Re: [802.3_4PPOE] IEEE802.3bt - End to End Channel Pair to Pair Resistan

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

Hi Yair,
I've inserted some comments below, and have attached a revised version with the Ad Hoc acronym updates.
Best Regards,
Ken

On 6/23/2014 6:06 PM, Darshan, Yair wrote:

Hi Ken,

Thanks, Please see my comments. I'll appreciate if you can address it prior the meeting so we can save time.

Slide 2: Please use the ad-hoc Acronyms as agreed on Annex J1 page 60 at the ad-hoc material. I am referring to bullet 2 when you say "P2PRunb". Do you mean End to End P2PRunb ?

Ken: Yes. (Addressed in the revised slides.)

The End to End P2PRUnb is system requirement that will generate eventually the current unbalance in addition to voltage unbalance PSE and PD. This parameter will not be in the spec. It is a guide line or "system" spec to us.

Ken: I understand that End-to-End P2PRunb is not intended to be a specification in the standard

The parameters that will be specified are:

PSE PI Resistance and voltage unbalance.

PD PI Resistance and voltage unbalance.

Channel P2PRUNB.

Ken: The point of the presentation is that PSE PI P2PRunb is not the best method. The derived equation accurately describes PI contribution to End-to-End P2PRunb, whereas PI P2PRunb does not.

Slide 3: No comments. It is correct. Furthermore I am going to propose to specify minimum resistance to PSE PI and PD PI per my presentation conclusions sent to the reflector and to the team regarding the channel use case analysis results. This is I believe supported and resolve the last two bullets in this slide.

Ken: Adding a minimum resistance does reduce worst case End-to-End P2PRunb, however the inequality still exists.

Slide 4: I like the direction. It supports slide 3. It is also similar to the analytical way that I have shown in the last adhoc meeting (on slide 14 of our next meeting material) which is:

I understand that the Kpd part and Kch part was merged to Y term. Is this correct?

Ken: I used the resistances in the End-to-End P2PRunb equation and solved for Rpse_max. (Kxx factors were not considered but may be accounted for in the answer, depending upon how they are defined.)

Slide 5. Looks good. Please show equation derivation to justify the method. It will helps also to those who interested to use it and run simulations to test the concept.

Ken: Will address in the meeting.

Slide 6:

3^rd bullet not clear. (Don’t otherwise provide a balancing technique)

Ken: If there is a balancing mechanism, such as active balancing, where simple Resistance and Voltage measurements don't describe the circuit behavior, or even fail based upon such behavior, then a resistance and Voltage requirement and test is not appropriate.

5^th bullet: Why "4 pair powering shall be sourced from a single DC supply" is Important if Voltage difference is specified?

Ken: Any source has an internal resistance; if more than one port exists on the PSE, then sourcing to other ports will cause a Voltage difference that is not present in a compliance test. I suppose a single-port PSE designed with two power supplies could be an exception.

Bullet 6: It is implementation issue. We cannot require it. You need to translate your intention to voltage, resistance, current to allow/disallow what you believe is best and we can discuss it.

Ken: It would decrease non-linearity with respect to PI unbalance. If not a requirement, it could be a recommendation, along with recommending matched diodes in a PD.

Bullet 7: Yes agree. I am proposing the same for PSE and PD i.e. to specify Rmin.

Annex – derivation of equations: Please supply next meeting calculated examples so we can verify in simulation. Please use only adhoc numbers so we can use the same data base and Christiam and me can both check it.

Ken: I don't have time to do this prior to the meeting... I'm also not clear on which numbers to use.

Example at the last slide: The point is clear. Please use the adhoc data base from the same reasons above.

Ken: This slide is only an example demonstrating the fact that PI Runbalance is not equal to PI Runbalance contribution to End-to-End P2PRunb. The derived equations cover this ok, so there's no need to discuss it.

Many thanks

Yair

From: Ken [mailto:ken_bennett@xxxxxxxxx]
Sent: Monday, June 23, 2014 9:21 PM
To: Darshan, Yair; STDS-802-3-4PPOE@xxxxxxxxxxxxxxxxx
Subject: Re: [802.3_4PPOE] IEEE802.3bt - End to End Channel Pair to Pair Resistance Unbalance Adhoc - Meeting #8 material

Hi Yair, All,
I've attached a new revision of the slides for the meeting.
Best Regards,
Ken

On 6/22/2014 8:03 PM, Darshan, Yair wrote:

Hi all,

Please find meeting material for Tuesday June 24, 2014. The focus is planned to be on slides 6-10, 13,14, 16-30.

I'll appreciate if you review the slides per the agenda prior the meeting.

Ken: You are scheduled to present your material from last time. Please resend it to the reflector/team.

Regards,

Yair

Darshan Yair

Chair

Power over HDBaseT Subcommittee

HDBaseT Alliance

Chief R&D Engineer

Analog Mixed Signal Group

Microsemi Corporation

1 Hanagar St., P.O. Box 7220
Neve Ne'eman Industrial Zone
Hod Hasharon 45421, Israel
Tel: +972-9-775-5100, EXT 210.

Cell: +972-54-4893019
Fax: +972-9-775-5111

E-mail: <mailto:ydarshan@xxxxxxxxxxxxx>.