Modified sine-wave in video systems

A high frequency superimposed / modulated on a low frequency cannot pass through a transformer.

This makes sense. Perhaps you can use some sort of isolation transformer or a noise filtering circuit in your installations. Your problem might lie with the quality of the UPS itself. The high quality UPS may be using a high enough frequency to convert DC back into AC that it is inducing an RF signal on your power input.

Another thing you could try is a reactor as found on Variable Frequency drives. This would be placed between your hot/common (or hot/hot as your country may be) in order to reduce the noise. Not sure how this would do with a video application, but it works acceptably in motor power leads.

Find an Amatuer Radio Handbook 5 or more years old.

There is a lot of filter theory (as well as other Electronic theory) and you can solve virtually any problem like this.

They are great reference manuals. The problem with the newer ones is they are limited in scope and are meant to be a lead in to sell you the same info found in separate books, thus parting you from more of your money.

You can then construct the filters you need easily.

The company that provided the UPS commented that the IGBT Converter opperatates at or near the same frequency as a camera signal. They have dispatched technicians on two separate occasions to install ferrite beads on the power lines in and out of the UPS these were installed with all of the conductors for power going through 3 and 4 beads at once and only making one pass. I have seen these installed within the devices we install but the conductors always make two passes through the beads before going to the device power supply. Any comments on that? At this point the only "fix" that we can figure is replacing the UPS units with a different model that uses a different converter. This will be expensive and irratate the customer to no end.

What kind of interference are you experiencing? I've run into situations were the video had a rolling noise bar across the screen. In each case, this was traced back to an undesired ground current traveling on a coax shield. Looking at it with a harmonics meter, the signature was similar to the switching power supply in the video unit. We generally fixed the problem by fixing or verifying a proper ground path, providing a better ground path, and/or putting a isolation transformer in the video coax.

We have a "bloom" and some jitter on the screens that goes away when we plug the power supplies into any ol wall outlet. The cameras are coming back some on coax and some on twisted pairs with NVTs. Some active and some passive receivers.