DIY Interconnects review

Category: Cables

In an attempt to live up to a comment I made on a thread that was discussing interconnects and the over inflated costs, I agreed to do some research and try to build my own interconnects. I could then use fact in my discussions of the value of the manufactured cables. I did intense research for about four months, reading as much as I could on Audiogon, Audio Asylum, web sights dealing with designs by Chris VenHaus, Jon Risch and Allen Wright. I also attempted to go to all the manufacturers web sites and read what they had as a philosophy. Please forgive me for the length of this post, but it was important to me to be thourogh in my methodology so as to help those reading this understand the mission.

It should come as no surprise to most of you, there are as many opinions on what makes a good interconnects as there are companies. Dispite this fact, most every interconnect made today both profesionally and DIY have common methodogy in construction. Most every design I read about starts with a core, has any number of wires coming in a wide array of materials that are wound around the cord. Then the dielectric is created, again using most any material one can dream of. This is followed by most companies and DIY’s using some type of EMI/RFI sheilding. Then some type of fancy casing is used so that the cable looks nice. The casing is generally connected to the connector with shrink wrap and often some type of reinforcement to allow for the users who pull out their interconnects by the wire and not the plug.

Great I thought! Clearly these are all logical components that are being used, and of course I figured there was some science behind all this construction. In fact Cardas is using the Golden Ratio (first dirived in Ancient Greece) so this must be good. Not only is every wire praportional to the other, but the shielding (all nine layers) are also applied using this ratio. Now don’t get me wrong, The Golden Ratio is often used in my profesion of Architecture, it was conceptualized my DiVince the was incorporated in Michelangelo’s St. Paul’s Cathidal, and it has been used ever sense, but audio cables???

Well I took all the ideas I could find and went off to my (three) local Radio Shacks to buy their entire inventory of gold plated RCA connectors. I also picked up a better solder station, a good de-soldering devise, copper wire of every size as well as robbing my local hardware stores of foam backer rods, Teflon tubing, Teflon tape, cottom rope, poly-something tubing and any other thing that looked cool. Total cost of components was roughly $300 ($20 each pair of interconnects) and enough stuff to supply every friend I know with interconnects.

So my first generations started with the pholosophy that more is better, so I built some interconnects using 14 ga wire twisted around a 1/2” teflon tube. I tried one two and three wires per positive and negitive run in equal and unequal combinations. I then built the same concepts using 18ga, then 22ga, 24ga, 26ga, 28ga and finally 30ga. I then repeated the process using a 1/2” Teflon and 1/2”, 3/8” or1/4” cotton rope. On these mock ups I wanted to hear the wire and core intraction with no other variable so I used only cotton thread to hold the bare wire to the core. In each case I listened for a short time to get some concept of what these designs were doing, but only in comparison to one another. I was not yet listening for musicality nor even technical aspects, only an overall concept. I often dragged my friends into this to get a second opinion, but really this was easy to distinguis.

FIRST MYTH EXPOSED: WIRE DOES MAKE A DIFFERENCE! Do not allow anyone to tell you otherwise. If they clain wire is wire, then they either have no experience with a decent system, or no experience trying cables.

It became clear to me that the thicker wire was woolyer , fuzzier and slower as a general response. I quickly drew my attention to 22ga and smaller wire for the next round of experiments. This round was bringing in a preconceved notion of my own that goes against vertually every cable design I researched. Almost without exception the only concept explored by manufacturers and DIY’s alike is twisting the wire around the core. My friends and I theorized that this allows the cable to be more flexable without fatuiging the wire. (like a spring) Spacing suggestions very, but almost without exception the cables are twisted. I used to own Nordost Valhalla cabling for about five years. The speaker cables are made up of forty very fine conductore that run parrallel to one another. This made me explore running my interconnect conductors parrallel as well. So in my next set of mock ups I tried variouse combinations of cable sizes and wire numbers either twisted or straight. Straight won every time! This was not subtle, it was very clearly less congested, and smoother sounding. The winding tended to make the sound feel constraned and mechanical. My personal explaination is wires running parallel on a two dimensional plane have only one chance to interact. Twisting around a core creates a three dimensional interaction and in sometimes many points of interaction. Someone once said, “keep it simple” and I think this advice fits pefrectly here. It also became clear that a single conductors in each direction was prefered every time over multiple conductors. (One conductor could be multiple stands bound as one conductor) The multiple conductor mock ups tended to be edgier or brighter, so I ended up focused on single conductor designs with 30ga or 28ga positive wires and 20ga to 24ga negitive as my prefered sound. I also determained I much prefered the cotton core to the Teflon.

SECOND MYTH EXPOSED: TEFLON SOUNDS BRIGHT! There has been quite a bit of discussion over the past couple years regarding the posibility that Teflon is adding a brightness, or a ringing to the sound of cables. This was bourn out in my mock ups to hold some validity; so next up was determaining what the best dielactric might be. I tried Teflon plumbers tape, cotton, paper, electrical tape and even wool on bare wire, teflon cased wire and enameled wire.. Again I tended to like the cotton, and it became even clearer that Teflon was an issue. Next step was to determain sheilding (if any) Now I know much has been talked up by manufacturers about sheilding, and I supose if you live next to a radio transmition tower… but if you look inside your components, wire is never sheilded, and these wires are running in, out and around transformers, power supplies and digital curcuits. I can not think of a harsher enviroment for EMI/RFI than this. So I must admit I entered this stage with serious reservations. I tried woven steel shielding, woven copper sheilding, copper foil wraps, aluminim foil wraps as well as some mock ups based on the Cardas Golden Ratio. Every one of these had some influence on the sound, but in my opinion always negitive. Shielding is a very dificult thing to do in a repeatable fashion for hand made cables, and I suspect one large area of cost for cable manufacturers. I ended up with no sheilding as my prefered direction.

Next up, two parallel wires run at different spacing. I decided on my own that even consistant spacing was important, so I tried a number of methods to build up a larger cotton core. This included multiple cotton ropes tied together, then in some cases covered in a cotton jacket with wire applied to that. Then another layer of cotton jacket… What a hassle, and clearly not repeatable. I did determain that their was a limit to spacing. Wires further than 7/8” apart began to lose cohenrency and naturalness to the sound. Less than 5/8” sounded brighter and more conjested. I settled on 3/4” spacing as the perfect compromise.

THIRD MYTH EXPOSED: INTERCONNECTS DO NOT NEED A CORE! I must be among the dumbest men on this world, luckly I have a lot of company. It finally hit me, interconnects can be flat. This was ground breaking for me. I’m sure everyone else already knew this, but hang in there, I’m a bit slow. I had purchased this 9mm (approx. 3/4” wide when layed flat) cotton sleeving from Audio Consulting. (Way too expenxsive) It comes as a flat sleeve, so I simply took my two wires, now in cotton jackets and sewed them into each folded edge of this tube. This was a sinple easy way to get a constant and repeatable 3/4” spacing for my wires. It also meant the least dieletric interface with material, leaving the majority of the dielectric as air. Air of course is the best dielectric, but bare wire hanging in space equidistant apart is hard to do, so given the makeup of cotton weave, it allows for about 90-95% air, with very little of the cotton even touching the wire. Perfect, NOT! Ever try to attach a cotton sleeve to an RCA connecter with two super fine wires holding it all together? The issue became painfully clear as I began breaking 30ga wire trying to devise a solution.

FOURTH MYTH EXPOSED: THE SCIENCE AND LOGIC OF A CABLE DESIGN IS OUT THE WINDOW THE MINUTE WE TRY TO MAKE THE CABLE INDISTRUCTABLE! Just take a moment and page through your favorite audio magazine and look at all the cables. Heavy outer jackets made from Kevlar, carbon fiber, Poly-something-man-made, indestructible Teflon uncoated… You get the point. Now look at the interface between this less than desirable dielectric jacket and the RCA. Heavy rubber shrink wrap as a minimum is used, often in layers. Air has a dielectric value of 1.0, Teflon is 2.0-2.3, (I know, they say 1.2 but I’m using a scientific table, not “what they say”) cotton is 1.3-1.4 and rubber is at best 3.0, Poly-something –want-a-cracker is 3.0-4.0 and so on. This means they claim to go through all this science and end up using some less than perfect materials because they need to, in order to make them stand up to us, the guys who pull out our interconnects by the wire. “OH YES YOU HAVE TOO, WE HAVE ALL DONE IT!” And yes, it makes sense that any manufacturer would build the product to withstand some abuse. So despite all the careful planning, the connector and wire interface is an issue.

