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

Re: [8023-POEP] The IEEE 802.3at, "Transformer and Channel" ad hoc will have a conference call Friday May 2, 9AM PST: materials



Hi Fred and all,

 

Here some answers to your questions:

 

   o What open circuit inductance reduction has a significant impact on realizing PoE plus magnetics?

       Yair: Realizing PoE magnetic is not an issue IF there is no constrains (size, power dissipation etc.).

            You have to specify first if you can accept magnetic core with a bit higher

            size (in one dimension) or not.

            The answer will change If you must stay with the magnetic mechanical size as been used

            in 802.3af or if you allow increase in one dimension.

    It is possible to generate quantitative expression in which you can get an idea of how the magnetic core  volume is changes as function of Inductance with and without DC bias.

 

   o How significant to realizing PoE magnetics is adding a PD surge allowance of Ppd_ave x 400/350?

Yair: The surge is ~14% above Icable. If Icable is 600mA then the additional current is 14%*600mA=~86Ma

           If Iunb/Icable is K (currently K=3% but we know it is higher..) then Iunb=K*86mA and Ibias will be K*86/2=K*43mA.

           For 1% imbalance we will get 0.43mA increase in Ibias. If Iunb is 10% (not 3%) then it will increase Ibias by 4.3mA only..

           Not allowing Surge current will create a bigger problems. If we specify that Ipeak=Icable (No surge allowed) then the PD designer will be forced to design application that consume power much lower then 25.5W as permitted with 600mA (we had 29.5W with 720mA). The power will be lower by 10-15% which is typical (low cost) accuracies and tolerances of power circuits so actual power will be reduced to 20-21Watts which is not worth the efforts we have made in this project.

This fact affects the magnitude of applications that we can support market size etc. Compare to the magnetic issue this is a bigger issue.

We have spent on this question significant time in 802.3af and 802.3at and I believe that our salvation will not come by going this road…

            

   o Are you aware of PD customers that use the PD surge allowance?  If so provide details.

Yair: Yes I am aware and data was supplied By me and by Thong. The relevant Market is Security cameras in which requires between 1 to 3 surge current pulses for a duration of 3-15msec each.

Alternate solution of using big capacitors at PD input is not cost effective.

We already discuss it also.

 

   o Quantify how TX operating inductance changes with bias current.

Yair: With typical magnetic core for 802.3af devices the approximate dependence is as described in the attached file " Inductance vs Idc at 70degC"

LM

LM

LM

 

@ 0C

@ 20C

@ 70C

Idc [mA]

887.184

925.134

791.8612

0

900.9577

939.2934

798.8184

2

896.0665

934.8246

798.8922

4

873.063

912.2101

786.0833

6

834.4384

873.7973

756.0426

8

783.9898

823.2257

707.7034

10

725.9821

764.6484

643.9903

12

664.4104

702.0251

570.8641

14

602.5403

638.6603

495.1902

16

542.7428

577.0133

422.7294

18

486.5417

518.7177

357.1593

20

434.7694

464.7141

300.1214

22

387.7518

415.4209

251.787

24

345.4749

370.8966

211.4861

26

307.7169

330.9707

178.1839

28

274.1428

295.3422

150.7724

30

244.3687

263.6464

128.219

32

218.0013

235.4992

109.6273

34

194.6622

210.5237

94.24901

36

174.0001

188.3652

81.47304

38

155.6965

168.6986

70.80672

40

139.4669

151.2312

61.85529

42

125.06

135.7024

54.3031

44

112.2548

121.8819

47.89768

46

100.8582

109.5674

42.43665

48

90.70167

98.58136

37.75727

50

81.6381

88.76854

33.72811

52

73.53935

79.99312

30.24261

54

69.81628

75.95669

28.67594

55

66.29365

72.13625

27.21392

56

62.95968

68.51925

25.84814

57

59.80336

65.09392

24.57093

58

56.8144

61.84927

23.37537

59

53.98315

58.77496

22.25513

60

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


      Generic curve for a PoE and PoE plus TX inductance value vs current at a few temperature ranges.

      I will do this if transformer vendors submit curves for PoE and PoE plus transformers.

 

Yair : For 802.3at devices, I need to know the mechanical constrains as explained above.

   o What affect will varying TX and RX inductance have on data communication?

Yair: At frequencies above 1MHz: If you reduce the Inductance from 350uH to ~200uH, the effect will be negligible è0 (Tested data)

For lower inductance few things more need to be checked and verified.

At frequencies below 1MHz down to 100KHz. From analyzing the System Transfer function per the Midspan adhoc work it looks that we have a lot of margin. Details at the next Midspan ad hoc meeting.

   o Is there an AC and DC component to the affect bias current will have on system data integrity?

Yair: The question is not clear to me. From PoE point of view the ac component frequency on the DC current is much lower then the data frequency. As a result the data transformer response to the PoE current change immediately i.e. DC bias is as seen by the transformer follows PoE current at "zero" time.  

- Discuss how PHY baseline wander correction can be quantified in terms of reduce OCL requirements for 100BASE-TX.

