Thrax

Thrax Amplifiers

Thrax Spartacus Power Amplifier

e69c6e1545ca3a0d945ecab5cc1b8b7f

The Thrax Spartacus Power Amplifier is a two stage shunt regulated directly heated triode power amplifier featuring one of the simplest possible single paths with absolute zero feedback. (wire-vacuum-wire-vacuum-wire).

To achieve simplicity while delivering exemplary performance each device i​n the signal path had to be the best we could lay our hands on. The distortion of triode vacuum tubes compared to solid-state devices (transistors) when used for voltage gain is a few orders of magnitude lower by default. A special breed of tubes called Directly Heated Triodes (DHT) deliver an extra order of magnitude of better distortion performance compared to indirectly heated ones, but not all DHT's are the same. Emission Labs manufacture some of the very best tubes ever made. They were the obvious choice for this daunting task. Even then we decided to go further and run two of them out of phase in an attempt to cancel some of the negligible distortion they would generate. For this to work we had to design an even more linear driver stage that would supply the two output tubes with exactly the same signal in phase for one and out of phase for the other. The choice was clear-use another DHT. The tube used produces enough power and voltage swing to accomodate a two-fold overdrive and positive grid drive if needed. Ultra linear DHT tubes have considerably low voltage gain and require high input voltage for full power. We implemented an input solution with the cleanest possible voltage gain - a step-up transformer.

As simple as the signal path is, the rest of it is not . Spartacus has one of the most sophisticated and complicated power supplies. We use a very well-known but rarely used arrangement. To reduce power supply noise instead of the normal arrangement of diode bridge rectifier and electrolytic capacitor bank we use a choke input full wave tube rectifier arrangement. This practically eliminates switching noise and anything that comes before the choke.

The other beauty of this topology is that the choke keeps the voltage on the capacitor bank quite constant. In our case this allows for a considerably small capacitor bank opening and the possibility to use highest quality film capacitors for the job.

​​​A power supply operates best when the load it's presented is constant. Following that thought, both stages of the amplifier were designed to use the same type of regulators namely, the constant current sourced shunt regulators. A constant current source is a device that passes the exact same amount of current no matter what happens on either end. This is followed by a shunt regulator that will try and keep a constant voltage across itself by changing its conductivity no matter what. At the voltages we operate in, this has become possible only very recently.

We use a special technique to supply the heaters (filaments) the tubes. Following a separate winding, rectifier and capacitors is a voltage regulator supplying an electronic choke isolating the filament from the DC source and counteracting modulation of the heater current by signal going through it. It is complicated, but small things make a big difference.​

​​As tubes age and cannot be perfectly matched we have designed a circuit that keeps the current in both output tubes the same. High power versions of the current sourced shunt regulators are used for both stages. We use a microprocessor control to make sure all is switched in the right sequence and that all separate systems function as expected.

The chassis to house all of the above had to be a masterpiece itself. Milled from solid plates of different thickness 6061T651 grade aircraft certified aluminum are used for the construction. We have the power supply mounted on a mid-level plate of this structure with the control board on top, and the transformer and choke hanging below, bringing the center of gravity low enough so the signal transformers can be mounted on the top plate to avoid magnetic interference. The top and mid plates link the front and back forming a protected cavity housing all the sensitive electronics. There are no screws visible anywhere on the chassis. And the chassis feels like a solid block.

​Thrax ​Heros Class A Hybrid - Tube/Fet Amplifier

be46376c039531d03a1088eec961e91d

The Heros are an unusual two stage hybrid transformer loaded amplifier. Based on our design concepts and experience it features a tube input stage providing the necessary voltage amplification followed by a transformer loaded cascaded jfet/mosfet output stage taking care of current gain. No solid state amplifier designer would attempt to use a transformer for speaker output as the whole concept of solid state amplifiers is to avoid it. The numerous benefits from using an output transformer are usually ignored mostly due to cost reasons. Yet the result speaks for itself.

​Almost double the efficiency of a normal Class A design, no DC on output and many other benefits are taken advantage of in our implementation. Like a fully differential (balanced) architecture using an input phase splitting transformer and summing back at the push pull output transformer. This solution completely floats the inside of the amplifier isolating it from any external influences.

