Grounding 24v AC IP cameras?

[musher 0]

Hi,

 

So I'm looking at an altronix 24vac power supply and it comes packaged with a 2 prong plug (no ground) for powering the unit.

 

Most of the reading I've done seems to indicate that folks pull 18/2 or 16/2 to power cameras--no ground wire.

 

Looking at the wiring instructions for a Sanyo HD3500, I see that it specifies attaching a ground wire when powering with 24V ac.

 

I've a couple of questions.

 

Is it usual for the 24vac IP cameras to require a ground? I'm guessing this is a safety issue in case one leg gets shorted to the case? I didn't think code required a ground for 24vac, but I could be ignorant.

 

My first inclination is to sell my spool of 16/2 and find some 16/3 so I can home run the ground wires as well. I can install a grounded plug on the altronix unit and bolt a ground bus to the case of the power supply if I do that.

 

Do I need to do this, or can I ground the cameras locally where ever is convenient? Not sure if this would be a problem for an IP camera.

 

I don't think the camera cases will be grounded by the installation method (not bolting them to grounded steel or anything like that)

 

This is in my home, but it is multiple buildings (3).

 

I'd appreciate anyone who could help me better understand the issues here.

 

thanks.

[thewireguys 3]

Why not just use POE?

[musher 0]

A fair question. I considered poe, but I need the heaters in the cameras. IEEE 802.3af only guarantees 13W more or less at the device.

 

One of the 3 cameras I'm messing around with is an ACTI-7411. The spec sheet indicates that the camera heater is not available when using POE (fan only). Being in Fairbanks, AK, I'm pretty sure I want the heater outside.

 

I had settled on 24v to minimize the effects of voltage drop on the runs and was planning to use a 24-12vdc converter for the ACTI (which doesn't do 24vac) at the camera end.

 

The Sanyo, I was just going to feed with 24vac.

 

If I'm wrong about POE, that would simplify things considerably.

[thewireguys 3]

Yea Sanyo's need extra power for the heater/blowers and to get extended temp range with ACTI you need extra power. Your other option is to use housings that are powered with POE, more money but easier and faster to install.

Edited January 23, 2011 by Guest

[shropna 0]

If you use transformers (not a switching power supply) , no grounding necessary. Also, no grounding to camera is better than grounding to two different (different than control room equipment ground, as example) grounding points.

[musher 0]

This is more or less something I'm doing to stay off the streets and out of the bars, so installation time isn't a critical factor on this project. I did want to install things more or less correctly though.

 

What do you think about the 24vac grounding?

 

I suppose another possibility is if I could locate a 24->12 power adapter that could do 2 amps, I could run 16/2 and drop to 12vdc for the sanyo as well. It needs 20W to run the heater, so 1.7A or thereabouts for 12vdc.

 

That would produce a little extra heat in the installation as well--not a bad thing up here.

[musher 0]

Hi Shropna,

 

I guess we cross posted.

 

The altronix is a transformer based ps.

 

You think the best solution is to drop the ground wire entirely then?

[shropna 0]

You think the best solution is to drop the ground wire entirely then?

Best way, if you will use grounding, run ground wire from "central point".

[Soundy 1]

I think most analog "box" cameras I've seen have a ground screw, but I've never used it... few dome cameras do, of those I've used, other than the Panasonic SuperDynamic models, but I've never used those either.

 

Quite simply, don't worry about it. My understanding is that it's included for the sake of obtaining UL certification, but really isn't needed in most cases.

[shropna 0]

I think most analog "box" cameras I've seen have a ground screw, but I've never used it... few dome cameras do, of those I've used, other than the Panasonic SuperDynamic models, but I've never used those either.

 

Quite simply, don't worry about it. My understanding is that it's included for the sake of obtaining UL certification, but really isn't needed in most cases.

Hm. As an "opposite" example, Bosch cameras (one of my favorite, excellent quality, expensive analog cameras), mains power AC or DC doesn't have grounding Also, mounting screw hole is insulated from camera body

Thing, why

[musher 0]

thanks for the input.

 

Regarding the poe enclosures you mentioned, thewireguys, can you mount these dome cameras inside them? Do you just move the guts over or what? If you're willing to point me at a particular product, I'd be interested in looking at them.

