|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
205.175.226.34
In Reply to: RE: Belden posted by Scholl on March 14, 2015 at 07:27:51
Belden attributed the difference in CMRR to certain colors being microscopicly different in diameter, thus affecting the mutual coupling in the twist.The differnce was on the order of 2dB or so (IIRC), and the best they could twist the pair at that time was on the order of 66dB.
I don't generally freak out about balance/unbalanced, it doesn't matter in practice. Unbalanced with an input transformer outperforms electronically balanced, period.
Edits: 03/14/15Follow Ups:
In the 90s I worked in a large data center were we ran about a mile of belden data twist a month terminating most of it by hand. Some of the pairs seemed twisted tighter than others. Some guys didn't pay attention to the pairing but it still worked.
testers measure unbelievable small differences. Even tensioning of wire bundles can affect impedances and the newer machines can actually pinpoint where the issue is. Sharp bends likewise affect signal transfer.
Much of this has become obsolete with the cheaper computer chips. The computers can correct the signal coming through the wire quite accurately, so that superior materials and construction is not as important as it used to be. Just send critical test signals down the cable and then the computer will correct for any anomalies
Just curious, do you recall the difference in diameters?
2 Db is not necessarily a small difference, particularly over longer distances.
IIRC, insulation is pulled over conductors by running them through a vat and through an over sized die. The size of the die regulates the diameter of the insulation as well as centers it.
Of course, a molten insulation means that a variation of insulation thickness is bound to occur, since the molten insulation has to be maintained within a critical range in order to flow smoothly. One would think that any unevenness would manifest in all colors, though.
In my experimentation, I have noticed, particularly in using a thermal stripper, the properties of different colored insulation is quite different from each other. Apparently the color dye lots has a greater affect than most of us would realize. Indeed if the insulation is stiffer, twisting would be more difficult.
I recall reading somewhere that insulation is usually extruded over the wire as the wire is pulled thru a die with the nominal diameter of the insulation. Because changing colors within one extruder is messy and wasteful, an extruder/die combination is devoted to one color or small group of colors. Insulation thickness variation results from the use of different dies for each color and the mfg variation inherent in the production of the dies. IOW, the insulation thickness variation is not directly related to color but is a result of production process variation in turn resulting from the decision to minimize color changes in the production equipment.
Looked for a link that references this but couldn't find it.
With today's capabilities, one would suspect fairly tight tolerances. You know CNC with laser cutting heads.
Insulation die tolerances should at least match the actual wire dies.
Still consider stranded copper. The distance between conductors will or should vary considerably. This would particularly be true with mixed gauge conductors (Kimber, Cardas).
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: