Multichannel Speaker Cables

I've found plenty of discussion about speaker cables, but not much about cables for speakers that have each channel driven separately.

I've designed and built a pair of active three-way speakers. I'm using a miniDSP to generate the crossovers and I feed its analogue outputs to the multichannel inputs of a 7.1 receiver via DIY Canare star-quad interconnects. These are unbalanced, although this could conceivably change sometime in the future.

I now intend running six "two-conductor" speaker cables from the AVR's speaker outputs to a wall plate (these cables will be 5' long) and then under the floor through my crawlspace to the speaker positions (a further 15' and 25' of cable). Three cables will supply three channels on one speaker and three on the other.

I would like to hear your collective wisdom with respect to how I should construct and configure these cables. With six two-conductor cables bundled together for 20', should I expect inductive interference?

I could build each cable by sandwiching two 16 gauge "zip cord" cables together, twisted and encased in braid, to make a DIY 13 gauge star-quad cable, but I would have to do this six times. I don't mind doing this if there would be a benefit, but would the six star-quads interfere with each other anyway?

Are there viable alternatives? Would six runs of 14 gauge plain Jane cable be good enough anyway? Since most of the cable will be in a crawl space, I'm not concerned about anything being plenum-rated.

Thanks for your help with this!
 
Dont know if youve read any cable threads but most end up nasty. First, do you believe fancy, expensive cables make a difference? Most people that do wont believe the physics that tell them different and they wont even try a blind A/B comparison, which I would suggest you try first. Get some one to randomly swap 2 different cables and see if you can consistantly tell which is which. As far as the gauge goes, At 25' I would use the 14 gauge zip for the woofers and the 16 for the mids and tweeters.
 
Most people that do wont believe the physics that tell them different and they wont even try a blind A/B comparison, which I would suggest you try first. Get some one to randomly swap 2 different cables and see if you can consistantly tell which is which
:eek:
Try under the floor - spaced - tangled
Try on the floor - spaced - tangled


Try six "two-conductor" - try twelve "single cable"

Try directly from the amp - try amplifier + wall socket

Try amplifier + wall socket -under the floor - spaced - tangled

Try directly from the amp 3' groovy- separated
Try directly from the amp 20' long

Try suspended from the ceiling - parallel - spaced - twisted -tangled

Try suspended from the ceiling - parallel - spaced - twisted -tangled + binding posts on speakers

Try suspended from the ceiling - parallel - spaced - twisted -tangled +with no binding posts on speakers

Try 14 -16 - 18-20 AWG cable tangled -parallel wall socket/ no wall socket

but DON't put 'em under the floor
:hypno1:
 
cbdb, I'll take physics over snake oil and fairy dust every time. A/B testing is hard to do objectively and it can be hard to reach a solid conclusion if the differences are subtle and one's acoustic memory is short.

I agree with your recommendations with respect to minimum gauges. As it happens, I found a good deal on some good quality 12/2 copper and was intending using three runs of that to each speaker...until I started thinking about electromagnetic interference.

Pico, for test purposes, I have odds and ends of cable meandering across the carpet. The sound quality is very good, but I'm concerned that replacing them with nice, tidy bundles will lead to some deterioration as a result of EMI. I think my wife will object to hanging them from the ceiling, except perhaps at Christmas if I hang lights from them.

So what does the science tell us?
 
An option is to use an 8 contact SpeakOn connector. One cable per speaker. Split the cable at the amplifier end. The engineers at Benchmark think SpeakOns sound better plus they are inexpensive and lock in place.

That's what I'm using to get from each speaker to the wall. I bought a 5' made up cable and cut it in half. I'll use one piece for each speaker, running from a SpeakOn socket on the speaker to another one on the wall. Incidentally, that was a C$75 cable, plus I need extra connectors and sockets!

Once I'm in the wall and under the floor, it really starts getting expensive. A 50' made-up cable is about C$260; a pair of 25' cables would run me C$300. Just bare 13 gauge 8-conductor cable is C$3.80/ft. In contrast, I bought a 250' spool of 14 gauge 2-conductor for about C$90.

SpeakOn connectors make a very clean and simple connection, but the cost adds up very quickly. I can hold my nose and swallow the cost if it would really make a difference, otherwise it's much cheaper to run plain Jane stuff where it's out of sight.

Nonetheless, I much appreciate the suggestion. It's easily something that I might not have considered.
 
An option is to use an 8 contact SpeakOn connector. One cable per speaker. Split the cable at the amplifier end. The engineers at Benchmark think SpeakOns sound better plus they are inexpensive and lock in place.

Speakons are good if you have to repeatedly connect/disconect speakers, like PA gear thats moved around. Sound better? Than what, how? If its the engineers they should have some data.
 
Just looked at there website, no data, just the marketing blurb to sell the speakons. If you can turn the binding posts so they clamp the wire, its the same conection as in the speakons, but you get rid of the 2 of the 3 conections you have using the speakons, so how is 3 connections ( inside the male speakon, inside the female, and between the 2 speakons ) better than 1 ?
 
