Re: CISPR22 Edition 5 SLCE and Ethernet

[emc-p...@ieee.org]

John,
 
Many thanks for helping getting a reply form the source. I mostly appreciate
it.
 
Neven

-- Original message -- 
From: John Woodgate  

> In message 
> <062220060603.17444.449A32C1000E06B844242207020853CECE020A900A02@comc 
> ast.net>, dated Thu, 22 Jun 2006, neve...@comcast.net writes 
> >AFAIK, there is no spec on imbalance of (each diff pair of) a port, and 
> >I will check it with my local IEEE Ethernet spec specialists. I don't 
> >know if there is a spec on the cable, but I will try to find it. I'd 
> >really like to see if the requirement from the standard has any basis 
> >in the actual real-world measurements. 
> 
> Specially for you, I asked my standards committee colleague Martin 
> Wright, who is Chairman of CISPR/I, and his reply is: 
> 
> It is quite correct that CISPR 22 is trying to emulate the performa! nce 
> of cabled installations. The effect of installations on LCL remains an 
> open question. 
> 
> At the time that this part of CISPR 22 was written, many attempts were 
> made to obtain data on LCL of installations from telecomms and ITE 
> installers but the relative performance of installations from an LCL 
> perspective was (and still is) seen as 'sensitive' data and was never 
> released to the standards bodies. 
> 
> The LCL values in CISPR 22 were based on a series of measurements made 
> by Telstra in Australia along with private inputs from some of the 
> telcos represented in the working group (NTT, BT, FT and Telia). This 
> data was circulated to both ITU and ETSI with a question 'Does the LCL 
> shown in these curves form a reasonable representation of the installed 
> LCL of the cabling base in your country' and this did elicit some 
> responses. The curves were then re! -aligne d with this information. 
> 
> From the perspective of the UK, I know that there was a major analysis 
> of the Telecomms network at this time which showed that very few lines 
> were worse than the LCL specified. This result implies that the radio 
> spectrum will receive better protection (in most cases) than these 
> values would suggest. This has the corollary that the emissions from 
> the system could potentially be higher than the limits allow with no 
> adverse affects. 
> 
> Against this background we also need to consider the wireline proposals 
> where the radio users are pushing for limits at least 20dB below the 
> CISPR limits in order to better protect the radio spectrum. It would be 
> very difficult to write a valid definition of wireline which did not 
> include Ethernet, 10base T etc. 
> 
> These two facts (maybe) imply that the CISPR limit is an OK compromise ?! 
&g t; 
> The current/voltage measurement in CISPR 22 comes from a completely 
> different approach where the interest is in the power emitted, wherever 
> it comes from, which also has much support in CISPR I. This only 
> consider LCL in the indirect way that some of the power emitted will 
> come from the LCL conversion. 
> 
> As a final point the issue of LCL specification is certainly not closed 
> in CISPR I. There is a current active liaison with cable and 
> installation groups (TC46X, ISO/IEC JTC 1 WG 25 and so on) to better 
> understand the LCL and installation issues. 
> 
> I hope this helps to give the background to the requirements but please 
> note that I am condensing nearly 15 years of development into this short 
> email so may well have missed something significant. 
> 
> Regards 
> 
> Martin Wright, Chairman CISPR I 
> 
> Contact details fo! r Marti n are at:http://tinyurl.com/loy3x 
> -- 
> OOO - Own Opinions Only. Try www.jmwa.demon.co.uk and www.isce.org.uk 
> 2006 is YMMVI- Your mileage may vary immensely. 
> 
> John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK 
> 
> - 
>  
> 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 emc-p...@ieee.org 
> 
> 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 emcp...@ptcnh.net 
> Mike Cantwell mcantw...@ieee.org 
> 
> For policy questions, send mail to: 
> 
> Jim Bacher: j.bacher! @ieee.o rg 
> David Heald: emc-p...@daveheald.com 
> 
> All emc-pstc postings are archived and searchable on the web at: 
> 
> http://www.ieeecommunities.org/emc-pstc 

-  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 emc-p...@ieee.org 


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 administrator

Re: CISPR22 Edition 5 SLCE and Ethernet

[emc-p...@ieee.org]

In message 
<062220060603.17444.449A32C1000E06B844242207020853CECE020A900A02@comc
ast.net>, dated Thu, 22 Jun 2006, neve...@comcast.net writes
>AFAIK, there is no spec on imbalance of (each diff pair of) a port, and 
>I will check it with my local IEEE Ethernet spec specialists. I don't 
>know if there is a spec on the cable, but I will try to find it. I'd 
>really like to see if the requirement from the standard has any basis 
>in the actual real-world measurements.

