My power meter project has 7 serial ports to evaluate during EMC testing to
EN61000-6-2.
I am in the process of creating a Test Mode for the ports in software to help
in the evaluation and to reduce auxilary devices.
I was looking for some reflections or advice on how you all attempt this.
Key to me is in keeping it simple and to minimize the auxilary equipment and
the number of connections that must leave a test chamber.
My Com ports are on two pug-in cards in the same base rack like housing.
Card one
* 1 10/100 base T Ethernet (RJ45/ Cat5 UTP/STP)
* 1 Proprietary twisted pair multipoint master/slave interface (modem
like)
* 2 RS485 ports - Modbus (Master/Slave Multipoint)
* I also have a separate display module that talks off one of these ports
Card two
* 1 10/100 base T OR via software control 100 Base Fx Ethernet interface
(share same Net Phy, only one interface works at a time)
* 1 RS485 ports - Modbus (Master/Slave Multipoint)
* 1 RS232 port - Modbus (Master/Slave Multipoint)
* 1 internal modem to Plain Old Telephone System
*
Note that the internal modem and RS232 port share the same UART under
mux
control (only one works at a time)
Test approaches:
The RS485 ports are schematicaly identical but of course have slight
variations in layout and one is on another plug-in card.
EN61000-6-2 has some provisions for testing one representative circuit out of
a group - do you see how I could reduce my testing with this clause?
Perhaps eliminate the second RS485 port on Card one?
I have two ports which have a time multiplexed data interface which I could
flip back and forth.
Do you see this as practical or would I have to run the tests an additional
sweep with each port engaged?
This is not bad for EN61000-4-4 but a real pain for EN61000-4-3.
Self Test
On the RS232 Port I could connect Tx to Rx over a ~3M cable, grounding the
shield at the far end to represent the CM impedances.
I could circulate bytes in and out and latch on errors with XOR and parity
framing check.
This would eliminate any auxilary equipment and keep the foot print down for
prescreening in our GTEM.
I could also arrange for the Display to talk to all 3 RS485 ports with one
cable - similar idea.
Also the Ehternet ports could be talking to each other...
Test with 2 meters
I could interface 2 meters to talk to each other - one as master one as slave
and then test both.
This would give good real world connections. I could put ~3M cable between
each.
This would be a large set up to test with - during emission testing I would
have two units radiating......
Leaving test chamber.
I would like to use my FO Ethernet as the means of leaving the test chamber
but then would be ignoring the multiplexed 10/100 base T port on the same Net
Phy.
This leaves the other Ethernet port needing som Aux equipment to talk to...
not good.
I could also interconnect the two 10/100 base T ports together and then leave
the chamber to monitoring PC with any other unpaired port.......
Well you can see that I have my hands full. Difficult to get a tight set up
for quick dwell times.
I copied work - please respond to this address as well with "respond to all"
if possible.
I get this list at home and would like to see your resonse at work too.
[email protected]
Thanks in advanced
Chris Wells
Eaton Electrical
---------------------------------------------------------------- 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/listserv/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:
Richard Nute: [email protected] Jim Bacher: [email protected]
All emc-pstc postings are archived and searchable on the web at:
http://www.ieeecommunities.org/emc-pstc
