Simulations are tricky.
You've got to be aware of at least two primary things:
1) What are the models used by the software?
What are the assumptions about parasitics,
ground planes, a trace?
2) What is the actual number crunching doing?
Is it crunching out something very carefully based
on Green's Function, or some programmer's idea of
what a transmission line should be?
In other words, one shouldn't JUST plug and chug.
You have to be at least aware of some things, in
some cases aware of A LOT of things, about simulators.
Unfortunately, a lot of us don't have the time nor
the money to get an M.S. in electromagnetics.
One of the continual points brought up with EMC
people is being able to get a bunch of data about
"something over here" (in this case data points
in or around a product) that will *reliably* help
you predict what will happen "over there" (at the
test antenna of an FCC site) up to 10 meters away
from the product.
There are alot of arguments pro and con about this and
we had quite a lively one at the last EMC symposium.
Majority seems to favor the cons, i.e. people tend to
doubt being able to reliably and repeatedly predict
different structures (like the case of your product)
other than well known and predictable antennas.
Sorry for being "Mr. Gloom and Doom", but I tend to
agree. This is from theoretic and actual hands
on experience with hard data.
Simulators are advancing in techniques, but until they
get any better AND cheaper, you'll have to live with
maybe up to a 10 dB variance between your results pitted
against an FCC test site. And 10 dB over the limit can
turn into a nightmare very easily.
Immunity is quite another matter also.
Yea? Well, what matters are those Doug?
Glad you asked.
Suppose for the moment you have to contain RFI from a
5 volt DC circuit inside your product. Some will say
that as long as the product does well with emissions,
it will do well with immunity.
Not so fast Sherlock.
ESD levels deal in the thousands of volts. So, in case
we have a 5V source on the inside of your product and in
another case we have a 5,000 volt source on the outside.
A difference of the two by three orders or magnitude.
Comparison of the two is almost like apples and oranges.
Yes, a buttoned up product that does well in emissions
can't hurt, but there's no guarantee.
Guess I really didn't answer your question, but
hope my rambling helped.
Regards, Doug
At 09:35 AM 10/23/98 -0500, Dave Artman wrote:
>Hello
>
>Am looking for a software package that can simulate RFI susceptibility, we
are the victim.
>Specifically would like to test pcb layout vulnerability and what nets or
areas are being affected.
>A plus would be optimizing features to help solve susceptibility issues.
A second plus would be
>an interface with Accel EDA.
>
>We currently use a typical RFI setup, signal gen. and antennas at various
distances and
>orientations etc.. Prototyping our pcbs is becoming more expensive so am
looking for cost
>effective alternatives and possible decreased design time.
>
>Any help or leads would be appreciated.
################################################################
DesignerCouncil E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
################################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
To subscribe: SUBSCRIBE DesignerCouncil <your full name>
To unsubscribe: SIGNOFF DesignerCouncil
################################################################
Please visit IPC's web site (http://www.ipc.org) "On-Line Services" section for additional information.
For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312
################################################################
|