Specially for you, I asked my standards committee colleague Martin 
Wright, who is Chairman of CISPR/I, and his reply is:

It is quite correct that CISPR 22 is trying to emulate the performance 
of cabled installations.  The effect of installations on LCL remains an 
open question.

At the time that this part of CISPR 22 was written, many attempts were 
made to obtain data on LCL of installations from telecomms and ITE 
installers but the relative performance of installations from an LCL 
perspective was (and still is) seen as 'sensitive' data and was never 
released to the standards bodies.

The LCL values in CISPR 22 were based on a series of measurements made 
by Telstra in Australia along with private inputs from some of the 
telcos represented in the working group (NTT, BT, FT and Telia).  This 
data was circulated to both ITU and ETSI with a question 'Does the LCL 
shown in these curves form a reasonable representation of the installed 
LCL of the cabling base in your country' and this did elicit some 
responses.  The curves were then re-aligned with this information.

 From the perspective of the UK, I know that there was a major analysis 
of the Telecomms network at this time which showed that very few lines 
were worse than the LCL specified.  This result  implies that the radio 
spectrum will receive better protection (in most cases) than these 
values would suggest.  This has the corollary that the emissions from 
the system could potentially be higher than the limits allow with no 
adverse affects.

Against this background we also need to consider the wireline proposals 
where the radio users are pushing for limits at least 20dB below the 
CISPR limits in order to better protect the radio spectrum.  It would be 
very difficult to write a valid definition of wireline which did not 
include Ethernet, 10base T etc.

These two facts (maybe) imply that the CISPR limit is an OK compromise ?

The current/voltage measurement in CISPR 22 comes from a completely 
different approach where the interest is in the power emitted, wherever 
it comes from, which also has much support in CISPR I. This only 
consider LCL in the indirect way that some of the power emitted will 
come from the LCL conversion.

As a final point the issue of LCL specification is certainly not closed 
in CISPR I.  There is a current active liaison with cable and 
installation groups (TC46X, ISO/IEC JTC 1 WG 25 and so on) to better 
understand the LCL and installation issues.

I hope this helps to give the background to the requirements but please 
note that I am condensing nearly 15 years of development into this short 
email so may well have missed something significant.

Regards

Martin Wright, Chairman CISPR I

Contact details for Martin are at:http://tinyurl.com/loy3x
-- 
OOO - Own Opinions Only. Try www.jmwa.demon.co.uk and www.isce.org.uk
2006 is YMMVI- Your mileage may vary immensely.

John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK

-

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 emc-p...@ieee.org

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   emcp...@ptcnh.net
 Mike Cantwell   mcantw...@ieee.org

For policy questions, send mail to:

 Jim Bacher: j.bac...@ieee.org
 David Heald:emc-p...@daveheald.com

All emc-pstc postings are archived and searchable on the web at:

http://www.ieeecommunities.org/emc-pstc

Re: CISPR22 Edition 5 SLCE and Ethernet

[emc-p...@ieee.org]

AFAIK, there is no spec on imbalance of (each diff pair of) a port, and I will
check it with my local IEEE Ethernet spec specialists. I don't know if there
is a spec on the cable, but I will try to find it. I'd really like to see if
the requirement from the standard has any basis in the actual real-world
measurements.
 
Neven
 

-- Original message -- 
From: John Woodgate  


> 
> When you do that, you see the combined effect of the cable impedances 
> and the impedances of the port at the far end. So the LCL can be 
> calculated from the specifications (if they exist) of those impedances 
> or equivalent data. Maybe that is what was done in order to get the LCL 
> figures specified in CISPR 22. 


-  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 emc-p...@ieee.org 


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 emcp...@ptcnh.net Mike Cantwell mcantw...@ieee.org 


For policy questions, send mail to: 


Jim Bacher: j.bac...@ieee.org David Heald: emc-p...@daveheald.com 


All emc-pstc postings are archived and searchable on the web at: 


http://www.ieeecommunities.org/emc-pstc 

Re: CISPR22 Edition 5 SLCE and Ethernet

[emc-p...@ieee.org]

In message 
<062220060528.732.449A2A99000B117902DC2200745672CECE020A900A02@comcas
t.net>, dated Thu, 22 Jun 2006, neve...@comcast.net writes

>However, the issue is not the imbalance specification looking into the 
>DUT connector (test port, diff pairs). The problem that I see here 
>is that CISPR22, 2005 is trying to emulate the imbalance of the 
>"Typical installation", and that I do not find any measurement data on 
>that, at least not for the CAT3-7 cabling. I do find the LCL (i.e. 
>balance) vs. frequency information in standards, but absolutely no 
>published data on LCL of "typical installations" to support them. So, I 
>don't know where these values in the standards are coming from.

