A New Interconnect Design

[beau2317]

A new interconnect design : Verumecce (which means "behold the truth")

NOTE: If you cannot hear the differences between interconnects, or for some reason believe that they all sound the same, please take your arguments to another thread. There are plenty of people who can hear the differences and that is who this is aimed at. I'm not selling anything here, just sharing a design and my experiences and I do not want this to turn into an argument about whether cables sound different. /NOTE

INTRODUCTION
I'd heard differences between interconnects in the past, but I'd never heard them make a huge difference. Until now. Here's the story..
I first came across "active" interconnects when I bought a second hand set of "Aural Thrills Active Palladium" interconnects several years ago for about $100.

These sounded interesting but I didn't use them long-term. I examined how they worked, and it was a simple system. The signal and earth cables were braided in Teflon tubes. The shield was put at a potential (ie voltage) from earth.

But, they sounded pretty much the same with or without the battery.

Recently I pulled these out again and got thinking about how to make a good set of active interconnects. I made the assumption that one way that dielectrics cause problems is due to "dielectric hysteresis" - by this I mean some non-linearity when the dielectric has to change polarity.

This sort of thing is documented here: Dielectric - Wikipedia, the free encyclopedia
The correct term is "dielectric relaxation".

In a normal interconnect, this happens every time the AC music signal crosses zero (i.e. basically all the time). A bit like a class B amplifier suffers bad crossover distortion.

I decided that you would like to have the signal conductors inside an electric field that was much stronger than the music signal, so that static field polarizes the dielectric. This would be the cable version of a class A amplifier.

After some thought I realized that a coaxial cable could provide an ideal electric field, if the core and shield were energized. The problem was that you needed to somehow get the signal wires in between the two poles.

How I did that was to disassemble the coax cable; wind bare wires around the dielectric core using a double helix. These were covered with Teflon tape, then the aluminum/Mylar shield was rewound over the top, and finally the copper braided shield over the top.

A voltage is applied to the shield and core, with the mid-point of the voltage being referenced to earth. This means that the signal conductors are floating at half the potential. So with 4 x 9V batteries you get 36V between the center and shield, with the signal sitting at about 18V.

The overall effect is to strongly bias the dielectric so the signal cannot cause a polarity switch.

And it seems to work. Comparing them to my previous interconnects was like listening to a different recording, with much more information coming through. My initial version used some quite ordinary wire, but later I used Mundorf Silver/Gold wire, which gave some improvement.

Later I took the palladium wire from the Aural Thrills cables and used that as the "hot" conductor. This produced a further improvement. The most noticeable thing was the treble, which took on a more powerful and lifelike sound (e.g. snare, cymbals). And on further listening, improvements could be heard elsewhere. However, palladium is very expensive - one quote I got from a metal dealer was around $200 for one meter of 30awg wire! (The original Aural Thrills I bought contained two meters of wire, so they were a great deal!)

I'm sure that I could improve these interconnects more; they are only my second or third attempt and have not been refined. They use cheap TV coax and cheap RCA plugs. Improvements could be expected from using good quality Teflon foam coax and better quality plugs. However they are good enough to sound quite stunning on my system.

I should also say that I am not interested in selling these interconnects (but having said that, already friends of mine have blackmailed me into making two sets for them).

CONSTRUCTION DETAILS

Parts needed :

1) Some coax cable. Best is to get one with a solid copper core, aluminized mylar cover and braided copper shield. You may need quite a bit more than the lengths you intend to make, as you need to strip down and re-assemble, and this doesn't always work first time, espeically putting the copper shield back on. But this is cheap so not a big problem.

2) Plumbers Teflon Tape

3) Wire. Ideally Mundorf Gold/Silver 0.5mm, or otherwise any good thin gauge (22-30swg) wire. Best to use uninsulated

4) Heatshrink in various sizes

5) Some Techflex to cover the whole lot

6) Phono plugs

7) DC line plugs and sockets (similar to what is used on wall warts)

How to make:

1) Strip the insulation from a length of coax that is about 1.5 times the length of the IC you want to make. You can normally do this by cutting the insulation for a few inches, then pulling the center so it cuts the rest of the outer insulation. My coax looked like this:



Remove the copper braid carefully. Best to run your fingers from one end to the other so it bunches up. Save this as you need to put it back later. It looks like this:



Also strip the mylar shield and put aside.



