DIY Speaker cable (found a great deal)!!!

[runninkyle17]

Okay,

So in the search to be different and find something very unique I have decided to make my own speaker wire. I am going to be doing a Cat5 design (there is a link to the instructions in my sig). I have been looking around for a good deal on some Belden Cat5 cable (westlake electronics had good deals but I wanted to see if I could find anything better). Browsing through ebay I came across a guy who has a lot of High Quality Belden cable (part numbers 1700A and 1701A, I am going for the Plenum 1701 myself). Anyway, I am going to buy 100 meters (328 ft) from the guy for $30 plus shipping!!!! That is an amazing bargain. So if anyone is interested in making their own speaker cable this is a great cheap way to do it (if you have the time). I am going to get some heatshrink from the local Home Depot (they actually have a good price on some quality stuff) and I will be buying GLS Locking Bananas to square off the connector part of it (if you have not heard of these bananas they are really great and amazingly strong).

So this is my current project until I have enough money saved to get myself an SVS or HSU sub.

 

[philh]

Just for the fun of it, made my own cables. I saw that design and decided it was too much work. Stripped the jacket off three cables and did a simple braid on three cables. If nothing else, it's a pretty cable It works, and I think made a slight improvement. I had a bunch of CAT 5 laying around, so the price was right.

 

[runninkyle17]

I am not really worried about how much work it is going to be. I study all day long anyway, so I will just make the cable while I study. I am happy to hear that it worked well for you with just the three braid design. I will use the 9 pair (3 braid) for the highs and the full 27 pair (9 braid) for the lows. I am going to bi-wire my front speakers just to see if it will improve the sound any. Thanks for the comment.

 

[j_garcia]

IMO, braided cables are a gimmick. It doesn't really add anything to the mix. That is really cheap though, so I don't see any reason not to do it.

I can tell you already that biwiring will not improve anything. Biamping is a different story.

 

[runninkyle17]

I am not sure if the braided cable will improve the sound or not, but I know that having the large gauge cable will help a bit. My buddy has some very nice Paradigm speakers and he has them setup with braided Cat5 cable (only a 3 braid pattern) and they sound great. He had them setup with some 14/4 cable from Belden and to be honest the braided cable sounds better (in my opinion). As for the biwiring, I am new to the biwiring concept so I figured I would give it a try. I am buying enough cable so if I decide to biamp later I will be able to with not much trouble at all.

P.S. If I were to biamp, what would you recommend on a second amp to run the lower end of my front speakers?

 

[Hi Ho]

I "made" my own speaker cables too. I went to Radio Shack, bought some 14g zip cord, stripped the ends, and connected them. It sounds just fine. It cost me about $10. I would be willing to bet that if I had a "premium" brand cable or some other DIY that the sound would NOT be affected. That's just me though.

The only time I have ever heard a difference in speaker wire is when I connected my 4 OHM Advents (as an experiment) with some super cheap and super thin cable. That sounded terrible. Then I tried some 16g wire, it sounded fine. Next was 14g, no difference. 12g, again, sounded the same as the 14 and 16g.

 

[sts9fan]

I don't really care if DIY cables make a difference. i enjoy making them and how they look

 

[edwelly]

runninkyle17
Where did you get your cable? I went on ebay and didn't find anything this good... Thanks!!!

 

[corey]

There's at least a hint of snake oil on Chris VanHaus's site. He says:

Give them a while to break-in before giving a seriously critical listen, though. They age nicely.

On the other hand, Audioholic Mudcat tested a bunch of speaker cables & his cat5 cables all came out on top.

I like this concept for 2 reasons - the cables will look nice & I like the idea of using something from my business life for my hoby.

My plan is to make a variant of these cables. Chris VH did 3 strand braids of twisted pairs to get 27 pairs per cable. Mudcat just braided cat5 with covers. I'm planning to strip the cover off 4 lengths of cat5 & do a round 4 strand braid with them. I did a test braid with 2' lengts & it looks good.

I doubt that they'll be any better that 12 AWG wire, but Mudcat's data seems to say they should work OK, & the colorful cat5 braid will look nice.

