Hacker News Reader

Comment by PaulWaldman | Hacker News Reader

My understanding is that PoE is typically 48V and 10g Ethernet is something under 4V.

I'm not up on the latest NFPA codes, but how much of a hazard is 52V? I'm also guessing the quality of the cable is a factor.

View on HN

You can start a fire with a 24v car battery (and dry tinder, and skill). However, PoE switches only supply power after successful negotiation with the client. You would have to be _seriously_ unlucky to get a cable intact enough to support a connection but damaged enough to start a fire - I don't think I could do it on purpose, even if I had a few days and a stack of spare cables to try.

I'd be more concerned about the power supply (between the wall and the router) than the 48V lines. Would definitely want to be sure that was good quality.

Or a 9 volt some steel wool and not much skill

I feel like that's cheating; steel wool will burn if you so much as look at it funny.

The normal way to do this is to get a dodgy CCA cable with insulation that is not riser rated, and run a bundle of them together into a wall.

IMO it’s less about the voltage and more about the wattage. A modern PoE spec supporting wattages ~100W is more than enough to start a fire. Currently there is a gap in the electric code (NFPA70) as it doesn’t really address this risk in the (vague) section on low voltage wiring. But building codes are written in blood. I’m sure it’s just a matter of time till some PoE cable starts a fire and then the code will get amended.

One might also consider the quality of the power supply itself, and how much of a fire risk it is.

Here is an (unrelated) PoE injector catching fire: https://community.ui.com/questions/Litebeam-5ac-gen2-poe-ada...

It's a different type of device, but my point here is about mains voltages.

Reply on news.ycombinator.com