AVS › AVS Forum › Audio › Receivers, Amps, and Processors › DIY Audio *analog* cables
New Posts  All Forums:Forum Nav:

DIY Audio *analog* cables  

post #1 of 91
Thread Starter 
Title: DIY Audio *analog* cables

So OK, AVS prompted me to check some related/recent threads:

http://www.avsforum.com/avs-vb/showthread.php?t=726473
http://www.avsforum.com/avs-vb/showthread.php?t=1004642
http://www.avsforum.com/avs-vb/showthread.php?t=993697
http://www.avsforum.com/avs-vb/showthread.php?t=951839
http://www.avsforum.com/avs-vb/showthread.php?t=833379

...which I did, even though the last 2 are digital.

Anyway, I got started yesterday (it's early morning now, too late for any thinking that I made a mistake!) by placing an order with Allied for some strange speaker cable.

Odds are this thread is gonna be a lumpy one, with me taking most of the lumps.

For speaker wire I went an order with Allied for 1) $121.46 of 100' of Belden 9913 coax, 50 ohm impedance, thick 10 gage solid center copper, braided shield (10% off deal thru Sept.30) 2) $9.44 of 20' Alpha .5" black heat shrink tubing and 3) $4.52 of 100 count of 3M 10-12 AWG spade lugs (overstock deal)

I've seen the ssabripo thread (1st one I provided above), and it started off on interconnect wire (refs that he gave on DIY speaker wire showed up later in that thread.)

Up front he started off with Belden 89259 RG-59 75 ohm impedance, a stranded plenum (expensive), which looks questionable to me as an interconnect cable.

I've liked solid connector, even for speaker wire, and even in a mixed format like an RG-8 such as Belden 9913. Mixed meaning solid center with braided shield.

But I'm mixing the whole speaker and interconnect thing.

9913 gets me started for speaker wire, and my past experience with solid connector, for both speaker and interconnect, has been positive. I know, I know, Beldon 9913 is solid for center (10 AWG) and braided (tinned copper) for the shield. I'm not sure if that matters, to the extent that I'm willing to go with 9913 as my speaker wire.

I'll think it through as I go.

I mean, it's not that much money, and my time doing the soldering for the lugs, as well as replying here, is gonna cost me a lot more.

Interconnects is what actually got me started on this trek as I swore that I'd not spend another $70+ for a 1.5m interrconnect (AudioQuest Sidewinder), but my head is still spinning on this issue. hifisponge has provided a number of refs on this to me very recently, and I'm still reading/thinking. For the moment, I'm still thinking of buying 500' of RG-8X for interconnect. (currently ~$350 at Allied depending on which you get) and Canare RCA interconnects.

I'll do responses as I can.

So far I'm hoping that my Belden 9913 will show up by next Monday.

re RG-8 as speaker wire, it was mentioned in Audio Critic issue 16, which can be downloaded at www.theaudiocritic.com

There's a lot of useful stuff there, which was part 2 of 3. Interconnectors were in issue 17, with write-in comments in 16, 17, and 18.

Aczel is roughly 82/83 now, so I suggest you d/l all of it now, as well as the 3 recent html webzine writeups

FWIW Aczel offers a very different view of the whole hifi scene. He and Gordon Holt (Similar age; I'm not even sure I've got the Holt last name right; the originator of Stereophile a long time ago), both get my vote.

Well... ... there's a lot of huff and puff and gets this lumpy thing off the dime.

Cheers
post #2 of 91
Good luck and post pics. I wish I documented my speaker wire construction.
post #3 of 91
You're going to be running the RG-8 in pairs for the speaker wire, no? Also, what kind of distances are you talking about?
post #4 of 91
Thread Starter 
Quote:
Originally Posted by Chu Gai View Post

You're going to be running the RG-8 in pairs for the speaker wire, no?

I'm presently running a single piece of 14 gage zip cord to each speaker.

