Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

RE: [802.3af] Late comment- ungrounded? leakage current

You must evaluate the leakage current requirements for any product that does
not have a path to "ground". If the PSE is an isolated switching converter,
there will be capacitive coupling, via the transformer, to any exposed metal
that the end user can touch. Granted that from a low voltage source such as
the POE provides (15W), the leakage current may be small, it may still be a
safety concern. This is even more important in hospitals, etc, with patient

There are definite safety (UL, CSA, etc) limits for leakage currents
depending on the equipment safety classification. 

My 2 peso's.

Ed Walker   
Analog Product Specialist 

Texas Instruments Incorporated 
HC66 Box 203 
Mountainair, NM 87036 

INTERNET: ed_walker@xxxxxx 
Office = 505-847-0576
Fax = 413-280-0812  

-----Original Message-----
From: Yair Darshan [mailto:YairD@xxxxxxxxxxxxxx]
Sent: Monday, January 13, 2003 3:15 AM
To: 'Geoff Thompson'; stds-802-3-pwrviamdi@xxxxxxxx
Cc: thompson@xxxxxxxx
Subject: RE: [802.3af] Late comment

Geoff and all,

The reason that there is no specification for tying one of the PSE outputs
to the ground is the fact that in the early days of the standard (802.3af)we
got the conclusion that it is best to leave the outputs floating!.

It was once offered by Steve Jackson/Jennifer Rasimas/Nortel on July 10,

to tie one of the leads to the ground however it was voted down from the
following reasons:

1.      In environment A, when single power supply is used to feed multiple
ports, the effective way to avoid
        current ground loops is by keeping the PSE port floating or DC
isolated from system/chassis ground along with the PDs.

2.      If you tie one of the PSE outputs to ground, it means that you tie
the MDI leads to ground.
        This connection violates IEEE802.3 specific requirement which
specify 1500Vac or 2250Vdc between all MDI leads to frame ground/chassis and
repeater circuits.

3.      If you tie one of the PSE outputs to ground, it means that the port
is no longer ac balanced and it may affect the RF parameters.

4.      In any case if eventually we will find good reasons why we should
tie one of the PSE outputs to the ground than it must be the positive
        lead as it is being done for 100 years or so in the telecom field
however in the telecom infrastructure we didn't had the RF issues and
      the IEEE802.3 requirements to consider..

From the above reasons I do not agree to change the standard as offered by

If any one think differently please lets discuss it over the reflector and
not at the meeting in order to save the time.



Darshan Yair
Chief  Engineer
PowerDsine Ltd.  -  Powering Converged Networks
1 Hanagar St., P.O. Box 7220
Neve Ne'eman Industrial Zone
Hod Hasharon 45421, Israel
Tel:  +972-9-775-5100, Cell: +972-54-893019
Fax: +972-9-775-5111

E-mail: < <mailto:yaird@xxxxxxxxxxxxxx>>. 
< <>>

Notice: The information in this email and in any attachments is confidential
and intended solely for the attention and use of the named addressee(s). It
must not be disclosed to any person without authorization. If you are not
the intended recipient, or a person responsible for delivering it to the
intended recipient, you are not authorized to and must not disclose, copy,
distribute, or retain this message or any part of it.

-----Original Message-----
From: Geoff Thompson [mailto:gthompso@xxxxxxxxxxxxxxxxxx]
Sent: Monday, January 13, 2003 6:48 AM
To: stds-802-3-pwrviamdi@xxxxxxxx
Cc: thompson@xxxxxxxx
Subject: [802.3af] Late comment


During the meeting in Vancouver I asked the group's indulgence with respect
to acceptance of yet another comment from me.

There seemed to be general agreement that the comment addressed an issue
that was important enough that it should be accepted.

The comment:
I can find no place where there is a specification as to which side of PSI
is tied to ground (i.e. negative ground vs. the positive ground usually
found in mid-fifties British sports cars with Lucas electrics.) I request
that a specification be added that dictates which side of the supply is

This documents my request to add this comment to the database and address
it during the Santa Clara/January meeting.



| Geoffrey O. Thompson                    |
| Vice Chair,  IEEE 802                   |
| Nortel Networks, Inc.  M/S: P79/06/B04  |
| 4655 Great America Parkway              |
| P. O. Box 58185                         |
| Santa Clara, CA 95052-8185  USA         |
| Phone: +1 408 495 1339                  |
| Fax:   +1 408 495 5615                  |
| E-Mail: thompson@xxxxxxxx               |
| Please see the IEEE 802 web page at     |