[Ulas]

I have two versions of ICs made with conductors of Cardas tonearm wire (33awg copper litz) terminated with Eichmann RCA plugs. I prefer either of these ICs to any of the 20 DIY ICs I have tried as well as commercial offerings from cheapskate to cost-no-object.

Both are low capacitance measuring around 3 pF/ft. Pictures are posted at http://www.audiocircle.com/index.php?action=gallery;area=browse;user=1053 and brief construction details follow. These ICs are unshielded and are not be suitable for use in high EMI/RFI environments.

Version 1: The wire is woven into a strand of polypropylene hollow braid (water ski rope). This IC is very robust (it’s made of rope!) but it looks kind of funky.

For each IC you will need: One length of 3/8” polypropylene hollow braid; Two lengths of wire, each 1.075 times the length of the rope plus 4 inches; One pair of RCA plugs.

Remove the blue plastic barrels from the RCA plugs; remove the setscrews from the barrels. Thread the barrels on the ends of the rope.

Using the flame of a match, melt and fuse the strands of polypropylene at the end of the rope. When done correctly, this will form a collar that will just squeeze through the threaded end of the barrel but not pass through the narrow end. (The fusing operation may require some practice.)

One wire is woven in and out exactly following the path of a right-spiraling strand of polypropylene; the other wire follows the path of a left-spiraling strand. After weaving, except for its size and color, the wire should be indistinguishable from any of the polypropylene filaments. It is easiest to start weaving from the middle of the rope and work to each end where wire is tucked into the center of the braid and passed through the hole in the fused collar.

Tin and solder the wires to the RCA plug. Work quickly and use a RCA jack as a heat sink; too much heat will melt the plug.

Screw the barrel onto the plug; insert and tighten the setscrew. The fused polypropylene collar provides strain relief against pulling and the setscrew prevents twisting.
 
Version 2: The wire is suspended in a lattice of Teflon tubes. This IC looks exotic and it is more fragile than the other.

For each IC you will need: Four lengths of 15awg Teflon tubing, each 1.035 times the length of the finished IC; Two lengths of wire, each the same length as a tube plus 4 inches; One pair of RCA plugs.

Thread each wire through an empty tube. Plat the tubes in a square sinnet such that the two wire-filled tubes crisscross each other in the same plane while the empty tubes crisscross in an orthogonal plane.

Start with the four tubes alternating filled and empty and repeat the following sequence until done.

Take the leftmost tube and cross it OVER its TWO neighbors.
Take the rightmost tube and cross it OVER its ONE neighbor.
Take the leftmost tube and cross it UNDER its TWO neighbors.
Take the rightmost tube and cross it UNDER its ONE neighbor.

Tin and solder the wires to the RCA plug. Work quickly and use a RCA jack as a heat sink; too much heat will melt the plug. A few dabs of hot glue applied to the tubes where they exit the barrel will secure them to the plug. Finish it off with some heat shrink.

[Occam]

Very, very cool!

Is that ski rope hollow on the inside? I'm having a hard time visualizing its construction, even looking at the pictures.

[Ulas]

Yes, it is hollow. I think that’s why it is called “hollow braid”  

Polypropylene rope is readily available in two forms. The yellow hollow braid I used can be found at Wal-Mart, Home Depot, etc. packaged in a variety of diameters and lengths. It is popular for use as a water ski towrope because it is strong, shock absorbent, and it floats. The hollow braid is also available in other colors; such as black, green, orange, etc. from marine supply houses that cater to commercial fisherman. I also use hollow braid to protect my litz-wire speaker cables rather than TechFlex. Polypropylene is a much better dielectric than Nylon, or even Teflon. It is less expensive, too.

The other form is a rope-lay construction, commonly black with an orange stripe, and often sold as “truckers rope” because it is used as a tie-down. It is not useful for use in making interconnects.

[beat]

Those look great Ulas,
the version two reminds me of something I have seen before...Is that your own braid design? Does it matter how tight you get it? Just curious. I'd like my next IC project to involve some braiding of some sort.
Cheers,  Beat

[JoshK]

Ulas, love your waterski rope IC idea!  that is great!  I am a former water ski addict (3time/week year round in WA state).   Seriously, very ingenious.  You are not the first to say that Polypropylene is a great dialectric, can say I have tested that out, but I believe it to be so coming from the sources I have heard it from.    You should really try polymenstride though.     (a throwback joke at Response Audio Mods).

[Ulas]

Thanks, Beat. The plat is a square sinnet. The tightness is dictated by the thickness and flexibility of the strands.

As far as I know, my design is original. Ray Kimber, of Kimber Kable, has a patent for the use of the four-part square sinnet with speaker cable, but his design employs four conductors: mine uses only two with two empty tubes to maintain the geometry. What is it with these cable companies who trademark and patent every thing under the sun?

In my experience, tightly braided ICs, such as the braiding used by Chimera Labs and others, is detrimental as it increases the capacitance. Chimera’s ICs measure a whopping 46pF/ft.

[corwin99]

Yes.. tightly braided ICs definately have a higher capacitance than loosely braided or unbraided ICs, however, they will also have a lower inductance... with a looser braid you do lower the capacitance, but you raise the inductance... what a dilemna.  

