KSTR
Major Contributor
Amir could test the intrinsic "cross-current susceptibility" of a device by impressing a current (with the 600Ohms output of the AP) between the relevant connectors of the device, doing a frequency sweep and measuring the DUTs output with shorted (or zero signal in case of DAC) input -- also for cables, testing shield impedance --.I would insist on a goal of "unmeasurable ground loop". One should be able to make it as good to stay below noise spectrum of the components used, i.e. under intrinsic noise/hum. It is possible, but usually only in case you have the whole chain under design control.
The device's own idle noise impressed between different ground points could also be measured and rated. For instance, the 8kHz packet noise (or other USB/internal activity noise) of DACs often appears as a voltage differential between the USB ground and the audio jack's ground, with the device fully floating (USB isolator required).
Last edited: