[johnjen]

I dunno…
It always kinda struck me as fairly straight forward.
 
There simply couldn't be ANY metal wire connecting the 2 'devices' together in any way in order to claim true galvanic isolation.
I mean that is what "galvanic" is all about right?
The minute voltage differential between to different metals when they touch.
And if they don't touch then true Galvanic Isolation has been attained…
 
At least that's what works for me.
And when I added a FMC (Fibre Media Converter) between my dac and my computer (along with an AOIP box in the middle) there was a noticeable change and in the ways I define as 'Better' at that.
And all it does is electrically isolate my dac (with an RN3 in the middle) from my Mac, via an optical link.
Oh, and it cost ≈$135 for this experiment.
 
Was it a Huge change?
No, but it is a ≈15-20% improvement (just above the ToP (Threshold of Perceptability)), and when added to all the other similar (10-25%) improvements, well, it all adds up after a while.
 
JJ

 

[boushh]

...  
There simply couldn't be ANY metal wire connecting the 2 'devices' together in any way in order to claim true galvanic isolation.
I mean that is what "galvanic" is all about right?
The minute voltage differential between to different metals when they touch.
And if they don't touch then true Galvanic Isolation has been attained…
 
...

 
 
That is what transformers are for
 
Simply connect the output (cable) of device A to one Side anf connect the other side of the transformer to the input of the second device. You could also use an opto-isolator (chip) or in your case an external opto-isolator.

 

[Baldr]

   
 
That is what transformers are for
 
Simply connect the output (cable) of device A to one Side anf connect the other side of the transformer to the input of the second device. You could also use an opto-isolator (chip) or in your case an external opto-isolator.

Amen.  "Galvanic Isoation" needs qualifiers and additional descriptors.  "Transformer coupled" is specific.

 

[boushh]

I'did not say it's easy
Finding transformers for the right frequency/bandwidth/resistance is a science in itself and mostly its expensive, but it should be possible.

 

[45longcolt]

A Schiit Shed?

No, a Schiit Sched has a crescent moon on the door...

 

[Oveja Negra]

  No, a Schiit Sched has a crescent moon on the door...

Also, the door should be positioned in the opposite side

 

[johnjen]

   
 
That is what transformers are for
 
Simply connect the output (cable) of device A to one Side anf connect the other side of the transformer to the input of the second device. You could also use an opto-isolator (chip) or in your case an external opto-isolator.

Yeah transformers can do this and ethernet uses them for just this purpose.
But there still is wire that connects one end to the other.
 
With no wire there is NO chance of galvanic interaction, at all.
 
And using my example of inserting a fibre link in an ethernet run, which results in an audible change, tells me that it may not be the transformer coupled signal wires themselves, but perhaps the wires used for the ground and/or the power connections, as the pathway for these issues.
 
And really I doubt that galvanic action, alone is the only, let alone even the primary form of signal contamination.
But that a wire directly connects one device to another in one way or another seems to be a key factor here.
 
It also seems odd that toslink - optical SPDIF doesn't fair better than it does.
Perhaps it needs better transceivers or less susceptibility to noise/jitter or some such, because it potentially should be 'better', except it isn't, at least in my experience anyway.
 
JJ

 

[KoshNaranek]

I suspect it is the poor quality Toslink connectors that are the problem. AT&T connectors have higher bandwidth.

 

[Letmebefrank]

I suspect it is the poor quality Toslink connectors that are the problem. AT&T connectors have higher bandwidth.

 
My Asus Xonar DX sound card has a high bandwidth optical transmitter, so no problems with toslink here. I do use USB more often just for the ease of use though.
 
Furman PST-8D got rid of all electrical noise in my desktop system, I highly recommend it for anyone experiencing electrical noise in their setups.

 

[johnjen]

I suspect it is the poor quality Toslink connectors that are the problem. AT&T connectors have higher bandwidth.

Perhaps that is it, but the Lifatec Toslink cable I have seems to be a rather precise cable and it faired no better than a $20 radiocrap cable, which suggests to me that it is the interference within the optical stream or perhaps it’s a signal timing issue, or noise in the optical couplers, or some such.
 
And that a gigabit FMC does manage to pass much greater bandwidth and do so quite easily it seems.
And perhaps it’s the duplex function, since as the FMC uses 2 pieces of fibre instead of one.
 
In any event the complete electrical isolation of a major noise source (my Mac) from my dac seems to make a contribution to 'Better' SQ.
 
JJ

 

[boushh]

Problem with Toslink is imho it has not advanced over the years, because it is "working" for the masses. If you look at the developement process of laser and fiber used for "ethernet", you can see what is possible. But that was payed because  IT needed faster connections.
 
Simply have a look at an Toslink "laser" and a laser for Ethernet. Toslink is polymere optical fibers and that simply means limited bandwith (have worked on an project for using the material for networking, biggest problem was good receiver/sender "chips").
 
As mentioned above the Toslink connector also is a big limit. For really better quality, Toslink needs an redesign. (imho that wont happen, because it "works")
 
Have a look at this picture
 
https://en.wikipedia.org/wiki/File:Optical_fiber_types.svg
 
There you can see, the smaler an optical fiber is, the better is the transmission. Because you have lesser reflections. Toslink uses fiber more than 1mm wide, that simply limits its possibilities to "relativ" low bandwidth. "Real" optical fiber is smaller by the factor of >100.

 

[KoshNaranek]

Problem with Toslink is imho it has not advanced over the years, because it is "working" for the masses. If you look at the developement process of laser and fiber used for "ethernet", you can see what is possible. But that was payed because  IT needed faster connections.

Simply have a look at an Toslink "laser" and a laser for Ethernet. Toslink is polymere optical fibers and that simply means limited bandwith (have worked on an project for using the material for networking, biggest problem was good receiver/sender "chips").

As mentioned above the Toslink connector also is a big limit. For really better quality, Toslink needs an redesign. (imho that wont happen, because it "works")

Have a look at this picture

https://en.wikipedia.org/wiki/File:Optical_fiber_types.svg

There you can see, the smaler an optical fiber is, the better is the transmission. Because you have lesser reflections. Toslink uses fiber more than 1mm wide, that simply limits its possibilities to "relativ" low bandwidth. "Real" optical fiber is smaller by the factor of >100.

Thank you, that was very informative. I also agree that there will most likely not be a change from Toslink, though I hope a switch to optical ethernet connectors and fibre is in the future.

 

[Clemmaster]

It's called ST Fiber. According to Jason, Mike was the first one to use this technology in DACs in the late 80s.
Too bad it was dropped by most manufacturers.

 

[Baldr]

  It's called ST Fiber. According to Jason, Mike was the first one to use this technology in DACs in the late 80s.
Too bad it was dropped by most manufacturers.

AT&T made the xmtrs and rcvrs.  The fibre required hand lapping and polishing at the time and was slow to make.

 

[johnjen]

  AT&T made the xmtrs and rcvrs.  The fibre required hand lapping and polishing at the time and was slow to make.

  It's called ST Fiber. According to Jason, Mike was the first one to use this technology in DACs in the late 80s.
Too bad it was dropped by most manufacturers.

 
A friend of mine has a Theta dac with the optical connection.
Unfortunately he no longer has the optical cable so he's limited to SPDIF which isn't in the same league.
 
And it was a tiny diameter cable which is probably why it was so highly rated at the time.
 
JJ