Yair: I don’t know. We need PHY experts on this matter. The fact is that if we add to the channel additional inductance i.e. reducing the total equivalent inductance by 50%, it doesn’t affect the BER.

 

Hope it helps.

Yair

 

 

Darshan Yair
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,

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

 

E-mail: <mailto:ydarshan@microsemi.com>.  

 

 


From: owner-stds-802-3-poep@IEEE.ORG [mailto:owner-stds-802-3-poep@IEEE.ORG] On Behalf Of Fred Schindler (frs)
Sent: Friday, May 02, 2008 5:05 AM
To: STDS-802-3-POEP@LISTSERV.IEEE.ORG
Subject: [8023-POEP] The IEEE 802.3at, "Transformer and Channel" ad hoc will have a conference call Friday May 2, 9AM PST: materials

 

 

--------------à>>  Meeting materials are attached.  This is a 1 hour meeting that may go to 90 minutes.

Hello,

The IEEE 802.3at, "Transformer and Channel" ad hoc will have a conference call Friday May 2, 9AM PST. This meeting may run until 6 PM.

Proposed Agenda

- Cover Patent policy

- Q & A

   - Q & A

   o What open circuit inductance reduction has a significant impact on realizing PoE plus magnetics?

   o How significant to realizing PoE magnetics is adding a PD surge allowance of Ppd_ave x 400/350?

   o Are you aware of PD customers that use the PD surge allowance?  If so provide details.

   o Quantify how TX operating inductance changes with bias current.
      Generic curve for a PoE and PoE plus TX inductance value vs current at a few temperature ranges.
      I will do this if transformer vendors submit curves for PoE and PoE plus transformers.

   o What affect will varying TX and RX inductance have on data communication?

   o Is there an AC and DC component to the affect bias current will have on system data integrity?

- Discuss how PHY baseline wander correction can be quantified in terms of reduce OCL requirements for 100BASE-TX.

- Update on progress

- Agree on a system model for current unbalance calculations.

- Use IEEE 802.3 requirements and legacy system data to refine the model used.

- Use the refined approach to model IEEE 802.3at current unbalance.

Thanks,
Fred

PS Some of this discussion will be based on materials previously presented.

http://grouper.ieee.org/groups/802/3/at/public/mar08/schindler_2_0308.pdf

 

Please review the IEEE-SA PatCom slides http://standards.ieee.org/board/pat/pat-slideset.pdf prior to the start of our meeting.

 

______________________________________________________________________________

Frederick Schindler has invited you to a Cisco Unified MeetingPlace Conference

 

Date/Time:               MAY 2, 2008 at 9:00AM America/Los_Angeles

Length:                  90

Frequency:               1

Meeting ID:              131904160

Meeting Password:       

 

Global Access Numbers:

http://cisco.com/en/US/about/doing_business/conferencing/index.html

 

    San Jose, CA: +408.525.6800    RTP:              +919.392.3330     

    US/Canada:  +1.866.432.9903    United Kingdom:   +44.20.8824.0117

    India:      +91.80.4103.3979   Germany:          +49.619.6773.9002

    Japan:      +81.3.5763.9394    China:            +86.10.8515.5666

 

TO ATTEND A WEB AND VOICE CONFERENCE:

 

CISCO INTRANET ATTENDEES

Join the Web & Voice Conference*

1. Go to http://meetingplaceinternal.cisco.com/join.asp?131904160

2. Enter your CEC User ID & Password then click OK

- Accept any security warnings you receive and wait for the Meeting Room to initialize

3. Click on CONNECT from the Meeting Room to join the Voice Conference portion of the meeting

 

EXTERNAL ATTENDEES - Outside the Cisco Intranet**

Join the Web & Voice Conference*

1. Go to http://meetingplace.cisco.com/join.asp?131904160

2. Fill in the My Name is field then click Attend Meeting

- If you have a CEC User ID, click on the Cisco icon

- Accept any security warnings you receive and wait for the Meeting Room to initialize

3. Click on CONNECT from the Meeting Room to join the Voice Conference portion of the meeting

- Note: Guest users will see a link to the Global Access Numbers.

 

*If this is your first time attending a Web Conference, disable any pop-up blockers and visit http://meetingplace.cisco.com/mpweb/scripts/browsertestupper.asp to test your web browser for compatibility with the Web Conference.

**Not all meetings are scheduled to allow external attendees into the Web Conference portion of the meeting, if the URL does not work, please follow the Voice only Conference instructions below to attend.

 

TO ATTEND A VOICE ONLY CONFERENCE

1. Dial into Cisco Unified MeetingPlace (view the Access Numbers and link above)

2. Press 1 to attend the meeting

3. Follow the prompts to enter the Meeting ID 131904160 and join the meeting

 

SUPPORT

Information about this Conference: Contact Frederick Schindler, 14085259859

Cisco IT Support Center: Attend the Voice Conference and then press #0 on your phone keypad

 

Inductance vs Idc at 70degC.xls