I ended up trying a few designs; the one I settled on (if I felt the need to address this issue) was some very puffy cotton “piping” from a fabric store. (Thanks to my friends wife, I have not been in a fabric store before, there are some very interesting people hanging around in them) I used a ¾” rope cut in half (length wise) and put the flat cotton interconnect in-between the halves. I then wrapped a thread around the rope to hold it together. At the RCA I shrink-wrapped the last four inched of cotton rope to the RCA. This allowed for the rubber to remain as far from the wire as practical. It looks a bit goofy, but it’s remarkably strong. I did not feel the need to go to this extreme and instead pulled the 9mm cotton sleeve over the RCA and put a piece of electrical tape around it. Half the tape’s width hitting the RCA and half hit the cotton, I figured electrical tape was at least twenty times thinner than rubber shrink wrap, so it should have one-twentieth of the impact to the sound.

SO NOW WHAT? Well the design is complete, now I needed to determine the wire material and what RCA’s to use. For RCA plugs I tried Audio Note, Cardas, Connex, DH Labs, Eichmann, Monster, Neutrik, Radio Shack and WBT. There is not a great deal of difference between the Cardas, DH Labs, Neutrik, Monster and Radio Shack. The Audio Note and WBT were a bit more natural and certainly less congested, but in further testing I could not find a repeatable difference between Radio Shack’s $5.00 gold plug and WBT’s $70 connector. To my surprise the Eichmann was quite a step up. The sound opened up with clarity and speed not found in the others. (I should add that I used both the Silver Bullets and the standard copper connectors. I could not hear a significant difference between the $20 and $50 connectors) If anything I may prefer the copper, a bit more natural and not as bright. I guess it all makes sense that the Eichmann sounds different in that they are the only one with a different design. They use a pin for both the positive and the negative connection, so rather than a whole circle of signal connection on the negative it comes to single points. I assume this makes for a cleaner, more effortless signal transfer??? At any rate the Eichmann was my choice. I ended up with one set of silver and one copper out of default. If I had not already owned the silver I would have used copper for both sets of interconnects and saved a few bucks.

Selecting wire was a bit tricky, in that it can cost a lot of money and there are still a lot of variables. Rather than go through the entire process I will skip to the wire I settled on and hope you believe if it’s possible, I tried it! As a very quick overview I’ll make some general observations. Copper tends to be full, rich and makes for a nice neutral sound. Silver adds PRaT and dynamics not possible with copper. Silver also adds clarity and extension not possible with copper. Silver is also brighter, edgier and at times has an artificial tonality on the top end. Gold was the best of the copper and silver without the bright thin signature of silver, and the clarity was even better than the silver. I attempted to use silver on the negative run in order to keep the cost down, but the improvement between gold and silver is substantial, and worth the expense (my opinion) I did not try stranding silver and copper together nor did I try silver coated copper, (like Nordost) so there is plenty of room for experimentation in these areas.

I ended up using all gold wire, and because of cost I made some assumptions (meaning I did not try every combination possible) For the positive run I ended up using three strands of 99.99% pure 30ga gold wire (not twisted but loosely bound together in a 2mm cotton sleeve) and for the negative conductor I settled on one strand of 24ga 99.99% pure gold wire. How I concluded this set up was some trial and error and some faith on the articles I had read that fit my experience to this point. If I had the money I would have tried a cable using three 28ga for the positive run, I suspect this would be even better, but I used what I had already purchased (bulk purchases lower cost substantially) If someone want to buy me wire…

Much has been said about gold having a significantly lower conductance than silver and copper, but given the distance of one meter, it really is not an issue!!! Gold has a very low resistance as do the others, so my selection was based on sonic quality rather than scientific characteristics. The side benefit (but not my reason for choosing gold) is it does not tarnish and therefore sonically degrade like silver and copper.

Sonically gold was clearer, smoother, richer and more natural sounding than the other two. People have claimed gold is slow or too warm. Neither of these comments matches my experience. One reason for using 30ga was the pace it produced in my trials. This held true with the gold as I used it, and as I commented on in the paragraph above I suspect 28ga gold would continue to have excellent PRaT. I also suspect anything heavier would indeed tend to slow and warm beyond my goals. The interconnect I built is every bit as dynamic and fast as my Valhalla’s were, but with a much fuller and richer top end. I have not experienced the warmth issue either; in fact I believe the 28ga gold positive strands would be a possible improvement if it did add additional warmth. As of this writing I am still trying to determine if this is necessary, a bit more warmth would be welcome, but it might be too much if I tied to achieve this. Over the past two weeks I have spent over six hours every day enjoying music I have not pulled out in years. In ever case, I have no thoughts of what more to do or what is lacking. No fatigue at all and so musical! So altering this design might only degrade the success, but it might be better. Money would answer this…

SOOOOOOO you ask. How does it sound? Well thank you for asking, but as often with me, before we go there let me add a couple more comments.

As I said earlier, I owned the Nordost Valhalla cables for many years. Prior to settling on them I auditioned most every medium and high end cable made at the turn of the century. It is no exaggeration to say I became an expert on cables during a two year period of testing cables. I also learned there are a lot of cables being sold today that are inherently flawed sonically. Over the last six years the amounts of cables that have come to the market have increased many times over, and my personal experience has proven that most of these cables are no improvement over their predecessors. As the sonic quality has improved very little, the costs have climbed to an astonishing degree.

This is not to say that there are not some truly magical cables that have been developed. This topic has become the source of heated discussions on this site as well as Audio Asylum. I have heard so many people who call everything they do not agree with snake oil or an illusion. These people I have turned my back on for comments like this neither are mot worthy of answering nor correct. Cables make differences, and I have a very extensive background to back my claims. When I hear someone say they would never pay more than $100 on cables, I roll my eyes. When I hear people claim they can build a cable for $100 that matches the performance of, (fill in the brand) I get physically sick. These comments are made from people who either have very limited experience or a system that is not capable of appreciating what is possible. In fact this entire experiment began from comments like these. People claiming they can build a cable as good as (fill in the blank) are endless. Let me say hear and now, they are right. I too experienced the level of AudioQuest, Cardas, JPS Labs, Kimber Kable, Shunyata or any other brand you prefer. This is not too difficult, but is that the goal, to match the sonic quality of inferior or flawed cable designs? If that is the goal, I can offer any number of design options that will match that of the mass market cable companies. In fact if the goal is to match the performance of 90% of the cables available, then I suggest you can do this for $25.

My goal here, and in my discussions of this topic is to match or better the best Interconnects available. I have invested countless hours and money to build the best system I can afford and/or justify. I spend the time needed to fully understand the topics I discuss here and I offer my opinions to educate not brag or be the top dog. My ego is fine without this audio site, and therefore I enter this hobby for the enjoyment of music. I would hardly classify myself as an equipment junky or an ill informed audio nut.

My personal experience says that Nordost Valhalla’s are the quickest, most transparent cable made. The bass is so highly defined and the impact of the bass is beyond reproach. The downside of Valhalla is it tends to be thin and bright on the top end. My Reference cables today are Kubala-Sosna Emotion. These are the most natural and neutral cable in all frequencies that I have experienced. They are the best overall performer I have ever heard, the super black background, perfect extension and definition are beyond reproach, but they do not carry the slam and dynamics of the Nordost. A truly magical cable is the Purist Dominus line (both Rev B & C, with fluid or Ferox) of interconnects. The midrange and upper bass are so rich and believable they force you to overlook the lack of clarity on the top end and the lack of bass compared to the best in this area. These cables cast a three dimensionality to the soundstage like nothing I have experienced before.

I bring this up not as some have accused me of bragging, but to help you understand what I was attempting to match in this experiment. I wanted the speed, dynamics, bass and slam of the Valhalla along with the naturalness and blackness of the Emotions and the colour, hues and dimensionality brought out in the Dominus. My goal was to combine the three best I have owned. Along the way I discovered the formula for Valhalla’s detail and speed. I discovered a midrange equal to the Dominus and I found a naturalness of Kubala-Sosna but combining these attributes and then bettering it was not as easy.

So here’s the deal, and a source of disagreement with many at this site. People say:

“Cables do not make a difference.” This is wrong, naive and foolish and clearly not arguable in a logical discussion.