I'll run a single piece of RG-8, which is 50 ohm impedance, to each speaker.

It hadn't occurred to me to run 2 pieces of RG-8 to each speaker, if that's what you meant.

Geez, LOL!

I mean, you gotta draw the line somewhere.

Quote:


Also, what kind of distances are you talking about?

About 25' to the most distant surround speaker. My SS amp is off center at the front of the room.

I plan to run 25' to each surround, so I'll have to coil one up, at the back where I have room for it.

I'll only run the RG-8 for the 5 speakers, and 75 ohm stranded wire to my 2 subs. (smaller and a lot easier to deal with than the thick, stiff RG-8.)

Cheers
post #5 of 91
Well, this is RG8...



Unless you run two cables for each speaker, you're looking at using the braid as one of your connectors, no? Also with two cables you'll have to keep them close together otherwise the inductance goes to shitz. Compared to something like zip cord, the capacitance is signficantly higher. You may, or may not be aware that a number of years ago some companies came out with wire such as you're envisioning (Polk comes to mind). What they found is that due to the excessive capacitance, many amps readily went into oscillation sometimes with fatal results leading to some lawsuits. Now, that may not happen with yours, but... With a solid center conductor, you're looking at an incredibly inflexible wire with a poor bending radius. I seriously doubt that fella at Bryston would make the same recommendation today.
post #6 of 91
Thread Starter 
Quote:
Originally Posted by Chu Gai View Post

Well, this is RG8...



Unless you run two cables for each speaker, you're looking at using the braid as one of your connectors, no? Also with two cables you'll have to keep them close together otherwise the inductance goes to shitz. Compared to something like zip cord, the capacitance is signficantly higher. You may, or may not be aware that a number of years ago some companies came out with wire such as you're envisioning (Polk comes to mind). What they found is that due to the excessive capacitance, many amps readily went into oscillation sometimes with fatal results leading to some lawsuits. Now, that may not happen with yours, but... With a solid center conductor, you're looking at an incredibly inflexible wire with a poor bending radius. I seriously doubt that fella at Bryston would make the same recommendation today.

Nice coax pic.

Odds are that it is you who's not done enough homework, and not the Bryston guy (who was mentioned in the speaker cable writeup in issue 16 of theaudiocritic).

Belden's spec for a single RG-8 cable (solid center Belden 9913) to each speaker has very good numbers for impedance, capacitance, resistance and Velocity of Propagation.

There's no mention of inductance. Since the 2 conductors are further apart with large RG-8, than with ordinary 14 gage zip, I'd think that the inductance would be somewhat higher with RG-8. The open question on inductance is: how important, if any, is inductance to SQ in a high end sound system? FWIW about 20 years ago (when running Quad electrostatic speakers) I made speaker wire using Radio Shack 18 gage solid coated with a thin shellac. Took an ordinary drill and carefully twisted it together. Checked it with an ordinary volt/amp meter. Sounded OK but not necessarily better, and I grew nervous about blowing out my amp if I so much as breathed too hard on that wire, so it didn't stay in the setup too long. So there are practical limits as to just how far I'll go. But I bet a nickel it is further than most.

Also, running 2 RG-8 cables to each speaker would change the numbers, and my hunch is not for the better.

Agreed that RG-8 is a clunky cable. Stiff and slightly over .4" in outside diameter. And you don't want to do any hard bends with it, or step on it either since that can alter the specs. Plus maybe going with a solid center makes no difference in SQ. I like solid wire for the hot connector, so I'm going with solid center both for the speaker wire (which is ordered) and also for the interconnect wire. Probably Belden RG-8X with a solid center (Allied has two, one at something like $340/500' and the other something like $380/500'; both .24" in OD with 15 gage center).

When I get around to taking a much closer look at Canare male RCA connectors, I may go with one at 13 gage center if that will work OK with one of the Canare RCA connectors.