In a 'typical installation', the cable is connected to a port at each 
end.

>Again, I am not talking here about imbalance at the DUT connector, but 
>imbalance that appears looking into the cabling from the DUT.

When you do that, you see the combined effect of the cable impedances 
and the impedances of the port at the far end. So the LCL can be 
calculated from the specifications (if they exist) of those impedances 
or equivalent data. Maybe that is what was done in order to get the LCL 
figures specified in CISPR 22.
> 
>The required level of imbalance for the ISN, combined with the max 
>possible (although not always used) amplitude of 10 BaseT (5.6 Vpk-pk) 
>makes me uncomfortable. On the other hand, if one uses the method with 
>current and voltage clamp, then there is no requirement for LCL, since 
>the measurement is done on the cable without any ISN.
> 
>In my opinion, manufacturers of Class B products that use Ethernet, 
>e.g. laptops, PCs, SOHO routers/switches) should be concerned by the 
>still (in my opinion) not well defined requirements.

-- 
OOO - Own Opinions Only. Try www.jmwa.demon.co.uk and www.isce.org.uk
2006 is YMMVI- Your mileage may vary immensely.

John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK

-

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 emc-p...@ieee.org

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   emcp...@ptcnh.net
 Mike Cantwell   mcantw...@ieee.org

For policy questions, send mail to:

 Jim Bacher: j.bac...@ieee.org
 David Heald:emc-p...@daveheald.com

All emc-pstc postings are archived and searchable on the web at:

http://www.ieeecommunities.org/emc-pstc

Re: CISPR22 Edition 5 SLCE and Ethernet

[emc-p...@ieee.org]

John, thanks for the explanation to David, you are right in what you wrote. I
was out the whole day so I am catching up now in the evening.
 
However, the issue is not the imbalance specification looking into the DUT
connector (test port, diff pairs). The problem that I see here is that
CISPR22, 2005 is trying to emulate the imbalance of the "Typical
installation", and that I do not find any measurement data on that, at least
not for the CAT3-7 cabling. I do find the LCL (i.e. balance) vs. frequency
information in standards, but absolutely no published data on LCL of "typical
installations" to support them. So, I don't know where these values in the
standards are coming from. Again, I am not talking here about imbalance at the
DUT connector, but imbalance that appears looking into the cabling from the
DUT. 
 
The required level of imbalance for the ISN, combined with the max possible
(although not always used) amplitude of 10 BaseT (5.6 Vpk-pk) makes me
uncomfortable. On the other hand, if one uses the method with current and
voltage clamp, then there is no requirement for LCL, since the measurement is
done on the cable without any ISN.
 
In my opinion, manufacturers of Class B products that use Ethernet, e.g.
laptops, PCs, SOHO routers/switches) should be concerned by the still (in my
opinion) not well defined requirements.
 
Neven
 
- Original message -- 
From: John Woodgate  


> I'm not sure what you mean by 'an impedance imbalance between the two 
> ports'. We were not discussing impedance mismatch but mode conversion 
> from differential to common (leading potentially to emissions from the 
> cable) or vice versa (leading to a lack of immunity). 
> 
> The functioning of balanced circuits in not radiating the signal and 
> rejecting common-mode disturbances depends on the impedances at each end 
> of the cable, from each conductor to a common reference point, being 
> closely equal. Any voltage between the common reference points is then 
> the input to a balanced bridge, and none of that voltage appears 
> between the signal conductors. And vice versa, of course. 
> 
> While transformers CAN be made with excellent symmetry and very closely 
> equ! al impe dances to the common point, they are not always so designed, 
> and it would be appropriate for the system standard to specify the 
> permitted unbalance, in one of the many ways that can be done. I think 
> this is what the original enquirer is asking about. 
> -- 
> OOO - Own Opinions Only. Try www.jmwa.demon.co.uk and www.isce.org.uk 
> 2006 is YMMVI- Your mileage may vary immensely. 
> 
> John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK 

