Tuesday, November 24, 2015

Heretical Digital Cable

Why Heretical?

Every digital audio coax (SPDIF) cable out there claims that it is "75 ohms" (whatever that means). The Mad Scientist Heretical Digital makes no attempt to conform to this standard. This post explains it all..

What is this "75 Ohms"

In the SPDIF standard (Sony/Phillips Digital Interface Format), the coax electrical interface is defined to have a characteristic impedance of 75 ohms.

Characteristic Impedance however is not so easy to understand or picture as regular impedance, say like a loudspeaker is 8 ohms, or an amp might have an input impedance of 50kohms. These are simple resistances, and you can measure them easily with a multimeter.

On the other hand, Characteristic Impedance is all to do with transmission lines. You also can't measure it with a multimeter. Here is the formal definition, from Wikipedia:

The characteristic impedance or surge impedance (usually written Z0) of a uniform transmission line is the ratio of the amplitudes of voltage and current of a single wave propagating along the line; that is, a wave travelling in one direction in the absence of reflections in the other direction. Characteristic impedance is determined by the geometry and materials of the transmission line and, for a uniform line, is not dependent on its length.

So why does that matter?

The crucial point is this : If you have a 75-ohm transmission line (like a digital cable) and you have some part of the system that is NOT 75-ohms, then you get signal reflections. The signal can bounce up and down the cable, in some cases dozens of times. These reflections can and do alter the original signal in hard-to-predict ways. This mechanism is the underlying cause of jitter in digital cables. And jitter is one of the chief culprits of 'digital nasties'. Not the only culprit, for sure, but a very important one.

There have been articles written about the optimum length for a digital interconnect, making it the right length so the reflections avoid the critical points where the detector is trying to detect a transition. However you need to know things like the rise time and propagation time to figure out the correct length for your system; as far as I can see there's no length that would work for everything. The effect that jitter produces is a horrible mixture of noise and distortion - not easy on the ear at all. But it accounts for the differences in sound of digital cables.


The amount of jitter that high quality digital clocks produce is getting to be very small - measured in a few tens of picoseconds. Now that figure is meaningless to most people, me included. So how's this : Light travels about 3mm in 10 picoseconds. And as you may know, light is really really fast. This small amount of jitter can be dwarfed by other sources like cables.

Heretical Digital Cable

The Mad Scientist Heretical Digital takes a different approach. It does not attempt to conform to a 75-ohm characteristic impedance. It does however have a significant amount of resistance as the main conductor is made from treated carbon fiber.

Two things happen with a carbon fiber conductor:

  • Skin Effect is very very small compared to metal conductors. The square waves that make up the digital data are sent at a  few megahertz - fairly slow by digital standards. However, the harmonics that make up the square wave go much higher, into the tens or even hundreds of megahertz. At these frequencies, copper has a skin depth of a few micrometers. By comparison the skin depth for carbon fiber is still as few millimeters.

This is important as the correct transmission of all the component harmonics is crucial for the correct transmission of the whole wave.

  • Resistance soaks up the reflections. This is probably more important than skin effect. The resistance of the conductor, being of similar magnitude to the 75-ohm loading means that reflections are not going to be able to do much damage - they will be turned into heat.

If you imagine the digital link being like a light tube, with flashes of light being the data pulses; A normal cable has silvered parts and so you get glare and reflections. Our Heretical Digital is like filling the tube with slightly darkened glass - only the bright flashes get through, reflections are absorbed.

The Heretical Digital Cable has a resistance of about 37 ohms - half the 75 ohm characteristic impedance. But all SPDIF inputs are terminated with 75 ohms. This means that a reflection that reflects off the DAC end of the cable/plug will travel towards the source where it can interfere with the data. But the resistance will tend to turn the energy into heat, so reflections are dissipated very quickly.

What it takes to develop cables - a small selection of prototypes

What Actually Happened

This all might sound like it was planned. But that's not quite how it happened. As is often the case in science, an ad-hoc experiment showed some interesting results which lead to further research, theorizing, more experiments and so on.

After the carbon fiber interconnects were developed, I thought I'd try one as a digital cable. I wondered whether it would work at all and I wasn't expecting it to sound good.

But it did sound good, which surprised me. So a new project was born. I needed to figure out why this was producing such good results. Also I needed to figure out what the best sounding design was.

It turns out that a much simpler design than YANAM/TORFORB is needed. Out with the multiple bundles of carbon fiber, now just a single, thicker conductor is used. Also a cheaper plug seems to work better.

As with the interconnects, I was trying to make the best digital cable that I could. It would also have been ideal if I could come up with a range to suit all pockets. But the result of the listening and design process showed that the simple one is the best. If anyone wants a "signature" version, I can sign it with a silver pen if you like ;)

How Does It Sound

Usual disclaimer about how I am biased, etc. But those that know me might want to take note..

