[Bosk]

Hey guys.

I'm planning on getting started making my own DIY cables for the first time, and am trying to decide what kinds of cables to build.
I need a pair of .5m interconnects & 1m speaker cables for my computer-based speaker system, which consists of the following gear:

Chaintech AV-710 > Optical cable > Lite DAC-AH > Trends T-amp > Fostex 126e bookshelf speakers

Right now my thinking is to make the interconnects using solid-core silver 28AWG wire in cotton jackets, using some reasonably high-endish Cardas or Vampire RCA's. I'd love to use WBT silver RCA's but they're just too much money.
For the Speaker cables I'd like to go with some kind of heavy-gauge copper wire in teflon jackets, maybe stranded because solid-core would be too inflexible. I plan to use the same kinds of bullet plugs they use in Nordost cables, the ones that are made from a single piece of metal and compress down as they're inserted into the outlets.



What sort of geometry should I be aiming for with these cables?
For example, braiding seems to be popular but seems like a lot of work and requires more wire. I've also noticed there's various types of braids.
I want to go as minimalist as possible - no techflex or anything else the cables don't absolutely need - both for the purposes of cost saving and also because I'm a great believer in KISS. (Keep It Simple Stupid)

Also, is there a simple method for testing weather or not I need shielded interconnects in my setup?


Thanks very much for your advice, it's much appreciated.

 

[dura]

I prefer braiding on the IC's; the sound of braided cables always seems more open to my ears, compared to coax. Braiding also provides shielding; because the individual wires go in and out the braid there is not much antenna-working ot pick up interference. Solid silver will give you a very pleasant clear yet smooth sound; at least my Kimber Silver Streak does.
Those Nordost plugs are nice indeed. Also consider Eichmann's bulletplugs for the IC's, though they are a pain in the *ss to solder for someone like me who is not very experienced.

 

[Bosk]

Cheers for your response dura, it sounds like braiding is definatly worth the trouble!

 

[qusp]

yep x2 on the braiding and eichmann bullets.. they are a bit tricky to work with but they are really great connectors. most places allow you to add metal housings too, which look nicer and are more durable. when soldering them its best to connect them to a RCA female to act as a heatsink. that way if you take a little too long, you dont melt the dielectric. vampire are also good but both vampire and cardas are very bulky. so with the eichmanns you want to get the eichmann bullet bayonet plugs. they are the ones that expand like the ones you mention. good luck its great fun and very rewarding.

 

[Juaquin]

From an engineering standpoint, transmission lines (cable) geometry is one of the most important aspects of the line. Of course, the effects are debatable for a line of this length. For interconnects, I would probably braid it (if building your own wire), since the extra length won't be much of a problem. For longer wires (1m+), I would probably just twist it, then maybe put a sleeve over it.

 

[linuxworks]

Quote:

Originally Posted by Juaquin /img/forum/go_quote.gif From an engineering standpoint, transmission lines (cable) geometry is one of the most important aspects of the line. Of course, the effects are debatable for a line of this length.

no, there is no 'debate'. it is simply NOT a transmission line. we are talking AUDIO FREQUENCIES here.

there is not now or never was any debate about this. check with any EE. audio frequency is anything but 'hard' and there is no transmission line effect here. none.

its all in your head.

braiding for audio is only going to make things LOOK nicer and that's about it.

 

[digger945]

*crickets chirping*

 

[Juaquin]

Quote:

Originally Posted by linuxworks /img/forum/go_quote.gif no, there is no 'debate'. it is simply NOT a transmission line. we are talking AUDIO FREQUENCIES here. there is not now or never was any debate about this. check with any EE. audio frequency is anything but 'hard' and there is no transmission line effect here. none. its all in your head.

Any conductive line is a transmission line. To what magnitude it is effected by the equations of transmission lines, yes, is dependent on the frequencies and the length of the line - and the geometry and material. Like I said, for audio applications at these lengths, you aren't going to see any difference in geometries (or even materials, really). However, as an Engineer you can never say there is NO effect - you can, however, say the effect is negligible in such-and-such situation, like this one.