“Cable manufactures are overcharging for their cables.” Yes they are crazy expensive but I can not say they are over charging. My simple little cable takes hours to build, and if a guy is to market, travel, inventory, purchase the equipment needed to produce a repeatable product… then I’m not sure they are so inflated. Yes I agree it’s silly, but my design has about $655 in materials, (approx. $300 in 30ga wire, $285 in 24ga wire, $40 RCA’s and $30 in cotton sleeves) maybe eight hours, so what is that worth? If any comment is true it would be the low and mid priced cables is the scam. Charging a few hundred dollars for a $25 value is the rip off. The top end is closer to a bargain. So yes they are all pricy, buy not out of line if a business is expected to make a profit.

“People buy expensive cables to brag or whatever.” Give me a break; people buy expensive cables to improve the sound. Not everyone wants to do what I just did, and yet they deserve the best sound. It is not the consumer’s fault that it costs a lot, we continue to demand more from our systems, and the manufacturers continue to push the designs. It’s just a fact, cables cost money to build.

“You can get the same sound from a DIY cable at a fraction of the cost.” Absolutely correct, I have tried most every cable made over the years, and I can unequivocally state that in my experiments I would be willing to guess I matched most every cable made. That is not to say this is a good thing, but just that I too can build glare, brightness, poor imaging, bloated bass, slow methodical pace… I can also build bloom, warmth, blackness and silence, natural… I can build it all, good and bad, so? So what does that prove? Nothing to me. If I had not heard the Nordost Valhalla, and the Kubala-Sosna Emotion and The Purist Dominus I would have no way to judge what I was building. If all I had heard was a Cardas Golden Reference, then I guess I would be satisfied when I had duplicated it. It is not right to chastise those who continue to explore the possibilities just because someone thinks cables are too expensive.

We can not judge life without experience. How do you know you’re happy if you have never been sad? How can you judge success until you have failed? You can not, and it hold true in audio too. Without having heard most every cable made, I could not properly judge the results, and thus I could not have ended where I am. So for those who choose to pock the fire, fine, but it is you that is missing out, not me…

Enough soap-box, now let’s talk about the sound.

The sound is quite close to a perfect blend of the three best cables I know of. This design is stunningly fast, excellent slam, highly emotional, very full and rich, fully open and extended. The imaging is tight with great depth, dimensionality and soundstage. The background is more liquid and blacker than any cable I have experienced. The clarity is the best I have ever heard. There is a tonal richness that is hard to believe. Piano is tonally perfect although one of my local A’gon friends wants a bit more bloom around the notes. The colour and hue around all stringed instruments is so perfectly natural and portrays an endless palate of tones.

To this point I still had one complaint, the high frequencies although fully extended and extremely clean, had a slight haze remaining around the highest frequency notes. This haze was not apparent when compared to the other cables; it only became apparent when I tried one more component in my design. I used a Bybee Slipstream Purifier ($75 each) on the positive conductor at each RCA. (Four total or $300 on one pair of interconnects) When I compared the one with the Bybee vs. the one without the haze became apparent. When I added Bybee Slipstream Purifier to the second set of interconnects the top end remained extremely transparent but now without the slight edginess first sensed. The entire presentation took on yet another dimension of realism. This got me to the goal; I now have bettered every single aspect of my system, but all at once! The final produce costs me almost $1000 in materials.

So in the final analysis

As many of you know my health continues to be a problem and time is all I do have. Not only that, but as many also know I have been forced to keep down sizing my system repeatedly over the years. This experiment is successful enough that I will be selling my manufactured interconnects and will be able to pay a few more medical bills!!! As for speaker cables and power cords I do not believe it is as beneficial to build my own, at least right now. My lessons learned in this project do not apply to either the power cords or speaker cables, so it would take another extensive research and a lot more money. My next project is to research power conditioners. This is where I think the true robbery is taking place! So if you care to contact me about the interconnect project or my coming research I would be happy to share what I know

Below are links to some of the research and material suppliers I used. I hope this thread will help settle some of the controversy about cables, and possibly help some find a better cable than they ever dreamt of.

Chris VenHaus's DIY Silver Interconnects: The original web site for the DIY interconnects
John Risch's DIY speaker cable and interconnect web site
Allen Wright
Allen Wright’s web site
Make Your Own Audio Cables: some reading on DIY cables
Radio Shack Magnet Wire ICs: great sounding diy interconnects on the cheap
The $2.99 Silver Wire Trick: a quick rundown of DIY interconnects by the late Dr. Harvey Gizmo
The Science of Cable Design;
Wire Gauge Calculator
Interconnect and Speaker Cable Design:Part IandPartII at
Dielectrics Constants
Dielectric Constant Reference Guide

VH AudioSupplies, wire, RCA, casings...
Reference Audio Mod’sAudio Consulting wire and cotton sleeving
Parts ConnexionHuge audio parts supplier [url/]Tweek Geeks[/url]ERS/ RFI absorbent sheets
Home Grown Audio: Supplier of Teflon insulated silver wire, braided wire, and silver cable terminators
Myron Toback, Inc. Supplier of uninsulated 30 gauge fine silver wire
McMaster Carr:Supplier of Teflon spaghetti tubing for insulating bare wire
A-M Systems, Inc:Supplier of Teflon insulated silver wire
Michael Percy Audio. Supplier of all types of high end, exotic, and esoteric components
Parts Express Suppliers of various wire and cable terminators as well as other components
Scientific Instruments Services, Inc.Gold and silver wire.
Surepure gold, silver, copper wire
JD this is another of your amazing posts!
I've had a basic clue regarding what you were up to (via the email) so when I read the title (in forum-summary) I figured this one was yours. Then as I scrolled through, my suspicions were confirmed before I ever read your name at the end. Thanks for sharing your experiences; this is very inspiring. This information belongs in the archives for sure. Now to scroll back up & finish reading!
What a wonderful journey you have been on! Thanks for all of the information and insight. When will we see your product offering?
Excellent write-up JD! I enjoyed reading about your thought-driven experimentation.

I have similiar results to back you up on the usage of gold wire and Eichmann bullet RCA connectors. I made my own IC, cloning a Dual Connect IC, using a pair of 30ga. 99.99% pure gold wires per leg, inserted in a pair of teflon tubes. I also used a third teflon tube which remained empty, and braided the three tubes together, and termintated the braids with the Eichmanns.

The ICs sound excellent, tonally spot on, quite fast, huge soundstage, and in general, similiar to your results. They may not be the "airiest" though, so perhaps I will give the cotton tube thing a try, and use three conductors for the positive, and one for the negative. Regardless, I am sold on using gold wire and Eichmanns, and yes, unfortunately these are expensive to make.

I'm just confused on the 9mm diameter cotton tubing your are using. Isn't that only .35" when layed flat? Did you mean 19mm tubing by chance? Then it would make sense. I would love to try the Bybees too, but alas out of my jobless budget at this point!

Again, thanks for a wonderful writeup.
Post removed 
While it's great to read about someone reading / learning and doing on their own, i'm sad to learn of your illness JD. Needless to say, i wish you all the best!!!

Other than that, most of what you've been playing with has to do with impedance, EM ( electro-magnetic ) fields and velocity factors. As far as impedances go, what you've found to work well in your system may / may not be universal for others. This has to do with the stability of the circuitry and the nominal impedances of the mating components being used. Not all components share the same stability or mating impedances, so results can vary here.

The nominal impedance of the cable is directly affected by the geometry of the conductors, which also directly effects performance in the area of EM fields. As you mentioned, shielding also effects the sound. Most of the time, this will be in a negative manner due to the improper implimentation of said shield. That's because the shield alters not only the nominal impedance of the cable, but also the EM field AND the velocity factor. As i've mentioned before, shielding IS beneficial so long as it is properly implimented. Having said that, i've also stated that i don't know of any commercially available cabling that is properly shielded : )

As far as the velocity factor goes, this has to due with the conductivity of the signal and how fast / slow the signal travels through the cabling i.e. how long it takes the signal to go from point A to point B through the cable. Once again, this has to do with impedance, geometry, materials used, etc...

Not only do impedance bumps ( small changes in conductor geometry / spacing ) alter the load that the components see, but it also alters the velocity factor in that area too. As such, you can have a cable that has non-linear electron flow ( in terms of speed ) in certain areas of the cabling. Not only can this happen at various lengths, but it can effect certain frequencies more than others.

As such, there's not one aspect of cable design that dictates how things will turn out, but a combination of factors that work together as a whole. You seem to have juggled quite a few variables and come up with something that you like, which is fabulous. I encourage others to try this and i applaud JD's willingness to both share what he has learned and the motivation to undertake such a project !!!

