Michael et al,
That's always good news about a robust design, and setting higher test
standards than the minimum. My previous background was R&D QA testing, but
that's not my role here.
IMO more than 30 meters is a lot of separation. (Except of course for
hazardous applications.) If sensors and instrumentation must be over 30
meters apart, it creates new problems - possibly unsafe ground potentials.
We've seen recent postings on ground potentials, and again as long as
copper remains a popular way to interface equipment over long distances.
We have a new product that allows a low-cost non-industrial desktop PC to
control a slave CompacPCI array of instrumentation over a single FO pair.
It allows many channels of data to be collected, and then processed
remotely on a USD$500 PC. IMO the cost savings of the cheap PC (cheap
peripherals, plus having flexible upgrade options) buys a fair amount of FO
cable. We charge USD$600 for 100 meters of FO cable, or one can spend
USD$500 on our isolated serial port and then buy copper cables, that is, up
to the usable limit of RS-232/485.
But like you say, there will always be some need for copper cables, but
hopefully fewer as technologies move forward. Bluetooth or some other
wireless widget may some day make FO an anachronism. (Not even trying to
touch on overall system reliability issues here.)
Out of a catalog of hundreds of products, I have a single product that
suffered field failures from a blown serial I/O port that R&D could never
debug until we started doing the EFT test for CE, which reproduced the
failure. The fix was simply masking off paint from under the GPIB D-shell
connector.
I'm definitely NOT claiming our folks design any better (and I'm holding
back some opinions on this one), it just could be that we haven't focused
on an application that might be far more demanding than how our products
are generally used.
I'm trying to understand this situation. When I started here in 95
(leaving the ITE Class B EMC world) my expectations were for a flood of
nightmarish field problems. It hasn't happened. We only tested to the
light industrial immunity and Class A emissions. Perhaps, in this age of
operating systems that crash for unexplained reasons, designers and product
managers are viewing the few EMC related hardware failures with less
concern than with software/firmware failures that impact our users
regardless of their EMC environment.
In other words, to management our anecdotal reports of EMC problems seem
tame in comparison to software problems, or is my experience really all
that unique?
Eric Lifsey
Compliance Manager, National Instruments
USA 512-683-8474, Fax 512-683-8880
Michael
Taylor To: "'[email protected]'"
<MTaylor@hach <[email protected]>,
[email protected]
.com> cc:
Subject: RE: LMC With Regard To
Surge Test
12/21/2000 Cable Length
09:08 AM
Great idea - - However the whole purpose of equipment designed for -
"Electrical Equipment for Measurement, Control & Laboratory use ...." is to
measure & control. This can be most difficult to do if your sensors and
control points are forced to be located in the "building". In the case of
some of my company's products you most definitely want the process away
from
the building, unless you enjoy the smell of sewer lines & treatment
plants".
We, as a mater of policy, surge all lines over 3 meters in length. We have
found by sad experience that failing to do so invites customer field
"problems" and warranty issues. By hardening our I/O circuits we ensure
correct operation of our products even when our customers place one of our
sensors / devices next to a (several) hundred HP motor. In my opinion, We
should be testing to the realistic conditions our products will encounter
in
the real world. By doing so, my company has lowered customer complaints
and
warranty issues to the lowest in our industry segment. A dollar saved in
warranty is a dollar earned in profit. The net result of this is my
company
is healthy and I have a reasonable expectation of continued employment up
to
retirement (only 8 more years - yeaah).
Best regards to all for a great holliday season.
Michael Taylor - NCE
Principal EMC Engineer
Hach / American Sigma divisions
Danaher Corp.
-----Original Message-----
From: [email protected] [mailto:[email protected]]
Sent: Wednesday, December 20, 2000 2:37 PM
To: [email protected]
Subject: LMC With Regard To Surge Test Cable Length
EN 61326-1 Product Family,
Laboratory, measurement, and control equipment within the scope of EN
61326-1 does not require surge testing until reaching the definition of
Long Lines (see clause 3.6) which is NOT 10 meters, but does include lines
that can leave the building.
This is an important point. By instructing the user to not wire the
interface outside of a building and limiting the maximum cable length,
surge testing can be dropped from your test plans. Even better, it also
encourages the customer to seek a better solution to connecting buildings.
Copper based networks or industrial I/O busses should avoid being routed
outdoors in favor of a media converter allowing the use of a purely fibre
optic cable. This dramatically reduces the threat of a lightning induced
RF field, plus yields other odd benefits such as elimination of ground loop
potentials.
Proponents of trying to meet the surge requirement on copper interfaces are
encouraging the use of copper wiring between buildings which, in my
opinion, is risk to both connected equipment and anybody in contact with
the equipment at the moment a surge (or direct strike) occurs.
Best Regards,
Eric Lifsey
Compliance Manager, National Instruments
USA 512-683-8474, Fax 512-683-8880
-------------------------------------------
This message is from the IEEE EMC Society Product Safety
Technical Committee emc-pstc discussion list.
To cancel your subscription, send mail to:
[email protected]
with the single line:
unsubscribe emc-pstc
For help, send mail to the list administrators:
Jim Bacher: [email protected]
Michael Garretson: [email protected]
For policy questions, send mail to:
Richard Nute: [email protected]
-------------------------------------------
This message is from the IEEE EMC Society Product Safety
Technical Committee emc-pstc discussion list.
To cancel your subscription, send mail to:
[email protected]
with the single line:
unsubscribe emc-pstc
For help, send mail to the list administrators:
Jim Bacher: [email protected]
Michael Garretson: [email protected]
For policy questions, send mail to:
Richard Nute: [email protected]
