Just so you don't think I've forgotten. I pointed this same issue out
with the possible new requirements for conducted emissions tests for I/O
cables. If there ain't a problem quite screwing around fixing it. If the
current test method - in this case limit - is addressing the problem
then leave it alone and work on more important issues. There is
non-compliant equipment out there, but I haven't seen any date that says
the bulk of this equipment is non-compliant either because of the test
methodology or because of inappropriate limits. Unless research shows
that to be the case, and I'll admit from where I sit I don't have a lot
of access to the data, leave it alone.
I believe the change to the C63.4 tests and the movement upward movement
of the test frequencies is a reasonable example of change. Having cables
in a multitude of arrangements, dangle here and there, and the increase
in processor speeds necessitated change, but I don't see it here.
If it ain't broke don't fix it.
Just FYI - we don't go into production until I have a reasonable level
of margin and we have a design goal of meeting class B even when we
currently are only required class A. I mention this only to assure you
that I am not against the requirements or trying to slide a bunch of
questionable equipment onto the market. I simply believe if it ain't
broke don't fix it.
Gary
-----Original Message-----
From: Jon D. Curtis [SMTP:[email protected]]
Sent: Tuesday, October 27, 1998 5:37 AM
To: Chris Dupres; [email protected]
Subject: Re: EMC limits.
Passing is passing. The limit is passing. Curtis-Straus will
write you a
report which says you pass if we observe zero dB or greater
margin. Just don't
expect to pass with great certainty on the next sample.
I would strongly suggest that you consider requiring a margin if
the product
will evolve and need to be retested or if you sell to a systems
integrator who
may have his/her system tested. But it is up to you. 4 dB is a
nice margin.
Now for the interesting comments:
Consider that some are lobbying for a new limit which is lower
than the old
limit by the uncertainty of measurement. There has always been
an
uncertainty. In fact, it's been going down as the techniques of
EMI
measurement become better understood. And emissions problems
have been fairly
benign from equipment tested to the limit with the older
measuring techniques
(those with high uncertanty).
What this discussion needs is a quantification of the old
uncertainty. For
argument let's say it's 7dB. If we want to take uncertainty
into account in a
pass/fail decision, we should compare the emission to the limit
plus 7dB minus
the actual uncertainty.
In other words, discussions of uncertainty should be viewed as a
way to raise
the limit while maintaining the protection objectives of various
regulations.
Or, uncertainty was never considered before, so consideration of
it now without
a limit increase is in fact a limit decrease. Since the
existing limit has
been working to control interference for some time now, any
decrease is likely
to be a brake on economic growth (it'll cost more to comply)
with little
benefit to the community (no reduction in non-existant emissions
problems).
Jon Curtis.
Chris Dupres wrote:
> EMC Folk.
>
> I have been reading all the learned submissions of what
constitutes the
> acceptable emission limits for EMC purposes. Most of you are
very clever,
> very technical, and I'm in awe of all of you.
>
> But there seems to be a bit of a missed point here. EMC in
Europe relates
> to the EMC Directive, which was born of the SIngle Market
arrangements
> between Euro States, and which were born of the Treaty of Rome
way back
> before my kids were born.
>
> The ultimate purpose of EMC Standards/limits is to provide a
level trading
> platform for Euro countries, so that all conditions are equal
in the market
> place, and that no-one can steal a lead over someone else by
dropping
> technical standards and therefore saving costs and putting
cheaper goods on
> the market. It follows that perhaps we should look at these
limits in the
> same way that the packaging industry looks at filling cans and
bottles, or
> the way car drivers treat speed limits. i.e, that the EMC
emission limits
> are a target in absolute terms, and if you can show honest
intent in
> achieving them, then the legislation has achieved it's aim.
>
> If I carry out an honest emissions test on a piece of
equipment, and the
> graph is below the line by the thickness of the pen, then I
believe that
> the spirit of the EMC Directive has been met. If this
acceptance level was
> an absolute amount, such as money in banking, then I would
allow a % for
> measurement error, but it isn't, it's an objective. No-ones
head is going
> to explode if the emissions are 0.5dB over limit, and in all
honesty
> dropping the emissions by 0.5dB can usually be achieved by
moving a cable
> or snapping on a ferrite sleeve. Hardly enough to change the
whole balance
> of trade in Europe is it?
>
> So, if I carry out properly conducted tests, with the
equipment working
> normally, and it shows emissions right on the limit, then I
think the EMC
> Directive has been followed, and the equipment can be CE
marked with
> honesty and placed on the market.
>
> Just a tuppence worth (what's THAT in Euro's?)
>
> Chris Dupres
> Surrey, UK.
>
> ---------
> This message is coming from the emc-pstc discussion list.
> To cancel your subscription, send mail to [email protected]
> with the single line: "unsubscribe emc-pstc" (without the
> quotes). For help, send mail to [email protected],
> [email protected], [email protected], or
> [email protected] (the list administrators).
--
Jon D. Curtis, PE
Curtis-Straus LLC [email protected]
Laboratory for EMC, Safety, NEBS, SEMI-S2 and Telecom
527 Great Road voice (978) 486-8880
Littleton, MA 01460 fax (978) 486-8828
http://www.curtis-straus.com
---------
This message is coming from the emc-pstc discussion list.
To cancel your subscription, send mail to [email protected]
with the single line: "unsubscribe emc-pstc" (without the
quotes). For help, send mail to [email protected],
[email protected], [email protected], or
[email protected] (the list administrators).
---------
This message is coming from the emc-pstc discussion list.
To cancel your subscription, send mail to [email protected]
with the single line: "unsubscribe emc-pstc" (without the
quotes). For help, send mail to [email protected],
[email protected], [email protected], or
[email protected] (the list administrators).
