Speaker cable recommendation

[Fishjie]

Hi there

 

Sorry for the newbie question, I’m working on my next upgrade on speaker cables.

I’m currently using Michi X3 fo and connecting to KEF R7(non meta).

About 3-4 meters apart and bi-wire sound better to me.

what will you suggestion on speaker cables?

[THOMO]

I can sell you that length of new Amber 8 cable which is a very good biamping cable [has four conductors].

Audiophile Melbourne retailed that for $100 a metre but would could sell you that length for $30 a metre including shipping. [8 metres for $240].That would be bare wire terminated.

Edited August 19 by THOMO

[Wimbo]

I made up some cables recently out of CAT6. Bit of work but impressed me and I'm using it now. Single core X 16 each cable. One is 6 metres long  and the other 5 metres . I've been using single core cable on and off since the 80's.

Speakers are around 2.5 metres apart (measured  inside to inside and with an impressive sound stage and centre image. Might be worthwhile if you have the time.

Edited August 19 by Wimbo

[PeterB7858]

 

1 hour ago, Wimbo said:

I made up some cables recently out of CAT6. Bit of work but impressed me and I'm using it now.

 

I made some cables last year out of solid core CAT6. 4-braided pattern per side (i.e. 32 x 23AWG wires per side i.e. 16 +ve, 16 -ve per stereo channel; equivalent to 11 AWG per side).

 

I thought the sound stage and imaging were very good as well. Seemed a bit shy in the bass, though. I might go back to them and do some tweaking, perhaps take the PVC cover of each cable and then braid again?

 

You can buy the CAT6 solid core from Bunnings...

Edited August 19 by PeterB7858
Added info

[andyr]

1 hour ago, Wimbo said:

I made up some cables recently out of CAT6. Bit of work but impressed me and I'm using it now. Single core X 14 each cable. One is 6 metres long  and the other 5 metres . I've been using single core cable on and off since the 80's.

 

As you say ... a fair bit of work (all that wire stripping!   ).

 

But you have confused me with "Single core X 14 each cable"!  Given a 'jacket' has 4x twisted pairs = 8 wires in it, do you mean:

• 16 wires to each spkr
• or you have - for each spkr - 14 wires going to the '+ve' terminal and 14 to the '-ve' ... ie. 2 separate "cables" of 14 wires each, going to each spkr?

 

[Wimbo]

1 hour ago, andyr said:

 

As you say ... a fair bit of work (all that wire stripping!   ).

 

But you have confused me with "Single core X 14 each cable"!  Given a 'jacket' has 4x twisted pairs = 8 wires in it, do you mean:

• 16 wires to each spkr
• or you have - for each spkr - 14 wires going to the '+ve' terminal and 14 to the '-ve' ... ie. 2 separate "cables" of 14 wires each, going to each spkr?

 

Yes. 16 wires  between each speaker and amp, so two runs per speaker. Pardon the confusion. 

[Wimbo]

1 hour ago, PeterB7858 said:

 

 

I made some cables last year out of solid core CAT6. 4-braided pattern per side (i.e. 32 x 23AWG wires per side i.e. 16 +ve, 16 -ve per stereo channel; equivalent to 11 AWG per side).

 

I thought the sound stage and imaging were very good as well. Seemed a bit shy in the bass, though. I might go back to them and do some tweaking, perhaps take the PVC cover of each cable and then braid again?

 

You can buy the CAT6 solid core from Bunnings...

This is Certech 23AWG 4 pair individually stripped and crimped as 8 cables. No solder.

[andyr]

15 minutes ago, Wimbo said:

Yes. 16 wires between each speaker and amp, so two runs per speaker. Pardon the confusion. 

 

Sorry, W - I'm still confoosed!  

 

For each spkr, you run 2 'jackets' (8 wires each) - one for the '+ve' run and one for the '-ve' run?

 

If so ... then this will give a cable that is low in capacitance but high in inductance.    And low inductance is good for a spkr cable.

 

Can I suggest you should try connecting all the full-colour wires of each twisted pair together for the '+ve' connection ... and all the striped wires together for the '-ve' connection.  This will substantially increase capacitance - but substantially decrease inductance - which is what you want.

