I replaced several analog cameras on a takeover job last year. At one of the camera locations, the image drops out once every 2 or 3 months. Power cycling the camera restores the image.

The camera is wired up with siamese RG59 and 18/2 that runs about 80' from a brownstone basement (where the power supply and DVR is located) to the outdoor camera location.

I've tried several different power supplies and several different camera models to no avail.

All camera models were from Wirepath. Wirepath tech support feels that there is likely a charge build up on the cabling, and that the image drops out once the charge reaches a certain point. There are 5 other cameras on the DVR, and none of them have any issues. So, the source of any charge build up would presumably be particular to the problem camera.


Two questions:

1. Exactly what sort of tests should I do on the 18/2 and RG59 to try and detect any electrical anomalies?

2. What can I do in the way of grounding, etc., to help prevent any electrical anomaly from impacting the camera?

Much appreciate any thoughts.

Is that power supply separate from the others? What else is on that circuitthat when turned on may be affecting the camera power?

When it does out is it completely black or is it video loss?

Of the several cameras you tried, did you just swap cameras with one of the other snap cameras that may be from the same batch? You might want to try one from a different manufacturer.

At one point, the camera was sharing a 12V power suppy with another camera. At this point, it has its own dedicated 12V power supply fed from the same surge strip as the other power supplies.

The failure is video loss (as opposed to completely black.)

BTW, disconnecting and reconnecting the RG59 does not restore the image, nor does rebooting the DVR. Only power cycling seems to restore the video.

I agree that a camera from another manufacturer may handle the site conditions better, but if i can get a better handle on the root cause I'd like to.

Just like Zohan said, can you swap the camera with a known working one. Which is one of the other cameras on site.

How about switching that camera's wires at the DVR to a different input? Of course re-labeling it in the DVR has to be done.

Or worse case, swap the cabling from DVR to the camera with a known working cable(s).

Have you looked at the entire run, are there any pinches, sharp 90º turns or a large coil of the cable?

KOT

Yes, we've done all combination of swapping the cable end points, and the "known good" DVR ports, cameras and power supplies all failed to work on the bad cable.

I don't know anything about the path of the bad cable. We didn't run it, and have found it difficult to get information about the wire path given the structure of the brownstone.

I know my original questions don't reflect the only way to go about trouble shooting this, but I'd really appreciate if anyone has direct answers.

I'm having a little trouble with the "charge buildup" cause. In any case one could protect against charge buildup by connecting a resistor across the coax. The value doesn't matter much as long as it is not close to 75 Ohms. A few Kilo Ohms would offer great charge buildup protection without significant signal loading.

[I don't know where this "charge buildup" cause comes from. I had an HDMI cable rep tell me how his little box would prevent charge buildup. His claim was that it's the "charge buildup" that causes so many HDMI problems. Fortunately, I was at a trade show and could simply walk away and did not need to kick him out of my shop. Possibly this is a misunderstanding of an "induced strike" associated with a nearby lightning strike.]

Are you certain that the camera and the DVR inputs are OK?

Have you checked the coax? A jacket break could cause the cable to take on water.

"Image drop"? Is this a loss of sync? All noise?

As far as testing the wires is concerned, check for conductivity end to end, and shorts from each conductor to its mate and to ground. While you are testing for conductivity, physically disturb the cable along its complete length. Does the end to end resistance seem reasonable?

A possibility: There could be a break in the coax or a failure at one of the ends. In any case an oxide build up at the break inhibits the signal. A surge associated with restarting the camera punches through the oxide coating and temporarily bridges the break. It's a fragile repair and a change in temperature or physical trauma can break the connection. [repeat the repair] This is a very common failure mode for speaker relays and poorly crimped connectors. Tracking these down can be frustrating because the slightest physical trauma can "repair" the break.

In order to rule out this sort of situation, pass some current through the loop. While a video signal could pass through a fragile point and the resistance might seem OK, the video signal is very low current.  Half an Amp or an Amp brief test current will not hurt a healthy cable, but it will probably blow out an accidental, fragile repair. Once you have blown out the "repair" a TDR tester can determine the cable distance to the break.

Since you are not responsible for the original installation, there might be an unauthorized splice hidden somewhere. I encounter these regularly. These electrician grade splices will hold together for a year or two, then I'll get a call.

+1 on Buzz's thoughts. I chased a similar intermittent CCTV problem for hours and finally found an F connector 1/4 turn loose on the backside of a keystone jack. A bit of contact cleaner and a wrench solved the vexing problem for good.

Thanks, these thoughts are helpful. I'm looking to gather as much intelligence as possible before going on site again, so further thoughts would be appreciated as well.

From all you have done, it sounds like you have without a doubt narrowed it down to a cable problem. But a problem on the coax (induced voltage possibly) or a problem on the 18/2 (broken insulation, causing corrosion on the conductors mid run). Just to name a few of the possibilities.