Cheers
post #7 of 91
So, what are you going to do then? Just use one coax to each speaker with the center conductor going to one spade and then bunch up the braid, which will be a signficantly different gauge to the other spade? That braid is copper and not aluminum I take it. As to how important the inductance is, that would largely depend if your speaker has a falling impedance curve with rising frequency. Only a few out there do that. VOP is pretty much irrelevent for the frequencies you'll be dealing with. It's not that your approach won't work, it's just that for myself, I don't see it conferring advantages.
post #8 of 91
Thread Starter 
Quote:
Originally Posted by Chu Gai View Post

So, what are you going to do then? Just use one coax to each speaker with the center conductor going to one spade and then bunch up the braid, which will be a signficantly different gauge to the other spade?

The resistance given, by Belden, for the center and outer braid are almost identical.

Quote:
That braid is copper and not aluminum I take it.

Tinned copper, 95%, as well as a foil wrapper.

The center is pure bare copper, not anything odd like copper coated steel.

Quote:
As to how important the inductance is, that would largely depend if your speaker has a falling impedance curve with rising frequency. Only a few out there do that. VOP is pretty much irrelevent for the frequencies you'll be dealing with. It's not that your approach won't work, it's just that for myself, I don't see it conferring advantages.

The advantage that I see is that it's likely to be as good as one can do for speaker wire, and only cost me $135.42 plus tax and shipping for all 5 speakers.

That's in comparison to buying into some of the outrageous high end speaker wire to the tune of thousands of dollars.

I'm a practical guy, and my time is still worth $500+/hour.

Same goes for outrageous high end interconnect prices, which is what finally got me off the dime.

Cheers
post #9 of 91
It should be about as good as any other approximately 16 AWG loudspeaker wire. IMHO, compared to your 14 AWG zip cord, using coax for loudspeaker wire is a lot of bother for no actual benefit.
post #10 of 91
I don't know man. 100' feet of regular old 10 or 12 gauge is pretty cheap. But, if you're a solid core type of guy, go for it. BTW, you can get 'enameled' wire (the stuff they use to wind motors and alternators) pretty darned cheap.
post #11 of 91
Well this is interesting as most DIY guys like myself prefer to use CAT-5 instead of coax for speaker application. Are you going to be measuring the capacitance, resistance, etc?
post #12 of 91
Quote:


It should be about as good as any other approximately 16 AWG loudspeaker wire. IMHO, compared to your 14 AWG zip cord, using coax for loudspeaker wire is a lot of bother for no actual benefit.

I agree, all you need is the proper AWG for the distance you are running.
post #13 of 91
Thread Starter 
Quote:
Originally Posted by Veda View Post

Well this is interesting as most DIY guys like myself prefer to use CAT-5 instead of coax for speaker application.

I've seen a number of them. e.g.:
http://www.audioholics.com/reviews/c...-cable-faceoff

Quote:


Are you going to be measuring the capacitance, resistance, etc?

No.

Why should I?

It's standard Belden 9913 which Belden gives accurate specs for, and which I'm using as is.

Cheers
post #14 of 91
Quote:
Originally Posted by habnab View Post

No.Why should I?It's standard Belden 9913 which Belden gives accurate specs for, and which I'm using as is.Cheers

That is if you're using 2 cables to each speaker, which comes back to Chu's concern. I didn't see you answer his question but I might have missed it. So is it gonna be a total of 4 cables?
post #15 of 91
Quote:
Originally Posted by DougWinsor View Post

I agree, all you need is the proper AWG for the distance you are running.

Except that even in same AWG size cables, they measure differently (refer to Audioholics). But they won't have a big audible difference anyway so...
post #16 of 91
Thread Starter 
Quote:
Originally Posted by Veda View Post

That is if you're using 2 cables to each speaker, which comes back to Chu's concern.

Maybe I'm missing something?

How would *you* connect 2 RG-8 cables to one speaker?

Quote:


I didn't see you answer his question but I might have missed it. So is it gonna be a total of 4 cables?

