case the noise
near the processor (where the majority of the decaps were changed) went
down as predicted, however the noise at another place on the board a
few inches from the processor went up.
Why ?? Because by making the impedance at location A very low by selecting
caps to provide excellent bypassing over a certain range of frequencies,
the impedance at location B about a quarter wavelength increased in value.
(a quarter wave impedance inverter). (It's kind of like a perforated
container full of jelly. Push in at one location and it pops out somewhere
else :) By changing decoupling caps to decrease RF levels at one location
on a board you may actually be making matters worse at some other location.
Quieting the entire board may not be a trivial exercise. It can involve
adding additional decoupling caps at locations where they didn't exist before.
So the bottom line is: It is relatively easy to make the measurement,
but interpretation of the results may take a bit of thought.
-Ray Anderson
Sun Microsystems
>
> To: "'EMC - PSTC Forum'" <[email protected]>
> Cc: Roman Seifert <[email protected]>
> Subject: Probing power plane with analyser.
> From: "Cameron O'phee" <O'[email protected]>
> Date: Thu, 28 Sep 2000 11:59:24 +1100
>
>
> Hello All,
>
> I have been asked by one of our design engineers if I can measure the RF
> signature on the 3v plane on one of our PCB designs. The purpose of this is
> to compare it to proposed changes for cost reduction, ie removal of bypass
> caps. I would assume I need some sort of CDN but I have no designs for this
> application. Does anyone have any ideas on how I can safely probe a 3v DC
> power rail with an EMC analyser (HP8591E)?
>
>
> Regards,
>
> Cameron O'Phee.
> EMC & Safety Precompliance.
> Aristocrat Technologies Australia.
>
> Telephone : +61 2 9697 4420
> Facsimile : +61 2 9663 1412
> Mobile : 0418 464 016
>
>
> Thanks.
> Best Regards,
> Barry Ma <[email protected]>
> ANRITSU www.anritsu.com
> Morgan Hill, CA 95037
> Tel. 408-778-2000 x 4465
> _______________________________________________________________________
>
> Free Unlimited Internet Access! Try it now!
> http://www.zdnet.com/downloads/altavista/index.html
>
> _______________________________________________________________________
>
>
> **** 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
> ****
>
**** 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
****
This archive was generated by hypermail 2b29 : Tue May 08 2001 - 14:29:36 PDT