-  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 emc-p...@ieee.org 


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 emcp...@ptcnh.net Mike Cantwell mcantw...@ieee.org 


For policy questions, send mail to: 


Jim Bacher: j.bac...@ieee.org David Heald: emc-p...@daveheald.com 


All emc-pstc postings are archived and searchable on the web at: 


http://www.ieeecommunities.org/emc-pstc 

Re: CISPR22 Edition 5 SLCE and Ethernet

[emc-p...@ieee.org]

In message 
<231db4b2f2634a4c94d87d89a4fa81d201542...@ny13ex011.global.ds.honeywell.c
om>, dated Wed, 21 Jun 2006, "Sterner, David [S&FS]" 
 writes

>If a pair of lines connects only two ports, each of the two ports is 
>100 ohms impedance, and the cable is uniform, please explain how the 
>Ethernet line can have an impedance imbalance between the two ports?

I'm not sure what you mean by 'an impedance imbalance between the two 
ports'. We were not discussing impedance mismatch but mode conversion 
>from differential to common (leading potentially to emissions from the 
cable) or vice versa (leading to a lack of immunity).

The functioning of balanced circuits in not radiating the signal and
rejecting common-mode disturbances depends on the impedances at each end 
of the cable, from each conductor to a common reference point, being 
closely equal. Any voltage between the common reference points is then 
the  input to a balanced bridge, and none of that voltage appears 
between the signal conductors. And vice versa, of course.

While transformers CAN be made with excellent symmetry and very closely 
equal impedances to the common point, they are not always so designed, 
and it would be appropriate for the system standard to specify the 
permitted unbalance, in one of the many ways that can be done. I think 
this is what the original enquirer is asking about.
-- 
OOO - Own Opinions Only. Try www.jmwa.demon.co.uk and www.isce.org.uk
2006 is YMMVI- Your mileage may vary immensely.

John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK

-

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 emc-p...@ieee.org

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   emcp...@ptcnh.net
 Mike Cantwell   mcantw...@ieee.org

For policy questions, send mail to:

 Jim Bacher: j.bac...@ieee.org
 David Heald:emc-p...@daveheald.com

All emc-pstc postings are archived and searchable on the web at:

http://www.ieeecommunities.org/emc-pstc

RE: CISPR22 Edition 5 SLCE and Ethernet

[emc-p...@ieee.org]

In a message 21-Jun-06 Mr. Woodgate stated "I don't understand why you
have sent this information. I've no doubt 
it's all true, but you haven't quoted any of the message you are
responding to, and you haven't commented on the permitted unbalance of
the port impedances, which is one subject under discussion.

If a pair of lines connects only two ports, each of the two ports is 100
ohms impedance, and the cable is uniform, please explain how the
Ethernet line can have an impedance imbalance between the two ports?
Given the requirements of IEE 802.3, a port imbalance is inconsistant
with transmission line theory.

My answer is for the Ethernet communications lines only (tx, rx);
termination imbalance is quite possible in the case of 'power over
Ethernet', where DC travels over the CAT-n wires not used for
communications.

.
Honeywell
David W. Sterner
Quality Assurance Engineering
Security & Custom Electronics
165 Eileen Way
P.O. Box 9035
Syosset, NY   11791
Phone: (516) 921-6704 x6970
Fax # (516) 364-6953
david.ster...@honeywell.com
 
david.ster...@ieee.org
  
Electromagnetic Compatibility
IEEE 05888136

-

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 emc-p...@ieee.org

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   emcp...@ptcnh.net
 Mike Cantwell   mcantw...@ieee.org

For policy questions, send mail to:

 Jim Bacher: j.bac...@ieee.org
 David Heald:emc-p...@daveheald.com

All emc-pstc postings are archived and searchable on the web at:

http://www.ieeecommunities.org/emc-pstc

Re: CISPR22 Edition 5 SLCE and Ethernet

[emc-p...@ieee.org]

In message 
<231db4b2f2634a4c94d87d89a4fa81d201542...@ny13ex011.global.ds.honeywell.c
om>, dated Wed, 21 Jun 2006, "Sterner, David [S&FS]" 
 writes
>Per ANSI/IEEE 8032.3x Ethernet Cat 5, 6, and 7 cables have only two 
>ports, additional ports are forbidden.  Both ports are transformer 
>isolated for a 100-ohm impedance termination.  Communication to other 
>network locations is always through an active device (e.g. hub, switch, 
>router).