As a side note, i would recommend sticking with conductors that are no heavier than 20 gauge at the very heaviest. That is, use conductors that are smaller in gauge ( higher in number ) than a 20 gauge ( 20, 22, 24, 26, etc... ) at the max. Use solid wire, not stranded. Pay attention to the nominal impedance ( spacing between conductors ) and stay on the inductive rather than capacitive side. Conductor length for each run of conductors should be as close to identical as possible. Using multiple conductors for each polarity makes this harder to achieve, not to mention trying to maintain a consistent nominal impedance via keeping all of the conductors far enough apart from each other in an equi-distant fashion.

If you follow those basic rules, regardless of the type of conductor or dielectric used, even your worst cables will end up sounding pretty decent. Maybe not perfect for your system or to your personal tastes, but at least pretty decent and nothing that you couldn't live with. Obviously, the better the materials that you use and the more consistent that you are in assembly techniques, the better the performance potential.

Keep up the good work JD !!! Sean

PS... Both Bob and i have cable burners and would be willing to "burn in" a set of these for you. You might be surprised at the "before / after" difference that one can hear when doing this.
Welcome back my dear neighbor. Your most informative posts have been greatly missed the last few months.

I am the "bloom" guy JD mentioned. When he had the Nordost cables last year, his system was way too analytical for me. I knew something was possible with what he had achieved, but the edginess on the top was too much. It was clearly the Nordost that were not to my liking.

JD then started a transition to the KS Emotion. Tonal coherency became incredible as did resolution. So many top-end systems impress, but so few resolve the performance like JD's. What he had gained here, and during a system down-size to boot, is only due to the hard work and passion of an obsessed perfectionist.

I still pushed JD to try and get a little more dimensionality and decays other than for me to donate my Callisto! And he clearly got some of this magic with the Dominus but with the Dominus went away some of the very very low-level details that so few of us ever get a chance to hear.

This week JD invited me to hear his new DIY cables. When I visit JD, it is the always same 3-4 tracks - it's like a trip to Arkansas. I am almost a Lucinda fan now!

To discuss the amount of information that is now present that was not before would require a novel. Massed strings in particular are wonderful. The degree of delineation is a breakthrough achievement. This last time I could truly appreciate what JD has finally pulled off .... and for so much less expense than many of us have done.

I am eager to hear his cables in my system. Perhaps this will be the birth of the JD Signature series. When JD and I have some definitive information to share on this, I will add it to this thread.

This is an awesome post. The insight and knowledge gained is priceless. Hats off JD. Well done!!
Super job and a great read some very interesting and informative conclusions......again great job!
Thank you for sharing this with the rest of us JD. This passion involves a big learning curve for most of us, and your write-up was very very helpful, and carries a lot of credibility.
I suspect most of those who claim that cable doesn't make any difference simply do not have resolving enough systems, have not explored the issue directly, or haven't yet learned to listen. I hope your review will open some ears for the better. Best of luck to you.
Top notch reading, I am sure you also had goodly amounts of the hidden joy to be found in our audio passions.
I to am sorry to hear of your malaise,and wish you Godspeed, and many more hours of enjoyment pursuing your interconnect nirvana.
You have now awakened my do it yourselfers curiosity.
Well, JD, thanks once again, a legendary posting! (It's been a long time since your "Winter's lessons" -- remember that post years ago?).

I have been experimenting with diy IC's along similar lines -- but in no way as thorough as you.

Finances permitting (one day) I will be kindly requesting you to make me yr ICs (I need sets of two -- 2 sources, bi-amping) -- rather than try to make them myself.

I will pm you with some ideas from a recent (lightweight)search on power.
Wow, not only an awesome and informative post from JD but it also dragged Sean back to the forums. Great to have you back Sean and nice job JD, enjoy the cables!
JD, first, I'd like to say thank you, and second, to wish you better health soon. What I think you've discovered is what most cable manufacturers already know. While there are certainly some that strive to produce quality products at different price levels, many are really only producing one or two very good cables, and the rest of their lines are just money-makers cashing-in on their reputation. This of course is a common business model, but it doesn't make you feel better if you've bought a $300 cable that is essentially the same sonically as a $50 DYI job. I've gone through much of the painstaking research as well, but on power cables instead of IC's. And some of the conclusions you've reached hold true in that area as well. But since I'll be introducing my own line of power cables soon, I won't get into that, as I don't want to be seen as pimping my products here in the Agon forums.
Just a quick thought about the costs involved in cable production; I think folks should be aware that the costs you see in a retail environment don't reflect at all the costs major cable companies see. Part of the reason you see so many 'twisted' geometries is that cable is far cheaper when manufactured in very large quantities, and the twisted design can be churned-out easily by the companies that produce these cables for most of your audio cable vendors. Of course, even at a low per-foot price, buying 30-40 thousand feet of cable is still a large investment, and must be repeated for each product in the seller's lineup. So this does make for a substantial investment, with an equal risk if the cable isn't favorably reviewed by one of the glossy mags. Of course, if the cable(s) is a success, then the company will use up it's stock, and it's NEW MODEL TIME!!! Which in most cases is just as well, as we Audiophiles are a fickle lot, and need new products to drool over and fantasize about. (Some of us at least, be honest). Unfortunately, not all of these new models will be an improvement over their predecessor, but merely different, or sometimes (ick!) worse. (But usually 'prettier').
Okay, so I expended alot of words to say 'caveat emptor'.
Thanks again JD, you've made it that much easier for us to trust our ears, and ignore the marketing hype.
I want to thank you all for the extremely kind words and compassion. It is people like you who make this site work and keep us coming back. Sadly I continue to get so frustrated with the management at times and the often blatant self promotion by people who choose to cheat the unaware audio lover. Of course I also get very frustrated with the comments from people who proclaim to be experts but share no evidence to back them up.

Because of these experiences I have left the site once again, but wanted to share my experience with this project. I felt the knowledge I gained was too valuable not to give to my friends here at Audiogon. I also hope that some of the mean spirited will simply learn from this and with time grow up. I miss you all, but I simply can not physically deal with the anger, amoral and self promotion I continue to witness. I pray it does not destroy what was an incredible community, but I suspect a lot of the damage has already been done.

So onward…

Hello Bob my wonderful friend. Your email has been very goofy, and therefore I’m not sure you received my email that discussed my life after my “near death/seeing the light” event. I hope you got it, if not I can try to re-send it. I get delivery delay messages often writing you.

Viridian, you summed up perfectly why I share this experience.

Zargon, no such luck. I have no intention of marketing this, but given the information shared, someone might. I might consider building some for my friends however. Working with 30ga gold wire is no picnic, and it too frustrates me, is their a theme here? I must be becoming a very frustrated person…

1markr, 9mm diameter 2πr = circumference, or 28.26mm. Laid flat it’s 14.13mm or .56inch. I suspect the way I’m using the sleeve it’s wider than 9mm, and my ¾” is outside dimension not inside, so you are correct to point this out. Measured on the outside it is ¾” maybe inside it’s 5/8”??? I will try to post pictures on my system page soon. I’ll be sure to notify you on this thread. Thanks for getting the clarification.

Sean, WHAT THE HELL ARE YOU TALKING ABOUT? Just kidding… I always am fascinated with your mind! Without question you have more knowledge and more understanding of this hobby than any other person here. My problem is I am 100% self taught from my Jr. High Electrical/science education and therefore a lot of the science is beyond my knowledge. I know just enough to get in trouble, but in all sincerity I become confused very quickly. What I would like to know from you is; what is the correct shielding method?

Hi Greg another of my long and cherished friends, thank you. The Winter Lessons is many peoples favorite too. I went back and look it over last year. It’s amazing how far we have all come since the early years of A’gon. This site has helped so many of us!

Hi Arni, yes the learning curve is huge and often the teacher is the one who shouts the loudest. This has always been a problem at this site, but not as bad as AA. I figure the best education is a complete one and I believe others can figure out the truth if they are allowed to hear real experience and not some who claims…

Tplavas, good luck with the PC project. Your comment about allowing ourselves to trust our ears is true, but not easy. It is hard not to want to hear what is making the waves. As I stated, it is not possible to know what is best if you have not tried them all. Then we are able to make sound judgments. I feel this is an important part of our hobby, and a large part of the growing curve. We (the A’gon community) need to learn to stop criticizing people for learning. Just because one person doesn’t understand it doesn’t mean everyone else are fools. If we can expedite the process through sharing knowledge and help lead people in productive directions great.

