Oh man - wrote more than I expected:
I have a nagging question for all you C5D owners who have connected your C5D between your computer and a power amp (or other grounded audio device; not headphones):
Do you experience ground loop interference?
(by the way, I had JDS Labs convert my C5D's mini jack analogue input to a line level output from the DAC section of the C5D).
Unfortunately, I experience very annoying ground loop noise between my computer and power amp when connecting either the C5D's line out or HP out to my power amp. Just as much interference as when I send audio from my computer's headphone output to the power amp. Normally, this would not surprise me because USB power is notoriously noisy, and there is always a potential for ground loop between a computer's USB ground and a power amp.
But, the C5D is supposed to be galvanically isolated via an ADuM3160 chip (drawing below; isolation on all USB pins). In other words, there should be no (or drastically reduced) ground loop because the ADuM3160 physically disconnects all pins (power, ground, & data) from the computer's USB out and the C5D's input to the DAC. In fact, I notice the same ADuM#### family of USB isolation chips is used in many of the USB isolation devices marketed as ground noise killers, and sensitive USB device protectors.
I have spent the past few days researching different methods of of fixing my ground loop interference between my computer and power amp (C5D in between).
- Quick note: I understand there are preventative measures to avoid ground loops; I am grounding to the same outlet, correct polarity, interconnect cables have adequate ground conductivity. My ground loop is parasitic between my computer and power amp.
I am only interested in implementing physically separate grounds between my USB interface & power amp. The 3 most credible methods of isolating my grounds I've found are
- supplying clean 5v dc power to the C5D, e.g. via regulated linear transformer powered USB hub;
- using a USB isolator - ironically most using the same ADuM family of chips as in the C5D;
- using a line level audio transformer between my C5D and power amp.
All 3 methods separate the ground between my computer, C5D, or power amp, in order to eliminate ground loop interference. Each has its pros and cons (I would probably choose a quality audio input transformer because I have a few different unrelated ground isolation uses for it). The only thing I can't wrap my head around is the C5D (should) already have galvanic isolation between the USB input and rest of the device. Which makes me pause before I spend additional dollars on similar devices. I expected the galvanic isolation of my C5D to at least attenuate ground noise compared to connecting the headphone out of my computer to the power amp, but the noise is just as bad either way.
Other sanity checks:
- Above ground loop noise is present both when C5D is in charge or battery mode
- Nexus 5 > C5D > power amp: absolutely no noise (as expected; only 1 ground via power amp)
- Plugged in laptop > C5D > power amp: absolutely no noise
- Computer > C5D > headphones: absolutely no noise (as expected; only 1 ground via computer)
- Computer TOSLINK > receiver: absolutely no noise (TOSLINK = galvanic isolation / ground isolation; physically separate grounds)
- Computer TOSLINK > receiver > power amp: small amount of noise (slight ground interference between receiver & power amp)
Based on my sanity checks above, I can only conclude my C5D is not at all galvanically isolated. It seems to be merrily sharing ground between my computer and power amp. I lent my multimeter out a couple weeks back thinking I wouldn't need it any time soon (D'oh!). I will check for continuity between input & output grounds on C5D when I get it back. Can someone please help by reporting their experience with ground loop interference (or lack of) specifically when connecting their computer > C5D > grounded audio device (power amp, powered monitors, receiver, etc)?
Note: aside from the above, I LOVE the C5D. If it never claimed to have galvanic isolation via the ADuM3160, I would have no confusion as to why I am hearing buzzing and interference. But since it does claim so, I am a little perplexed. Am I misunderstanding the fundamental implications of C5D's galvanic isolation? I think not?
Drawing of the ADuM3160:
Edited by pixelcity - 5/30/14 at 9:47pm