I don't know if anyone has talked about what these are doing, but, I will and I'm sorry if I'm repeating information.hellomrbike wrote:Would a ground loop noise isolator on the 3.5mm output side also do the trick? e.g.Peter - IK Multimedia wrote:Same here. And I have an Ebtech unit like that in my home studio since one circuit definitely had hum and it was easier than getting someone to work on the wiring (I rent a house). Works like a charm for years now.Scotty wrote:USB noise in my experience cannot be solved by rf solutions like ferrite clamps and shielded cables. The USB noise is usually coming from within the computer (both hum and sometimes even hard drive chatter) and not being picked up by radio frequencies. There is no harm in trying that solution - it won't make it worse but my experience with this issue is a USB isolator is needed.
I had USB hum using specific laptops in the past and I was only able to solve that issue with the Hum Eliminator gadget. This is not a USB isolator. The hum was being picked up by the audio interface and could be heard on speakers and headphones.
https://www.amazon.com/Ebtech-Hum-Groun ... T4ZADSWQ4X
https://www.amazon.ca/AUKEY-Ground-Isol ... B01M1671I0
My suspicion is that this is equivalent to the modified TRS cable, but I'm not quite confident. Since USB isolators cost around 3x more than an audio ground loop isolator, I'd like to make sure I'm exhausting the simpler options before going with the more expensive (albeit probably most correct) choice.
These isolators, e.g., ebtech, are a fancy ground lift. There is a resistor in parallel with two back to back diodes. Normally there isn't enough current flowing at sufficient voltage through the ground conductor to cause conduction through either diode. In the case of a fault, which is the reason for the ground connector in the first place, the larger voltage would turn on the diodes causing current to flow. Hopefully it's not so much that it would take the diodes out, thus breaking the ground connection.
The low voltage USB noise won't cause the diodes to conduct. The parallel resistance allows the ground to remain connected, but with reduced current, breaking the loop. Should a fault occur, the diodes turn on. They should be of high enough current such that they will allow a fault condition to trigger the breaker in your house.
So, in terms of hum, they do the same thing as a cheap ground lift plug that you can get at the grocery store, that you should never use, even to test to see if that's the solution, because DANGER DANGER DANGER!
The thing is, there is some question as to whether the ebtech boxes pass code as, from what I've read, they violate the principle that the ground should be connected directly to the ground connection. They also seem quite small so I'm not sure how large the diode is in there. Finally, should a diode fail open, the box provides no fault protection.
The trick is to break the loop. Often DI boxes (with transformers) will also solve this problem. It depends on how many loops you have.
On Edit: Minor corrections to current vs voltage so as not to mislead. Silicon diodes need about 0.6V to conduct. The USB voltage is much much lower which is why a failure would cause the diodes to conduct and the USB noise does not.