The very 1st step in troubleshooting this was testing voltage on the cable.
18/2 = Take your multi meter and test what voltage you have at the source (beginning of the run). Then test at the camera and see what your voltage drop is. On 18/2 going just 80 feet, with a Camera drawing 400ma at 12vdc. You should have no more than a .4-.7 volt drop (depending on conductor temperature and who's volt drop calculator you use).
So if you start out with 12.6vdc at the source and you are only getting 11.5vdc at the camera - then you could have bad splices or corrosion between here & there.

RG59 = Disconnect the coax at each end. Use you meter to see if there is any DC voltage on the cable. Also check for AC voltage. You will have some 'ghost voltage' that will show up with both ends being open on that coax. But it should not be more than a volt on DC or a few volts on AC. This ghost voltage comes from many things (electrical induction nearby, radio waves from the air, etc...) - 80ft is a long antenna.
If you find several volts of DC or 10's of volts AC, then there is a problem.

That's because it is pure BS suggested by someone with no clue about electricity.

A 75 ohm load won't allow charge buildup, so the only way it could happen would be for all of the tested cameras to have open inputs, or the hot or shield of the cable are open. There would never be an image in that case. So it's not that.


Or re-entering this reality.


But the 75 ohm load of the camera input cannot possibly be all capacitive, especially down at sync frequencies.


If you google charge buildup, a couple of the autofills are "charge buildup between plates capacitor" and "charge buildup in insulators." Charge buildup is a high impedance, VERY high impedance, phenomenon, and all realistic loads in our cameras and power supplies will short out any kind of such buildup.


I'm not criticizing Brad's answers here, just adding to them:

Measure the voltage at the camera with the camera connected, and at night if the camera has IR LEDs.
On 18/2 going just 80 feet, with a Camera drawing 400ma at 12vdc. You should have no more than a .4-.7 volt drop (depending on conductor temperature and who's volt drop calculator you use).
Using the numbers given and consulting resistance of 18 ga wire, I get 0.4 volts drop.

Disconnect the 18 gauge, twist the wires together at one end, and measure the resistance of the run. It should be about 1 ohm. Be careful that you know the reisstance the meter shows when it's actually shorted.

Then short together one end and check the resistance of that, too.

And be sure to carefully inspect all interconnections. That loose connection reported earlier is a classic thing to overlook.

True, but some units are rather sloppy about proper termination.

Can the cameras use 24VAC as well as 12VDC? Although it's only 80', 24VAC travels over longer distances without voltage drop and if the line voltage drops too much and then something causes more draw that normal, it could put the cameras below their threshold for usable voltage. Also, since you didn't do the initial installation, you might want to check all of the power wire terminations.

You have a point in suggesting 24 volts, but it's not the point you say it is.


Not true. This would be magic. There is voltage drop.
However, IF (and I don't know if this is true) a 24 volt camera consumes the same amount of POWER as a 12 volt camera, then:
The current at 24 volts will be half the current at 12 volts.
So the voltage drop (current x resistance) will be half at 24 volts what it is at 12 volts.
So the percentage of voltage lost on the run will be 1/4 the amount at 24 volts that it is at 12 volts.
So, yeah, you might get better performance with 24 volts. But how does a 12v/24v camera switch from one to the other? Does it somehow raise its input impedance so any of this will matter?


This is exactly the problem you'd have due to normal voltage drop, too. All that theory and percentage stuff doesn't matter if the drop is enough to be too little to operate the camera.

And none of this explains the camera working for a while and then stopping. The issue is the "something" mentioned above.


Heck yes!

I forgot to include 'as much'. Sue me.

What do you mean by "But how does a 12v/24v camera switch from one to the other? Does it somehow raise its input impedance so any of this will matter?"? The input is designed to receive either and nothing is required to use either. Have you not used cameras with dual voltage input? They're great- you connect them to either and do nothing else.

If there's a power hog in the area that starts a particular operation only on selected days once a month, it's possible. I would monitor the voltage and check for peaks/dips over the typical shutdown period but since it's a job takeover, I would go through and re-terminate the connections.

Not needed. Just looking for technical info to be technical. If a beginner read that, they might believe it. Help 'em out. Be accurate.


I've used dual voltage cameras. You're right -- you plug 'em in and they work. That reveals absolutely nothing about how they work, so asking how they work is relevant when the existing camera is failing or on the edge of failure.

Thinking more about it, a couple of scenarios result in 24 VAC probably being a better bet. (I'll leave out the next ten paragraphs for those who want a definitive statement, not proof or understanding.)


What, you mean like the neighbor turning on the smelter and dropping the AC voltage throughout the neighborhood?


Power monitors COST!
And, well there's this:

So such monitoring involves a time period of at least three months where you could not bother to monitor for the first, say, six weeks. There is no "typical shutdown period."


True.