You missed it, but I'll say it again.

I'm going to run 1 RG-8 coax cable to each speaker.

i.e. using the 10AWG solid center conductor as the hot wire ("red"), and the outer braid as the return wire ("black").

Cheers
post #17 of 91
Quote:
Originally Posted by habnab View Post

Maybe I'm missing something?

How would *you* connect 2 RG-8 cables to one speaker?
Cheers

Just don't use the braids on either one.
post #18 of 91
Quote:
Originally Posted by habnab View Post

i.e. using the 10AWG solid center conductor as the hot wire ("red"), and the outer braid as the return wire ("black").Cheers

And you're saying the braid will have the same spec as the center conductor? If 10 gauge cables can have different capacitance etc what makes you think the braid will be the same as the center? Belden didn't give "accurate specs" for the shielding braid as it is not intended to carry the signal. Even if you follow the audiophile way of thinking, the braid would be the last thing you use as it is outside the insulation. Most people would use 2 center conductors (2 cables) that are already well sealed to each post. Food for thoughts.
post #19 of 91
Thread Starter 
Quote:
Originally Posted by Veda View Post

And you're saying the braid will have the same spec as the center conductor?

The braid appears to me to be about 11 or 12 gage.

Quote:


If 10 gauge cables can have different capacitance etc what makes you think the braid will be the same as the center?

With both center and braid carrying the signal, it functions as a system.

The main specs Belden gives, are for the system.

Quote:


Belden didn't give "accurate specs" for the shielding braid as it is not intended to carry the signal.

As far as I know, most interconnect coax cable has the shielding braid carry the signal on the return.

Quote:


Even if you follow the audiophile way of thinking, the braid would be the last thing you use as it is outside the insulation. Most people would use 2 center conductors (2 cables) that are already well sealed to each post. Food for thoughts.

It's a different system when you do that.

Cheers
post #20 of 91
Thread Starter 
Quote:
Originally Posted by habnab View Post

Maybe I'm missing something?

How would *you* connect 2 RG-8 cables to one speaker?

Quote:
Originally Posted by Chu Gai View Post

Just don't use the braids on either one.

I have no clue what the specs would be for using the cable that way. I mean, cable functions as a system, and most of the specs are system specs.

It's even possible that it might be dangerous for some amps.

So be careful.

You were the one that mentioned that possibility, early in this thread.

Cheers
post #21 of 91
I'm sure using just one coax cable for the speaker wire will work habnab. It just seems to me anyways as not the right stuff for the job in question. What I do find as peculiar is that you say you're a solid core copper kind of guy yet only the center conductor meets your criteria while the braid, well, that's just a bunch of tiny wires woven together. I didn't see any spec's whatsoever on Belden's site for the effective gauge of the shield. As far as interconnects go, for nominal lengths, you can get away with just about anything.
post #22 of 91
Quote:
Originally Posted by habnab View Post

I'm going to run 1 RG-8 coax cable to each speaker.
i.e. using the 10AWG solid center conductor as the hot wire ("red"), and the outer braid as the return wire ("black").

This seems majorly F'ed-up, as you're using a 50 ohm transmission line to connect a sub-1 ohm amp output to a speaker at a few ohms. I think you'll immediately fry the amp, or attenuate a major part of the signal, or something. And I don't know what the capacitance is, but that could be a problem as well.

This would be a good case for an expert on this topic to pipe-up.
post #23 of 91
Thread Starter 
Quote:
Originally Posted by habnab View Post

I'm going to run 1 RG-8 coax cable to each speaker.
i.e. using the 10AWG solid center conductor as the hot wire ("red"), and the outer braid as the return wire ("black").

Quote:
Originally Posted by bmwf1fan View Post

This seems majorly F'ed-up, as you're using a 50 ohm transmission line to connect a sub-1 ohm amp output to a speaker at a few ohms. I think you'll immediately fry the amp, or attenuate a major part of the signal, or something. And I don't know what the capacitance is, but that could be a problem as well.