2) Now comes the wire wrapping. Cut a length of wire that is about 30cm longer than the IC you are making. Tape one end to the inner of the coax, leaving about 1/2 inch of the coax inner sticking out. You also need to leave 2-3 inches of wire so you can terminate to a phono plug.

Wrap the wire around the coax inner. Aim for a pitch of 2-3 inches. When you get to the other end, tape it.



Now tape a second length of wire to the other side of the inner, and wrap this to produce a double helix. Don't overlap the helices, but don't worry unduly if the wires touch in places. We'll sort that out in the next step.



3) Wrap with the plumbers tape. This can be a frustrating step.. Start by wrapping a few turns at one end to get the tape fixed. Then work your way along the cable, about 2-4mm pitch. As you go, keep an eye on the wires and if they get close to touching, adjust as you go. When you are done it looks like this:



4) Now wrap the mylar shield back. You will probably need the shield from a longer length of coax than the length of the IC you are making.



5) At one end of the cable, cut a short length of the inner off so you expose the center conductor. We will need to connect a wire to this later on.



6) Now the fun part.. You need to put the copper shield back on. Slide in on carefully and it's easy.

7) At the end where you exposed the copper center conductor, leave some spare outer copper shield so you can connect a wire to that. Solder wires onto each and cover with heatshrink.



8) Now you just need to terminate each end with an RCA plug. I did this by using teflon tube that was just slightly larger than the wire I used. This insulated the wire from the end of the cable up until it was soldered inside the RCA plug. There's no strength in the cable like this however, so you need to cover with heatshrink to give it strength. The final thing:



BATTERY BOX
I started with a simple set of batteries, but developed the circuit by ear until I had this:



I use a total of 6 x PP3 9 volt batteries. But I found it sounds pretty much the same with 4 x PP3. You can use any batteries you like.

You can connect all of your interconnects to the same battery box. Just parallel the outputs.
The earth connector should be connected to a chassis earth somewhere.
I put mine in a small plastic box:



Leave the battery box connected all the time. There is no current flow to speak of, so the batteries will die of old age.

HOW IT SOUNDS
Well of course I am biased, but I like the sound of the Verumecce Interconnects very much. You may find that it improves with a few days use. Someone I know is writing a full review.
But in my opinion, the sound is very detailed but also very smooth. Soundstaging is pinpoint. The overall effect is one of realism, which is what we are all after I think.

Enjoy!

[analog_sa]

Thank you for making the effort to post such a detailed description. I will certainly give your recipe a try.

Originally Posted by beau2317

NOTE: If you cannot hear the differences between interconnects, or for some reason believe that they all sound the same, please take your arguments to another thread.

Let's see if this spell works. Not holding my breath.

[jameshillj]

So, basically you have twisted 2 new signal conductors (signal & return) around the centre core of the Coax cable, insulated them with a cover of Teflon plumbing tape, replaced the "twisted" metalised foil wrapper (mylar sheild) and then added the copper mesh over the top of that and fitted some outer cover over the lot, yes?

The two new "exotic" conductors are then connected to the RCA phono plugs, as per normal, making it a pretty good IC on it's own, batteries or not.

Then the central coax conductor is charged statically (+ve?) by connection to the battery via the 2k2R resistor and the outer copper braid (and metalised mylar sheild) is charged in opposite polarity by connection to the battery negative via another 2k2R resistor, yes?

The overall effect is to strongly bias the dielectric so the signal cannot cause a polarity switch

And both the central coax wire and outer copper mesh are isolated from each other, apart from the capacitor across the battery supply together - just making a static charge across them?

Okay, a couple of questions

Where does that Earth connection at the centre point of the batteries get connected to - Chassis earth on one of the component chassis, a seperate earth point via a power plug, etc?

Higher voltage (54 volts) didn't provides any better results?

Could you use seperate Teflon sleeves for each of the signal wire's insulation to avoid playing around with that damn Teflon plumbers tape?!

It is certainly a different approach and relatively easy to try.