I simply suggested braiding for interconnects because it looks nice, and twisting for longer runs just to simplify things - from a SQ standpoint I acknowledged that there was no tangible difference.

I completely agree so I'm not sure what you're arguing about. Oh, and I am an Electrical Engineer (well, an undergrad getting ready to graduate next year), so I do understand the basics here.

 

[tomb]

Test an amp with interconnects using a nice braid vs. single lead wires stretched out all over - there's a difference. Take the stretched out wire jumble and put it in a steel tin - still not much difference. Shorten the leads to minimum and/or braid them, and there's less "hash" in the noise and distortion graphs. This is pretty easy to see and do with RMAA, too.

 

[linuxworks]

I will do that test, tom! I am expecting an audiophile m-audio firewire 'rmaa test box' in the next day or so. I've been wanting to get something serious for test gear for a while now.

if I see a difference, I'll admit such and report back; but I cannot imagine audio frequencies having 'transmission line' effects. impedance? swr? loss due to skin effect? hardly!! just NOT at audio frequencies.

but I will give the test gear a try and see what happens. I'll use my best amp (ppa) as the test bed and then try some braided wire as well as coax and even straight hookup wire.

do you have such tests you can post?

one last (key) question: tom, are you saying you can HEAR any diff or only that some very sensitive test gear can measure it?

I'm serious. I've actualy done far more work in RF than I have in audio and its always been my understanding that RF 'matters' and audio does not when it comes to this stuff.

 

[tomb]

Quote:

Originally Posted by linuxworks /img/forum/go_quote.gif I will do that test, tom! I am expecting an audiophile m-audio firewire 'rmaa test box' in the next day or so. I've been wanting to get something serious for test gear for a while now. if I see a difference, I'll admit such and report back; but I cannot imagine audio frequencies having 'transmission line' effects. impedance? swr? loss due to skin effect? hardly!! just NOT at audio frequencies. but I will give the test gear a try and see what happens. I'll use my best amp (ppa) as the test bed and then try some braided wire as well as coax and even straight hookup wire. do you have such tests you can post? one last (key) question: tom, are you saying you can HEAR any diff or only that some very sensitive test gear can measure it? I'm serious. I've actualy done far more work in RF than I have in audio and its always been my understanding that RF 'matters' and audio does not when it comes to this stuff.

There is always a difference between what you can hear and what you can measure. However, if all we were interested in was only what we could hear, we wouldn't be building half the stuff we build.

Proper braiding minimizes interference by optimizing the ground sink around the signal wires. If this wasn't true, then grounds would make no difference at all, neither would shielding. Yet, we know this isn't true. We also know that running lengths of signal wire willy-nilly in different lengths and different directions ends up in worse signal quality than wires that are run properly.

Do your tests, but I'm not sure I understand why there's a question about this.

 

[nikongod]

I dont belong in this discussion.

 

[qusp]

sorry but I doubt the appropriateness of the audiophile by M-audio for this test. the audiophile itself doesnt test so good. its hardly high end IMO sorry but in my experience the m-audio products are very consumer grade. thats their niche. all the same it may well still be able to find some differences. how accurately rendered those differences are is another question.

 

[linuxworks]

some people use that firewire box for testing. serious testing.

its already better than 99.9% of our ears. it may not be good enough to test 'test equipment' but its fine enough to be BETTER than our ears.

so my reasoning is: if the equipment can't 'see' any diff and its better than my ears - the end effect is inaudible.

have you seen braided wiring in pro audio studios? I'm not aware that pro audio guys buy into this. if they don't, why should I?

 

[Pars]

Quote:

Originally Posted by qusp /img/forum/go_quote.gif sorry but I doubt the appropriateness of the audiophile by M-audio for this test. the audiophile itself doesnt test so good. its hardly high end IMO sorry but in my experience the m-audio products are very consumer grade. thats their niche. all the same it may well still be able to find some differences. how accurately rendered those differences are is another question.

Amb uses the same box for RMAA testing, as do I (based on his recommendation). Where did you see reference to it not testing so good as you put it? When I have run loopbacks on it it has tested quite well. For example, The β22 Stereo Amplifier, including the loopback baseline testing just the box itself.