I don't understand why you have sent this information. I've no doubt 
it's all true, but you haven't quoted any of the message you are 
responding to, and you haven't commented on the permitted unbalance of 
the port impedances, which is one subject under discussion.
-- 
OOO - Own Opinions Only. Try www.jmwa.demon.co.uk and www.isce.org.uk
2006 is YMMVI- Your mileage may vary immensely.

John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK

-

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 emc-p...@ieee.org

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   emcp...@ptcnh.net
 Mike Cantwell   mcantw...@ieee.org

For policy questions, send mail to:

 Jim Bacher: j.bac...@ieee.org
 David Heald:emc-p...@daveheald.com

All emc-pstc postings are archived and searchable on the web at:

http://www.ieeecommunities.org/emc-pstc

RE: CISPR22 Edition 5 SLCE and Ethernet

[emc-p...@ieee.org]

Per ANSI/IEEE 8032.3x Ethernet Cat 5, 6, and 7 cables have only two ports,
additional ports are forbidden.  Both ports are transformer isolated for a
100-ohm impedance termination.  Communication to other network locations is
always through an active device (e.g. hub, switch, router).  

The only variation is full-duplex communication where 4 lines are active and
connexion is to a switch (hubs do not support full duplex).  In full-duplex
both ports have two transformer terminations, one for transmit and the other
for receive.

Ethernet adapter cards (NICS) support full duplex, but negotiate to half
duplex if connected to a hub.

.
Honeywell
David W. Sterner
Quality Assurance Engineering
Security & Custom Electronics
165 Eileen Way
P.O. Box 9035
Syosset, NY   11791
Phone: (516) 921-6704 x6970
Fax # (516) 364-6953
david.ster...@honeywell.com
 
david.ster...@ieee.org
  
Electromagnetic Compatibility
IEEE 05888136

-

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 emc-p...@ieee.org

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   emcp...@ptcnh.net
 Mike Cantwell   mcantw...@ieee.org

For policy questions, send mail to:

 Jim Bacher: j.bac...@ieee.org
 David Heald:emc-p...@daveheald.com

All emc-pstc postings are archived and searchable on the web at:

http://www.ieeecommunities.org/emc-pstc

Re: CISPR22 Edition 5 SLCE and Ethernet

[emc-p...@ieee.org]

In message 
<062120060433.25342.4498CC271E7162FE2207020853CECE020A900A02@comc
ast.net>, dated Wed, 21 Jun 2006, neve...@comcast.net writes
>I am looking for the actual data supporting the chosen LCL 
>characteristics for the CAT5, 6 and 7 ISN per the standard. 

Yes, and I think that data (or a specification for the balance of the 
cables) may be in the standards I mentioned. Of course, you are also 
concerned with the impedance balance specifications at the cable ports, 
and I suppose those are specified in the Ethernet standards, which are 
not, of course, IEC standards. I believe they are very large, but I 
don't know where you get them from.
-- 
OOO - Own Opinions Only. Try www.jmwa.demon.co.uk and www.isce.org.uk
2006 is YMMVI- Your mileage may vary immensely.

John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK

-

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 emc-p...@ieee.org

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   emcp...@ptcnh.net
 Mike Cantwell   mcantw...@ieee.org

For policy questions, send mail to:

 Jim Bacher: j.bac...@ieee.org
 David Heald:emc-p...@daveheald.com

All emc-pstc postings are archived and searchable on the web at:

http://www.ieeecommunities.org/emc-pstc

Re: CISPR22 Edition 5 SLCE and Ethernet

[emc-p...@ieee.org]

John,
 
I am looking for the actual data supporting the chosen LCL characteristics for
the CAT5, 6 and 7 ISN per the standard. According to the described intent of
the standard, the ISN characteristics are such to purposely introduce
imbalance, supposedly to replicate the LCL characteristics of a "typical"
network installation, but I do not find any published data on the LCL of the
"typical" CAT5, 6, and 7 installations.
 
The method of the measuremtn is described in ITU-T G.117 standard. I hope that
there is a considerable work somewhere available that supports the chosen
CAT5, 6, and 7 LCL characteristics, but I also fear that the information is
not available, and I believe it should certainly be available for a standard
with such a broad consequence for the industry.
 
