Originally posted to the ASEE internal Information Technology blog by Sean Stickle:

As discussed at lunch, I have long believed the Polycom Power-over-Ethernet cable to be a fiction: at best, a misunderstanding on the part of the sales agent; at worst, a ploy to make extra money off customer ignorance. So I decided to take apart one of the magic cables and see what mysteries it held. Unfortunately, the cables didn’t show up with the phones. So I dissected the next best thing: one of the power cables that was bundled with the phones.

Here are the interesting results of my investigation.

Warning: these pictures show graphic dismemberment of a relatively expensive power cable. No otters were harmed during the filming of this investigation.

I began by stripping off the plastic sheath that protected the four pairs of wires.

Then I began to remove the hard plastic shell around the part of the cable where the wire from the electrical cord connects to the cable.

At this point, we can see four wires connecting from the Ethernet cable to the power socket, on a small printed circuit board.

I had to clip those wires to continue removing the plastic casing. Having noted what connected where, I cut them, and was able to see more of the delicate internals.

A little more work with the exacto knife, and I was able to remove all the black plastic from the bottom of the circuit board, revealing the wire traces and a basic circuit.

With yet more work, I was able to remove the top part of the plastic case, revealing two inductors (removed in this picture) and the rest of the circuit.

Topside:

Underside:

The circuit, once exposed, was easy to figure out.

In fact, this is exactly the wiring standard for 802.3af Power Over Ethernet. It was exactly the opposite of surprising. This was exactly how it should work. The power is delivered over the spare set of wires.

But our new HP switches (2650-PWR and 2626-PWR) are 802.3af Power Over Ethernet compliant. So why, why, why didn’t the phones work when we plugged them into the switches? We should have been able to plug them in, and the phones should have powered up immediately. But we could never get that to work.

Many furious minutes later, after downloading and reading HP’s Power Over Ethernet Manual Supplement, the answer became clear:

Our switch delivers power over the data wires. The phones look for power on the spare wires. The two don’t fit together.

And there is the explanation for the mysterious, fabled cable of Polycom. It must be a cable that converts “power over data wires” to “power over spare wires”. And that is a sufficiently complicated operation that I can see whay they cost $40. And they’d be hard enough to make that there’s no reason for us to try to make them ourselves.

So, a little investigation led to understanding. Now we understand the phones and the switches better, and understand why we need these magical cables that we are still waiting for, and understand how the whole voice network operates with all the pieces plugged into each other. Not a bad result for a few hours of dissection.

A quick picture that demonstrates how power is delivered over the data pair of a CAT5 cable, and then extracted:

This is basically what the magical cables from Polycom will have in them, I warrant. A small transformer to separate power from data.

This entry was posted on Wednesday, June 6th, 2007 at 10:41 am and is filed under How Things Work. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.