Welcome! Need support, you got it. Or share your ideas and experiences.
Return to Planar Speaker Asylum
69.255.242.123
I have Maggie 1.6s driven by a Bryston 4bST. I am thinking of making Cat 5e based DIY speaker wirehttp://www.audioholics.com/techtips/setup/interconnects/DIYSpeakerCablesp1.html
What will be better better choice - wire with lowest inductance or lowest capacitance? I will make equivalent 12 guage wire since my run length is short - approx 8 feet.
Follow Ups:
another problem with cat5 is each pair is a different length. cut a 1 foot length of cable then untwist and measure the wires. i think you will find quite a difference from the logest to the shortest.
Two questions for Hemholtz:1. For a 10ft or so length would the difference in lengths be significant?
2. Can the relative length of the +ve and -ve be effectively equalized by choosing the pairs accordingly?
Low inductance is what you are trying to achieve for best sound. However, you don't want to do it at the expense of excessive capacitance. I have tried CAT5 cables (but not on my Maggies), and they sound nice, but definitely not neutral. They were smooth, but a bit dark in the microdynamics (although I didn't have the teflon variety -- I think that could have been important).I think that high capacitance speaker cable is like playing with fire. It can sound great until you go too far, and start changing the sound due to the interaction of the high capacitance with the amplifier. Or worse yet, make the output unstable.
Well designed cables have geometries that minimize inductance without raising capacitance too high. These include quad-star geometries and such. You can do okay with twisted pair designs, but my feeling is that the thin wire in CAT5 requires too may pairs to get a low resistance, and this in turn results in too high a capacitance to be sure of a neutral sound.
I wouldn't use the 27-pair or 18-pair CAT5 recipies. I think the capacitance is too much, unless you have short runs (like 8 feet or less).
Lowest capacitance is crucial if your amp cannot cope with a high-cap cable (however, for only 8' I doubt the total capacitance will be a problem).Otherwise, go for the lowest L possible as it's inductance which acts to oppose current flow in the speaker cable.
12g total wire thickness will give you nice low R.
With Cat5, there's another choice apart from the 1-5 alternatives used in the article (and much easier to make than braiding all the individual wires, as some people do) ... use 4 runs of Cat5 cable per speaker (so you have 16 wires for '+' & 16 for '-') but use all 16 of the solid-colour wires for '+' & all the striped wires for '-'.
Within the Cat5 jacket each solid-colour wire is twisted with the corresponding striped wire - so you get maximum C and minimum L! :-))
Oh and I would suggest using Cat5 cable which:
a) is solid core, not stranded, and
b) has teflon insulation round the wires (Belden 1585a) ... or even teflon as the jacket as well (1585lc).Regards,
Andy thanks - what's the advantage of solid core over stranded?To simplify - I was thinking of twisting the four runs.
Yeah, twisting the four runs would be fine ... as you are using each twisted pair as '+' & '-', you've already got the tightest coupling you can get between the wires ... so how you "lay" the 4 jackets will not make any difference to L or C.And re. solid-core vs. stranded ... that's just an obsession I have! :-))
Regards,
This post is made possible by the generous support of people like you and our sponsors: