A collegue of mine insists that it is impossible for a port to have gigabit speed while simultaneously using 802.3af PoE.
Can anyone please confirm or deny?
There's a round of beers running on this one.
-RFH
A collegue of mine insists that it is impossible for a port to have gigabit speed while simultaneously using 802.3af PoE.
Can anyone please confirm or deny?
There's a round of beers running on this one.
-RFH
PoE works just fine on Gigabit speeds. Most of Cisco's phones are Gigabit capable and are powered using PoE.
Your collegue would be wrong. Done all the time and has been out for a couple years now. In the Cisco world the 7941G-GE and 7971G-GE phones are gigabit, both of them use PoE. Pretty much all switches from the 3560 on up offer a version of Gig PoE, ie 3560G-24PS, 3750G-24PS, etc.
snipped-for-privacy@j33g2000pri.googlegroups.com...
I think he is saying that the port will then not push gig throughput, not that PoE doesn't run on gig ports or something. Can anyone with an IP phone run a ftp test to a local server and see the throughput they get?
messagenews: snipped-for-privacy@j33g2000pri.googlegroups.com...
I believe he is refering to how in one mode of PoE it uses the "unused pairs" but in Gigabit there are no unused pairs so how could it work.
The third-pair differential signal and the fourth-pair differential signal are used by GigEthernet. Power is fed to the transformer center tap, so is a common-mode DC level on the third-pair, and a different common-mode DC level on the fourth pair, and the power adds nothing to the differential signal, so does not interfere.
The Ethernet standard allows up to (?) 500V of DC on the twisted pair wiring, I seem to recall.
Look up 'phantom power'; this was a trick of similar sort played with microphone preamplifiers, going back to vacuum tube days.
Phantom power is still very much alive and well in professional audio circles.
And amateur!
Sam
In what way?
martin
He's wrong. Gigabit ethernet works nicely full speed with PoE.
Sure. I just happen to be setup to run IPerf and JPerf on gigabit in my palatial office:
The test server is a Dell Optiplex 755 machines (2.6GHz Core2Duo,
1333Mhz FSB, 2GB DDR2, SATA-II HD, etc) running XP SP3 with the lastest band-aids. The test client is a Dell SX-280 (P4 2.8GHz, 533MHz FSB, 1GB PC5200, SATA HD, etc).The power injector is a Phihong POE125U-560-8:
The Gigabit switch is a Linksys SLM2008:
I disabled QoS in the switch for testing. There's about 3 meters of CAT5e patch cable in between.
On the server: | iperf -s -M 100000 -w 64K -l 24K which means: Packet size = 1514 bytes Window size = 64KBytes Buffer size = 24Kbytes These are not the defaults as gigabit requires some tweading to get full speed.
On the client: | C:\\>iperf -c 192.168.1.100 -M 100000 -w 64K -l 24K | ------------------------------------------------------------ | Client connecting to 192.168.1.100, TCP port 5001 | TCP window size: 64.0 KByte | ------------------------------------------------------------ | [1908] local 192.168.1.101 port 2362 connected with 192.168.1.100 port 5001 | [ ID] Interval Transfer Bandwidth | [1908] 0.0-10.0 sec 1.09 GBytes 921 Mbits/sec | [1908] 0.0-10.0 sec 1.06 GBytes 915 Mbits/sec | [1908] 0.0-10.0 sec 1.08 GBytes 927 Mbits/sec
Ship the booze to the address below.
messagenews: snipped-for-privacy@j33g2000pri.googlegroups.com...
Take a look at the standards doofus.
It is becoming common in a lot of audio equipment for performers at all levels.
Condenser microphones (one of the two main types) require a power source. They are either fitted with a battery (which needs to be checked and renewed) or are phantom powered. Similarly DI boxes (devices which match different signal levels and impedances) are often powered and again it's convenient not to require a battery. I think I've noticed other things being phantom powered recently but those are the traditional things.
Sam
Ah, that makes perfect sense as the source of the OP's coworker's disagreement: in one mode 802.3af uses the spare pairs, in the other mode
802.3af uses the data pairs (in common mode, with power being taken off the center taps of the powered device's isolation transformer). The "spare pairs" arrangement would also work for 1000baseT, as long as the same sort of circuitry is used on those no-longer-spare pairs as on the data pairs. But a midspan PoE injector probably won't do that unless it's explicitly designed for GigE use, so the OP's coworker might well have experience with GigE devices that don't work on PoE ports.Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.