(Quick answer: there is something funky with the SPDIF connection from a Roku... I wrote the below and then towards the end I changed to a different source, an OPPO player with its SPDIF out and all the problems you see described below are seemingly gone. This is with the same cable. What could be the source of the problems described that is due to the Roku? What on the SPDIF input could cause all the problems described below? Never mind that I want to use my Roku I want to understand where the crazy DC offsets could be coming from for future reference... Please do read the below and offer advice)

I have the original Buffalo v1.1 (firmware 1.0.3), Volumite 1.0. I am taking SCA/SDL from the board along a twisted pair of wires to the Volumite and I am taking GND/VCC for the Volumite from the GND/VD terminals on the Buffalo (also twisted). I am using SPDIF from a Roku as input (all dip switches are in the UP position and the SPDIF switch is in the ON position). The sound is beautiful and clear notwithstanding all the DC offset issues described below.

When adjusting volume I can sometimes hear a hum (not 60Hz; something higher). When I tweak the volume VERY carefully I can get to a position where the hum stays on and is reasonably loud (not faint; standard volume loud). It feels like there are some magic positions on the dial where this sound can be heard. Mostly one just wipes over it and one hears but a fraction of a moment of this sound. Turning off the DAC and turning back on I can sometimes still hear the sound (sometimes not). In other words, it does not seem to get cured by a forced reset... I moved the wires around since I read about EMI issues, but that seems to have no influence from what I can tell.

Since my adventure with the IVY I have taken to measuring the DC offset. That's the other surprise. The volumite control goes along with widely varying DC offset (this is both on the input to the IVY and on the SE as well as XLR outputs). With hitting pause on the Roku (signal is locked but there are no "sound" bits; only silence) I can see varying DC offsets depending on how I turn the volumite. There appears no logic to it. It can be as low as < 1mV and as high as 980mV(!!!). It's different in both channels with no rhyme or reason and it's just plain all over the map.

My current setup is still at the "breadboard" stage in that everything is mounted on a open piece of plywood and mains earth is not connected anywhere.

I noticed all of this when I noticed my loudspeaker cone moving in and out in slow swings as I adjusted volume (basically large DC attenuated by the amp input cap).

I disconnected the volumite completely (dangling wires on the Buffalo board). Ok this is crazy. I disconnected the input (automute light is on) and I see 6mV and 10mV DC offset on the SE outputs on the two channels. Then I connect again the paused Roku; I get a Lock and now I see about 8 resp. 10mV DC between the +/- terminals on the output of the DAC board (and input of the IVY) but on output of the IVY I see 980mV(!!!) on the SE on one channel and 860mV on the other. WTF? I turned the box on and off (with some moments in between) with the same result.

Ok. I am totally lost. What could be going on here?

(and this is when he tried this with the Oppo player connected and found all the above issues to have magically disappeared... WTF?!)

Dear Russ,

Yes. In the meantime I have plugged a different Roku (I have several) in and the same issue is there. I have also changed the cable with no change in behavior.

Here is the thing. It's SPDIF. The sound is clean and clear and beautiful. The digital bits are not corrupted. I am now thinking this is maybe a grounding issue. What else could it possibly be?

The bigger issue for me is that there is some equipment that will make the Buffalo DAC/IVY put out significant DC. That worries me and I desire to understand the fault. The Roku is a stock unit. No messing with anything and I have hooked it up to a good number of different commercial DACs and this is the first time I am having a problem.

Any ideas what I might look for to understand what is happening here?

Alternatively, any way to isolate the input to avoid this? The bits are clearly ok, so if I can maybe electrically isolate the source from the DAC I can get around this...

Peter

Dear Russ,

I finally got around to trying this. No cigar. The SPDIF signal is fine. I took a careful look with my scope. Nothing weird on there. I then decoupled with 100nF caps both wires (100nF in series with signal as well as ground). No difference. I analyzed the strange hum sound some more that I get in certain positions on the volumite (recall that for some positions of the volumite I get a hum on the output). It is a 92.69Hz triangle(!) wave. So no relation to multiples of 60Hz...

Yes, it has to do with the source, but there is nothing peculiar I can find about this source relative to others. The SPDIF is clean from what I can tell.

What else could I possibly try?

peter