 

Are they 803.2at to support enough power to run a heater/fan in addition to the camera or are they just more efficient/less power?

[thewireguys 3]

thanks for the input.

 

Regarding the poe enclosures you mentioned, thewireguys, can you mount these dome cameras inside them? Do you just move the guts over or what? If you're willing to point me at a particular product, I'd be interested in looking at them.

 

Are they 803.2at to support enough power to run a heater/fan in addition to the camera or are they just more efficient/less power?

 

 

No these POE housing will not work with dome cameras sorry.

[MarcoZandrini 0]

You think the best solution is to drop the ground wire entirely then?

Best way, if you will use grounding, run ground wire from "central point".

 

Central point " title="Applause" /> !

[Dirk_D 0]

Leave the FG, or floating ground terminal unused. You should be fine with your 18/2 siamese cable.

 

The poster who's recommending the central grounding is basing the recommendation on keeping the ground potential the same throughout the system, & it's good advice in both my field experience & understanding of the theory.

 

Having a central power supply (fused box preferably @ head end) seems to fix any ground potential / loop issues.

[Soundy 1]

One central ground, or a "star" grounding topology, CAN help with noise and ground-loop issues... SOMETIMES. Sometimes the extra grounds just create other problems.

 

If you want to plan ahead, use shielded 18/2 for your power and just leave the shield disconnected *unless you find you need it*.

[Dirk_D 0]

One central ground, or a "star" grounding topology, CAN help with noise and ground-loop issues... SOMETIMES. Sometimes the extra grounds just create other problems.

 

Like what? Are both ends of the cable grounded in your scenario?

 

Can you name a single instance where it would be a better idea (using analog cameras in a single building with under 800' runs) to have random transformers plugged into random outlets powering random cameras?

[Soundy 1]

One central ground, or a "star" grounding topology, CAN help with noise and ground-loop issues... SOMETIMES. Sometimes the extra grounds just create other problems.

 

Like what? Are both ends of the cable grounded in your scenario?

If they're not connected at both ends, then they aren't grounding anything, are they?

 

Can you name a single instance where it would be a better idea (using analog cameras in a single building with under 800' runs) to have random transformers plugged into random outlets powering random cameras?

Well first of all, if you're talking analog cameras (this thread is referring to IP cameras), they're already all grounded to a common point via the signal lines.

 

Beyond that, there's nothing WRONG with powering cameras from individual transformers, be they all near the DVR, or separate locations closer to the cameras.

 

But all this has nothing to do with my statement: I come from a studio background, and I've seen every conceivable type of grounding layout for equipment, including having *everything* lifted and isolated except for the signal grounds... and I have seen instances where a star ground for everything actually CREATES noise problems. There's no one explanation for it - sometimes there are environmental reasons that are beyond your control, sometimes it just takes one oddball piece of equipment.

 

I'm not saying that star grounding NEVER works... I'm saying that it's NOT a universal always-works setup. Sometimes additional grounding helps... sometimes it doesn't make any difference... sometimes it creates more problems than it solves. My experience in CCTV, it simply not necessary 99% of the time.

[Dirk_D 0]

If they're not connected at both ends, then they aren't grounding anything, are they?

 

It sends the EMI on the cable to ground, I thought. Grounding both ends can cause radio stations to be picked up on line level audio signals IME.

 

oops on the analog, I missed the thread's title.

[Soundy 1]

If they're not connected at both ends, then they aren't grounding anything, are they?

 

It sends the EMI on the cable to ground, I thought. Grounding both ends can cause radio stations to be picked up on line level audio signals IME.

Well, we're not generally dealing with audio on this forum, are we?

 

In any case, the suggestion to use shielded 18/2 was simply just for cost and convenience...

 

As far as audio, well, pretty much every XLR-terminated mic cable I've ever seen had grounds connected at both ends, save those that were poorly soldered, or one or two that had the grounds disconnected specifically to provide a ground lift to address those odd, pesky, unexplainable noise issues

[Dirk_D 0]

OT:

Problems that I've ran into with distributed audio in this scenerio, which also deals with what we thought were ground loops were say an audio amp / sat receiver on one end of the building, & the phone switch at the other. Some of the mics we still deal with are grounded at the amp, the ground *is* terminated @ the mic, but it's only the cable that's grounded when checked with my meter (hot/ground/common -or- rd/bk/shld). These mics are also they type that are commonly connected w/ XLR connections, not that it matters.

