Speaker wire construction

jazzerdave · « **on:** 9 Mar 2012, 08:25 pm »

I'm going to attempt to make some speaker wire out of 28 AWG silver wire. I have a little over 1,000 feet of bare silver and 1000 feet of ptfe tubing (24 AWG). I'm planning to use 30 8ft wires per run. I've made interconnects in the past with the same type of wire, but making a three-way or even 6-way braid is pretty easy to figure out. There seem to be so many different ways I could arrange the wires, but I can guarantee I won't be attempting a 30-way braid.

So, I could just twist them all in the same direction. I could make twisted pairs and then twist the pairs in the opposite direction, then twist those pairs in the original direction, and follow this pattern until I've got one cable. I could divide the conductors into groups of ten and then do a simple three-way braid with those. I would go on, but I think you get the point. I really don't know which way to go with this. Any advice would be appreciated.

Thanks.

JohnR · « **Reply #1 on:** 10 Mar 2012, 10:19 am »

3x3x3 would get you close enough.

You have my admiration if you manage to make them.

Speedskater · « **Reply #2 on:** 10 Mar 2012, 01:13 pm »

Would that be equal to something smaller than 16 AWG ?
That is 15 Send conductors and 15 Return conductors per speaker cable.

jazzerdave · « **Reply #3 on:** 10 Mar 2012, 04:28 pm »

Sorry I wasn't more clear Speedskater. I'm talking about 30 for each post. 30 wires for the signal and 30 for the return. What would you think about twisting 10 wires together and then braiding those groups?

srb · « **Reply #4 on:** 10 Mar 2012, 06:14 pm »

30 X 28ga = 13ga which sounds good. It would be difficult to theorize the outcome, but if I had a capacitance meter I would probably make a few 1 foot samples with different twist groups and braids and see which one had the lowest measured capacitance.

Steve

Speedskater · « **Reply #5 on:** 10 Mar 2012, 09:04 pm »

That's better, just a littler smaller than 13 AWG, but then again it's silver! You are going to intermingle the Send and Return strands aren't you? With everything else being equal, it's a trade-off between capacitance and inductance. More inductance then less capacitance and vice versa.

jazzerdave · « **Reply #6 on:** 12 Mar 2012, 12:09 am »

I don't really have enough of the pure silver wire to experiment, so I'm trying to get my ideas straight ahead of time. I will have plenty of other wire to practice the pattern I plan to use, but I would guess that the measurements I would get wouldn't necessarily transfer well (with different conductor and dielectric material).

I could make a braid like this:

using 10 twisted wires for each leg of the double-run three way braid (if someone has a better name for that, let me know).

cheap-Jack · « **Reply #7 on:** 12 Mar 2012, 04:28 pm »

Hi.

Quote from: jazzerdave on 9 Mar 2012, 08:25 pm

I'm going to attempt to make some speaker wire out of 28 AWG silver wire. I have a little over 1,000 feet of bare silver and 1000 feet of ptfe tubing (24 AWG). I'm planning to use 30 8ft wires per run.

I could divide the conductors into groups of ten and then do a simple three-way braid with those. Any advice would be appreciated.

Pure silver is the metal with highest electrical conductivity, lowest contact resistance, & highest thermal conductivity on earth. Excellent used for musical signals transmission used in interconnects & loudspeaker cables.

FYI, the 6 pairs of audio interconnects I am using for years now are ALL 4N purity AWG#22 solid silver conductors of German origin I DIYed. Sound superb!

Back to yr question:- The book shows solid wire of AWG#28=159 circular mils. For loudspeaker cable, I'd sugest AWG#12 (same gauge for my discrete bi-wiring loudspeakers). So #12=6,529 cir. mils. To build a #12 cable, you need 41 x#28 silver wire bundled together.

You can braid 3 wires to form one bundle. So #12 will need 13.66 bundles or 14 bundles.

Braiding will decrease the self inductance of each wire substantially. The disadvantage is twist braiding will increase the effective capacitance of the whole cable bigtime. Excessive capacitance & inductance of the loudspeaker cable can affect the stability of the driving amp. Be careful!

c-J

"If the facts don't fit the thoery, change the facts" quoted Albert Einstein.

jazzerdave · « **Reply #8 on:** 13 Mar 2012, 11:49 pm »

OK, whether wise or not, I've decided to add some more wire to the project. I'll end up somewhere between 10-12 AWG. I'm not quite sure yet. I'm going to acquire a full extra 1000 ft, but I may set some aside for interconnects rather than use all of it. The final determination may come down to whether or not it will remain flexible enough at larger gauges. I'm going to be making test runs with some bulk magnet wire I have.