The Heretical Digital Cable astonishes me. I've tried a good number of digital cables in my time, conventional and not-so-conventional. This one shines, being the best one I've heard. Not "best for $99" but "best for any money". The things that leap out at me are:

  • analog-like sound but with digital crispness - so a lush sound but the leading edges are still fast
  • very pure treble with lots of air and very fast top end
  • wide and fast dynamics
  • intelligibility - I'm hearing lyrics much more easily. Also makes complex passages easier to parse

Thursday, November 19, 2015

What's a Torforb?

TORFORB is our top-of-the-range interconnect. This is the story of their development.

There are actually three models of interconnect, all related by the common carbon fiber signal conductors they use. All three have identical signal conductors, but vary in their construction and earth conductors.

Why Carbon Fiber?

I first came across carbon fiber as an interconnect conductor in the 1990s when I lived in Holland. Local star Van Den Hul released the FIRST cable, a pure carbon cable. I heard some at a friend's place and went out and bought a set. They sounded very smooth and musical.  In the 90s, digital nasties were more prevalent than they are today, and the fact that the Van Den Hul FIRST rounded off the music somewhat was a good thing.

Although they are very musical, they paint a romantic picture, a bit like old-school tubed gear. But they also don't render the top end and 'air' very well. So they would not be competitive today.

I started experimenting with carbon fiber early this year. It became quickly apparent that there was a special sound to carbon fiber, but I ran into the same sort of issue - rounded off highs. But it turns out that this was quite easily solved. I won't go into exactly how. But the final signal conductor consists of two bundles of fibers, different sized. I also found that the fibers used were important and found a company that used high quality Japanese fibers as the base.

So now the job was to make a product. I had been using a simple copper-occ single strand for the earth conductor. I figured that trying to use carbon would not give good results due to the high resistance - you really want the earth conductor to have low impedance.

One interesting thing about carbon fiber as a conductor is that it's immune from high frequency skin effects. Why so? Well if you calculate the skin depth for carbon fiber you find it's several inches, mostly because it's not a metal and has high resistance.

Development continued with trials of various earth conductors, using different materials, constructions and geometries.  At some point I tried using copper foil as earth conductor and got a real performance boost. I had to try silver foil..

You should see the large pile of prototypes I have here..

After lots and lots of testing, I found that there was a number of designs that I liked, some simple, some more complex. In order of improving sound quality, I ranked the designs:
  • Copper OCC wire
  • Cross wound copper OCC and silver OCC wires
  • Copper foil with copper OCC wire
  • Copper foil with cross wound copper OCC and silver OCC wires
  • Silver foil with copper OCC wire
  • Silver foil with cross wound copper OCC and silver OCC wires
To make matters more complicated there was also a hierarchy of wire types - I found at least three types of OCC wire that I liked. The most expensive copper OCC wire is about 10x the price of the cheapest OCC wire, same with the silver.
Then there is silver/gold foil. I wouldn't say that this is better but it has a slightly different sound, a touch warmer.

This was turning into a problem. There were far too many options here. So I decided that there should be three models, with maybe some options. 


Around this time I decided that TORFORB would be a good name for the top of the range. It stands for "Too Rich For My Blood", the name of a Patricia Barber track that is a regular test track of mine. So I went searching for other Patricia Barber songs names that could be abbreviated, and came up with ATOH ( A Taste of Honey) and YANAM (You & the night & the music).

The ATOH is the entry level interconnect, and that uses the simple copper OCC earth wire, using Neotech teflon insulated wire.

YANAM uses copper foil and cross wound copper OCC and silver OCC wires, using Neotech wires.

And TORFORB uses silver foil (or optionally silver/gold foil) with cross wound VH Audio foamed dielectric occ copper and silver wires.

Around this time we were also developing our Magic Tubes. These worked well in many places including interconnects. So I had the idea to include a pair of Magic Tubes with some models.

I had been attaching the Magic Tubes using sticky tape or Blue Tak. Clearly this would not work for a product, so I came up with the idea of the wooden "Audio Block".

The reason I call them Audio Blocks : They are made by a local company, Woodzone,  who make things from NZ Native timbers. When I went to pick up the first batch, the girl in the office was making an invoice for me, and says to me "What are they called?" - I gave her a blank look and said that was a very good question, no idea.  We stood around for some time pondering this, and someone suggested Audio Blocks, so the name stuck.

They are made from NZ native Rimu wood, and laser etched. The YANAM and TORFORB models come with audio blocks and magic tubes; they are an option on ATOH

How They Sound

Of course I am biased, but my take on them is that they all sound somewhat different to normal wire interconnects. Hard to put into words, but there's a rightness to them that makes many other interconnects seem wiry, distant or colored. They also do dynamics very well - interconnects that can make you jump at the loud parts.

Although there is a distinct improvement as you go up the range, there is also an undeniable family sound. They share much more in terms of sound than they differ. 

The best way to hear how they sound is to try a set. We do 30-day money back returns, but I can't see anyone returning them.