The design being fully differential is immune to power supply noise, but still we use our flagship current source shunt regulator for the input stage.

​The power supply for the output stage uses a new breed of rectifier diodes using a new material called Silicon Carbide. They have no reverse current and switching loses.

When properly applied the rectifier produces no EMI/RFI in the chassis. The result is an amplifier free of grain and frizziness, giving new meaning to transparency and tone.

Operating all devices in Class A produces a fair amount of heat. The design of the chassis is such as to cope with the heat and yet to provide vibration free structure for the components. All the electronics are enclosed in separate compartments as not to interfere with each other.

Blending this with stylish aesthetics, produced a beautiful sculpture - to be known as the Heros.

Operating all devices in Class A produces a fair amount of heat. The design of the chassis is such as to cope with the heat and yet to provide vibration free structure for the components. All the electronics are enclosed in separate compartments as not to interfere with each other.

Blending this with stylish aesthetics, produced a beautiful sculpture - to be known as the Heros.

Thrax Teres Transformer Coupled Hybrid Power Amplifier

574dface2ba38f461ee5d5dcd55e21ae
14e992863dcc2bc210e71f34dc668a8a

​​Teres is based on the architecture of Spartacus. Single ended tube input stage loaded with a phase splitting transformer controlling 2 separate single ended amplifiers out of phase for output. Tube clarity and tone with might and power. Unusual but true..

The C3g tube handles all the voltage gain needed. It uses a new generation or our choke rectified shunt regulated power supply. providing a stable noise free operating point.

​The output stages power JFET cascodes form a bridge that when unbalanced by the input signal via the 2 transformer windings restores balance by adjusting their current conduction through the load. The 2 output stages actually enclose the load in their control loops and try to correct each others errors reducing the distortion to negligible amounts. Leaving only the “character” of the input stage.

To reduce the amount of heat generated the amplifier is biased so only the first few watts are in true class A. Music has a crest factor of 10 on average. This means that most of the time we listen with 1/10th of the peak levels. For most users this translates to under 5W. Peaks could reach over 100 watts during playback, but only for few milliseconds. For those peaks the amplifier switches to class B using only one leg of the output stage capable to deliver up to 250W/8ohm and 350w/4ohm.

Not sharing a common reference point apart from the load itself makes the second leg correct for any error produced by the first one. Result is an amplifier capable of extreme power wile delivering exquisite smoothness and low level resolution.

The driver stage is isolated by the interstage transformer and is not influenced in any way by the output stage maintaining tone colour and resolution at all powers and loads.

​Having the output stages floating around the load prompted for the use of a unique biasing technique powered by solar cells. This was the quietest galvanically isolated power we could come up with. Using opto-electrical conversion we control the bias by adjusting light intensity in the control circuit.

Teres is an example of modern technology solving age old problems allowing us to go a step beyond.

The physical appearance of Teres is typical Thrax understated elegance. Exquisite finishing ads to the presentation of a solid block of metal like a bigger brother of Heros.

Thrax Pre-Amplifiers

Thrax Dionysos Line Level Pre Amplifier

c139e435abd6610554aa7c1031cf7c7d

The Thrax Dionysos Line level Pre Amplifier is a single ended single triode stage transformer coupled remote controlled line level amplifier. The line level amplifier (or control amplifier as some call it) is still the centerpiece of each high-end audio system. It is responsible for source selection, signal conditioning and volume control. It has to match the output of your source to the input of your power amp. It’s output stage must be able to drive any power amplifier through any length of cable with optimal quality via balanced and unbalanced lines.

​The input of a pre amplifier has to provide optimal loading for the source component to develop its full potential, allowing it to extract the maximum amount of information while keeping all noise picked up on the way out of the signal chain. During our research and evaluation of the various approaches we settled for the most obviously "perfect" one - a transformer.

​Advantages: Inductive loads are very well received by almost all output topologies. It provides complete isolation from the rest of the system. It stops EMI/RFI and ensures truly balanced differential input. To be fair, transformers have disadvantages - expensive, and few know how to make them properly.

Our solution is fundamentally different and as simple as it gets. We use a transformer with various winding ratios and we switch between them. In the above example, it will operate as a gearbox, transferring all the power of the engine at all speeds. In other words CONTROL.