 

[Mudcat]

Speak of the Devil

Attached are my latest creations made with 4 CAT 5 cables braided in an interlocking chain weave. I used both regular PVC and Teflon cables for these pictures so that they show up well. I haven't made any measurements yet, as I am working on a similar weave with both 6 and 8 cables so that a 10 foot 4 cable is roughy equivelent to 13 AWG (the cross sectional area is equivelent to 12 awg but you need 25% more cable for a specific length, i.e. 12'6" for a 10 foot finished length), 6 cable = 11 AWG, and 8 cable = 10 awg.

Anyway check out the pictures, there are two types of braid, although when using all teflon cable, the patterns will be indistinguishable.

 

[corey]

If I'm looking at this correctly, Mudcat's numbers seem somewhat conservative. The table I found listed 24 AWG wire as d=0.51054 mm, which would give a cross section of 0.2047 mm2. Using 4 lengths of cat5 gives you 16 wires per conducter, with a combined area of 3.2754 mm2. The table lists 12 AWG as d=2.0523, which is an area of 3.3113 mm2, so you'd be within 1% or so.

Mudcat, is this math correct?

Attached is a sample of the cable I have in mind. It's not a real tight braid, so the finished length is 10% less than the cut lenghth.

 

[Mudcat]

If I'm looking at this correctly, Mudcat's numbers seem somewhat conservative. The table I found listed 24 AWG wire as d=0.51054 mm, which would give a cross section of 0.2047 mm2. Using 4 lengths of cat5 gives you 16 wires per conducter, with a combined area of 3.2754 mm2. The table lists 12 AWG as d=2.0523, which is an area of 3.3113 mm2, so you'd be within 1% or so.

Mudcat, is this math correct?

Yes, but the cross sectional area is just a starting point.

Lets work in circular mils for cross sectional area please, it's just easier for me. I didn't do the math in my previous post (but after doing it here I've edited my previous post), so lets do it now,


example:

Say you have a 10 awg zip wire like Sound King. By definition, 10 AWG is 1 ohm/1000 feet (actually it is 0.9989, but we'll use 1 here), therefore, 10 feet os 0.01 ohms. The cables resistance is 0.02 ohms (there and back). Now if that zip wire was braided to another zip wire just like it, it's resistance would be at best:

1/R=1/R₁ + 1/R₂

1/R = 1/0.02 + 1/0.02

1/R = 50 + 50

So R = 0.01 ohms, but the wires are braided, so that two ten foot lengths braided together make an 8 foot cable (actually, depending on how your braid is dictates the length, a 4 wire french braid will be about 9' 3", the braid I used in the pictures shown require 25% more than the finished length). Now your resistance is still 0.01 ohms, but over 8 feet instead of 10 feet. So your unit resistance went from 0.001 ohms per foot to 0.00125 ohms per foot, more like a 11 awg cable.

Example 2:

For the cables I pictured in my previous post, there are 4 CAT 5 cables each consisting of eight 24 awg wires.

24 AWG has a xsection area of 404.01 cm and a base DCR of 0.02567 ohms/ft.

16 x 404.01 = 6464.16 ~12 awg

1/0.02567 = 38.956
16 x 38.956 = 623.296
1/623.296 = 0.0016 ohms/ft

0.0016 x 10 (length of cable to start with) = 0.016 ohms

0.016/8 (length of finished cable) = 0.002 ohms ~13 AWG

Attached is a sample of the cable I have in mind. It's not a real tight braid, so the finished length is 10% less than the cut lenghth.

This is Chris VanHaus's version. I do not know if he is the originator, but he certainly made it popular. He frequents the cable asylum over at AA, you'll never find him here at Audioholics. This is a good cable but has very high capacitance. I believe Gene tested something like this a long time ago, I know he tested another popular DIY cable (John Risch's cross-connected coax). I'm also working on a coax version of my own - but no details yet.

 

[corey]

Thanks, Mudcat.

