I've got a question on the best way to ground an APOX passive volume control. I'll say what the signal chain is and what I am considering doing, and perhaps someone will have some feedback.
I'll be putting the APOX volume board in a separate chassis, galvanically isolated from the controller.
Source: floating (double-insulated), balanced
APOX passive attenuator: floating (double-insulated), balanced 5K
Active crossover: floating (battery), balanced
Amps: Mains grounded case, balanced
I plan on putting four attenuator channels (L, R, C, Sub) in one box. I will use a DPST switch to allow choice of input or output ground as shield. (mainly for future flexibility)
My question is, how should the PCB shield ground and chassis grounds be used? I am wary of simply connecting these to the cable shields because of "pin 1" problems. Yet if I do not do so, they will floating...unless I connect them to the mains ground. Also, since there are two seperate pairs of channels a choice will need to be made about which one (if either) to connect to chassis ground.
Next question would be: it is typical to AC-couple the "floating/disconnected" end of a shield cable when there is no DC ground path. But in this case you would essentially be decoupling noise into your signal ground...is that a good idea?
I have a Tek mainframe and can measure very low levels of noise, but I'd love some feedback so as to reduce the number of variables. I've also noticed that noise problems can vary hugely day-to-day, so test results are sometimes unreliable.
I'll be putting the APOX volume board in a separate chassis, galvanically isolated from the controller.
Source: floating (double-insulated), balanced
APOX passive attenuator: floating (double-insulated), balanced 5K
Active crossover: floating (battery), balanced
Amps: Mains grounded case, balanced
I plan on putting four attenuator channels (L, R, C, Sub) in one box. I will use a DPST switch to allow choice of input or output ground as shield. (mainly for future flexibility)
My question is, how should the PCB shield ground and chassis grounds be used? I am wary of simply connecting these to the cable shields because of "pin 1" problems. Yet if I do not do so, they will floating...unless I connect them to the mains ground. Also, since there are two seperate pairs of channels a choice will need to be made about which one (if either) to connect to chassis ground.
Next question would be: it is typical to AC-couple the "floating/disconnected" end of a shield cable when there is no DC ground path. But in this case you would essentially be decoupling noise into your signal ground...is that a good idea?
I have a Tek mainframe and can measure very low levels of noise, but I'd love some feedback so as to reduce the number of variables. I've also noticed that noise problems can vary hugely day-to-day, so test results are sometimes unreliable.