Interconnect Directionality


Have I lost my mind? I swear that I am hearing differences in the direction I hook up my interconnect cables between my preamp and power amp. These are custom built solid core silver cables with Eichmann bullet plugs. There is no shield so this is not a case where one end of the cable’s shield is grounded and the other isn’t. 

There are four ways ways to hook them up:
Right: Forward. Left: Forward. 
Right: Backward. Left: Backward
Right: Forward. Left: Backward
Right: Backward. Left: Forward. 

There is no difference in construction between forward and backward, but here are my observations:

When they are hooked up forward/backward there appears to be more airy-ness and what appears to be a slight phase difference. When hooked up forward/forward or backward/backward, the image seems more precise like they are more in phase. The difference between forward/forward and backward/backward is that one seems to push the soundstage back a little bit while the other brings it towards you more. 

What could possibly cause this? Does it have something to do with the way the wire is constructed and how the grains are made while drawn through a die? Am I imagining this? Have I completely lost my mind?
mkgus
Post removed 




jea48
3,144 posts
10-22-2019 9:40am
"And where did I say in any of my responses to you anything to the contrary?" 

- mainly your statements were gobbly gook, so I just did my best with what I was working with.

"And that relates to wire directionality how? Are you saying because of your statement, flipping an analog or digital coax cable end for end can’t possibly make a difference?" 

"Again the shield of a coax is also the signal ground and is connected at both ends. Your above statement does not apply then does it?"

- Again, the outer conductor is not really a shield, not in the traditional sense. It does not "shield" the inner wire as most think, but the overall construction does reject noise, similar to how a twisted pair is not shielded but has inherent noise rejection due to the construction.

- Outside of a single ended shield connection, I am saying we are talking audio frequencies for analog signals, and either low frequencies, short lengths versus frequencies/edge speeds, and/or effectively time-non-dependent digital signals, such that a simple bulk-model, independent of direction is going to more than suffice. Even coax for 100GHz signals, where timing is truly critical, is bidirectional, because the impedance is controlled along the length. Transmission line effects don't come into play at audio frequencies. Skin effect (inaudible anyway), works the same in both directions. Short of intentionally adding unidirectional circuitry to the cable, direction is not going to matter for analog.  For digital, at least w.r.t. audio, slow edge speeds coupled with tolerable matching are such that any induced jitter is near nill, and coupled with the variation in the two directions is effectively 0, and in any competent DAC is going to suppress jitter by 10's of db in addition, if not buffered such that any jitter on the incoming signal has no impact on the output. 

Uh, we aren’t discussing shielded cables. Hel-loo! I’m afraid someone doesn’t yet know what directionality even means, whilst arguing so strenuously. I won’t mention any names. I suspect this is probably a case of can’t see the forest for the trees. Perhaps something more uh, disturbing. 😫
Uhmmmm your friend jea48 was, but if you aren't going to invest the time to read all the comments, what is the point of responding?   We are talking about the directionality of fuses and your erroneous claim that the fuse only impacts the current in one direction (take your pick  either positive or negative phase of the wave), not both directions .... which is simply not true.

geoffkait17,612 posts
10-22-2019 12:29pm
Uh, we aren’t discussing shielded cables. Hel-loo! I’m afraid someone doesn’t yet know what directionality even means. I won’t mention any names. I suspect this is probably just a case of can’t see the forest for the trees.

Post removed