[Ulas]

Corwin, what’s the dilemma? As a professional audio cable maker you should know better than I which parameters are the most important for different interfaces. With a high impedance circuit like interconnects, where there is negligible current flowing, maintaining a low inductance is not important. But maybe you know something nobody else knows.

BTW, did you ever measure the capacitance of your ICs?

[corwin99]

Ulas,
I apologize for butting in your thread... i realize that my input was not welcome. So much baiting... sorry I just don't want to go there. Cable philosphy is a dangerous subject.

And yes, I have measured the capacitance of my cables.

[Ulas]

Corwin, it’s not cable philosophy, but well-established cable science. It is my understanding that capacitance is the dominant parameter in ICs whereas inductance is dominates in speaker cables and power cords. But I could be wrong. So, if you know something different, please enlighten me.

I don’t mind you butting in. Although I started this thread to share my DIY IC designs, I don’t own it. If you have questions or something of value to contribute, you are welcome to participate. Even if you just want to spout nonsense, I can’t stop you. If you, as a cable maker, believe lowering the inductance of ICs is important, then, by all means, lets hear your ideas and we’ll discuss them.

I asked about your capacitance measurements because I don’t see them on your web site.

[ctviggen]

When you say "dominates," what do you mean?  Do you mean that it's best to minimize capacitance for ICs?  Or that techniques for making ICs cause capacitance to be the highest important contributor (certainly not the highest value -- resistance I assume is a lot higher than capacitance) to impedance?

[cryotweaks]

Great job!

I have an idea for you to consider.  Instead of nylon rope, try cotton.  Cotton has a lower dialectric constant  than even Teflon!

On another note, I also noticed in one of your photographs that your preamp(?) has a ground connector.  It might be interesting to make a shielded cable and instead of attaching the shield to the signal return, you would attach a drain wire to the shield and ground it to your chassis connector (or better yet, house ground).

Cheers

Mike

[corwin99]

Ulas, like you said, the generally accepted cable "science" is that with interconnects a low capacitance is most desirable, and with speaker cable a low inductance is most desirable. However, in my tests of cables, I have not always found this to be true... additionally, I have also found that cables with nearly identical LCR values can sound very different. I do not know why this is the case, as cable "science" says otherwise. But this is my experience. The reason for not including measurements on the site is because I feel that they are misleading in demonstrating the performance of a cable based on my observations.

I have not done nearly as much testing with speaker cables, so I can't say the same about LCR values as they affect speaker cables.

FWIW, i also prefer a looser braid for most of my cables, though this wasn't always the case, which is kind of why i jumped in with the inductance. I thought that may have been the factor in my given case. I have also found that a tighter braid rejects noise(EMI/RFI?) a little bit better, however not enough in a noisy area.

[Occam]

Ulas, like you said, the generally accepted cable "science" is that with interconnects a low capacitance is most desirable, and with speaker cable a low inductance is most desirable. However, in my tests of cables, I have not always found this to be true... ...

Corwin,

I find your use of the word science in quotes to be absolutely charming. Science would require an understanding of how measurable parameters such as inductance, capacitance, and impedance quantitatively effect frequency response. This can be demonstrated by a few simple calculations. Should I assume that 'science' does not?
I do agree with you that cable perfomance might not be fully charaterized by either science or 'science'.

Would you please show by formulas and calculations specifically what impact differing amounts of capacitance, inductance for an arbitrary impedance would have on frequency response?

This is the Lab Circle. Specifically, it is my Circle as I am its moderator. As a vendor, I hold you to  higher standards and expectations than I expect from a hobblyist. If you are unwilling or unable to provide a more detailed (mathematical) explanation, after referring to those parameters but notihing of substance to back it up, please take it elsewhere.

[Ulas]

Ctviggen,
The first one. Typically, IC designers strive to minimize capacitance and speaker cable designers strive to minimize inductance. There is a see-saw relationship between inductance and capacitance, as Corwin said, but it is not a concern for ICs. It is a big issue for speaker cable where a high capacitance can create problems with some amplifiers.

There are important trade-offs and dilemmas in IC design but capacitance verses inductance is not one of them. A biggie is the decision to shield or not. Others are aesthetic concerns, such as, appearance, flexibility, strength, robustness, etc. I made an IC with a single strand of 40 micron, 6N copper. I thought I had engineered a secure mechanical connection between the RCA plug and the tubing that held the wire, but the cable didn’t even survive the trip from the workbench to the stereo.

I tried a few ICs where the fine conductors are spiraled around a core and then mummy wrapped with Teflon tape. In addition to being difficult to make, they were stiff and unreliable because of the weak mechanical connection between the tubing and the RCA plug. I developed the Version 1 IC to solve those mechanical problems. The result is a strong, flexible, mechanical base that will support any fine wire in a fixed geometry. I tried fine silver wire, 30-awg magnet wire, cryoed CCC wire, and my preferred tonearm wire. I tried them with the wires parallel (spiraling in the same direction) and crossing (counter-rotating spirals).