​The output has to provide an absolute optimum driving signal for the following power amplifier's input stage. This would include: low impedance, DC free, truly balanced, isolated from noise sources and the other components in the system. The solution yet again is a transformer. As a bonus it is the perfect load for vacuum tubes.

​The power supply of Dionysos uses a very well-known but very rarely used arrangement. We use a C core transformer with a gap in the core to prevent any saturation of the core from distorted mains or a DC component in the mains. As the power transformer is in the same case we chose a design that has two identical sets of windings (essentially two transformers) on the same core. This way all the nasty magnetic and electrostatic leaks cancel each other out, and it does not influence any of the other components in the chassis.

​To reduce power supply noise further instead of the normal arrangement of bridge rectifier and electrolytic capacitor bank we use a choke input full wave tube rectifier arrangement. This practically eliminates switching noise and, anything that comes before the choke stays there. The other beauty of this topology is that the choke keeps the voltage on the capacitor bank pretty constant. In our case, this allows for a considerably small capacitor bank, opening the possibility to use vibration resistant military paper in oil capacitors for the job that would last a lifetime.

The chassis to house all of the above had to be a masterpiece itself. Milled from solid plates of different thickness, 6061T651 grade aircraft certified aluminium are interlocked together to form a ring ​like structure.

We have the power supply mounted on the left sidewall of this structure and the signal transformers on the right side. Two semi-flexible bars on which the motherboard and its heat sink are attached, which then support this ring. The top and bottom plates link all sides of the ring with the flexible bars
and the feet made from carbon doped plastic on flexible screws. There are no screws visible anywhere on the chassis. And the chassis feels ​like a solid block.

The ergonomics. It does not matter how advanced a technology is, if it is not useful. We have tried to anticipate all the possible needs of our clients. The Dionysos has 4 inputs on RCA connectors and 2 inputs on XLR connectors. All inputs are balanced differential unless the grounding option is engaged on the RCA inputs. As we switch the ground plane when switching inputs there are no hum loops and induced ground noises, allowing for large setups with multiple sources powered from different mains outlets in the house. It applies for the outputs as well. For those that use bi-amping we have accommodated 2 sets of outputs active simultaneously. It might be considered legacy but we have implemented a buffered tape output that could be disengaged when not in use. We also have a level set function programmable on any input allowing the pre-amplifier to be in the loop of home cinema processors when the music and cinema systems share amps and front speakers. Controls are as simple as possible and self-explanatory, but we have also added an absolute phase switch with memory for each input. All functions are available on the supplied remote.

Thrax Orpheus Phono Pre Amplifier

652fe23dbeb99457b515ed790d2ac9e7

The Orpheus phono pre amp with LCR RIAA equalizer is designed to amplify the delicate signal of your phono cartridge without loosing any of it and to apply the RIAA equalization. To achieve this our design accommodates for both MC and MM type cartridges trough the use of a built in Lundahl cobalt amorphous step-up transformer for the MC types. It is switchable for low and high impedance providing for appropriate load and gain to the wide variety of cartridges.

The input stage is a super low noise D3A German tube used as a triode. It provides the necessary gain and output impedance to feed the custom passive LCR RIAA correction.

The equalization is done with high precision custom made Sowter coils wound with Cardas wire and Jupiter BeeWax impregnated paper capacitors. As the equalization is passive the next stage is another super low noise tube (a Russian triode similar to WE417) loaded by a Hashimoto transformer to bring the signal up to the necessary level and provide complete isolation of this sensitive circuit from the following components.

The equalization is realized through the use of constant impedance filters. This means that the input stage feeding it sees a constant load. This making its behaviour predictable at any signal level. The filter is out of any gain loops and sits between the two gain stages protected from any unpredictable influences. To make it properly, very precise high quality inductors are needed. The signal is in series with them so high wire quality and magnetic material with lowest hysteresis are used. This solution was made popular years ago by a Japanese transformer manufacturer called Tango. Due to the difficulty in manufacturing consistently and the high cost associated, penetration of the idea was very limited to just a few manufacturers. Yet this was enough to start a worldwide movement worshiping it. We decided to go a step further, making it from discrete components allowing us to tune it in context of the whole system.