Speedskater · « **Reply #9 on:** 14 Mar 2012, 12:43 am »

Remember that cable length is the bigger part of the equation. Not much difference between short 13/14AWG silver cables and long 12AWG copper cables.

jazzerdave · « **Reply #10 on:** 14 Mar 2012, 12:54 am »

The ultimate goal is 8ft cables (give or take).

cheap-Jack · « **Reply #11 on:** 14 Mar 2012, 02:15 pm »

Hi.

Quote from: jazzerdave on 14 Mar 2012, 12:54 am

The ultimate goal is 8ft cables (give or take).

Why 8ft run only, no longer or shorter? What "ultimate goal" is it?

c-J

jazzerdave · « **Reply #12 on:** 14 Mar 2012, 04:56 pm »

c-J - I believe you've read enirely too much into the phrasing "ultimate goal". The "ultimate goal" is just the length of cable I need. I could have said the end result or desired length. I also could have said "the ultimate goal of this project is to create a set of 8ft cables". I was responding to this from Speedskater:

Quote

Remember that cable length is the bigger part of the equation. Not much difference between short 13/14AWG silver cables and long 12AWG copper cables.

I was only stating the length of cables that I planned to make to get Speedskater's opinion on length vs AWG.

cheap-Jack · « **Reply #13 on:** 14 Mar 2012, 05:08 pm »

Quote from: jazzerdave on 14 Mar 2012, 04:56 pm

c-J - I believe you've read enirely too much into the phrasing "ultimate goal". The "ultimate goal" is just the length of cable I need. I could have said the end result or desired length. I also could have said "the ultimate goal of this project is to create a set of 8ft cables". I was responding to this from Speedskater:

I was only stating the length of cables that I planned to make to get Speedskater's opinion on length vs AWG.

Yes, shorter the better for any audio interconnects, including loudspeaker cables.

c-J

jazzerdave · « **Reply #14 on:** 14 Mar 2012, 05:29 pm »

If I could make it work, I'd be making shorter cables, but I've got some placement constraints where 8ft is about right. I can almost get 2 meter cables to fit, but I like some slack for safety/versatility purposes.

cheap-Jack · « **Reply #15 on:** 14 Mar 2012, 06:17 pm »

Hi.

Quote from: jazzerdave on 14 Mar 2012, 05:29 pm

If I could make it work, I'd be making shorter cables, but I've got some placement constraints where 8ft is about right. I can almost get 2 meter cables to fit, but I like some slack for safety/versatility purposes.

Don't forget to have any floor running interconnects, & loudspeaker cables LIFTED off the floor to avoid RFI, statics & floor vibrations.

I used semi-rigid foam blocks to lift up the whole runs of loudspeaker cables at equal spacing. NO no metal lifters please.

I saw in the most hi-end audio boutique I visited recently, similar semi-rigid foam blocks were used to lift up the loudspeaker cables for demos. Each such foam block lifter asked for $40.00 greenback.

Yes, shorter the audio cables the better. FYI, the 4N silver interconnects from
my TT to one of of my tube phono-preamps gets only 1 foot long "tailor-cut" length !!

c-J

Speedskater · « **Reply #16 on:** 15 Mar 2012, 01:00 pm »

"j_dave" for your short 8 foot cables I think that 13/14AWG silver will be fine for almost all speakers. I don't think that another 1000 foot roll is needed. Oh what speakers do you have?

jazzerdave · « **Reply #17 on:** 15 Mar 2012, 02:21 pm »

I'm using Decware ERR's powered by a highly modified Dynaco ST-70. The ERR's are a very simple 8 ohm load.

After starting on prototypes, I'm finding that I'm running into some possible issues with stiffness, so I'll working to see what I can do to alieviate that while still keeping inductance and capacitance low.

Speedskater · « **Reply #18 on:** 15 Mar 2012, 11:00 pm »

It's the very low Ohm ribbon speakers that need the biggest cables.

Chromisdesigns · « **Reply #19 on:** 15 Mar 2012, 11:24 pm »

Quote from: jazzerdave on 13 Mar 2012, 11:49 pm

OK, whether wise or not, I've decided to add some more wire to the project. I'll end up somewhere between 10-12 AWG. I'm not quite sure yet. I'm going to acquire a full extra 1000 ft, but I may set some aside for interconnects rather than use all of it. The final determination may come down to whether or not it will remain flexible enough at larger gauges. I'm going to be making test runs with some bulk magnet wire I have.

When you buy your wire, specify you want "dead soft" hardness. I.E., fully annealed. This will make your task much much easier -- take it from someone who has made crocheted silver cable necklaces! Even fine silver softens substantially when annealed. It's going to work-harden when you make it up into cable, of course, but the end result will be much more pliable than if you started with hard or half-hard wire to begin with.

Ideally, you'd anneal the finished cables, too, but you probably don't have the equipment you would need to do that.