 

 

 

[Wimbo]

46 minutes ago, andyr said:

 

Sorry, W - I'm still confoosed!  

 

For each spkr, you run 2 'jackets' (8 wires each) - one for the '+ve' run and one for the '-ve' run?

 

If so ... then this will give a cable that is low in capacitance but high in inductance.    And low inductance is good for a spkr cable.

 

Can I suggest you should try connecting all the full-colour wires of each twisted pair together for the '+ve' connection ... and all the striped wires together for the '-ve' connection.  This will substantially increase capacitance - but substantially decrease inductance - which is what you want.

 

 

 

Whats the insulation colours got to do with it A?

 

 

 

[andyr]

1 hour ago, Wimbo said:

What's the insulation colours got to do with it, A?

 

Simples, W!  

 

A network cable consists of 4x twisted pairs - ie. 8 wires; in the Belden 1585A which I use, these are coloured thusly:

• orange / orange striped
• brown / brown striped
• blue / blue striped
• green / green striped.

 

If you use 2x 'jackets' to go to a spkr (so 16 wires in all), there are 3 ways you can make these wires into a spkr cable:

1. one jacket (ie. 8 wires) for '+ve', one jacket for '-ve'.
2. two twisted pairs from each jacket for '+ve', the other two twisted pairs for '-ve'.
3. the solid-colour wire from each of the 8 tps for '+ve' ... with the striped wires for '-ve'.

 

Configuration #1 gives the least capacitance ... but the most inductance.

Configuration #3 gives the least inductance ... but the most capacitance.

Configuration #2 is in the middle for capacitance & inductance.

 

The best sound comes from Config #3 - having the least inductance ... as inductance acts as a brake on current.  However, for long runs, the cable's capacitance may not suit some amplifiers.  Hence one would use Config #2.  I myself use Config $1 for the 10-15m runs I have in my house, to ancillary spkrs in other rooms - but Config #3 for the main spkrs.

 

Hence my question to you; if you are using Config #1 ... then you owe it to yourself to try one of the other Configs.  The issue is ... can your amp cope with the capacitance of a 6m run of Config #3?  I would estimate this to be 2000-2500pFs; if it can - then go for it (to get the best sound!).  If it can't ... use Config #2.

 

[Wimbo]

I hope the OP understands all that.

8 individual strands (Jacketed) for the positive and  8 individual strands (Jacketed) for the negative. This is working fine.

 

[Wimbo]

10 hours ago, andyr said:

• one jacket (ie. 8 wires) for '+ve', one jacket for '-ve'.
• two twisted pairs from each jacket for '+ve', the other two twisted pairs for '-ve'.
• the solid-colour wire from each of the 8 tps for '+ve' ... with the striped wires for '-ve'.

Initially I thought you were "taking the p..s" but after doing some research, it seems that CAT6 through inductance creates a back EMF which causes less energy flow making the power amp work harder and get hotter. I'll check this today. I might even nik out to Bunnings and buy some larger AWG copper wire to compare Decibels at the same volume setting. I still find what you have listed above, hard to get my head around as the cables are running the same way but just terminated differently.  What I will maintain is that the HIFI attributes of the sound in this configuration creates more informative detail an I was pleasantly surprised.

[andyr]

10 hours ago, Wimbo said:

 

8 individual strands (Jacketed) for the positive and 8 individual strands (Jacketed) for the negative. This is working fine.

 

 

Of course it's "working fine", W - so does "figure 8" cord.  

 

33 minutes ago, Wimbo said:

What I will maintain is that the HIFI attributes of the sound in this configuration creates more informative detail and I was pleasantly surprised.

 

IMO - that's due to the use of individually insulated, solid core wires - rather than stranded wire.

 

33 minutes ago, Wimbo said:

I still find what you have listed above, hard to get my head around as the cables are running the same way but just terminated differently. 

 

Consider the construction of the Cat5/6 cable ... there are 8 wires in a 'jacket', consisting of 4x twisted pairs.  When you use one jacket for +ve and one for -ve (my Option #1) ... the +ve and -ve runs have some distance between them.