The power supply of Orpheus uses a very well known but extremely rarely used arrangement. We use a C core transformer with a gap in the core to prevent any saturation of the core from distorted mains or a DC component in the mains. As the power transformer is in the same case we chose a design that has two identical sets of windings (essentially two transformers) on the same core. This way all the nasty magnetic and electrostatic leaks are cancelled and it does not influence any of the other components in the chassis. To reduce power supply noise further instead of the normal arrangement of bridge rectifier and electrolytic capacitor bank we use a choke input full wave tube rectifier arrangement. This practically eliminates switching noise and anything that comes before the choke stays there. The other beauty of this topology is that the choke keeps the voltage on the capacitor bank quite constant. In our case, this allows for a considerably small capacitor bank opening the possibility to use vibration resistant military paper in oil capacitors for the job that would last a lifetime.

The ergonomics. It does not matter how advanced a technology is, if it is not useful. We have tried to anticipate all the possible needs of our clients. Orpheus has 2 inputs on RCA and one on XLR connectors. All inputs are balanced differential unless the grounding option is engaged on the RCA inputs. As we switch the ground plane when switching inputs there are no hum loops and induced ground noises, allowing for large setups with multiple sources powered from different mains outlets in the house. It applies for the outputs as well. We have provisioned for balanced and unbalanced outputs. Switches on the back, which program the unit, set the gain and type of each input. We have included an absolute phase switch with memory for each input.

Thrax Dac

Thrax Maximinus Audio DAC

db9e293ed0aaa4103a3189d778142acd

The Maximinus Universal DSP controlled discrete resistor ladder 32bit/384kHz Digital analog converter (DAC) is the simplest DAC concept with the most sophisticated front end. Maximinus uses a concept called R2R ladder. This is a resistor matrix that is switched for the various output levels outputting a fraction of an internal reference (much like a volume control). This concept was introduced to audio by Philips with the TDA1540 and the first CD players. It was followed by further developments that reached culmination with the sign magnitude Burr Brown PCM 1704 chip. All new development chips currently used by high end DAC manufacturers are delta sigma modulators. They use very sophisticated techniques to achieve very good parameters. Unfortunately never being able to match the sound quality of the multi-bit convertors mostly due to the inherent problems in the single-bit concept itself, forcing the use of filters following the DAC output.

To take advantage of the available resolution and bandwidth we had to implement a state of the art digital pre-processing. This is a suite of algorithms that would apply digital filtering and up sampling to the incoming data stream. After processing a 16bit 44.1kHz CD data stream it is converted to 32bit 352.8kHZ data fed directly to the DAC. This greatly improves low-level resolution and the sense of space. The processing is all user controllable and defeat able for purist non-oversampling use mostly used with high-resolution source files.

In order to bring the performance of multibit DAC’s to new levels the only solution is to build them from discrete components as there are no off the shelf chips good enough. This involves the selection of ultra precision resistors, thermally coupling them and building a very fast and sophisticated switching logic to control them. This comes at a substantial cost as one such resistor costs more than 12 Euro!! , but the result is an order of magnitude better performance than what is achievable by IC based dac's.

​Apart from taking the uncompromising approach of using discrete ladder technology we have taken exceptional care with the construction and operation of each circuit within ​the DAC. We use transformer or optical decoupling of all I/O ports, meaning that contaminated ground and other ​interfering signals don’t make it to the inside of the unit.

We have no output buffer or filter at the DAC output providing the cleanest possible output signal. We also use separate power supplies for (not separate regulators ​but physically separate transformers and all) the Converters, Clock, DSP and Control logic all with floating ground planes and our unique constant current regulator technology.

​Then the whole assembly is mounted in a solid aluminum case for vibration damping and EMI/RFI screening following the same construction concept as our preamplifiers.

​It’s a different league from all currently available high end DAC’s hence the name Maximinus.

