Stefan Brüns wrote: > On Samstag, 23. Januar 2021 18:52:50 CET Per Jessen wrote: >> Per Jessen wrote: >> > I have hooked up a headset to my Nanopi Neo Air, but the microphone >> > quality is severely impacted by electrical hum. I have tested it >> > and determined it is due to a power supply with no ground, i.e. the >> > typical 5VDC 2A with a two-prong Europlug. Using one of those >> > small power supplies in metal cabinets, with a ground wire, and the >> > hum is gone. >> > >> > The latter is not really an option, so what do people do instead? >> >> I have discussed it on EE stackexchange, and there is no doubt it is >> a ground loop (Brummschleife) issue > > There is no doubt this explanation is utterly wrong. You can't have a > ground loop without (multiple!) ground connections.
I don't even have a single one :-) * connect an oscilloscope ground to the ARM board, the hum goes away. * touch the metal part of the microphone with my finger, the hum roughly doubles. * use a small 5vdc SMPS with mains ground, no hum. > There are other possible explanations: > - badly shielded cables I have tried using a shielded balanced pair cable to the microphone, but the microphone capsule is sat in an old analog telephone receiver with an unshielded spiral cable anyway ... > - conducted noise from some other part of the equipment, e.g. monitor, > external amplifiers (AV Receiver), ... No monitors, no amplifiers - only the ARM board, the microphone, the wifi aerial and the powersupply. > - bad (unsufficient) power supply I have tried four different ones - Goobay 2A, another one bought specifically for a Raspi (2.1A), a large Ikea multi-charger 3400mA and even an Iqos charger. Same problem with all of them. Use one with ground, and the problem disappears. > You should try disconnecting all external equipment save the headset > and then verify the noise persists. That is what I have :-) I record a sample with 'arecord', then copy it to another box to listen to it. -- Per Jessen, Zürich (2.6°C)