 

In contrast, when you use one wire of each twisted pair for +ve and the other one for -ve (my Option #3) ... the +ve and -ve runs are tightly coupled; it is this which changes the capacitance and inductance of the cable - and hence the sound.

 

41 minutes ago, Wimbo said:

I might even nick out to Bunnings and buy some larger AWG copper wire to compare Decibels at the same volume setting.

 

Can I suggest a much better idea would be to buy some more of that Certech 23AWG 4 pair and make up a new set of cables to compare to your existing ones, constructed using either my Option #2 or #3.

 

[Wimbo]

6 minutes ago, andyr said:

Of course it's "working fine", W - so does "figure 8" cord.  

In some of my research, I found that too much Inductance can cause Oscillation in some Broad Frequency amplifiers. 

 

10 minutes ago, andyr said:

Consider the construction of the Cat5/6 cable ... there are 8 wires in a 'jacket', consisting of 4x twisted pairs.  When you use one jacket for +ve and one for -ve (my Option #1) ... the +ve and -ve runs have some distance between them.

 

In contrast, when you use one wire of each twisted pair for +ve and the other one for -ve (my Option #3) ... the +ve and -ve runs are tightly coupled; it is this which changes the capacitance and inductance of the cable - and hence the sound.

My cables are sheathed side by side. 

 

11 minutes ago, andyr said:

 

Can I suggest a much better idea would be to buy some more of that Certech 23AWG 4 pair and make up a new set of cables to compare to your existing ones, constructed using either my Option #2 or #3.

I've got a roll of it. Hence using it. But too much mucking around just to find out if Decibel levels change. If they do to a noticeable degree, I'll try your alternative configurations.

Thanks @Fishjie for your understanding here and I hope this conversation might help.

[Fishjie]

Thanks all for for your inputs.

I found with my system bi wire do sound better.

The amp are 1 meter and 3 meters apart from the speakers I’m trying to get both cable same length so ideal should be 4 meter each to have enough length

[andyr]

6 hours ago, Wimbo said:

In some of my research, I found that too much Inductance can cause Oscillation in some Broad Frequency amplifiers. 

 

Certainly, an excess of inductance can cause oscillation - back 20 years ago, I blew up the output transistors in a Harman Kardon amp twice ... when I connected it up to the 15m NAC5 spkr cables which I had going to another room, in my previous house.    Luckily, good ol' Trevor Lees allowed me to trade it in on a Naim amp, after I had had it repaired for the 2nd time!  

 

Also, too much capacitance can cause problems with amplifiers - hence not all amps can drive electrostatics, well.

 

6 hours ago, Wimbo said:

My cables are sheathed side by side. 

 

If you mean ... the 2 jackets are held close, side-by-side in a sheath - that's a good thing, as it means that capacitance is higher than it would be otherwise (if the 2 jackets were just laying close to each other).  Hence ... their inductance will be lower - which is a good thing!

 

6 hours ago, Wimbo said:

I've got a roll of it. Hence using it.

 

 

6 hours ago, Wimbo said:

But too much mucking around just to find out if Decibel levels change. If they do to a noticeable degree, I'll try your alternative configurations. 

 

Sorry, non comprende???

 

I suggest that making the same cable from a slightly thicker-guage wire won't change dB levels (that you can hear, anyway).

 

But reducing the inductance - using the same wire but using my Options #2 & #3 - will make a difference to your sound.

 

You get high capacitance - and, hence, the desired low inductance - by having the '+' and '-' conductors very close together.  You can't get conductors much closer together than a twisted pair ... about the only other possibility is ribbon cables, where the +ve and -ve ribbons are only separated by some insulating tape.

 

[PeterB7858]

@andyr

 

The CAT6 solid core four braid per channel that I made last year (see below) kept the blue PVC in place and I took individual wires from each of the 16 twisted pairs to create the +ve and -ve runs.
 

I can’t remember what “formula” I used to choose what individual wires contributed to each, though. I will pull them apart at some stage and check.

 

I gather you remove the outer jacket, Andy?

 

And, I can understand from your explanations above why that might be beneficial, but how did you keep the twisted pairs together?

 

Assuming I can keep all 16 twisted pairs successfully, does braiding them (as twisted pairs) add much to SQ? Any thoughts would be appreciated.