As for the issue of me building these for people who do not want to build their own, I would consider it, but I’m not sure anyone would like my hourly rate. The last commercial building I designed four years ago I charged $250/hour and the last residential was $160. Of course I was good at that and this is not Architecture, but I would need to get something. I have a very hard time putting a cost on my time, so we would need to figure that out. I would not want this to become something that created stress and take the fun away from my love. Of course money can always help relieve the pressure.

For those who are so kind in offering your love, my health is sadly something that does not improve. I had a massive heart attack nine years ago (I was 40) leaving me with only 20% living heart muscle. I was expected to die then and certainly within five years. As my cardiologist told me when I last asked her if this goes on indefinitely she told me she could not answer. If there was a study of 100 people with my damage, 100 would have died within three years. If the study had 10,000 then 10,000 would have died within five years. If a million, a million would be dead. I just keep going (like the Energizer bunny) with small setbacks yearly and a major hospital stay every 18 months or so. These are the expenses that have destroyed our finances. This is old news, but each year we add about $20,000 of debt to remain in the house we are. We will look at moving in another year when our disabled son finishes High School. That will allow us to pay the debt back. If I die Julie is in good shape financially, so our goal is to stay within some kind of budget, and not get into too much trouble.

We have sold virtually everything we own and my stereo remains my last hold out. Given the fact that some days I can not even take my twenty minute nightly walk, I am not able to do the physical sports I used to. I ended up having to quit designing three years ago, so my music and writing (I am working on my second book) are my escapes. (I was lucky enough to do some volunteer work in the Architectural drafting program at the local High School, coincidently my old High School and the program that got me started) I do not know what I would do without this hobby and the wonderful friends I have met through this site. This is what life is about…

Lastly thank you to the MANY friends who have written me emails. I will get back to you! You all give so much with your kindness.

JD: To make things as simple as possible, a shield should not interact with the normal rise and fall of the EM field generated by the active conductors within the circuit. Gregm alluded to this phenomena in another thread.

If the shield is placed in close proximity to the EM field that the conductors are generating, distortions are introduced at a non-linear rate. How severe these distortions are has to do with how much the shield changes the nominal impedance of the cabling. This may / may not be consistent depending on either the frequency and / or the intensity of the field that the shield is interfering with. The more effective that the shield is within the frequencies of operation, the further from the conductors it would need to be.

Since shielding is directly related to both coverage area AND material density, some forms of shielding are far more effective and / or ineffective than others at certain frequencies. As a general rule, foils are more effective at very high frequencies whereas stranded braid ( ala the tightly woven braided shield in good coax ) is more effective over a wider range, especially lower freq's.

Combining the two can offer greater advantages over a wider spectrum, but once again, there are many other factors that have to be juggled in order for them not to create as many problems as they may potentially solve. Physical size, high production costs and product flexibility also come into play, which is why many manufacturers / cable companies do NOT market optimally designed cables. Sean

Fascinating quest for the grail... I've tried just a few DIY designs myself, and have been particularly impressed by great-sounding DIY ICs and speaker cables made of copper ribbon pulled through clear heat shrink tube (5 mil x 1/2" for ICs). Just curious whether you experimented with ribbon conductors and if so, what you thought of them. I'm intrigued that one of your parts sources offers both silver and gold foil.

Thank you Sean, this jives with what I expected. One thing I had a very hard time finding was any definitive explanation of how far the effective EM field goes out. I suspect it's virtually limitless, but there must be a way to predict the point where another material (shielding) will interfere. Also I read that some people were advising using a small capacitor between the shielding and ground???

Also I will consider your idea of burn it. I had thought about it, just never got that far...

Hello Dave, I did contemplate ribbon, but admit I did not try it. The main reason I would have wanted ribbon (which I did) was to maximize the surface area (shin effect) and minimize the slow core response. Skin effect from what I read is only a controversial theory, but it made sense to me, so...

The reason I avoided ribbon was difficulty of handling. If I was to receive benefit I surmised I would be best using something thinner than .01" (30ga) to receive the benefit of less core area. I was considering a ribbon of .005" (36ga) or thinner, but I suspect that only a machine could work with something so fragile and prone to folds, brakes and kinks. If I was a manufacturer of some type and had access to the proper equipment I suspect a wire .04" x .005" (18ga x 36ga) would be an excellent starting place. How you could keep this flat (no kinks) and remain in a low dielectric constant environment would take some work. Perhaps setting the ribbon on a flat cotton sheet, then using a dry mount paper sheet followed by another cotton sheet. Then heating the "sandwich" to seal the ribbon flat into a cotton/paper dielectric. Paper has a 2.0 value, so this would remain better than Teflon and not have the downside.

Now after discussing this I wonder...

It might be worth a try???


The reduced skin effect compared to round wire is what attracted me to foil as well. .005" thick ribbon is pretty easy to work with and will not wrinkle when heat-shrink tubing is collapsed around it using a heated clothes iron insulated by a paper towel. Thinner foils can be wrapped in clear packing tape in lieu of heat-shrink tube. Copper ribbon should be dipped in polyurethane or lacquer to prevent oxidation. You could try it with/without the poly insulating jacket or with paper, cotton, or even just air dialectric (with just the urethane coating & loom separators to guard against shorts). Email me if you want more detailed information.

I'm not sure how my foil IC/speaker cables stack up against top-of-the-line commercial wires, but I have preferred the ribbons over Audio Consulting silver litz-in-cotton (ribbons have better PRaT, dynamics, smooth midrange, and bass), Goertz, MIT Oracle 2.1, Synergistics Resolution Reference, & Audioquest Diamond 2x. I'm using inexpensive McMaster-Carr electronics-grade copper-alloy foil of undetermined purity. You probably have sources for better metallurgy.


My complaint about ribbon IC's is their lack of ultimate weight. They seem a bit dynamically 'flat' as well. But then again, I haven't heard a perfect cable yet, so everything is a tradeoff.
JD, congratulations on putting forth such an exhausting body of work in search of the truths we all seek!!! In my estimation, this is among the finest two or three threads in the history of Audiogon. Boiling it all down has required me to take about a day to read, reread, digest, ponder, and start all over again several times to take it all in.

I dare say that in my opinion, this thread should be regarded as a seminal work into actually understanding cable, and the various technologies that make good cable good cable, poor cable poor cable, and extraordinary cable extraordinary.

You have done all of us audiophiles a tremendous service through your efforts, and I for one, will not soon forget this. We are all richer. Hopefully, those who provide cabling products for the rest of us will use it as a springboard in further advancing the craft.

hank you again! Your friend, humbly...
We are in your debt. May your health improve. You have certainly enriched my life. Many thanks.
JD: The EM field of a signal carrying cable will vary with frequency and amplitude. Using specific geometries can reduce the intensity of the field, allowing shields to be placed closer to the conductors. Only problem is, these geometries also alter the nominal impedance of the cable, which may end up being something other than desired.

Like anything else, you have to juggle specific variables in order to achieve specific goals, which are really nothing more than a personal set of compromises. This is much the same as system building, which also results in a personal set of compromises and standards.

Dgarretson: Depending on the specific design attributes of your home-brew foil cables, i have no doubt that they could achieve staggering results. By keeping speaker cable inductance to a minimum, offering a short, straight, single path that minimizes skin effect, it would be hard to go wrong. The only problem with most foils is the lack of surface area i.e. gauge, but this is only a problem with speaker cabling and not interconnects.

As far as interconnects go, that same short, straight single path that minimizes skin effect is still a winner, but you really have to play games with the nominal impedance. The nominal impedance is dictated by the spacing and geometry of the conductors used. Since different mating components will have different nominal impedances, the spacing that works "best" for one set of components may not work optimally for another set. Increasing or reducing the spacing of the conductors can be used to fine tune the nominal impedance, which will alter the transient and bandwidth characteristcs between the mating components.

On top of that, you also have to start worrying about EM / RF interference with this type of interconnect. The geometries that tend to produce the most desirable impedances tend to act as an efficient antenna. In order to minimize potential problems in noisy environments, you now have to introduce other variables that will require further experimentation and design challenges.

Either way, it is GREAT to see others reading, learning, doing and reporting their results here. Sean

Hi Trelja, it was you who I referred to in the original thread as the person who I was honoring when I committed to doing this project. You my old friend brought this to me, and my deep respect for you gave credibility to your words. I therefor approached this project seeking the truth, as I struggled to keep my preconceived opinions out of the process.

