[802.3af] FW: Isolation requirements for Power Over MDI - Need clarifications?
All,
I forward to you , questions regarding possible conflicts in isolation
requirements as required by the 802.3 standard.
I have already received comments from Rick and Don.
I believe that this issue is very important due to the fact that if it is
not clear it can cause system manufacturers (Switches, Hubs etc) problems in
the future.
I would like to share with you the question and the comments and if you
familiar with the guys that where involved in setting the standard please
forward this email to them.
Attached below the questions.
Yair.
************
802.3 standard specifies the isolation requirements between the MDI
leads
and the electronic hardware in the Switch/Hub.
1. In ENV A, according to paragraph 9.7.1, a 500Vrms isolation is
required
between MDI leads to Repeater protective ground.
2. However, in 23.5.1.1 there is a requirement of 1500Vrms isolation
between
Repeater circuits, Repeater frame ground and all MDI leads.
Question 1: Does (1) contradict (2) as for the isolation between MDI
leads
to frame ground.
Question 2: Now, in some Switches, the manufacture connect the PHY
common to
the chassis (protective ground?). Isn't it violates (2) ?
If we add to the equation that we connect a power supply output to
the MDI
leads, we need that the power supply output will be isolated from
the
chassis (protective ground) by 500Vrms or by 1500Vrms?
I fill that the issue is open to different interpretation and I
would like
to have your comments as for my understanding the current standard
or if we
need to explicitly specify the requirements.
Thanks
Yair.
************
> -----Original Message-----
> From: Stewart, Donald S (Don) [SMTP:dsstewart@xxxxxxxxx]
> Sent: г, аевеси 08, 2001 9:42 PM
> To: Rick Brooks; Yair Darshan; Avinoam Levy; Jetzt, John J (John); Peter
> Schwartz; R karam; scott_burton@xxxxxxxxx; Steve Carlson;
> brian_lynch@xxxxxx; Dave Dwelley
> Cc: Stewart, Donald S (Don)
> Subject: RE: Isolation requirements for Power Over MDI - Need
> clarificatio ns?
>
> Yair, Others,
> I (and John Jetzt) looked into the 802.3 spec over a year ago trying to
> make sense of this. It seems inconsistent. I was not involved in creating
> any of these clauses so what I provide here are the interpretations of a
> "reader of the spec".
>
> Environments A and B are specified within each of
> - 9.7 with Section 9 being "Repeater unit for 10 Mb/s baseband networks".
> This uses COAX in illustations but lists 10Base -T (twisted pair) in
> certain tables, and
> - 27.5.3 with Section 27 being "Repeater unit for 100 Mb/s baseband
> networks". This does not indicate specific media but refers to differnet
> PHY clauses including 100BASE-T4, so it applies to twisted pair cable.
>
> My assumption is that both simple repeaters and no-so-simple switches
> (that regenerate pulses from input port to output port) msut meet the
> "repeater" specifications.
>
> Both of these "A" and "B" isolation requirements are consistent. "A" calls
> for 500 vrms isolation from segment to protective ground of the repeater
> unit, and "B" calls for 1500 vrms between each segment and pretective
> ground and between each segment and all other segments.
>
> But the Physical Media Attachment spec differs from this. Clause 23.5
> (with section 23 being "Physical Coding Sublayer (PCS), Physical Medium
> Attachement (PMA) sublayer and baseband medium, type 100BASE-T4") calls
> for 1500v rms (or 2250 vDC or 10 2400v pulses ..) isolation between the
> DTE, or repeater circuits including frame ground, and all MDI leads. I
> interperet this to say "shall withstand 1500 vrms from any/all leads to
> frame ground".
>
> Clauses 24 and 25 address 100BASE-TX (Category 5 UTP) but do not provide
> electrical specifications directly. Clause 25 is structured as exceptions
> to a FDDI TP-PMD standard, ANSI X3.263. So I do not know what electical
> specs may be there or if they are important.