 

Regards,

Peter

 

 

[Wimbo]

2 hours ago, andyr said:

I suggest that making the same cable from a slightly thicker-guage wire won't change dB levels (that you can hear, anyway).

Didnt say that. Using normal high gauge speaker cable to compare to the supposed high inductance cable that you say I'm using atm. Never mind. It's all good.

[Wimbo]

51 minutes ago, PeterB7858 said:

Assuming I can keep all 16 twisted pairs successfully, does braiding them (as twisted pairs) add much to SQ? Any thoughts would be appreciated.

Havent tried that but it looks interesting

[andyr]

1 hour ago, PeterB7858 said:

@andyr

 

The CAT6 solid core four braid per channel that I made last year (see below) kept the blue PVC in place and I took individual wires from each of the 16 twisted pairs to create the +ve and -ve runs.

 

OK - a 4-braid per channel cable means 32 individual wires per channel, right?

 

Yes, the braiding of the 4 jackets keeps all the jackets together (which is good, in terms of low inductance) ... but I don't see it as doing anything more than simply using all the solid-colour wires in each jacket for '+ve' ... and the striped wires for '-ve'?

 

1 hour ago, PeterB7858 said:

I gather you remove the outer jacket, Andy?

 

Yes, I do - basically, because that jacket is PVC ... whereas each wire insulation is teflon (or some relation).

 

1 hour ago, PeterB7858 said:

And, I can understand from your explanations above why that might be beneficial, but how did you keep the twisted pairs together?

 

I enclose all the twisted pairs in a black woven sheath - see here, for a cable I made recently (8 wires for +ve and 8 wires for -ve):

 

 

 

1 hour ago, PeterB7858 said:

Assuming I can keep all 16 twisted pairs successfully, does braiding them (as twisted pairs) add much to SQ? Any thoughts would be appreciated.

 

I don't think you can get a tighter coupling between '+ve' and '-ve' strands - which produces a low-inductance cable - than by using the solid-colour wire of each twisted pair for +ve ... and each striped wire as -ve.

 

Therefore 'jacket braiding' is not needed.

 

[PeterB7858]

27 minutes ago, andyr said:

I don't think you can get a tighter coupling between '+ve' and '-ve' strands - which produces a low-inductance cable - than by using the solid-colour wire of each twisted pair for +ve ... and each striped wire as -ve.

 

Thanks for sharing your knowledge and thoughts, Andy. I appreciate it. 

 

So you just twisted the 8 twisted pairs together in the above cables? And used the black woven sheath to stop them unravelling?

 

We are away for the next week or so but I will return to this when I get back and see what I can come up with. 

 

I currently use Lenehan Foilflex speaker cables, and I think they are pretty good, but I'm always on the lookout for something to try up against them.  

 

I haven't used the braided CAT6 cables in the system since I got the Weston Topaz KT120 valve amp and the Lenehan ML2+R speakers, so I must give them a go.

 

I might have gone for overkill with 16 twisted pairs per channel , but I'll see what I can come up with when I strip the covers. You never know...

 

Thanks, again.

Edited August 20 by PeterB7858
grammar

[muon*]

Bettering the Foilflex speaker cables is a big ask IMO, will be interesting to say the least.

[PeterB7858]

8 hours ago, muon* said:

Bettering the Foilflex speaker cables is a big ask IMO, will be interesting to say the least.

 

Yes, I agree. I'm not expecting to. But it's nice to have a cheap DYI project to tinker with.

And you just never know...  

[muon*]

1 hour ago, PeterB7858 said:

 

Yes, I agree. I'm not expecting to. But it's nice to have a cheap DYI project to tinker with.

And you just never know...  

I agree, It's fun to tinker

[andyr]

11 hours ago, muon* said:

Bettering the Foilflex speaker cables is a big ask IMO, will be interesting to say the least.

 

Agreed Ian.  I will be most interested in Peter's results.  

 

11 hours ago, PeterB7858 said:

So you just twisted the 8 twisted pairs together in the above cables? And used the black woven sheath to stop them unravelling?

 

Not quite!  Once you strip off the jacket, each tp remains twisted; the woven sheath simply holds all the tps together.