My back-of-the-envelope calculation tells me that 10 BaseT Ethernet, with
maximum 5.6Vpk-pk amplitude, as defined per IEEE 802.3 standard is most likely
to fail ClassB limit with the CAT5 LCL as defined by CISPR22, 2005. Most
manufacturers do not generate as high voltage, but it is possible per the IEEE
Ethernet standard. Laptop/PC and other SOHO Ethernet-product manufacturers are
going to have hell of a time trying to make it through.
 
Regards, Neven

-- Original message -- 
From: John Woodgate  

> In message 
> <062020062152.8474.44986E288DE2211A2205886442CECE020A900A02@comca 
> st.net>, dated Tue, 20 Jun 2006, neve...@comcast.net writes 
> 
> >Does anyone have or know the sources (preferably published, verifiable, 
> >with the exact test method described) of the LCL (Longitudinal 
> >Conversion Loss, i.e. balance) of "typical installation" that support 
> >the LCL values required by CSPR22 for ISN for CAT5, 6, and 7 cables? 
> >(CISPR22 5th edition, 2005, page 51) 
> 
> Are you looking for a method of testing the actual cables for unbalance? 
> 
> I think you need to look at IEC standards in the IEC 61156 series, at: 
> 
> http://tinyurl.com/zarag 
> > If you click on the standard number, you get a detailed page about it, 
> and if you click on the 'Preview' icon you get the Table of contents, 
> the Scope clause and a few more pages of the standard to view for free. 
> -- 
> OOO - Own Opinions Only. Try www.jmwa.demon.co.uk and www.isce.org.uk 
> 2006 is YMMVI- Your mileage may vary immensely. 
> 
> John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK 


-  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 emc-p...@ieee.org 


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 emcp...@ptcnh.net Mike Cantwell mcantw...@ieee.org 


For policy questions, send mail to: 


Jim Bacher: j.bac...@ieee.org David Heald: emc-p...@daveheald.com 


All emc-pstc postings are archived and searchable on the web at: 


http://www.ieeecommunities.org/emc-pstc 

Re: CISPR22 Edition 5 SLCE and Ethernet

[emc-p...@ieee.org]

In message 
<062020062152.8474.44986E288DE2211A2205886442CECE020A900A02@comca
st.net>, dated Tue, 20 Jun 2006, neve...@comcast.net writes

>Does anyone have or know the sources (preferably published, verifiable, 
>with the exact test method described) of the LCL (Longitudinal 
>Conversion Loss, i.e. balance) of "typical installation" that support 
>the LCL values required by CSPR22 for ISN for CAT5, 6, and 7 cables? 
>(CISPR22 5th edition, 2005, page 51)

Are you looking for a method of testing the actual cables for unbalance?

I think you need to look at IEC standards in the IEC 61156 series, at:

http://tinyurl.com/zarag

If you click on the standard number, you get a detailed page about it, 
and if you click on the 'Preview' icon you get the Table of contents, 
the Scope clause and a few more pages of the standard to view for free.
-- 
OOO - Own Opinions Only. Try www.jmwa.demon.co.uk and www.isce.org.uk
2006 is YMMVI- Your mileage may vary immensely.

John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK

-

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 emc-p...@ieee.org

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   emcp...@ptcnh.net
 Mike Cantwell   mcantw...@ieee.org

For policy questions, send mail to:

 Jim Bacher: j.bac...@ieee.org
 David Heald:emc-p...@daveheald.com

All emc-pstc postings are archived and searchable on the web at:

http://www.ieeecommunities.org/emc-pstc

Re: CISPR22 Edition 5 SLCE and Ethernet

[emc-p...@ieee.org]

I didn't get any reply to my original question, so I'll try a little smaller
bite.
 
Does anyone have or know the sources (preferably published, verifiable, with
the exact test method described) of the LCL (Longitudinal Conversion Loss,
i.e. balance) of "typical installation" that support the LCL values required
by CSPR22 for ISN for CAT5, 6, and 7 cables? (CISPR22 5th edition, 2005, page
51)
 
I searched the IEEE publication archives, and couldn't find any reference. I
also Googled it with no success. The closest I get is articles that reference
the values from the standard (various revisions), but no actual data
supporting the chosen LCL characteristics of CAT5,6,7, ISNs.
 
Regards, Neven
-  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 emc-p...@ieee.org 


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 emcp...@ptcnh.net Mike Cantwell mcantw...@ieee.org 


For policy questions, send mail to: 


Jim Bacher: j.bac...@ieee.org David Heald: emc-p...@daveheald.com 


All emc-pstc postings are archived and searchable on the web at: 


http://www.ieeecommunities.org/emc-pstc