>
> My interpretation is that the "PHY" specification, i.e. 23.5's "1500 vms
> or 2500vDC or 2400v pulsed" supercedes or overshadows the Environment A
> 500 vrms level. This basically agrees with Rick's view, below.
>
> Just "one reader's" interpretation ....
> Don
>
>
> -----Original Message-----
> From: Rick Brooks [mailto:ribrooks@xxxxxxxxxxxxxxxxxx]
> Sent: Wednesday, August 08, 2001 11:49 AM
> To: Yair Darshan; 'Avinoam Levy'; Stewart, Donald S (Don); Jetzt, John J
> (John); 'Peter Schwartz'; 'R karam'; 'scott_burton@xxxxxxxxx'; 'Steve
> Carlson'; 'brian_lynch@xxxxxx'; 'Dave Dwelley'
> Subject: RE: Isolation requirements for Power Over MDI - Need clarificatio
> ns?
>
>
>
> Hi Yair and all.
> Yes, it certainly sounds a bit confusing.
>
> Section 9.7 seems to refer to simple repeater circuits from the days of
> coax (thinnet, thicknet).
> UTP cable was not yet a part of the spec. This section talks about MAU's,
> which attach to the coax.
> In this case the 500VAC applies.
>
> However, we are only doing power on the MDI (RJ-45), so section 23.5
> applies.
> So for 802.3af, 1500VAC is the isolation requirement from the MDI leads to
> frame ground.
>
> It should not matter whether or not the PHY itself is connected to chassis
> ground,
> that's why we have the transformers.
> Someday, there might be robust PHYs that sit on the wire without
> transformers,
> and we talk to them via some kind of HV capacitive coupling into the
> package or something, ...
> There is nothing in the spec to preclude that, but I'm not saying that it
> will happen.
> The bottom line is that the MDI leads must be isolated from frame ground.
>
> just one opinion...
> - Rick
>
>
>
>
>
>
>
> -----Original Message-----
> From: Yair Darshan [SMTP:YairD@xxxxxxxxxxxxxx]
> Sent: Wednesday, August 08, 2001 6:48 AM
> To: 'Avinoam Levy'; 'Donald S. Stewart'; 'Jetzt, John J'; 'Peter
> Schwartz'; 'R karam'; 'scott_burton@xxxxxxxxx'; 'Steve Carlson'; Brooks,
> Rick [SC5:321:EXCH]; 'brian_lynch@xxxxxx'; 'Dave Dwelley'
>
> Subject: Isolation requirements for Power Over MDI - Need
> clarifications?
>
> Hi guys,
>
> 802.3 standard specifies the isolation requirements between the MDI leads
> and the electronic hardware in the Switch/Hub.
>
> 1. In ENV A, according to paragraph 9.7.1, a 500Vrms isolation is required
>
> between MDI leads to Repeater protective ground.
> 2. However, in 23.5.1.1 there is a requirement of 1500Vrms isolation
> between
> Repeater circuits, Repeater frame ground and all MDI leads.
>
> Question 1: Does (1) contradict (2) as for the isolation between MDI leads
>
> to frame ground.
>
> Question 2: Now, in some Switches, the manufacture connect the PHY common
> to
> the chassis (protective ground?). Isn't it violates (2) ?
>
> If we add to the equation that we connect a power supply output to the MDI
>
> leads, we need that the power supply output will be isolated from the
> chassis (protective ground) by 500Vrms or by 1500Vrms?
>
> I fill that the issue is open to different interpretation and I would like
>
> to have your comments as for my understanding the current standard or if
> we
> need to explicitly specify the requirements.
>
> Thanks
>
> Yair.
>
>
>
> 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, Ext. 210. Cell: +972-54-893019
> Fax: +972-9-775-5150
> > E-mail: < <mailto:yaird@xxxxxxxxxxxxxx>>.
> > <http://www.powerdsine.com>
> >
> >
> >
>