This is a classic “Ground Loop”. In RCA type audio connections the shield (outside metal sheath) is assumed to be “Ground” (zero volts). The audio signal played through the following electronics is added to this zero for final rendering. If, for some reason, the grounds are different for each box in the system, which should be used as the zero reference for the final output?

We could have a similar issue discussing “how high is that picture?”. If you measure from the room floor and I measure from the sidewalk outside the building we will disagree on the measurement, but the picture has not moved.

The optical connection does not have a metal component, therefore there cannot be a system Ground Loop. (which requires a metal connection between units). You can also have local Ground Loops within each component. An analog of this would be you measuring the picture height from the top of the carpet while I measure from the wooden floor under the carpet. Ground Loops both internal and external are the bane of audio engineers.

In your case you may find that disconnecting the coax cable feed from the street will minimize or eliminate the Ground Loop. (Caused by a ground connection down the block at another house,)

Anyway, using the optical connection is a quick and easy potential cure for a system Ground Loop.

There is another thread that discuss the same components having an issue with very low line output level … so that there is almost no sound to amplify. I have a ZP80 with that problem, and I intend to check if the optical output may be a workaround for this as well. Stay tuned :-)

ZP80 & ZP90 have a history of being easy victims of Ground Loops.

Yes !!! … that worked too. My ZP80 phono output level is materially gone. But, the optical output still works. So, I have saved a ZP90 and a ZP80 this week, some money and some CO2.