I must confess I was surprised that I was able to discover such a good cable, and most pleased that it actually out preformed the top three I know of. This has started a cascade of thought in me. I went out and bought a book called, "Electrical Course For Apprentices & Journeymen." I figure it's time to at least understand the basics in that I now want to discover power conditioning, isolation and suppression. (So hopefully I will be able to carry out an intelligent conversation with Sean soon!)

Thank you Joe, you have awakened a very bored man's mind. This is a gift beyond anything you could have provided, for there is no worse punishment on this earth than being left without ambition.

I am very strongly considering building one more interconnect, this time using the 28ga positive. For what ever the reason, either burn in or(?) the cable has begun to sound overly bright. Every aspect is so perfect, but an edge has crept in over the past couple nights. I hope it's just part of a process of burning in and not an end result. At any rate, I should have an answer to this and other pressing questions (like war in Iran or North Korea, I just can not decide) by the time you are ready to proceed with your DIY project.

TA-DAAAAAAAH !!!! Persistence prevails!!!

Hopefully, this thread will encourage others to experiment and share their results. Kudo's once again to JD and all of the effort he put into not only undertaking this project, but also the persistence in posting this thread and KEEPING it posted. Kudo's to the others that joined in with supporting the free exchange of info here at Agon. Sean
Persistence indeed - you all need to witness this guy positioning a pair of speakers to the exact millimeter - persistence indeed.
Here's an experiment very roughly based on JD's design -- VERY successful!

Details: 24awg (0,4mm) enameled copper wire, each pole spaced 1,9mm apart (as per JD), standard WBT rca connectors; wire is soldered onto the rca, used pieces of cotton rope for strain relief at each rca.
Quick descriptive items: sonic clarity increased; dynamic impetus was better (or simply more evident). Another thing is, the musical presentation seemed more "sturdy" for wont of a better expression.

Note that I placed this wire between system volume control & top amp (i.e. the amp is fed a full-range signal BUT is cut at 80Hz on the speaker).

BUT, here's a singular occurence (as Sherlock Holmes would have said). Happy with what I'd done, I prepared a second cable. Same length (60cms) same wire -- but different rca connectors (good neutriks I had laying around). Placed this between cdp & volume.
Sound obtained absolute sonic clarity (I'm using this word as opposed to "congested, undecipherable lyrics...) i.e. the sounds (the loud ones along with the less prominent, etc) make excellent musical sense together...
BUT, my mid-bass energy was lessened...

OK, so the cdp feeds a full-range signal which then goes to two different amps. The cable probably created a hi-pass effect s/where...

Here's the "UNCANNY" bit: I changed places between the WBT terminated cable & the neutrik (no logical reason; I just didn;t know what else to do & I didn't have the equip to measure). The midbass energy returned; the other positive sonic effects remained however.

How's that for voodoo:) {It's not voodoo, obviously. I just haven't got down to measuring anything.

As an indication, the output impedance of my cdp is 21-30ohms across (constructor's data when cdp was new). Amp (top) is 10kOhm, bottom is 33kOhm.

Just a little story...
This thread remains a great read, thanks for JD and others who keep it intresting.
JD, thank you again for initiating this thread, and your persistence in having it reemerge. I think it speaks volumes to your methodical and results oriented nature.

Again, I feel this thread will come to be recognized as one of the truly seminal works regarding audio cable; one that will spur development in many a company that raises the level of performance we will all be fortunate enough to enjoy in the days ahead.

We are the beneficiaries of your hard work and dedication to the hobby!
Hello Greg,

I can not express strongly enough the clarity phenomena you brought up. You and I have both spent copious amounts of time over the past years learning the benefits of power conditioning, dedicated power, outlets, power cords, cables and isolation tweaks. With every step we took we would check back into Audiogon to share our experience. “Another veil was removed from the sound,” or “I could see deeper into the soundstage,” and “It was as if a haze was present that has been removed.”

I remember stating these things with complete amazement. It has been years since I removed the last layer of fog, and the sound became clear. The issue you have brought up is not this, we are not talking about seeing clearer into the … instead this is as if the focus on a camera lens was corrected. It is easier to hear the words and separate the voices of each instrument.

One example was from last night’s session. I was listening to Emmylou Harris “Spyboy.” I have listened to this disk well over 100 times and have always been bothered by the mixing process. On most of the cuts Emmylou’s guitar (a high frequency steel string acoustic) is in the left side of the soundstage, and on some cuts is hard left. It has always bothered me because her strumming is so light it is hard to find within the power of her band. If I am able to separate it, I can not hear it, meaning I can not tell what cords she is playing. With these DIY Interconnects I can actually separate the sound from the background. Her guitar is finally clearly an instrument and I can focus my attention to it. It is hard to explain, but it is clarity, not separation and not lack of background noise/fog. It is simply clearer.

This experience is repeated in every disk I play. I continue to hear the words better. The instruments are much easier to separate from the soundstage. Again not in the “old” cleared (less noise/fog) but just plane easier to see.

It is amazing, and impossible to properly describe.

Thank you Joe, Greg, Chad, Sean, (I tried to email you, not sure it got through) Albert and John for the continued support. It took all of us to make this happen.


PS the edgieness is gone
Gregm: Changing the spacing between conductors alters the nominal impedance. The further the gap between the conductors, the higher the nominal impedance. While a higher nominal impedance can act as a "buffer" to the source component loading into the cable ( think "cd player" ), and it doesn't really "hurt" the load component ( think "preamp" ), it can introduce other problems into the equation.

When spacing gets too wide, the inductance of the cable goes up. The higher the inductance, the more potential that there is for the cable to act as a receiving antenna, both in terms of RFI and EMI. Coiling excess cable up can also have a similar effect, as a coil is the most basic form of inductance.

On the other hand, using a cable with a very low nominal impedance can cause a component with a higher than average output impedance to become "loaded down". This not only causes excess current to flow, but can also limit the bandwidth capacity and linearity of the circuit. This can more easily happen with tubed gear, which typically has a higher output impedance. Conrad Johnson gear comes to mind here. Having said that, it can also happen with certain SS gear, especially when very high capacitance / low inductance interconnects are used.

High capacitance / low inductance occurs when you have two conductors in VERY close proximity to each other. The closer that they are, and the more surface area that is in close proximity between opposing polarity conductors, the lower the nominal impedance. As such, a single conductor round wire placed as close as possible to another single conductor wire won't have as low of an impedance as a flat or foil conductor placed next to an identical flat or foil conductor. The wire and foil / flat conductor might share the same overall gauge as a round wire, but due to the wider surface areas of the foil / flat conductor coming into "near" contact with each other, inductance is reduced and capacitance goes up.

Due to the differences in circuit design and impedances that we are dealing with, lower inductance is beneficial to speaker cables. On the other hand, lower capacitance is beneficial to interconnects. As such, using similar geometries and impedances in both interconnect and speaker configurations is a less than optimal approach. That is, cable geometries that are optimized for line level operation are normally quite different than cable geometries that are optimized for speaker level operation.

Honestly, i find it kind of funny that both Goertz and Nordost based their entire lines of cabling on a single yet opposing geometry. That is, all of the Goertz cabling uses a low inductance approach whereas the Nordost uses a low capacitance approach. This is why they can have such different sonic characteristics, even though most people think of them in both being "flat cable" designs. Even though both are "flat" ( per se ) conductors, they are electrically quite different. One will work best as an interconnect while the other will work best as a speaker cable.

There is obviously a gray area where impedances become too high or too low to work in either type of circuit, and some of that will vary with what specific gear is being used. Twisted pair conductors kind of fall into the "universal" category, as it can typically be utilized in line or speaker level applications with pretty good results. What the nominal impedance of the twisted pair is will boil down to the amount of insulation between the actual conductors and the amount of turns per foot that each design utilizes. Sean
Thank you Sean.
I still have to get down to getting a bridge of sorts to measure these JD I/Cs I made -- unless I figure an amateur way of measuring them connected?
Sean - I can bet I am missing something, but I will ask anyway; How do you know what kind of ICs your gear would work best with [high inductance vs high capacitance]?


PS: Sorry for the poor English-
Dbld: Very few components ( i really don't know of ANY in specific ) will work best with a high capacitance interconnect. Having said that, one can use a high capacitance interconnect to tame top end brightness and / or to tailor the response to one's own personal preference. This "softening" will also come at the expense of high frequency dynamics and "airiness", so there is a balancing act that one will end up playing.

Normally, one would want to choose an interconnect that had a nominal impedance that was AT LEAST as high as the output impedance from the source component. This can be found as the output output impedance of cd player as that "loads into" the preamp and / or the output impedance of the preamp as that loads into the amp. In most cases, this would typically be somewhere between 20 - 100 ohms with some exceptions ( like Conrad Johnson for example ) well above that range. Input impedances for most components range from as low as a couple thousand ohms up to several hundred ohms.

