And I want your opinion on this, as guys here are mostly into audio compared to RPi forum
As I’ve described on forum, this is my thought, do you think I’m right?
Problem is this isn’t interference made from static, electricity. Nope. I know how that sounds.
For example, if I’d use Lenovo’s 45W or RPi 5 USB-C charger (tho with it I get noise from grid/static) and set the JACK to sample rate 96kHz, buffer size to for example 256 and number of periods to 3 everything would be fine.
And when I use the DELL’s 65W charger which doesn’t have feritte on it, everything works even with 96kHz, 64 samples and number of periods set to 2!
USB audiointerface is RODE’s RODECASTER DUO, so yes, fairly known brand
So, not really sure what’s going on.
The only thing that might be causing this is that not enough power is given on starting the JACK and Patchbox’s module MODEP and then X-RUNS start and that’s why I get “metal”/robotic sound with other chargers.
Yes, I’ve got the DELL’s new charger today, 65W, and everything works great!
So, if anyone will work with audio, crucial real-time thing, as I’m also using a RT kernel for this purpose, be sure to buy DELL’s 65W charger!
The common noise on audio lines caused by PSUs is a high frequency whine, usually 30db down or more… quiet, but audible enough to be annoying. This problem isn’t from “static” or from “the grid” - it is from the high frequency shifting going on in the PSU itself, and it is leaking into the audio subsystem.
Another common problem is with audio devices that take power from a device they are also exchanging data with. Problem here is that the computer side rarely takes any pains to keep digital noise (again, high frequency) from leaking into the power pins in USB. And audio devices that are saving money by being powered from USB don’t generally do the isolation either. And while your audio interface has it’s own power… who knows how well they’ve kept the digital and audio grounds isolated.
Ground loops introduce another form of noise… but it is also very particular: frequencies of the line rate (50Hz or 60Hz) and it’s harmonics. Grounding the power supply may improve this, or may make it worse! Depends quite a bit on the other equipment and how the power supply is made.
BUT - none of this seems like what you’re hearing. Because if it sounds just like x-runs - it’s probably x-runs (or any similar form of buffer drops). That sound is very particular, and you don’t get it from the issues above.
So - I suspect what is going on is that the 45W charger isn’t supply enough power, and causing the RPi to drop clock speed at times. And thus the buffer under-run sound.
Here’s a simple test: Drop your Jack configuration to 48kHz sample rate. If the problem goes away - it had nothing to do with PSU design, electrical interference, or grounding issues.
We’ve previously seen some USB chargers impacting sound quality, such as Apple USB phone chargers. We’ve always recommended using the official Raspberry Pi’s power supplies, they work just fine.
Another idea of what may be going wrong with a high power USB C power supply - as the USB C device is supposed to signal the power supplier about its power needs, there might be some issue preventing them from proper negotiation, like it happened with Raspberry Pi 4 before: Raspberry Pi admits to faulty USB-C design on the Pi 4 - Ars Technica - it still worked just fine with their own power supplies.
Hi @Giedrius , as I’ve described, even RPi 5 original charger is causing this.
Another problem with chargers like RPi’s is that they don’t have grounding (for EU), so that’s causing another interference with audio other then the problem I’ve described with X-RUNS.
But yes, I’m also thinking that not enough power is delivered momentary with some chargers when MODEP boots up, and that causes X-RUNS that doesn’t know how to recover.
But yes, I’m using now DELL’s 65W charger and everything works perfectly!
