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

Re: [8023-POEP] 802.3af Minimum Signature Voltage



Joe,
 
The key is to use a depletion-mode N-chan MOSFET.  These switches are fully turned on at Vgs=0, so Rsig is seen at the RJ45 even when the PD is completely unpowered.  After the PD is fully powered, you turn off the FET (removing Rsig from the circuit) by making Vgs=5V or so.
 
Chip makers don't seem to be having trouble with this so far.  I don't think the spec needs to be changed.
 
Steve
-----Original Message-----
From: owner-stds-802-3-poep@xxxxxxxx [mailto:owner-stds-802-3-poep@xxxxxxxx]On Behalf Of Joseph DeNicholas
Sent: Wednesday, March 16, 2005 9:56 AM
To: STDS-802-3-POEP@xxxxxxxxxxxxxxxxx
Subject: Re: [8023-POEP] 802.3af Minimum Signature Voltage


Hi Jerry -
Thanks for your reply, I'm sorry I wasn't more clear about which MOSFET I was referring to.  Many PD ICs are putting a MOSFET in series with the 25k signature resistor to "shut it down" after signature mode is complete because it wastes power at 48V.  This MOSFET must be turned on for the PSE to be able to "see" the 25k resistor.  The MOSFET must also be capable of standing off high voltage as it will see the full supply voltage.  So, the PD is responsible for turning this device on at a low enough voltage such that the PSE sees the 25k resistor.  But, with 2.8V at the connector and a lot of drop caused by the input diode bridges, it doesn't leave a lot of voltage head room to turn the MOSFET on.  Even just another 0.5V would be extremely helpful.  2.8V doesn't seem to be an arbitrary number, seems as if that was selected for a reason and I was wondering what drove that voltage spec.
thanks,
Joe



"<Gerard> <Bachand>"
Sent by: owner-stds-802-3-poep@xxxxxxxx

03/16/2005 10:35 AM
Please respond to "IEEE 802.3 Power over Ethernet plus"

       
        To:        STDS-802-3-POEP@xxxxxxxxxxxxxxxxx
        cc:        
        Subject:        Re: [8023-POEP] 802.3af Minimum Signature Voltage



Joe,
The signature voltage is only used to validate the signature, not to directly turn on the series MOSFET.
 
In other words, the signature voltage is used by the signature validation circuit, at a low level.
Once the signature is validated, and after classification (if used), the MOSFET is turned on.
Remember, there's always full supply voltage available within the PSE circuitry to turn on the MOSFET.
 
Regards
 
Jerry
 
--
Gerard E. Bachand
Power Systems Engineer
7 Atwood Terrace
Cherry Valley, MA 01611
508-756-7738
bachand@xxxxxxx


-------------- Original message from Joseph DeNicholas <Joseph.DeNicholas@xxxxxxx>: --------------


Greetings -

I was wondering why the minimum signature voltage in 802.3af was placed so low, at 2.8V.  With the input diode bridge drop of 1.5V or even 1.6V at cold, this doesn't leave a lot of head room (1.2V)  for the PD ICs to turn a series high voltage MOSFET device on and get it into low RDS(ON).  This MOSFET disables the 25k signature resistor at high voltage to conserve power.  I noticed the majority of PSE ICs are supplying higher than that, around 3.5V or more, but I was wondering why the spec is so low when there doesn't seem to be a need for such low voltage operation when all we are trying to do is measure a resistor.  


Has any! one proposed changing this in PoEplus?


Thanks,

Joe