That's what I'm using to get from each speaker to the wall. I bought a 5' made up cable and cut it in half. I'll use one piece for each speaker, running from a SpeakOn socket on the speaker to another one on the wall. Incidentally, that was a C$75 cable, plus I need extra connectors and sockets!

Once I'm in the wall and under the floor, it really starts getting expensive. A 50' made-up cable is about C$260; a pair of 25' cables would run me C$300. Just bare 13 gauge 8-conductor cable is C$3.80/ft. In contrast, I bought a 250' spool of 14 gauge 2-conductor for about C$90.

SpeakOn connectors make a very clean and simple connection, but the cost adds up very quickly. I can hold my nose and swallow the cost if it would really make a difference, otherwise it's much cheaper to run plain Jane stuff where it's out of sight.

Nonetheless, I much appreciate the suggestion. It's easily something that I might not have considered.

Did the speakers come with speakons? I would use the 14 gauge (16 if you want for mid/tweet) and binding posts on the wall that screw down, then solder the wire to the back of the binding post. Speakons are also relatively large.
 
Last edited:
Did the speakers come with speakons? I would use the 14 gauge (16 if you want for mid/tweet) and binding posts on the wall that screw down, then solder the wire to the back of the binding post. Speakons are also relatively large.

The speakers are DIY. I decided to use SpeakOns because it's one connector at each end of the cable instead of six binding posts and there will be one cable across the floor instead of three. There will be no possibility of someone misconnecting and zapping a tweeter with a woofer signal. Also, the speakers have dual woofers, so with 8-conductor SpeakOns I have the option of driving them individually in the future if I were to build an 8 channel amplifier.

My SpeakOn wall sockets have solder tags for the under-floor cables. My wall plate at the amplifier end has screw-down/banana terminals. I'll be using the screw-downs, which look like they'll work quite well. I'll feed each conductor down a banana hole and then out through the screw-down hole. When they're all cinched down tight, I bind all the cables together with heat shrink so that they won't go anywhere if one or two should work loose somehow.
 
It's not just the final speaker cables that must be connected treble to treble, bass to bass etc.
Ensure all cables from source to speaker are clearly identifiable and/or impossible to connect wrongly. I'd mark each cable at both ends to ensure errors cannot occur even while making up the cables.
For me, it's very much the case; if it can be connected wrong, it will be connected wrong. YMMV.
 
It's not just the final speaker cables that must be connected treble to treble, bass to bass etc.
Ensure all cables from source to speaker are clearly identifiable and/or impossible to connect wrongly. I'd mark each cable at both ends to ensure errors cannot occur even while making up the cables.
For me, it's very much the case; if it can be connected wrong, it will be connected wrong. YMMV.

Ain't that the truth! Last week I tore down my test gear and reassembled it on new furniture. Just doing that I got two connections wrong.

I've just spent the morning running all the cables under the floor and was careful to mark each end of each cable as I cut it to length. I had my wife double-checking as I went.

It becomes particularly mind-boggling when working on the back face of a wall plate or connector and everything is a mirror image of the front face. I've just marked the back surfaces of my wall plates so that I get them the right way up as well as L/R the right way around.

Just to provide the ultimate test, the miniDSP outputs are labelled 1 to 8, I'm using DVD-A multichannel inputs on the AVR, the AVR outputs are labelled FL, LS, LSB and so on, and the SpeakOn connectors are marked 1+, 1-, 2+, 2-, etc. My 14/2 cable has black and red conductors, while the SpeakOn cable has black, white, red, green, yellow, orange, brown, and blue. I've drawn up a table for future reference!

When everything is ready, I'll start with just the woofers connected!
 

Attachments

  • labels.jpg
    labels.jpg
    81.4 KB · Views: 92
Twist each pair of conductors with different turns per foot. That minimizes crosstalk. Once the cables' resistance is 5% or less of the loudspeaker it is thick enough.

I do have special cables made up for me to do large installations. It takes a forklift to move the 1,000' spools!
 
Thanks for the tips. Those labels are a good idea. Too late for twisting - I've just finished and tested the installation, and in any case I used 14/2 cable runs which were in a round sheath and would have been hard to twist. I could space the cables an inch or two apart for some of the run through the crawlspace if that would help.

I measured the resistance of one conductor from the point of connection at the receiver to the point of connection at the speaker (so banana to SpeakOn) and got 0.3 Ohm. I had to use a long piece of speaker wire to extend my multimeter wires across the room. I used both conductors in a fairly long 16 gauge cable, which measured 0.2 Ohms. This seems suspicious to me...a double run of 16 gauge with higher resistance than one run of installed wire that has four additional connections en route? However, my meter has a resolution of 0.1 Ohms, so I'll put it down to limitations of accuracy.

Twenty minutes with a few test tracks suggested that there might be a small improvement in clarity and a slight reduction in sound stage. I credit the former to getting the speakers onto resilient mounts and the latter to different room positioning, although they could both be positioning. It's hard to A/B test this reliably and it could all be in my imagination, but some experimentation with speaker and listening positions should be worthwhile.

In the meantime, I'd welcome any comments on other reasons why the sound stage might have changed. I'll double check all my polarities, although I'm confident that they are correct.

Thanks for everyone's input so far...much appreciated!