 

We also deal with POS & OCS systems - they're spec'd to have iso grounds to prevent variable ground potential issues on the network & between devices.

 

The best illustration of the problem (variable ground potential) had a bunch of boats on a bunch of different waves, each boat was floating on a wave at a different height. This illustration was supposed to show the relative difference between the relation to ground on each circuit. A quick google search turns up nothing that I can help others see what I saw, but I'll post it if I come across it.

 

Early on (about halfway through my CCTV career about 9 years ago) I had issues with great pictures, but every so often, like every 30 seconds, a camera image would flicker. It took a while but we went with regulated power supplies. Over time, maybe just out of luck, I gradually switched over to individually fused power supplies. It works for me, it's cleaner, & it's a hell of a lot easier to help the customer cycle the power to see if it'll help a blacked out camera come back up before we send a tech.

 

I've also had issues with dealing with grounding issues, over cat5e, throughout an IR system on a hospital's IR repeater system that we put in for their ERs. There was severe EMI that was fixed on this somewhat large IR system by grounding the unused cat5e (approx 4) wires to either the chassis or earth ground (conduit) - I don't recall, that was >9 years ago.

 

So I've tried every conceiveable method on several different technologies, & sometimes it boils down to what would appear to be some sort of workable electrical principle dealing with EMI on the lines on STP & UTP, in Audio, Video, & Data lines. Take it FWIW. Or don't. I have more technologies where this works, weekly, but the bat signal is in the sky and I has ta go.

 

I suppose the only thing you absolutely should gather from my OT post is that grounding the cable @ one end is good, umkay - grounding both ends is often problematic. I haven't worked with baluns in this field yet, and I have absolutely no experience with IP cameras, so I'm afraid this is all I have to contribute on this topic.

[Dirk_D 0]

OT: Soundy, pull out your $10 harbor freight meter next time & see where the shield is connected to next time, chances are, it's not.

 

EOM - I'm out.

 

Have a great day guys.

 

[Soundy 1]

Heheheh, I've BUILT enough mic cables to know it's attached at both ends... generally to pin 1

 

In any case, back a little more on-topic: the one problem that's common with baluns and cheap 12V cameras is ground loops induced by the fact that the cameras use a common ground for power and video, and baluns basically put a coil inline with the signal ground on both ends, effectively creating one very long ground path, while you still have a separate shorter ground path through the power ground. AC and dual-voltage cameras avoid this because the power ground is separated from the signal ground by the regulator/rectifier (and in some older cameras, a transformer).

 

Back even more on-topic: none of this has anything to do with IP cameras, except to reiterate the point: grounding your cameras will MOST OF THE TIME make NO difference... if you DO have noise issues (which would probably show up as network dropouts on an IP camera), grounding MAY help... or it MAY make things worse. If you DON'T already have problems, grounding MAY also create issues. In short: your guess is as good as anyone's

 

My advice would be: don't ground the cameras by default. I've never done it, and I've never had an issue. Save it as something else to try IF problems arise... or of course, if local electrical codes require it.

[Tailbone215 0]

For IP cams just use shielded CAT5e cable. Worth every penny.

 

http://www.cablestogo.com/product.asp?cat_id=303&sku=27432

[Soundy 1]

STP is nearly twice the price of UTP... and the shielding doesn't do a lot unless you terminate with at least shielded connector. I've installed tens of thousands of meters of Cat5e (including one job that was a total of 30,000m) and never needed it... I don't see where it's worth the extra price.

[Tailbone215 0]

STP is nearly twice the price of UTP... and the shielding doesn't do a lot unless you terminate with at least shielded connector. I've installed tens of thousands of meters of Cat5e (including one job that was a total of 30,000m) and never needed it... I don't see where it's worth the extra price.

 

Had a job where nearby variable frequency drives were playing havoc with noise. This stuff did the trick. I agree, wouldn't use it on basic installs. Probably could have ran EMT, but that would have been miserable.