Why do digital cables sound different?


I have been talking to a few e-mail buddies and have a question that isn't being satisfactorily answered this far. So...I'm asking the experts on the forum to pitch in. This has probably been asked before but I can't find any references for it. Can someone explain why one DIGITAL cable (coaxial, BNC, etc.) can sound different than another? There are also similar claims for Toslink. In my mind, we're just trying to move bits from one place to another. Doesn't the digital stream get reconstituted and re-clocked on the receiving end anyway? Please enlighten me and maybe send along some URLs for my edification. Thanks, Dan
danielho
Gentlemen,

There are always Impedance mismatches.. the trick is
to minimize... usually we have tolerances at 10%...Even
for high speed stuff. Impedance mismatches do cause RF current problems BUT if terminated it will be minimized.
A 10% mismatch ( VSWR of 2.0 ) at RF frequencies ( which digital data is transferred via RF ) can begin to play MAJOR havoc with some circuitry. I know that text-book theory states otherwise, but you have to realize that much of that theory was based on tube circuitry that made use of matchingn networks and tank circuits. Most SS gear does not have this type of circuitry, so loading characteristics become far more critical.

The severity of how much the operation of a device would be affected obviously depends on the stability of the circuitry and whether or not the impedance mismatch was of a reactive or resistive nature. Not only would the sending unit have to deal with the reflections playing games with the output section, signal loss is increased. The resultant decline in detail is increased due to loading via the "back-pressure" within the transmission line. This is not to mention that line loss increases as VSWR increases, compounding the factors involved drastically. As such, ANY type of vswr / signal reflections tend to start a very drastic downward spiral effect once they come into play. This is FAR more apparent with SS gear, but then again, i don't know of any transport or CD player that uses a tubed digital output section.

As such, i would think that all of my experience with RF loading characteristics and impedance mismatches would directly apply to digital data transfer. I would suspect that the output section of most transports / cd players would not be very stable due to lack of a buffer circuit or complex impedance matching network. Therefore, ANY gains in terms of increased power transfer / minimization of reflections would be of multi-fold benefit for the aforementioned reasons. I have not verified this personally, but see no reason why this would differ from any other RF based SS circuit loading up. I really do want to start measuring and experimenting in these areas, but just have not had the time to do so. Sean
>
Sean,

I can tell you there is an 80-90% chance that the
last reply you sent went thru a OC192 backbone router that is running at 10ghz and the 10% mismatch rule applied to that product too and it works real good. We have to allow for material caused mismatches.. PCB fabrication is a major
contributor as is cable construction.

Sorry if im overstating the odvious BUT Its a fact of life...mismatches will always be there just like taxes.
Taxes are not "mandatory", nor are "mismatches" and "reflections". The more that one becomes familiar with and "works" the system, the less of each they may have to deal with. If one is really good and / or dedicated, it is possible to reduce any of the above to a point of negligence. Just because they are there does not mean that they should be "accepted".

Such is the case for signal reflection and vswr within a system. Why settle for 90% efficiency when one can have 99 - 100% ? The only difference is how much effort and money one is willing to invest to achieve that 10% improvement? When dealing with mass manufactured products, it is probably good enough. When dealing with small quantities of "high tech" products ( such as "high end" audio ), that is too much "waste" as far as i'm concerned. "Settling for" mediocrity or what is "good enough" never made me happy. Sean
>
Sean,

I need to talk to your tax guy :-)

I agree with you 100% its all about the benjamins$$$
I never said i like mismatches... Why do you think i
spent big bucks on my cables... But there are still
other problems on the chain and unless you youself
are designing AND fabricating the product including
the transistors you are at the mercy of ACCEPTABLE
tolerances. Welcome to the real world.. I dont like
it either BUT its reality.

Maybe you can explain to me how to get better impedance
control than what the fab houses can fabricate the product
for? In other words we design these "low tech" 48 layer
10 thousand connection PCB's running at 10ghz and 50 ohms dead nuts on... Its like having 10 thousand interconnects in a 5 sq foot area... What are we gonna do have kimber cable wire us up something?
We are forced to rely on material and fabrication tolerances. There are only two board houses IN THE WORLD that can fab that PCB for that"low tech" backbone router.

How would you do it?