Maker Pro
Maker Pro

Video Distortion Over Cable Distance

Last_Mavrik

Nov 15, 2012
9
Joined
Nov 15, 2012
Messages
9
Hello again all!

For those of you who helped with my previous questions, thank you! Our new camera system is up and running, with a few minor issues, but most of them easily addressable. This one however, not so much.

For those of you with no clue what I'm talking about, the system in question is basically this:

[Power Cable] --pwr120vAC--> [Transformer] --pwr12vDC--> [130ft 6-Conductor Cable] --pwr12vDC--> [UBEC Voltage Reducer] --pwr5vDC--> [Camera Assembly] --vid--> [Same 130ft 6-Conductor Cable] --vid--> [Monitor]

Simple, but here's the catch: The video signal produced by the camera is beautiful, if you hook it up directly to the monitor. Once you put the [130ft Cable] in the line, the video has minor diagonally scrolling lines of distortion. Still usable video, but not the high-quality you'd expect of a finished product.

So, I've troubleshot this extensively, and here's what I've found:
- Excluded all components from the system save the Camera, Cable, and Voltage Reducer. Direct 12v power from an external source and direct video feed to an externally powered monitor, no change.

- Move power to just behind the Voltage reducer, no real change so long as video is still being pulled from the opposite end of the cable.

- Power fed through the cable, video pulled from just behind the camera assembly (before the 130ft of cable) and the video gets white-washed.

- Put power in just behind the voltage reducer and take video directly from the camera, awesome video.

My most educated guess with all this is either video signal strength loss over distance or power bleed-over from the wires being in close proximity inside 130ft of cable. None of the 6 wires in the cable are shielded, just 22AWG stranded conductors with assorted colors of insulation and a thick outer insulation.

Q: Is there any way I can strengthen the video signal or clean it up before it hits the monitor?
Q: Would a Ferrite Ring be of any use on the video line?
Q: Is there any plausible solution that doesn't require replacing the cable with one that has a shielded Video line?
 
Last edited:

davenn

Moderator
Sep 5, 2009
14,254
Joined
Sep 5, 2009
Messages
14,254
130ft... yeah there will be significant loss of video signal
putting at least 1 booster halfway along the cable run will either solve or at least improve the signal
There are also video over ethernet units available, used to use them when I worked for a previous company. They are just a couple of small boxes a TX one and a RX one
about the size of a packet of cigarettes

Dave
 

Last_Mavrik

Nov 15, 2012
9
Joined
Nov 15, 2012
Messages
9
130ft... yeah there will be significant loss of video signal
putting at least 1 booster halfway along the cable run will either solve or at least improve the signal
There are also video over ethernet units available, used to use them when I worked for a previous company. They are just a couple of small boxes a TX one and a RX one
about the size of a packet of cigarettes

Dave

Thanks for the suggestion Dave,
Unfortunately, putting anything in the 130ft length of cable isn't possible. It's a Fiberglass-Rod Inner Core Push Cable that needs to retain it's integrity from end-to-end. The unit is designed for pipe inspection, and the camera end needs to remain as small in diameter as possible, so anything added to that end would have to be short, in-line design, and around an inch in diameter at best.

I have a few ideas, though I'm not sure if they'll work:
- As I understand it, with twisted pair lines, the 'strength' of the signal is proportionate to the electrical flow through both wires. The Voltage Reducer (UBEC) I am using right now is a 5v 3amp. While I'm aware that the camera will only pull as many amps as it needs, I've purchased a 5v 5amp one to see if the increased potential flow will have any effect.

- Next, I've ordered a few clip-on Ferrite Rings. While I'm aware that they're usually employed to help 'clean up' power by reducing EMF in the line, I haven't seen anything about their use with a video signal, so I'm experimenting there.

- Alternatively, a higher voltage camera would likely have a stronger return video signal and loose less over the distance, so all else fails I'll be looking for one. The down-side is our camera is an in-line PCB design made for Endoscopes, so it's tiny, and the housing is designed to fit it. Finding one of the same size that uses higher voltage may be impossible, and any others will require a complete re-design of the head.

Oddly enough, we don't have this issue with every assembly, despite them all using the same parts and being identical in every way save the occasional inch or two difference in wire lengths.
 

Electrobrains

Jan 2, 2012
259
Joined
Jan 2, 2012
Messages
259
Some comments, ideas and suggestions:

1. It almost looks like an engineering black-out to put a fast video signal in a long un-shielded cable together with power supply and whatever.

2. For a long distance video signal you would normally use a shielded cable with a well defined impedance and matched source and termination resistors. If you don't have that, you will get reflections that can totally distort the signal.
I guess that your 6-wire un-shielded cable has no specified impedance and will be susceptive to what is put on the other wires, how they are internally drawn, where the cable is placed, how it's folded etc.
There will be capacitive coupling between the wires resulting in noise and attenuation.

3. If you cannot replace your cable, you have to find some compromising solution. Here some suggestions:
Do you have free wires inside your cable? If yes, you could try some different combinations, like connecting more wires to ground, using several wires for the power supply or even placing the video signal on other or more than one wire. Also, exchanging some wires might influence the result.

4. I don't think the ferrite bead will help your video signal. It will rather limit down the bandwidth even more. But you could try to put the ferrite on the power supply wire(s) instead.

5. You could play around a bit with different terminating resistors, to see what happens (eg. 50 Ohm up to 1k).

6. There exist a lot of very good and cheap video buffers now. If you want to build your own one, you can for instance look at these types or search for "video buffer" on Newark, Farnell, Digikey etc:
OPA692 (good and cheap, but needs split supply)
LMH6559
They are small in size and do not need many other components. You could probably build a small amplifier in an hour or two on a small piece of bread board and place it as cable driver just beside the camera.

7. Twisted pair cables are excellent for differential signals (RS485, Ethernet, USB etc). I suppose your video signal is ground referenced though. To twist the video wire with a ground wire might help, but could also bring inductive and capacitive negative influences on the impedance, attenuation etc.
 
Last edited:

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
The industry impedance standard for CCTV is 75Ohms. The camera's output Z = 75Ohms. The cable impedance also equals 75Ohms. Finally the monitor's input impedance should be 75Ohms. Any mis-match from any of these three components can cause what you describe. Some monitors have a switch or keyboard option to switch the 75Ohm termination out of the circuit. This is done when there is more than one monitor. Only the last monitor on the line should be terminated in 75Ohms.

Chris
 
Top