Okay here I go starting another forest fire!
Before answering the question I will put in my two cents worth.
Regardless of what the test house does I would strongly urge you not to
accept equipment for production with a single sample test that measures
close to the limit. A onesy-twosy kind of thing is about as risky as you
want to get, beta sites for example. Even then you want to fix the
problem and consider retrofitting those onesy-twosy field units. As a
pragmatic matter I would recommend that you adopt at least a 3 or 4 dB
margin requirement. I have seen even more restrictive requirements but
generally somebody ends up paying for the extra margin.
Test houses - please read all of this before flying off the
handle. You are not being attacked - after reading sending this I will
put on my asbestos underwear and you can fling the pertinent citations
out of the CFR47, or CISPR documents at me, and I will change my ways.
Having said that, the FCC holds the manufacturer - or once in the field
the customer (read manufacturer) - responsible for the emissions not the
test house, and they do tend to forget that. What they are attesting to
in all of their documents etc. is that they have the capability in
equipment, processes and test personnel to make tests that the FCC and
CISPR would deem accurate and reliable.
The legal limits are the limits and there is no provision for
guard-banding by the regulatory agencies. The test houses cannot usurp
the FCC/CISPR's legal authority by imposing stricter requirements. Why
stop at 2 why not require 3 or 4 before you issue a report? I would
strongly suspect that the FCC/CISPR injected into the published limits
some comfort level of their own on the effects of having equipment a 0
dB. Years ago the VDE staff used to require 2 dB of margin and while I
never saw the legal requirement for it in writing they simply would not
accept the data.
The test houses have a reputation to protect. Its toward this end, I
believe, that they balk at providing a pass within some close proximity
to the legal limits. I support their hesitation but not there refusal.
They have a strong desire to never get overturned by and FCC audit. 2 dB
isn't going to allow them to sleep much better and I believe its a
backhanded complement to them that they have some resistance.
However, this is not what they are chartered to do. They are only
charted to make the measurements and compare them to the limits provided
by the legal authority. The more appropriate response to this problem is
for them to accurately describe the test set-up, etc, along with notes
in the report of the uncertainty of the test and any other concerns they
might have. Concerns such as 2 dB being with those limits of uncertainty
and that they no confidence that production lots will always meet these
limits.
They have now properly warned you of the consequences and clearly shoved
the whole responsibility back to where it properly exists - you! They
will have in their files the report including their concerns and can use
that to clearly defend themselves should you have any latter regrets.
You might not want your customers to read those kinds of comments but
that is the price for accepting the responsibility for being so close to
the limits.
It is interesting to note that if you chose a site that was not
qualified to run these tests and indeed the equipment actually did not
meet the limits there is no financial or legal risk to the test house.
It is your responsibility to insure that the test house is competent to
perform the tests.
The closest to a margin requirement that I know of is the 80/80
stipulation of CISPR. That still doesn't allow you to exceed the limits
for any single emission but by testing multiple units and applying the
statistical process to it they have gained some assurance that while the
five units might be very near the limits it is slightly more likely that
your production units will also meet the limits in the long run.
Actually, the 80/80 does applies for all tests not just those close to
the limit. not apply only when close to the limits but any test no
matter how much margin is provided during the first test).
I would have more faith with a test partner that works with you by
advising against using results that are so close than in working with
one that will pass anything you give them.
Damn the torpedoes fire away!
Gmac
-----Original Message-----
From: [email protected] [SMTP:[email protected]]
Sent: Monday, October 26, 1998 9:50 AM
To: [email protected]
Subject: Required Margin for EMI?
Hello all,
When doing EMI scans of ITE, I have always worked with a margin
of 2 dB
less than the actual limit. No test house I have ever been to
will give a
"meets the requirements..." without having this margin. Can
anyone point
out if this is an actual requirement built into any of the
standards, i.e.
EN 55022 or FCC Part 15, etc. and if so, where is it in writing?
I understand about measurement errors, product to product
differences and
repeatability of testing issues. Just want to know if the 2dB
margin is
actually written into law.
Thanks for your comments.
Scott
[email protected]
---------
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).