I strongly suspect that using RG-8 coax, as speaker wire, is safe. To my eye, there's nothing risky in the specs.

The capacitance is similar to the other cables in the review at:
http://www.audioholics.com/reviews/c...-cable-faceoff

Quote:


This would be a good case for an expert on this topic to pipe-up.

If I'd said anything too off base, they'd show up in mass.

Cheers
post #24 of 91
Quote:
Originally Posted by bmwf1fan View Post

This seems majorly F'ed-up, as you're using a 50 ohm transmission line to connect a sub-1 ohm amp output to a speaker at a few ohms. I think you'll immediately fry the amp, or attenuate a major part of the signal, or something.

No, as long as the signals are in the audio frequency range and not RF, and assuming the capacitance of the coax does not destabilize the amp.
post #25 of 91
Thread Starter 
Quote:
Originally Posted by Chu Gai View Post

I'm sure using just one coax cable for the speaker wire will work habnab. It just seems to me anyways as not the right stuff for the job in question. What I do find as peculiar is that you say you're a solid core copper kind of guy yet only the center conductor meets your criteria while the braid, well, that's just a bunch of tiny wires woven together.

You need both a hot wire, and a return wire.

That is a system.

Quote:
I didn't see any spec's whatsoever on Belden's site for the effective gauge of the shield.

Belden shows DC resistance for the braid that is a bit higher than that of the solid center. The solid center is 10 AWG, so all it takes to get a guess as to the AWG of the braid is to look at the DC resistance of 11 or 12 AWG wire.

OTOH and AFAIK, AWG is a physical distance measurement, whereas DC resistance is an electrical measurement, so even that comes up short.

Quote:
As far as interconnects go, for nominal lengths, you can get away with just about anything.

That's an interesting question. Namely: does coax cable, used for interconnect purposes, have more leeway with it's specs than coax cable used for connection to speakers?

Anyway, I'm comfortable with using RG-8 coax as my speaker wire.

Whereas I'll never again twist together shellacked 18 gage solid wire, for use as speaker wire, as I think that there may be some danger for the amp there.

i.e. danger beyond the issues of possible shorting with the abuse that speaker cable takes over time, depending on the specific amp it is used with.

I was perhaps lucky, in that my amp survived the experience.

One other option to approaching using 2 solid core wires twisted together is to buy some tubing that will just fit over one of them, and which will pretty much eliminate any chance of shorting. And maybe some tubing to go over the finished 2 conductor cable.

But I'd have no clue as to if the specs for such a cable would be safe for most amps.

The open question in all of this is does a solid core hot wire have any effect on the SQ?

That is the question that was carefully avoided in the writeup at:
http://www.audioholics.com/reviews/c...-cable-faceoff

For myself, I think that solid core may have some SQ merit for the hot connector.

I'm not all that worried about have a mixed cable. i.e. a solid center with a braided return.

Not that it matters to me, but AudioQuest also seems to share that view.

Anyway, I'm going with that for both my speaker coax wire, as well as my interconnect wire when I place an order for it in the upcoming week.

Cheers
post #26 of 91
Thread Starter 
Quote:
Originally Posted by habnab View Post

Anyway, I'm comfortable with using RG-8 coax as my speaker wire.

Finally got feedback from Allied on my order for RG-8 coax cable and spade lug connectors and tubing. Shows up Monday.

So I've got this weekend, or longer, to figure out which/what Canare RCA connectors will work with either RG-8X 15 gage solid, or 13 gage solid, and place an order for wire and at least one RCA connector.

Cheers
post #27 of 91
Thread Starter 
Quote:
Originally Posted by habnab View Post

Whereas I'll never again twist together shellacked 18 gage solid wire, for use as speaker wire, as I think that there may be some danger for the amp there.

For what it's worth, I give myself 5 attaboys for out of the box thinking here.