[beau2317]

The overall effect is to strongly bias the dielectric so the signal cannot cause a polarity switch

Yes that's correct.

Where does that Earth connection at the centre point of the batteries get connected to

Either will probably do the trick. Usually chassis earth or mains ground are normally at pretty much the same potential on most systems. For example, on my system, chassis earth is connected to mains ground, but signal ground is lifted slightly by some back to back diodes. The idea is to make it so the signal and earth potentials are roughly in the middle of the static voltage set up on the coax core and screen.

Could you use seperate Teflon sleeves for each of the signal wire's insulation to avoid playing around with that damn Teflon plumbers tape?!

I'm sure you could. I think the plumber's tape is good for some reasons - it's cheap and very thin, thinner than the walls of a teflon sleeve. I don't know how easy it would be to just wind teflon sleeves around the core and then manage to put the screens over the top.

I did an earlier implementation of the idea using air cored coax, where there are 5 air channels running along the length of the insulator. I pushed two wires through the air cores. This sounded pretty good, but I could not find any place to buy that now, so I came up with the double helix/teflon tape design I described here. And guess what..the new design sounded considerably better than the first one.

So I can certainly believe that this design can be improved upon.

[janneman]

Interesting design. I've had some interconnects with a small box on one end, forget the brand, and there was a battery in the box. Similarly to your idea but of course not symmetrical and the 1.2V battery certainly allowed for signal zero-crossing at high levels. I never heard any differences with or without the battery so I stopped using them. But yours might well be different.
I have two questions:

1 - In some systems (maybe most), there is quite some noise and junk on chassis/mains ground with reference to signal ground. I would expect even cleaner sound when using the signal ground for the midpoint.

2 - Have you done any measurements to see if it gives a measureable difference? I'm not looking for a measurement vs audible discussion, and you ignore this part of my post if you want. Just wondering.

jan didden

[SY]

Actually, since you've now added some strongly triboelectric materials and rather loose construction (capacitance will be modulated by vibration), you could very well have some audible differences, plausible even to skeptics like me.

[beau2317]

Higher voltage (54 volts) didn't provides any better results?

I tried both 36V and 54V. I couldn't reliably detect any improvement, but there might be a small improvement. Perhaps someone with a higher resolution system than mine might. (I use : Front end - Squeezebox into 60-chip DDDAC. Preamp : Aikido with 4x6SN7. Power amp: Moskido Class A hybrid - see the thread on this forum - aikido driving mosfets. Speakers : Magnepan MG3.6.)

1 - In some systems (maybe most), there is quite some noise and junk on chassis/mains ground with reference to signal ground. I would expect even cleaner sound when using the signal ground for the midpoint.

Yes I agree. But remember there is no connection between this earth and the signal. It's merely referencing the voltage.

Have you done any measurements to see if it gives a measureable difference?

No, because I don't have the equipment to do meaningful measurements. I'd be interested in measurements, but I don't know what you would measure. Simple things like freq response don't capture time domain things, for example.

I did measure the capacitance using my multimeter, and this was quite low, about 50pF per meter with no plugs on.

[beau2317]

Actually, since you've now added some strongly triboelectric materials and rather loose construction

I'm not following you here. What material is strongly triboelectric ? The three materials that are in contact : teflon, silver, palladium and polyethelyne and all quite close together on the tribolectric series. And as far as I can see, there is no rubbing or friction involved.

I would not call the construction "loose". The wires are held quite tightly between the inner insulator and the teflon tape. In my first design the conductors were "loose" but not in the current one.

btw, SY, I like your signature. I can think of plausible reasons why the basic design sound is good, but am completely mystified by why palladium wire sounds so different to silver. The difference in the treble is very easily noticeable.

[janneman]

Originally Posted by beau2317

[snip]Yes I agree. But remember there is no connection between this earth and the signal. It's merely referencing the voltage.
[snip]

Sure, but there's always capacitive coupling to the signal wires. But you probably would have heard it, if you had dirty mains/chassis grounds.

jan didden

[SY]

Teflon is HIGHLY trioboelectric. I posted an anecdote a few months ago about being badly fooled due to a Teflon-insulated silver wire interconnect and the triboelectric response it generated.