This is where things can get tricky. If you've got an output impedance of a preamp rated at 100 ohms, and you're using a "coaxial" based interconnect ( common design ), the nominal impedance of that cable is probably about 50 - 75 ohms ( give or take ). The end result of using this impedance mismatch could end up as ill-defined bass, poorer transient response, a reduction in focus and transparency, narrower bandwidth, etc... This would be most common on designs utilizing Integrated circuits with "measly" power supplies. This can also happen with tubed designs IF the cable is WAY low in nominal impedance and the tubed circuitry isn't all that stable and / or is also current limited.

As such, one can learn how to manipulate various component interactions by juggling input, output and interconnect impedances. When one hears drastic changes in sonics from changing interconnects, that's because they are using the cable as an impedance transformer ( along with some other variables ). Different cables combine with the various input and output impedances of the interconnected devices, therefore producing different electrical characteristics and circuit responses.

While the best design would be to have matching impedances through-out the entire system ( cabling and components ), this is also the most costly to build and toughest to design. It would also limit what components would work well with each other as this is FAR from "industry standard". This is true even though it is the most logical approach when it comes to electronics. As i've said before, many things in audio are "bass-ackwards".

As a side note, electrical characteristics are additive by nature. That means that if one starts off with a short cable that is capacitive ( or inductive ) by nature, adding more length adds more capacitance ( or inductance ). This is why many preamps and / or source components specifically recommend shorter interconnect lengths and / or the manufacturer makes a specific point of stating that their design is stable into any length interconnect. With the latter i.e. suitable for use with longer interconnects, this typically means that the output impedance of the device is quite low and makes use of a reasonably "beefy" power supply / current path.

There's a LOT going on here with tons of variables. This is why so many different opinions exist, even when trying the exact same cables in different systems. Each system / circuit has a nominal impedance and as such, some cables will improve / decrease performance characteristics accordingly. Better designs are less susceptable to cable changes, so long as the cables themselves aren't "crazy" with capacitance and / or inductance. Sean
Sean, thanks for such a thoughtful response.

I need to take some time and digest it now, but it lends credit to those who have sent me what they thought were great cables, which when installed in my system, didn't seem to make any difference.
I have received many requests for photos, so I have added five pictures to My System Please check them out. You will see I added four strands of cotton piping to the flat interconnect. I then housed it in a nylon flexible casing with shrink wrap to RCA's. Looks almost professional. NOT!

I have not tried cryo treating but I have considered it.
Thanks for a great thread. I have recently made a couple of skinny magnet-wire cables - 4 parallel conductors of 30 AWG radioshack wire sandwiched inside packing tape!- sounds pretty good to me - but then I've changed so many things lately its hard to say - sounds better than it has in years. A couple of followup questions:

I've read a few hundred articles and reviews on interconnects in the last fews months. Rarely is the location of the IC ever discussed - ie, are we talking about from source to pre-amp or pre-amp to amp? My homemade magetwires are running from source (DAC) to a tubed preamp - while I'm running some cheapo Tara Labs to the amp. Do most people run the same type of cable in both positions? Is there any thoughts/theories as to which type of cable sounds better in which position? Or which cable (source-out or pre-out) has more influence on the overall sound. Your testing was done on the source-out cables or...?

Issue #2 has to do with wire quality - specifically hardness/temper. I've read that the softer the wire temper the better the sound. (supersoft wire is obtained via "annealing" - heating up then cooling slowly) softer wire is said to have longer crystal structures - cryogenic treatment also affects metal crsytal structure. Silver wire DIYers mention tha you may specify temper/software when ordering silver wire from some sources but most wire sources do no specify.

So my point is - properly annealed/treated copper-wire might very well sound better than harder gold wire, and so on - in the interest of comparing apples to apples it would be beneficial to discuss the temper/hardness of wire in these comparisons. We need a true metalurgist to tell us the ideal way to build long crystal structure in wire - is it anneal then cryo or vice versa or what...?
Hi Gdoodle and Sean,

The Radio Shack magnet wire is the same stuff we used in our early prototypes. It does a very good job for the money. The enamel coating must be removed for soldering, a bit of a hassle, but not too bad. Best I could tell the dielectric constant for enamel was around 5.0, so this is the big downside to the magnet wire.

Your question on location is very good. I always tested the cables in both the CD/Pre-amp link and Pre-amp/amp link. In my system there was a slight difference in compression, but not worth discussion. I would always recommend your best sounding cable be tried in the CD output first, with the (garbage in, garbage out) theory in effect here.

More specifically however, this is what Sean has been trying to get us to understand. The actual design will vary depending on the specifics of the components. I would like to ask Professor Sean to walk us through this lesson using the specifications from my equipment.

Source: Sony SCD-1; Output level 2v RMS with load impedance over 10 K ohms.

Pre-amp: Aesthetix Calypso; Input impedance, 40 K ohm maximum input 3.5v RMS
Output impedance, 300 ohm recommended load is 10 K ohm minimum.

Amplifier: Plinius SA-102, Input impedance 47 K ohm, Input sensitivity 0.8v RMS

The gold wire has an impedance of .0012 ohms/cm or .017 over the length of the interconnect. Again the separation of the positive and negative conductors is ¾” and I would say this should be our variable for this exercise.

Sean, I’m not sure what else you may need to help walk us through this. If you can indeed help us learn this would be awesome!

This entire process has been fascinating and filled with many surprises along the way. Last night was no exception as my faithful friend lent me his ears once again for another listening session.

Up for audition last night was the very best of the Radio Shack magnet wire cotton rope design. (This was the last of the round/core design) At the other extreme in pricing was the Gold Wire/Eichmann RCA/Bybee Slipstream filter design in a flat cotton sleeve. (This is the final design that started this thread.) In the mid price catagory we had an all Silver Wire/Silver Eichmann in flat cotton sleeve as well as three cables that substitute masking tape for the cotton sleeve.

As a quick reminder, the masking tape design is using a 2” wide blue painters tape laid sticky side up with a ¾” manila masking tape laid down the middle of the 2” tape. (Laid length wise; sticky side to sticky side.) The ¾” tape is a simple measuring devise/spacer and could be ½” for a closer conductor design or 1” for wider, and so on. The building time went from five to eight hours for a pair to less than one hour, so it would be very quick and easy to test countless wire size/spacing options. Wire combinations were all copper, all copper with silver plate and all copper with gold plate. In all three designs the wires were (3) 28ga. Positive and one 24ga. Negative rune. This combination was arrived at from earlier tests (see bulk of the thread above) thus eliminating one variable. A lot more testing is needed here too. Finally n the ends of the masking tape interconnects were Radio Shack gold plated RCA’s soldered to the wire using WBT silver solder.

For this test we used RCA signal splitters on the outputs of the Sony SCD-1 CD/SACD player so that I was able to hook four pairs of interconnects into my pre-amp and switch between them with the remote. We started with the two all copper pairs, (one core design and one masking tape) the all silver with cotton sleeves and the silver plated copper (masking tape). One of us would sit with our eyes closed and listen while the other flipped inputs. Then we traded positions. We listened to a number of segments of very familiar music ranging from jazz, folk, rock and classical.

100% of the time the two all copper designs were eliminated. The edginess and semblance was quite noticeable, and very un-acceptable. We replaced the two copper interconnects with the gold plated copper (masking tape) and the all gold wire/cotton sleeve design with Bybee filters at each end.

Quickly we were able to eliminate the all silver as edgy and lack of midrange richness. The gold plated copper was next out as being a bit weak in the bass and treble. The midrange was pleasing, but not better than the final two pairs.

This left us with a roughly $1000 all gold design that takes about eight hours to assemble (and has broken five times at the solder joint over the past two weeks of testing) and the silver plated copper jewelry wire, Radio Shack RCA’s and masking tape dielectric with a total cost of about $25 plus one hour labor.

It began to be difficult to distinguish anything at this point (1 ½ hours of testing) but the differences were not consistent. My friend preferred the silver plated over the all gold on jazz; (Sarah K and Patricia Barber) where listening to folk (Natalie Merchant and Greencards) impressions were not at all conclusive, when about 50% of the times he flipped on his opinions. The rock (Stevie Ray Vaughan and Mark Knopfler) and classical (Minnesota Orchestra) were consistently the all gold pair won. For me I preferred the silver plated on the jazz (except piano). On the folk and classical I clearly preferred the gold (strings) and with rock I was not able to consistently pick a cable.

