AC Voltage Interference Over DC Control Line

Like JRaef said, ground the control line shield at one end only for low-freq interference (50/60 Hz). Twisted pair wire is also best for this type of installation. Since your wiring is already installed and you are looking for a solution for the installed system, you could try to raise your ground potential of the control signals. By raising it above the reference utilized by the AC circuits it would put the AC peak voltages under your control signals.

The chassis bonding scheme of the controllers and power supplies internal to those controllers with reference to the hotel bonding scheme (star or not) could also induce/reduce the interference seen by the control circuits.

I had a similar situation and the "control chassis" utilized a VME backplane which could have been bonded or floating (compared to reference ground). I changed the existing bonding scheme by isolating the power supply from the chassis and bonded the DC common/reference to the chassis ground. This brought the AC interference down below my analog (DC) signals.

The (problem with most) CAT cables today (is that they) are manufactured with bonds on both ends, which is fine for high speed/frequency communications, but older control schemes may be negatively affected, especially when used in concert with AC lighting.

The presentation was fundamentally incomplete. That indicates , that the OP has his work cut out for him. Worse, he might encounter resistance from some equipment makers to his findings. With that said, let's start.

1,. Proving, and measuring single point grounding of shielding is always a bitch.

1.1,. Maintaining it with generally uncooperative equipment suppliers is an absolute necessity, and a bitch of a problem on the square.

Having these out of the way, here are my concerns:

2,. TRUE differential signalling without shield has a 30-60dB dampening from an external aggressor signal, depending on the quality of the balance. Meaning, that outside of a powerful radio transmitter, you should be ok. But...

2.1,. If you take the twisted pair as one, the Common Mode signal is as big, as it would with a single wire.

2.2,. At this point, you need to consult the manuals of all connected equipments, to see which are suspect both on generating and handling badly on the receiving side the unwanted Common Mode signal. Common Mode Suppression factor on both sides chips is a good starting point.

2.3,. The impedance for the Differential Mode wanted signal is normally 50, 75, 150 Ohms. The Common Mode one is normally (if you are lucky) much higher.

2.4,. That means, placing a few hundred Ohms resistor to the ground from BOTH terminals drops common mode, without dampening differential mode significantly. It may drop it below the threshold the receiver cannot handle.

Let's stop here. As I said, you have your work really cut out for you. I do not, nor do you know, where the problem's origin will be found. But, I bet, you will be quite vigilant in the future about the introduction anything new.

Talk with you later.