The Version 1 ICs, as I make them, cost about $65 for a 1-meter pair. Most of the expense is in the Eichmann RCA plugs; the Cardas tonearm wire isn’t cheap, either. If you substituted Radio Shack magnet wire and RCA plugs (the ones that come 8 to a package and are colored red, yellow, blue, and white), and used a smaller diameter polypropylene rope (I think 5/16” or ¼” would fit the barrels of the RS plugs), you could make 4 ICs for around $10.

My Version 2 IC trades flexibility and strength for better control of the geometry and a slightly more transparent sound.

Cryotweaks,
It’s polypropylene, not nylon rope. The dielectric constant of polypropylene is similar to dry cotton. I don’t like cotton because it is hydroscopic which tends to destroy its good dielectric properties. Also, cotton rope stretches and is not very strong. I can’t imagine how one would weave fine wires into clothesline rope or how one could secure it to an RCA plug. If you can solve all those problems I’m sure everyone here will be interested to learn how you did it.

The ground wire you see is to ground the turntable. I don’t like shielded ICs and I don’t use them because I live where RFI is not a problem. The only electrical problems I have are getting power from a rural power grid and lightning. For the former I have a propane-fired, standby generator and for the latter I have a professionally designed and installed lightning protection system with spikes on the roof and numerous interconnected ground rods and copper plates buried around the periphery of the house.

Corwin,
Cable “science” does not say cables with nearly identical LCR values will sound the same. Where did you learn that? All cable “science” says is, all other things being equal, an IC with a higher capacitance will have more high frequency attenuation than an IC with a lower capacitance.

I don’t know how you can draw any firm conclusions about the effects of inductance in ICs when, as you pointed out, inductance and capacitance are intertwined and it’s very difficult to change one without affecting the other and still keep every other physical and electrical property the same. Besides, somehow I doubt you have the equipment or knowledge to make the required measurements.

[Occam]

Lordy.... are cable parameters like the weather? Everbody talks about it but nobody actually calculates what the real impact is? (on a purely objective level)

http://www.dact.com/html/ac_calculator.html
http://www.digitalprosound.com/2001/07_jul/features/resist3.htm
FWIW

Sooooo...... how much capacitance would have to be in that interconnect feeding a 47k ohm load to be down 3db at 40kHz?

[corwin99]

Ulas, you are right, cable science does not state that all cables that test the same will sound the same, I worded that wrong. My apologies. I meant to say that some EE's (on other forums mostly) claim that all cables with nearly identical LCR will sound the same because cable science dictates...

The equipment i measure with is an LCR meter that i purchased at the recommendation of Occam, the mod here (original meter did not measure L and was not precise enough).. no signal generator or anything fancy like that. I had tested different tightness of braids including unbraided and found that in some cases, a tighter braid sounded better, I thought it _might_ have to do with the change in inductance and capacitance. This is by no means scientific, and I do not have an EE background and cannot recite all the formulas and calculations, Occam. I respect that you would expect me to as a vendor, however I researched building cables by reading other people's studies about the effects of capacitance and inductance on cables and some experiences of different types of braids and materials used by other hobbyists and EE's. I decided to take my experiences and developed my line of cable from that... mostly using the examples and studies I read as a reference, and then listening to the differences as i made small adjustments to each to suit my tastes.

I did not realize that this thread was in the lab circle, my bad. Oh yes, I forgot to add... the reason for the quotes around "science" was because science is using theory to explain phenomena, but it doesn't always explain why cables can sound so different.

[neergan]

I use the VenHaus DIY Fine Silver Interconnects.  Instead of teflon tubing for the core, I used 1/4" cotton rope. First, I wound a layer of teflon plumbers tape around the rope. Then, 30 ga. .9999 silver wire (Michael Percy Audio) at 2" spacing, then two layers of teflon. I used Vampire 800/CB connectors (pricey, but less than Eichmanns). These are the best I've heard for anywhere near the price. Outstanding in every way. The only change I would make is to use heavier (yellow) teflon tape for the outside layer.

[Ulas]

Corwin,
According to you, some EEs (on other forums) claim that all cables with nearly identical LCR will sound the same because science says so. You say cables with nearly identical LCR sound different because your perception says so. I speculate that the EEs will say your perception is flawed because it doesn’t match their analysis. You say science is flawed because LCR measurements fail to predict your perception of how a cable sounds. I don’t think there will ever be reconciliation between those two extreme positions and neither side is doing anything to advance the science or the art of music reproduction.

The reason for not including measurements on the site is because I feel that they are misleading in demonstrating the performance of a cable based on my observations.

If you believe that, why did you invest in a LC meter? Why do you measure your cables? If you believe measurements have no value why did you bring up the subject in this thread?

[corwin99]

Ulas, yes... that is the typical argument.

As for why I purchased the LCR meter, I was curious. I had used my old Capacitance meter which did not have enough resolution to measure to 1 pF... decided that I wanted to know how my cables tested more precisely... because some other cable manufacturers listed these values. So I bought one and tested them to see if they were comparable... they were.

I do believe measurements have value, but I don't believe that they are absolute.