Hugh? $25 vs. $1000!

After my friend left, I spent a little more time listening without the splitter, meaning I needed to switch interconnects on the CD outputs. Here the opportunity to switch instantly eliminates the double blind concept, but the results we far clearer now. Without the degraded signal from the splitter, the gold was far richer, more ambient information and much richer midrange. Both had excellent bass definition and extension, and the two were not distinguishable in the treble.

I left the night knowing more designs were needed, and these should use the same RCA (Eichmann) connectors. This means the cost will raise from $25 to $45. I guess I will not be surprised to find a design the will better my all out assault gold interconnect. If this is true, this will be an extremely important discovery. As I have said all along, matching the performance of the majority of cables on the market is not hard, but my goal was to beat the best cables I have heard (Purist Dominus, Nordost Valhalla and Kubala-Sosna Emotion) The step from the majority of cables to the best is not huge for the casual listener, but if you want the very best results from your equipment you will need to rise above the normal cable.

In today’s marketplace there are countless cables in the $500-$1000 price range. Each one I have heard (and I have heard most) are fundamentally flawed in one or more aspects. Edginess, poorly defined bass, inaccurate PRaT, poor imaging, lack of inner detail, lack of dynamics… To find a cable that has all the attributes right has been a quest for me for the past seven years. Each of the aforementioned cables were far above the majority in the subtle yet important areas I listed above, but still I was not able to find one cable that did everything right.

The gold cable that started this thread is the best total combination I have found. Last night began to question that assertion, and I am now wondering what I will discover with the next round. And what if it was good wire (OFC/99.99% silver plate) and what about gold plated silver and what about tape and all gold and what if I used cotton sleeve on gold wire with tape outer construction, and what if the all gold design was 28guage instead of 30 gauge, and what about spacing and???

I ran out of money with the first “what if.” I need these answers, I need someone to try taking the next step with their money, or if someone wants to pay for the material for the cause of research I will send them back the test product.

I started this post stating that this has been a fascinating journey. Unfortunately I am flat out of funds to continue, so if people want my guidance on the next steps or want to fund this project let me know.

And no, I do not intend to start a business building these, so if you are concerned about funding my R&D, your not! You are funding an Audiogon community project, and it should be treated that way, at least that’s what I think.

I hope to hear others results soon.


my best wishes on your health, and my compliments on this excellent post.

question: was there any break-in period given for any of the cables? IME, copper breaks in the fastest, tinned / plated copper takes longer, and silver cannot (!) break in properly without a proprietary break-in device (though my experiences w/ silver are limited to teflon dielectric, and i plan on building a new set of ICs w/ silver using cotton dielectric).

i've experience w/ gold, but only on power cords. in general, i am very fond of their greater tonal color palette, but find them ever so subdued relative to "faster" metallurgys (EG: rhodium, palladium, silver). 99% of systems would do better w/ gold IMHO.

again, my highest compliments on this thread.
Hi Rhyno,

Thank you so much for your kind thoughts. I can not tell you how much your compassion means to me. I figure it's people who keep me going, so I thank you.

As for your question, itrn in is maybe the most important aspect of auditioning cables, and in my mind the least understood. Why does burn in matter??? I have no idea, I just know it does. Your description is good, copper burns in relatively quickly, where I too have felt silver never really comes around without a high voltage, high current burners help. Even then I find silver settles back away from full burn in over time, and requires another round of high voltage burning.

That being said; specifically with this project, I used my friends solid state system to do the initial burning. Cables that showed some promise received additional time on my system using the XLO and Purist burn in CD's. A dear Audiogon friend, Bob_bundus sent me his Mobi cable cooker and I have been using that since.

So the test cables all had some burn-in, but certainly not in a predictive or scientific fashion. As I stated above the pure gold wire became a bit bright, followed by a loss of bass. This is a common symptom of what a wire (and component)goes through during a burn in process. When the Mobi arrived I put the gold interconnects on the Mobi for a few days and the treble calmed back down.

Last night my system sounded the best it ever has, so I suspect the gold wire is still evolving. The bass is tight and deep, mid-range is highly textured and displays endless hues and tones. The treble is smooth and extremely clean. The bright or edginess is no longer noticeable. (EXCEPT THIS PAST WEEKEND)

Another wonderful Audiogon friend (Jafox) stopped over with a arm full of Stealth power cords. The top end cord "Dream" was quite amazing. The bass was super deep and the lower mid-range filled in with a weight and body not ever heard before on my system. The surprise was the treble. Completely extended with absolutely no glare, edginess or brightness. It was easy to raise the volume 25% without ever feeling the bite of the top end. It was absolutely perfect, with Lucinda William's live disk #1, song seven being used as a test. Lucinda has never been mistaken for lush, and her scratchy voice has always been a test for me in that I love her emotion and MUST be able to play her often. At the end of the song there is a series very sharp hits to a quite metallic instrument. The hits are edgy and meant to bite. Using the Stealth power cord the sound was excellent. It was just the right amount of metallic sound, not harsh, but wow, so colorful. Each hit has it's own tonality and texture, each time presented as a pleasing end to the song. I tell you this not to report on this amazing power cord, but to explain the edginess or brightness I have been sensing is coming from my power supply, not the interconnect.

This slight haze of electronic glare was never seen before the Stealth. Once it was "filtered" through the Stealth Dream power cord, the true extent of how amazing this DIY interconnect shone through.

So today I have three different DIY designs using silver and gold, or silver and copper burning in on the Mobi. As you stated, this is the material most needing burn in and so the results are not yet known.

What I can report is gold takes maybe 40 hours at most, and the changes are very small. This is a simmialar time frame to copper in my experience. This may also be due to the lack of dielectric/shielding materials. I suspect construction has as much to do with burn in as the actual conductor.

Thanks again for the comments,

I have a question on people's preferred termination method; crimped, soldered, or both? And if soldered, how important do you feel the "mechanical" connection within the solder joint is to the performance?

I know this has been debated before, but I don't have a clear answer and folks here seem to be bit more civilized than in other threads.
it depends. if the contact area w/ screws is large enough and materials (metals) are good quality (i.e. brass / silver / copper, not nickel screws or steel), then screw is fine. otherwise, i'd go w/ solder. i'd never use crimping.

Now, my experiences w/ a DIY xlr (sean, this is right up your alley).

i put together a balanced IC (1.5m), using 28ga high purity silver in a cotton dielectric, with + and - legs separated by 1/2" of cotton rope (no ground in use).

here's where it gets good: using a TG Audio high purity XLR as reference (incorporates a ground and has all conductors in close proximity..i.e. less than 1/8" separates all). i noticed the following between my einstein preamp (50r output impedance) and my mcintosh 501s (20k balanced input impedance):
1) with the TG cables in place, system dead quiet w/ volume control fully attenuated, and wattage meters on macs reflect 0.00
2) with the DIY homebrews, even with volume fully attenuated on the preamp, i hear some music getting through (VERY low volume, but its there), and i notice that the wattage meters reflect @ 50mv. (i am happy to note that they are not functioning as mini-antennaes however, a problem i had with other DIY ICs that didn't employ self-shielding (twisted pair) geometry).

once i turn up the volume, everything seems fine. no listening impressions until i give them 7 days on the MOBIE. and while i'm not about to say that the wattage meters on the Mcintosh amps are as accurate as what's on john curl's test bench, i would love to know some suggestions as to what is responsible for the IC-wattage experiences above. my guess is that, because the DIY design has its conductors further apart, it is high impedance / high inductance / low capacitance, and i'm simply letting more music through with less self-attenuation built into the IC geometry. --that is, if i understand sean's posts above.

Rhyno: You don't have a ground, so there is no way to have "zero" ( ground shunted ) volume. This has nothing to do with the spacing ( nominal impedance ) of the cabling that you are using.

The volume that you do hear is the difference in output levels between the positive and negative legs of the preamp. This tells me that the balanced circuitry inside the preamp isn't running in perfect symmetry. If they were in perfect "balance", there would be no output.

As far as cooking cables on the Mobie goes, you really need to leave them "burning" for at least 30 days uninterupted. I talked to Bob about this several times and we both came to the same conclusions about this. A shorter time may be beneficial, but it may not be quite as good as it can be. Sean
Rhyno - Why do you say no to crimping? The crimping I am thinking of in the technique used with the WBT-0108 RCA.

BTW - I am not making an argument, I am genuinely curious/interested.