I want to share with you a potential pitfall in conducted emission testing of
IT equipment using an ISN (or alternative) on an Ethernet connection.
The pitfall will have to lead to adjustment of the test requirements for
IT equipment in this set-up, and/or to a change in requirements for
modular DC power supplies.
Many (IT) equipment is offered with a modular power supply.
The equipment as a combination is being tested on the AC side for
conducted emissions, and in some cases also on the DC-side of the EUT
(depending on many factors) but never on the DC output of the power supply.
Now requirements exist for testing on Ethernet ports using an ISN or other
method. In one special case a part of equipment was tested on this port
with a laboratory power supply and a “pass” was concluded on the Ethernet
port. Later in the test procedure the laboratory power supply was replaced by a
client modular power supply, and the Ethernet connection was re-tested and
failed.
It was clear that the emissions were caused by CM switching frequencies from
the
DC power supply. In most equipment a CM coil is used at the DC input side of
the
DV power supply (for radiated emissions and immunity reasons).
These coils ore most often designed for the frequency range of radiated
emissions and
immunity and start filtering at 100 -500 kHz.
The CM SMPS frequencies pass through unattenuated ( or almost not). The
amplitude
of CM SMPS frequencies were up to 90 dBuV , and a significant part
of it will be present on the ground plane of EUT.
As most output transformers of Ethernet ports have their center taps at GND
via
a High Voltage 2nF cap any CM current will be coupled to the Ethernet cable
directly.
This made the test to fail.
Well the problems are:
- Testing the device without the actual power supply makes an
essential difference
in test results on testing a cable not expected to be impacted by that power
supply.
- The test setup of the conducted Ethernet emission measurement does
not
sufficiently describe the setup of “auxiliary” equipment
such as the modular power supply
but to connect it to an AMN (CISPR 22), especially the
requirement to test with
the supplied modular power supply.
- It was found that supplies with a grounded power supply cord have
more problems
then a Class II power supply with a 2-lead power supply cord (as expected, due
to the CM signal properties)
- Changing a power supply for an alternative (on defect), even if
approved for
IT use, will not guarantee continued compliance, as the DC properties
of power supplies are normally not tested. This is a frequently encountered
scenario, as most manufacturers will not even make the compliance link between
supplied power supply and supplied equipment , specifying only voltage
and current (and polarity) but not brand and type.
Even respected nr 1 brands do not make this relation in labeling or in the
manual
between the power supply and equipment sold with.
(unless a dedicated connector is applied).
Regards,
Ing. Gert Gremmen
ce-test, qualified testing bv
-
This message is from the IEEE Product Safety Engineering Society emc-pstc
discussion list. Website: http://www.ieee-pses.org/
To post a message to the list, send your e-mail to [email protected]
Instructions: http://listserv.ieee.org/request/user-guide.html
List rules: http://www.ieee-pses.org/listrules.html
For help, send mail to the list administrators:
Scott Douglas [email protected]
Mike Cantwell [email protected]
For policy questions, send mail to:
Jim Bacher: [email protected]
David Heald: [email protected]
All emc-pstc postings are archived and searchable on the web at:
http://www.ieeecommunities.org/emc-pstc
