The USB hub would just make the ground loop/inductive pickup larger.
Any ground loop current flowing through the analog audio signal ground (shield) causes a voltage drop across the shield's resistance (and inductance, if big enough/high enough in frequency) which acts directly at the input of the following analog stage. It in effect is wiggling the next stage's signal reference ground.
I've seen this lots of times in measurements (used to design and sell audio test measurement gear that used USB audio adaptors). Wish I'd save some plots, but they were always considered bugs to fix, not to cherish. Looked like spurious tones, probably from stuff going on inside the connected computer or it's power supply (laptop power supply blobs are often horrible, it seems). My device had an additional USB connected box (that handled analog switching, level calibration, mic bias and preamp, etc.), so that was another avenue for ground currents to run around. In later times we shipped with very short USB cables and 6 inch audio cables to minimize the likelihood of the problems. Audio exotica is of course worthless for this problem, though. Often running the laptop on batteries took care of the problem, sometimes moving power cords to a common outlet would help.
Any ground loop current flowing through the analog audio signal ground (shield) causes a voltage drop across the shield's resistance (and inductance, if big enough/high enough in frequency) which acts directly at the input of the following analog stage. It in effect is wiggling the next stage's signal reference ground.
I've seen this lots of times in measurements (used to design and sell audio test measurement gear that used USB audio adaptors). Wish I'd save some plots, but they were always considered bugs to fix, not to cherish. Looked like spurious tones, probably from stuff going on inside the connected computer or it's power supply (laptop power supply blobs are often horrible, it seems). My device had an additional USB connected box (that handled analog switching, level calibration, mic bias and preamp, etc.), so that was another avenue for ground currents to run around. In later times we shipped with very short USB cables and 6 inch audio cables to minimize the likelihood of the problems. Audio exotica is of course worthless for this problem, though. Often running the laptop on batteries took care of the problem, sometimes moving power cords to a common outlet would help.
Last edited:
I still put that in the realms of may rather than will.
Back to the ODAC which I quote because its cheap and measurements are available no PSU noise issues cos it was actually designed rather than boutiqued. It does however barf with one type of USB port. cheap powered hub fixes that. I was rather hoping the paranoia brigade would have done some research and found that, but they didn't.
The issue was low ESR caps on the host causing regulator oscillations. gnarly edge case but proves they exist and engineering can fix it rather than flooby.
Back to the ODAC which I quote because its cheap and measurements are available no PSU noise issues cos it was actually designed rather than boutiqued. It does however barf with one type of USB port. cheap powered hub fixes that. I was rather hoping the paranoia brigade would have done some research and found that, but they didn't.
The issue was low ESR caps on the host causing regulator oscillations. gnarly edge case but proves they exist and engineering can fix it rather than flooby.
Low frequency noise is rather more prone to spreading out, HF can be controlled by good engineering practices that are used in commercial stuff, ferites, pi filters all these are your friends in controlling interstage noise and interference, as I have said its one coin two sides... signal integrity/EMC
You are correct. I don't know what the hell I was thinking about. Maybe I need to stop posting when I'm tired.
I suppose maybe I was thinking about at the source, the reader. No error correction at the other end, as far as I know. There is some reclocking and clean up, but that's all. The HD-SDI video recievers I use everyday have adaptive EQ built into the chipset to condition the incoming signal. It's necessary to recover clock and date from such high bandwidth signals. Works great. Digital audio has nothing like that AFAIK.
Some of the cheap USB DACs I've used show a fair amount of noise and what looks like jitter. Better ones, like over $100, do not. I suspect the chipset much more than the cable.
FWIW, I used to have a very cheap, thin Toslink cable that sounded really awful. All the music was there, no clicks, pops or dropouts, but the sound was very sour and just plain wrong. Not a typical analog sounding defect. I kept it for years as a demonstration of the fact that digital can work but still be wrong.
Consider these three arrangements:
Situation 1: PC-> 10m corning glass usb cable -> yggdrasil DAC
Situation 2: PC-> 10m corning glass usb cable -> usb 2.0 powered hub ->Yggy
Situation 3: PC-> 15 foot micro center usb cable-> usb 2.0 powered hub ->Yggy
The glass usb cable has a built in receiver and transmitter.
some have described this as a 'single port usb hub that reclocks the signal'. I don't know how accurate that is, I am not an electrical engineer to actually have taken apart these things. My sense is that the noise in these three arrangements increases as you go down the list. But I don't know how jitter is affected and whether a 'de-crapifier' like the jitterbug plus REGEN would give a better signal.
I have a longish stretch between the pc and the audio stack so I need a longish usb connection. What are your thoughts?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.