If you don't have the means to demo this setup first, I would pass. Most products like this adhere to a 75 ohm standard. I would choose something other than the V-link. Keep in mind, though, I'm not an expert on this. 50 ohms may be fine. I would just err on the side of caution.
10 responses Add your response
Preferably neither. Frankly, I wouldn't use those two components together. And as far as I am aware, a 50 ohm output impedance is non-compliant with the S/PDIF standards.
I see that the manual for the V-link model you are probably referring to does indeed indicate a 50 ohm output impedance, which is very surprising to me. And a quick Google search reveals a number of people expressing concern about the resulting impedance mismatch, along with some not very conclusive indications that it MIGHT be a specification error.
If there is no other choice my instinct would be to go with 75 ohms, so that the match at the receiving end is as good as possible, and given the possibility that the 50 ohm spec might simply be wrong. But which is best could very conceivably vary unpredictably depending on the length of the cable and the unspecified risetimes and falltimes of the V-link's output signal. Preferably, as I say, I would avoid using that component altogether.
Edit: My response was composed before seeing ZD's. Great minds think alike :-)
Thanks guys - that was my thinking also
However, I do have these two components and they work extremely well together.
Perhaps it's a typo on the v-link spec? - I'll have to contact them
Perhaps it's the cables I'm using - which are far from standard 75 ohm cable specs
If you're interested in the cables I'm using I've posted details on Cheap Tweaks
02-19-15: Williewonka I don't know why this works - but it does...
It would be interesting to actually know what the scoop really is :-)
I would use 75 ohm cable, since reflection from the other end is possibly worse, as Al stated, but also because output RCA connector is not likely 75 ohm anyway.
I would use very short cable - as short as possible. Wire becomes transmission line when propagation time is longer than 1/8 of transition time. Assuming typical 25ns output transitions cable should be limited to propagation time of about 3ns, equivalent to about 2 feet (5ns/m). Since internal connections on both sides (including PCB traces) count I would limit it further to 1foot max. Otherwise I would use 1.5-2 m cable to ensure that reflections from the other end miss originating transition.
Here's the response from MF...
most digital parts I know of are 50 ohms (or 100 ohms balanced) to preserve square wave edges. That said at these relatively lowish frequencies (low MHz) such a mismatch is unlikely to cause problems unless going into extremely long cables.
I am extremely happy with mine - my Bifrost has never sounded so good.
And remember - Don't shoot the messenger :-)
Thank you kindly, noble messenger :-)
While Mr. Bingham is correct that a lot of digital signal applications involve 50 ohm parts and components and 50 ohm cables, that is irrelevant because S/PDIF is specified as 75 ohms.
Also, it seems that he doesn't recognize what the concern is. The concern is that signal reflections caused by the impedance mismatch will result in distortion of the signal waveform during transitions between its lower voltage and higher voltage states, and vice versa. If that distortion happens to occur around the mid-point of those transitions, the result is likely to be increased jitter (random timing fluctuations) at the point of D/A conversion within the DAC.
That is likeliest to be a problem if the length of the cable is in between being very short and fairly long, as Kijanki explained earlier. However, whether and to what degree a problem may occur is all somewhat unpredictable, because it depends on the happenstance of the unspecified transition times (risetimes and falltimes) of the particular signal, on the propagation velocity as well as the length of the particular cable, on the design and especially the jitter rejection capability of the particular DAC, on the degree of impedance mismatch between the cable and BOTH of the components it is connecting (there will inevitably be some degree of mismatch even when connecting a 75 ohm cable to a 75 ohm component, causing some degree of re-reflection of signal reflections caused by the 50 to 75 ohm mismatch at the other end, in turn resulting in multiple back and forth re-reflections), etc.
See this paper for further elaboration.
The bottom line, IMO: In any given setup there may or may not be a sonic issue resulting from this mismatch, but it cannot be considered to be a recommended way of assembling a system.