Weird AC Voltages in a PoE Camera System
We are switching the Gravesend Inn video systems over from an old, all-analog system to a modern, IP-based system with ACTI ACM-1431 cameras powered over the Ethernet line using PoE (Power over Ethernet). I got the cameras a few months ago and did a lot of testing with them on a plastic cart in the shop, and everything worked great. I wanted to use one for a show right at the end of the semester, and discovered that the ACTI mounting bracket had a convenient 1/2" hole that fit a lighting c-clamp perfectly. I clamped the camera up to our lighting grid, ran the cable down to a Netgear PoE switch in our sound booth, and plugged it in. It didn't work. I went back up to check for lose wires and so on, and couldn't find anything. Working with a student, we tested the cable, and then eventually plugged into another port on the switch, and it didn't work either. I disconnected the camera from the grid to get better access to the wiring, and we plugged it again (to another port) and it worked. Right about that time, I got a shock from the camera through to my other hand which was holding onto the grid. I knew that PoE runs at 48VDC, so I figured it was some issue with that and a loose wire. Then I surprised my student with a big spray of sparks as I went to clamp the camera back on to the (grounded) pipe grid, and of course, the camera stopped working again. I got out the voltmeter and found some very interesting things, which I have documented here in this video:
I forgot to mention in the video that I tried moving the cameras around to the various PoE sources but saw similar results. I also got the same thing when using D-Link PoE extractors (which will be our standard set up--we need the extra power to power a microphone at each camera position). Also, only the bigger Cisco Small Business switch shows a low-resistance path to the ground pin on the incoming power plug--all the others use external supplies that apparently float off the ground.
I assume most people don't see these kind of things because they have PoE devices that are never connected to ground, or have a plastic case. I guess I could go in and decouple the camera electronics from its case but I'm not too excited about that idea.
I think we'll be OK for the Gravesend Inn since I'm planning to use the Cisco Small Business switches you see at the right in the video, which show only a tiny (and seemingly reasonable) leakage voltage, but I'd like to get to the bottom of this. Does anyone know anything about why we'd be seeing 50 or 60V AC on a device that's powered from 48VDC? Please post a comment if you do!
John Huntington
NOTE: I have posted a followup with new information here.
Reader Comments (7)
Did you check what's coming in on the PoE lines from the switches and see how that compares to the values you're reading from the camera body? I know that a lot of PoE systems wander off of the official standard, and also that Cisco had at least at one time used a PoE pinout that reversed the polarity of the powered pairs. If the camera conforms to the old Cisco pinout and tied what it expected to be the ground wires to the chassis then I could see why a presumably older Cisco switch would not show the voltage to ground while the others would all show a significant potential to ground.
As for AC vs. DC, that I don't know, but again what's coming from the switches?
I graduated way back in 1966 and since I have not worked with electronics in over 40 years, this may be off, but doesn't the AC voltage seem to reflect pretty near the .707 RMS value (at 60hz) of the supplied 48 volts?
Perhaps it's possible the cameras are converting internally to AC for varying voltage requirements?
However, why someone would design a case that is hot would be beyond me.
MADE IN CHINA?
They do silly things like that over there.
:-)
That's nuts! And the difference from switch to switch and the injector wow.
I could surmise whats going on as well as you im sure, but I'd rather not. I am just thinking how insane it is to be like that and how unsafe or simply evil that is...
You'll want to check your mains power sockets, they may be reverse polarity.
Is this helpful?
I noticed you weren't using a splitter. According this document you may need a splitter..
PoE Info
http://docs.google.com/viewer?a=v&q=cache:p2dYa4PLmCIJ:www.acti.com/support_old/package/%257BA743C2DD-E5D8-465E-9B84-DEDA7959D0F.PDF+PoE&hl=en&gl=us&pid=bl&srcid=ADGEESh43hZnef5YXaXUJ9uLVSnRL9WMeqhhrWcZYLRtgQQWYq5GCmf0Q6gkzd2hlbH4c8VYUskNnoP-U1WO-PQIfjQwuorarsv-bLpiwn5TL9XFPCJxUcW_Yn9UAv6XGMVRlMM5hwWm&sig=AHIEtbSWBHGQIkO4lIr1wVf-ERU_VV7uyg
If you can't paste the link above
(1) Go to ACTi
http://www.acti.com/home/index.asp
(2) Type in search. PoE
(3) Top of the results should be (PDF) [support package]
<I>" PoE selection and install guide for ACTi products. PoE stands for (Power-over-Ethernet)."</I>
Or alternatively, just Google.. (w/o quotes)
PoE selection and install guide for ACTi products. PoE stands for (Power-over-Ethernet).
Thanks to everyone for the comments! We've had further discussions over on the show control mailing list here:
http://tech.groups.yahoo.com/group/Show-Control/message/15123
I'm also in communication with the chair of the IEEE PoE committee; he asked me to do a couple more tests which I will do tomorrow. I'll definitely post a followup if I find anything.
Also, to answer a couple points posted here:
Not sure about the P-P/RMS thing, but the main confusion to me is the DC source of PoE power and the AC I'm measuring. Most of the speculation I've seen seems to be based on the PoE receiver, or the grounding scheme on the camera. I'll check both tomorrow.
I checked the mains power and the ground, and also had the same problem on two completely different circuits. As I mentioned and link above, I also had the same problem using a D Link PoE splitter.
I'll post an update after I dig further tomorrow, thanks to everyone for the input!
John
Dear John Huntington,
Thank you for initiating this interesting discussion. The phenomenon of the voltage read on chassis of a camera is very clear to us and in order to avoid further misunderstandings, we recorded a short video clip for you:
http://www.youtube.com/watch?v=mw_tU40u0_M
To sum up:
(1) If PSE is not properly isolated then you might get the voltage read on the chassis of any IP camera (Pelco, Axis, Hikvision, ACTi ...). You will also get the same read on the chassis of the PSE device itself!
(2) There are 2 reasons of not having properly isolated PSE: 1. The AC cable of PoE switch power supply does not have a ground pin; 2. the ground pin exists, but it is dummy (nothing is connected to it within the PSE device).
(3) Properly isolated PSE has no voltage reads on the chassis of PSE device itself; Consequently, there is no voltage read on the chassis of the PoE supported IP cameras either.
I also have one small request, if possible. The way the current video of yours was shot might confuse and mislead lot of people who might conclude that "it is ACTi bullet camera's issue" without going deeply into discussion. This kind of rumor might unfairly damage the reputation of our brand and the position in the market. Therefore, would you kindly arrange the content or the conclusion of your study so that the visitors of this blog or Yahoo group would not think of this topic as an IP camera issue. We would appreciate your effort a lot!
Thank you so much!
Sincerely,
Ando Meritee
Manager of ACTi Product Support Division