The cable I have in mind is a cross between Chris VanHaus's and yours. The outer cover is stripped, like CVH's, but it's only a single braid like yours. The resistance seems tollerable, but I don't know what to do about capacitance.

In your "DIY Speaker Cable Face-off", the cat5-V2 seemed to have OK capacitance, and your top 3 were all cat5. Do you have an opinion as to wheather the high capacitance of the CVH cable is influenced more by the fact that it has multiple levels of braids, or that the covers are off?

Does how the wires are combined into 2 conductors factor into this? Your design uses all the wires from 2 cat5's for a conductor, the CVH cable calls for dividing the twisted pairs into solids & stripes.

Sorry for all the questions, but it's your excelent article that got me going on this in the first place.

 

[Mudcat]

In your "DIY Speaker Cable Face-off", the cat5-V2 seemed to have OK capacitance, and your top 3 were all cat5. Do you have an opinion as to wheather the high capacitance of the CVH cable is influenced more by the fact that it has multiple levels of braids, or that the covers are off?

Does how the wires are combined into 2 conductors factor into this? Your design uses all the wires from 2 cat5's for a conductor, the CVH cable calls for dividing the twisted pairs into solids & stripes.

Multiple Levels of Braids have the greatest effect. Then there is an added effect of all those wires close to each other. Not actually having made and tested CvH's version, I can only guess. We know what happens when I guess

Parallel capacitance is added regularily (the twisted pairs of wires in a CAT 5 cable are parallel capacitors) , unlike parallel resistance (¹/_{R_T}=¹/_R₁+¹/_R₂+¹/_R₃...)

Parallel Capacitance is C_T=C₁+C₂+C₃...

Series Capacitance is added like parallel resitance:
(¹/_{C_T}=¹/_C₁+¹/_C₂+¹/_C₃...)

With CAT 5, each twisted pair has a nominal capacitance of 17 pF/ft, so if you have a 10 foot cable, that 170 pF. since there are four twisted pairs per cable, thats a total of 680 pF. CvH's version has 27 twisted pairs and requires about 11 feet to make a 10 foot finished cable, so thats a total of 5049 pF. A capacitance this large can have a bad effect on some amps. You might want to PM Dan Banquer about what exactly in the amp design causes this. I don't do amps and wouldn't even guess at this one.

My CAT 5 cables (with the jacket on) are an order of magnitude lower, having only a couple of hundred pF total. Of course the inductance of my cables is higher then too, but not by the same amount.

Of course the reverse is true for inductance, which has a larger effect on what you hear.

In the near future, I'll be redoing most of the cables from the original article (except for the Monster cables since I sold those on eBay for almost what I paid). Mainly because I got better measuring equipment, and secondly, because I multi-strand braiding down pat. I might do twisted lays again such as V1 and V3, it will be a time issue.

 

[runninkyle17]

Here is what my final cables will look like. I am debating if I should wrap them in rechflex or leave them as is. If you like the pattern PM me and I will tell you how I did it. It is really very simple.

 

[Mudcat]

Definitely Cover Them In "Reechflex"

Here is what my final cables will look like. I am debating if I should wrap them in rechflex or leave them as is. If you like the pattern PM me and I will tell you how I did it. It is really very simple.

Two questions:
<OL type="1">
<LI>How did you do the braid?
<LI>Is this what Bip The Michelin Man's turds look like
</OL>

Sorry, I had to toss that last one in.

 

[runninkyle17]

Yeah yeah I know. They are not the nicest looking cables, but they are my first attempt at my own cable so I am fairly pleased with them.

I would have gotten different colors of Cat5 if I had realized that I did not want to mess with techflex.

Anyway, I braided four ten foot lengths of Cat5 cable (using three strands of Cat5 cable). Then I cut about a 14 ft length of Cat5, stripped off the outer covering and seperated the four inner strands. I clamped two of the white braided strands to a vice, along with two of the small colored twisted pairs (to give the final cable some color, these two twisted pairs were treated as one whole strand in the final cable). I then braided the three strands together with the small twisted pairs acting more as just a brace to hold the larger white braids together. The cable turned out very nice and in the end there is a total of 26 pairs of wire, so the final wire is around 9 or 10 gauge.

