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

Re: [802.3_4PPOE] Are diode bridges really needed (2).pptx



This is great topic and I’m glad we are having an ad hoc “ad hoc” on this. 


My take on this from an “interoperability standards” point of view is that market acceptance could be affected if things plug in and don’t work because of the cross-over patch cable situation (or any cable that crosses over).  This is ‘more invisible’ than for instance a ‘Y’ cable.  What I mean is the data would be fully functional if checked and there would be no splitter visible (for instance), yet the power path would not.


I would also be concerned that existing ‘blind’ injectors would have a much greater and very real ‘safety issue’.


To elaborate on a well know issue, today if a PD gets connected to a ‘blind power injector’ generally it will not be damaged, and will likely just work in many cases (regardless of the polarity of the non-IEEE injector).  As I showed in the last meeting, these blind injectors often have a wall supply adaptor, and a polarized power plug.  Specifically, wall adaptors are notorious for not being standardized for polarity at the DC plug end.  In the case where this adaptor is >> than LPS, without rectifier diodes or fuses, a PD TVS diode could get very hot.  The size of these TVS protection diodes is getting larger due to the increasing cable surge energies now being required by many customers.  I may try and do some experiments to see what would happen with a >>LPS power supply running into a TVS.  I suspect it will be more spectacular than a termination resistor but I plan to do both and let you all know. 


There is also added risk during debug in the lab to customers who often use bench supplies instead of PSEs when running and debugging their PDs.  Somehow I feel that there is more than a 50/50 chance it will be hooked up backwards… This is a secondary issue but still worth some consideration.


That being said, at this point, I am in favor of keeping rectifiers in bt PDs but I also understand the counter arguments and would love to see all of the points of view summarized and presented in Beijing.





Jeff Heath
Design Center Manager

Description: Linear Technology Corporation




402 East Carrillo Street, Suite D


Santa Barbara, California 93101








Thank you.


From: Lennart Yseboodt [mailto:lennartyseboodt@xxxxxxxxx]
Sent: Thursday, February 06, 2014 2:17 PM
To: STDS-802-3-4PPOE@xxxxxxxxxxxxxxxxx
Subject: Re: [802.3_4PPOE] Are diode bridges really needed (2).pptx


Hi George,

I believe Christian's proposal is not really aimed at 'getting rid of (all) the diodes'.
The proposal is to have a fixed polarity at the PSE side in all cases. Whereas a 'reversed' polarity on the data pairs would be allowed in 802.3at, it would no longer be allowed in bt.
That's it as far as the standard goes.

Because the polarity is now fixed, the PD vendor has the _option_ to choose not to implement a rectifier.
For most 'typical' PoE products today that would not make sense: the cost of a rectifier is very low, and supporting all cable configurations has real value.
I would expect that most devices would still implement rectifiers.

For high density PoE applications or applications at the high end of the power scale the situation is different: the rectifier is substantially more expensive and/or lossy and/or bulky.
Lighting is an obvious candidate: installation is done once, in bulk, so no chance of getting cross cables and the efficiency gain is about 1.4% on a typical full power PoE installation.
On large installed power, that is a very significant number.

Lennart Yseboodt
Research Scientist - Solid State Lighting
Philips Research

From: George Zimmerman [george@xxxxxxxxxxxxxxxxxxxxxxxxxxxxx]
Sent: Thursday, February 06, 2014 20:38
To: STDS-802-3-4PPOE@xxxxxxxxxxxxxxxxx
Subject: Re: [802.3_4PPOE] Are diode bridges really needed (2).pptx

Christina -

I would confirm what Keng-hua reports, that I know of many PHYs with the crossover correction, on both sets of pairs independently.  I agree that this feature in the PHYs works data-only, and will not effect powering.  However, because of this feature, we therefore do not see any trouble calls from mis-connected crossover cabling.  Therefore we do not seem to have a good way to see whether the two sets of pairs are crossed over independently in the field.


One reason that the feature is popular is that the equipment vendor no longer has to worry about the configuration of the wiring (with regards to crossovers).  Specifying a specific configuration for 4-pair POE applications seems to be a step backwards, and would lead us back to the days when the end-user would have to have multiple sets of patch cords handy when connecting a new link so that he could make sure he had the right crossover combination at the end point.  Remember that there are several connections in the link topology where a crossover may or may not occur (either by mistake or by design).




George Zimmerman

Principal, CME Consulting

Experts in Advanced PHYsical Communications Technology