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

RE: [802.3af] 45Kmin instead of 70Kmin ?

John, Roger, Dave
1.   The 70K was a number that was selected to ensure that if two PSE's are
connected together and one of them or both are
      powered from some reason, we will not have damage due to the high
     45K is still high impedance so I am not worried about it.
2.  The alternative of using the low impedance with a diode is OK however I
am referring to the high impedance 
     circuit that doesn't requires a diode and the set of numbers there are
not consistent to other definitions as I have
     described below.
3.  I totally agree with John that we should check if we do not get
detection when PSE is connected to PSE. 
     I'll wait until John will check it. I have looked at some of Rick's
papers and I have found that he use the 75K number 
    as a given number and not as a result which is a must to guarantee
something, I'll check it again.


> -----Original Message-----
> From:	R karam [SMTP:rkaram@xxxxxxxxx]
> Sent:	ג ינואר 08 2002 20:26
> To:	Dave Dwelley; Yair Darshan; 'Jetzt, John J'
> Cc:	'stds-802-3-pwrviamdi@xxxxxxxx'
> Subject:	Re: [802.3af] 45Kmin instead of 70Kmin ?
> Hi
> I am worried that this is a 'major' change we have 'messed' with this so
> many
> times that at this stage of the game, this would require revisiting the
> details
> all over...
> so unless we have some good data on this, i am against it, the Lucent
> hazard
> matrix that sold us this in the first place relied on the 70k number....
> my 2c.
> roger
> At 09:44 AM 1/8/02 -0800, Dave Dwelley wrote:
> >
> >Yair, John -
> >
> >As I recall, the 70k number was intended to prevent two PSEs from
> detecting 
> >each other if they were erroneously connected together. Rick showed a 
> >presentation in St. Louis (if I remember right) that showed a possibility
> >of detection if the source impedance was less than 60-some K. At the time
> I 
> >ran his numbers and agreed with them, but I'd have to look back at my
> notes 
> >to see how we got to to that number.
> >
> >We do have text that allows any source impedance if there is a diode in 
> >series (so it's always 1st quadrant). 45K works fine this way.
> >
> >I'd oppose this change, at least until I get a chance to look back at 
> >Rick's presentation.
> >
> >Dave
> >
> >At 01:54 PM 1/8/2002 +0200, Yair Darshan wrote:
> >
> >>Hi guys,
> >>We specified that the high impedance option for the detection circuit
> >>requires 70K min.
> >>We also specified that the voltage across the port is 2.8V to 10V.
> >>Now, if we want to get the 10V when we have the following conditions:
> 70K
> >>series resistance, Rsig=23.75K (25k - 5%)
> >>the probe source voltage need to be 10*(70k+23.75k)/23,75k=
> 10V*3.94=39.4V.
> >>We have limited the open circuit voltage to 30V max.
> >>So to get the 10V max we need to reduce the 70k min value to (30V*23.75K
> -
> >>10V*23.75K)/10V = 47.5K.
> >>Having 5% margin we get: 45K min.
> >>In case of PSE to PSE connection the worst case power dissipation over
> this
> >>resistor would be: 57^2/45k=72mW which is <125mW.
> >>
> >>Thus to keep all numbers consistent with the intentions I suggest to
> change
> >>from 70K min to 45K min.
> >>
> >>Do you see problems with this change?
> >>
> >>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, Cell: +972-54-893019
> >>Fax: +972-9-775-5111
> >> > E-mail: <mailto:yaird@xxxxxxxxxxxxxx>.
> >> >
> >> >