I am not going to use any techflex, but I will obviously put some heat shrink on the ends and terminate the cables with some angled locking bananas that I found here http://cgi.ebay.com/Set-of-8-High-Quality-Gold-Plated-Locking-Banana-Plugs_W0QQitemZ5830063185QQcategoryZ14993QQssPageNameZWDVWQQrdZ1QQcmdZViewItem.

I am going to stay away from spades because I really like the fact that the locking bananas give the cable a little extra support. Each cable probably weighs about 3 to 5 lbs for each ten foot length. I cannot wait to break them in.

 

[Mudcat]

I cannot wait to break them in.

What

Insert the screach the pod people make here. Also insert Edvard Munch's Scream painting.


I thought you were one of us. The Good, the rightous. Yeah I know, cables do require break-in, but it only takes a couple of milliseconds.

 

[edwelly]

One thing that I am still unclear on is "are you using 3 Cat5 cables for each set or each wire" - meaning 3 Cat5 = 1 speaker cable OR 3 Cat5 for a positive and 3 Cat 5 for a neg. -

runninkyle17 - I bought 328ft from the same guy. My stuff should be here end of week of first of next week... Thanks again!!!

 

[runninkyle17]

What

Insert the screach the pod people make here. Also insert Edvard Munch's Scream painting.


I thought you were one of us. The Good, the rightous. Yeah I know, cables do require break-in, but it only takes a couple of milliseconds.

I didn't mean "break them in" like you think I did. What I meant is that I cannot wait to get some good use out of them. I am sorry if I mislead you down the "snake oil" path. Hehe

edwelly - This is how I cut the cable:

1. Measure a 12 ft length of the Cat5 cable (this is the cable with the outer covering still over the four 24 AWG twisted pairs).

2. Once the 12 ft length is measured; measure and cut 11 more lengths just like the first one. For a total of 12 lengths of Cat5 cable (with the outer covering still over the four 24 AWG twisted pairs).

3. Take three lengths of the cut cable and braid them together (you will end up with about a 10.5 ft length of braided cable). I left 1.5 ft on each end of the cable unbraided so that I had plenty of cable to work with when I strip the ends.

4. Once you have braided the first length of cable you should have 9 lengths of 12 ft cable left. Take three lengths and braid them together like the lengths in step three. Now you have six lengths left; take three and braid them together and then braid the final three lengths together.

5. After all of this, you will end up with four individual braids of cable that are each around 10.5 ft long.

6. Next is the part where you actually strip the outer covering off of the Cat5 wire and use the small 24 AWG twisted pairs. The cable that I bought was white on the outside and on the inside consisted of four pairs of 24 AWG wire. The inside pairs were colored as follows: green-white, orange-white, blue-white, purple (brown)-white. The wire that I bought had bonded pairs so the only hard part about stripping the wire is seperating each of the bonded pairs. Since I was going to be using the small 24 AWG wire as a brace to hold the larger braided strands together I went ahead and cut a 15 ft length of Cat5 cable and stripped the outer white covering off of it to reveal the small paired wires inside.

7. So after I stripped the outer white covering off of the Cat5 cable, I seperated the small paired, colored wires. There were four of them, but I only needed one more strand to braid together my other larger braids (remember, the big white braids). So to add some color to the cable I chose to pair the green with the brown and the orange with the blue.

8. Then I put two of the large braided cables in my clamp and also fastened the two bonded pairs (I treated the two bonded pairs as one individual wire) in the clamp. Then I braided them and tried to make the braid as tight and neat as possible.

9. Once the cables were all braided I taped up the ends to fasten them together until I have time to strip off the white covering and seperate all the individual wires into two sets of wires (one set is the white wires and the other set is all the colored wires; for me the white wires are twisted together and this is the negative side and the colored wires are twisted together and this is the positive side).

I know that is very long and probably overkill for everyone, but I was bored. I hope this helps and if anyone has any questions just drop me a PM. I will post some pictures of the final product once I get all my connectors and heat shrink on the cables. Cheers!!