Which as far as I know is my own. I certainly never saw such an idea from anyone else before trying it myself.

My thinking of using RG-8 coax as my speaker wire is not my own, and is an idea that's been around for more than a decade.

But few, if any, seem to have actually have done it.

Cheers
post #28 of 91
I see no real advantage of using that Belden RG-8 Coax cabel. It may probably as good as any 10 gauge ordinary speaker wire. A good speaker wire made of OFC (Oxygen Free Copper) or UP-OCC (Ultra Pure Ohno Continuous Casting) copper is easier to use and manage to route the cable. The power (in terms of Voltage and Current) travels through the speaker wire is great enough so the shielding that you are referring doesn't seem to have any benefits of the such coax cable to use as a speaker cable. Parts Express sells Dayton 10AWG wire for about $100/100 ft.

The interconnects may benefits from Coax cables given that the signal travel in the interconnects are very small to have any EMI / RF effects on the cable. Shielding is more practical and a must for interconnects from RF / EMI / interferrences from other audio / video equipments.

But if you must really use the Coax for speaker cable, I don't see any reason why not. It's not gonna blow your amp or receiver as long as there is no short between the shielding and the center core.

"The open question in all of this is does a solid core hot wire have any effect on the SQ? "

How could a solid core hot wire (say 10AWG) can have a better SQ than the multi-stranded 10AWG speaker wire. What kind of magic can it do to improve the sound quality? I think it can't do any better than what Monster speaker cable can do for you because Monster speaker cable doesn't do any better than a probably made speaker cable does).

For all my multistranded 10AWG wire, I solder the bare wires before terminating the wire in the speaker connector so essentially a solid core 10AWG on both ends (wires in the insulation is multi-stranded wires though). The current will flow through multi-stranded wire just the same as the single wire as long as there is no break in any of the strands.
post #29 of 91
Thread Starter 
Quote:
Originally Posted by megasat16 View Post

I see no real advantage of using that Belden RG-8 Coax cabel. ... Parts Express sells Dayton 10AWG wire for about $100/100 ft.

The interconnects may benefits from Coax cables given that the signal travel in the interconnects are very small to have any EMI / RF effects on the cable. Shielding is more practical and a must for interconnects from RF / EMI / interferrences from other audio / video equipments.

But if you must really use the Coax for speaker cable, I don't see any reason why not. It's not gonna blow your amp or receiver as long as there is no short between the shielding and the center core.

"The open question in all of this is does a solid core hot wire have any effect on the SQ? "

How could a solid core hot wire (say 10AWG) can have a better SQ than the multi-stranded 10AWG speaker wire. What kind of magic can it do to improve the sound quality? I think it can't do any better than what Monster speaker cable can do for you because Monster speaker cable doesn't do any better than a probably made speaker cable does).

For all my multistranded 10AWG wire, I solder the bare wires before terminating the wire in the speaker connector so essentially a solid core 10AWG on both ends (wires in the insulation is multi-stranded wires though). The current will flow through multi-stranded wire just the same as the single wire as long as there is no break in any of the strands.

I find that cheap bare spade lugs work better than simple soldered ends.

$4.52 per 50 count of 3M 10-12 AWG spade lugs from Allied. (search for: 3M terminal block fork)

Cheers
post #30 of 91
Quote:
Originally Posted by megasat16 View Post

Parts Express sells Dayton 10AWG wire for about $100/100 ft. How could a solid core hot wire (say 10AWG) can have a better SQ than the multi-stranded 10AWG speaker wire. What kind of magic can it do to improve the sound quality?

He wanted to save money so he's thinking that it'll end up a lot cheaper than buying any 10 AWG wire. I think that's why he's doing this, though to me it seems a lot of trouble and the cost difference is insignificant. But heck, if it's part of the fun of this hobby...
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Receivers, Amps, and Processors
This thread is locked  
AVS › AVS Forum › Audio › Receivers, Amps, and Processors › DIY Audio *analog* cables