Subscriber Discussion

Video Ground Loop Puzzle

AD
Alan Dodds
Mar 01, 2015

Hi everyone.

I posted not too long ago some issues I was encountering at a local horse ranch in Scottsdale AZ. As it turned out, there was a bad gound in the electrical system that kept damaging a number of cameras. The electric company sent someone and found the issue which now appears to be fixed.

So I went ahead a re-installed new cameras and everything has been working fine since.

Except for one thing: a few cameras are exhibiting some significant ground loop video issue, and specifically 2 of them. So I was wondering if I could get some suggestions on how to resolve this in the most expedient and economical way (the usual here).

Here's the set up: this was an exisiting system that I have serviced and added cameras to for the last 2 years. 24 cameras, all analog, 2 DVR's (IC Realtime), a mix of RG59/18-2 and CAT5 wiring. A mix of 12VDC and 24VAC power supplies. Power supplies are not centrally located (a few were placed closer to some stalls and buildings).

The main issue: 2 cameras, sharing the same 24VAC ower supply located in the DVR room.

I have isolated the cameras from the metal structures they were mounted on which helped a bit but the rolling bars are still there. I tried lifting the ground of the DVR's and other power supplies located in the same room with no success ( which puzzled me). The wiring (16/2?) and CAT5 goes to a junction box and then is split into 2 different conduits to reach the camera locations.

The one thing I have discovered when troubleshooting is that if I disconnect any one of the 2 cameras in question from the DVR (bnc balun), the ground loop completely disappears on the remaining connected camera.

I appreciate any suggestion on the issue.

Thanks.

Alan

PS: The secondary issue involves 2 cameras sharing a local 12VDC power supply, CAT5 for the video. The video noise is intermittent, not fixed by lifting the ground on the power supply. I have not isolated the cameras from the metal square poles on which they are mounted (poles that are integrated into the property metal fence). This is less critical at this time since the issue is intermittent but I will get to it next.

MI
Matt Ion
Mar 02, 2015

If you'll recall my response to your other post, seeing this on 12VDC-only cameras with baluns is not uncommon when they share a PSU. Try putting each camera on a separate wall-wart. It's not typically a problem with PSU or camera grounding, but with the fact the cameras have two VERY-different-length ground paths via the shared PSU and the baluns. AC and dual-voltage cameras tend to avoid the problem because the power and video grounds are isolated via their internal power rectifier/regulator.

You might want to see if the noise on the AC cameras goes away if you unplug the problematic DC cameras....

(1)
AD
Alan Dodds
Mar 02, 2015

Thanks Matt.

I'll explore my wiring options further but for the most part I'm stuck with existing limitations.

Rewiring might be the final and only solutions so I can apply the fixes recommended.

I don't remember if I tested the AC cameras (which by the way are DC with added AC to DC converters) with the DC cameras disconnected. It would be interesting if there was some interaction between the 2 sets of cameras.

Alan

MI
Matt Ion
Mar 02, 2015

If you're using a single board regulator to power multiple 12V cameras off a 24V supply, it's still essentially the same thing as using one 12V supply for them, with the same potential problem - the shared power ground is where these issues CAN arise.

MI
Matt Ion
Mar 03, 2015

Adding to my earlier reply: test first by using two separate wall warts for the cameras - if that clears it up, you can just leave it that way, no major re-wiring required.

UI
Undisclosed Integrator #1
Mar 02, 2015

Must have same power from power company ( Phasing ) a, b, c

No Sub panel to main panel arrangements ( phase a one system , phase a on other system)

also common Bond on grounds , both systems must be bonded to ground s locally and between.

Must have wires separated from any induced power lines by at least 18", Min. 12 " if not possible

check if connections are taped , sealled , not near any metal to grounds

see previous discussions from ipvm on this issue

use ground loop isolation baluns.on the listed lines

AD
Alan Dodds
Mar 02, 2015

Thanks for the feedback, "A". The only way I could check off all the recommended points would be to rewire which in this case would be last resort due to how labor intensive it would be. That's why I'm looking for "tricks" or solutions to get around it.

I've tried ground loop isolation baluns but they had no effect. Maybe they were of poor quality, I'm not sure.

I'll be going back and trying various things soon again and will report if anything helped.

Alan

U
Undisclosed #2
Mar 02, 2015
IPVMU Certified

Alan, since you have now isolated the cameras from the metal structures, are you saying there is no connection to earth ground until back at the DVR or power supply?

You had mentioned possibly using lightning arrestors at the cameras, was this done?

Avatar
Mark Jones
Mar 03, 2015

I would do two things next. Take a 9" monitor to the site and start at the camera. Check the picture there, right at the camera. Work your way back to the DVR, one connection at a time. Check both sides of your video balun.

Then take a camera out to the current camera. Replace the current camera with a new, differerent model. Have someone at the DVR. Check the signal at each and every connection point. Move it all the way back until your new camera gives you a clean picture. That will be your trouble spot.

By using a monitor and a camera, you will find the issue. The instant the image clears, you will know it and that will tell you what you need to know.

BD
Bob Damrau
May 01, 2015

Sharing, grounding and AC hum are a bad mix.

Chokes, caps, isolation, shielding, etc are all script for troubleshooting.

Watch for workmanship and "Doooh, who did that" moments.

(1)
New discussion

Ask questions and get answers to your physical security questions from IPVM team members and fellow subscribers.

Newest discussions