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

RE: Candidate Detection Tolerance Allocation and Test Limits


I disagree, see below

At 12:09 PM 11/21/00 -0800, Schwartz, Peter wrote:

My preference would be the 5% detection tolerance, This eases the
requirements on the reference voltages (or reference currents) used by the
PSE during its detection process, while still allowing for some degree of PD
power-draw "enumeration" by resistor value.

On the second issue, I believe that power-draw enumeration is an issue
better handled in software than in hardware.  I don't believe that it is
economical, in the final analysis, to use a 100Mb/s Ethernet port to
trickle-charge a 12V or 24V exit sign battery,

First, no one is saying that it will be a 100 Mb/s port. It is more likely 10 Mb/s shared.
Second, we have seen some pretty compelling presentations that it is very likely to be more economical over time.

and - just speculating - I
don't even know if that would comply with code.

Current code perhaps but it is fairly clear that it is not legal to standardize implementations where performance can be used (the Hydrolevel case). I would expect code to migrate when proven product is available.

More importantly, some
devices may need to be given priority access to power (if the CEO wants to
plug his PDA onto the net, I don't want my .mp3 player telling him "no...").

That may be so but it is beyond the scope of the current project.


Peter Schwartz
Phone:  408.435.2460