Yes, I agree, and you even pointed out the clarification on “turn on/off”, as “- turn
PoE draw off – perhaps just having it say, ‘the PD shall turn
PoE draw on’ instead of the ‘PD shall turn on’ might be clearer.
Anyhow, just to be clear, which version of the proposal were you going to suggest?
The original with ‘should’? (which didn’t have half of the issues we discussed) or something else?
Principal, CME Consulting
Experts in Advanced
PHYsical Communications Technology
(PLEASE NOTE NEW EMAIL ADDRESS.
THE OTHER WILL STILL WORK, BUT PLEASE USE THIS FOR CME BUSINESS)
Dan Dove [mailto:dan.dove@xxxxxxxxxxxxxxxxxx]
Sent: Thursday, March 19, 2015 9:44 AM
Subject: Re: [802.3_4PPOE] PD Load turn on
I think you have it well stated.
On 3/17/15 5:47 PM, Koussalya Balasubramanian (kobalasu) wrote:
The conversation definitely seems to have taken a bigger scope. Thank you for all the discussions. I notice, there are 3 different points in the discussions
1) Clarifying that Voff applies even during power up and not only after power on has happened
- This was my original intention. This doesn't change the existing standard behavior at all – this only clarifies that Voff applies during power up as well (as few of us have heard
some customers arguing it doesn't). Hence it will not make any compliant PD non-compliant.
2) Dead band between Von and Vport_PD
3) Does existing IEEE802.3 Std 2012 text indicate PD shall never turn off?
- Original IEEE802.3Std 2012 language of "The PD shall turn on at a voltage less than or equal to Von"
--> does this mean PD shall never turn PoE draw off if it has local source?
The original solution from me suggestion was only for 1). The suggestion from Dan seems to cover 3) - but it becomes slightly more complicated for 2). Given there is a language of
"less than or equal to" may be the original language of less than or equal to Von suffices ?
Nevertheless, given we have a comment cycle coming up – face to face discussions might help here. I will go ahead and submit my original suggestion (only as a starter) which we can
play with more during comment resolution.
I think your point is good. A PD should not be mandated to accept power once the PI voltage exceeds Von, but rather it should be restricted from accepting power (beyond a signature/class current) until the PI voltage exceeds Von.
How about this...
"The PD shall
not turn on until a voltage greater than Voff and less than or equal to Von
or VPort_PD(min) whichever is greater, is present at the PI."
This would allow PDs to draw local power or draw from the LAN based on their own decision making process, but would prevent them from drawing power until the PI voltage meets the appropriate limits.
Nice hearing from you too. :)
On 3/17/15 10:55 AM, Michael McCormack wrote:
How does a PD ever turn off? In your efforts to be able to affix blame for devices that are non-compliant (trust me - idiots will still do idiotic things you will only now have two
things they fail to do properly) you have precluded PDs which allow and prioritize local power supplies.