Thrax Lyra/Basus Speakers

Thrax Lyra Speakers

"A Stunning Success!! "

6e3b6974ffcc96e6d5fb397531e8e065
7de78c8bb4c1fb85b4eaa5304a54c0a6

Loudspeakers are the final component in the reproduction chain. You will hear all kinds of opinions on the importance of loudspeakers in the replay chain pitched by gurus around the globe. Most leaning to the fact that loudspeakers don’t measure very well. They all have considerably high harmonic distortion, limited bandwidth and frequency nonlinearity. Those are the parameters that are easily measured even by amateurs. Other, just as important parameters, are directivity, stored energy, time alignment, those are discussed by connoisseurs and by marketing people and require a higher level of equipment and knowledge to identify. Things like signal to noise (yes speakers have signal to noise!), break up behaviour, diffraction interference and others are discussed only among people with true experience and understanding. To measure those you need special jigs and equipment that is usually purpose built.

Loudspeakers are the final component in the reproduction chain. You will hear all kinds of opinions on the importance of loudspeakers in the replay chain pitched by gurus around the globe. Most leaning to the fact that loudspeakers don’t measure very well. They all have considerably high harmonic distortion, limited bandwidth and frequency nonlinearity. Those are the parameters that are easily measured even by amateurs. Other, just as important parameters, are directivity, stored energy, time alignment, those are discussed by connoisseurs and by marketing people and require a higher level of equipment and knowledge to identify. Things like signal to noise (yes speakers have signal to noise!), break up behaviour, diffraction interference and others are discussed only among people with true experience and understanding. To measure those you need special jigs and equipment that is usually purpose built.

Main feature of the speaker system is a solid aluminum carved front panel incorporating a horn with specific directivity and frequency response as to perfectly match the directivity of the adjacent woofers. The geometry of the front plate is such that it is time aligned between the tweeter and woofers. We use a custom designed ring diaphragm compression driver. The driver features a 118db/watt sensitivity and a very low resonance allowing us to use it effectively to below 1500Hz. This driver/horn implementation allows for very high levels and low distortion as well as very low spurious noises. The magnesium diaphragm woofers are able to match the speed and clarity of the tweeter at the expense of a very nasty break-up mode in the upper midrange, so utmost care has been taken to make sure this resonance never gets excited, so the 3rd order crossover is 3 times lower then the nasty resonance.

​The speaker system features an uncommonly high sensitivity of 90db making it usable with lower power amplifiers or in very big rooms while bass extension is generous considering the size of the speaker and it’s efficiency. It sustains severe levels of abuse without giving in. The filters are constructed with resonance free foil wound inductors and premium Polypropylene capacitors. The all aluminium body is internally damped for any internal vibration and resonance. Styled in the Thrax family of products. The value of our speakers is in the knowledge required to make them not the materials used. Time precious materials lose their mythical properties in audio. Diamonds and silver are more expensive, so they must be better is the general assumption. (I really doubt, gold teeth are still going on in dentistry). Aluminum has 2 times lower resistance per unit weight for voice coils compared to silver or copper. If we talk speed this is it.

​As said music is sounds and silence. Everyone tries to improve the sound and we just improve the silence.

Wasn’t Thrax an electronics company? What do you know about speakers? 28 years ago, I designed and built my first speaker. Even then with a ribbon tweeter (by Decca) and with an Alnico magnet bass unit. Needless to say it did not sound that good. Main reason was …It was designed by the book. No thinking involved. Following the rules gets you a mediocre (average) result.

​During my studies at the Danish Technical University in Copenhagen on the subject of acoustics, coincidence introduced me to a character called Steen Deulund, and then I found I knew nothing.

A number of books later I regained enough confidence to point at a good, or a bad design. Than I found out how little people know about good speaker design, and certainly could identify a good one.

Leap forward 20 years to an argument in Sofia about tweeters used in high end speakers and which sounds better accompanied by demonstrations and piles of marketing bla bla. Until an older gentleman comes to me and said “you should hear what Plamen has done”. Never payed much attention to this as Plamen has been doing cinemas and studios for as long as I can remember, but then again curiosity prevailed. A quick listen to a prototype set of speakers using a ring diaphragm compression driver loaded with 2” short (that is very short) horn, was a smack in the face.

The experience was backed by a very thorough paper written on the subject of intermodulation products of tweeters and sub harmonic distortion acting as contaminants to the background and reducing signal to noise of the speaker system. Apart from the usual break up modes there was excursion induced nonlinearities, cavity resonances, magnetic modulation, diffraction and other subjects covered.

But the proof of the pudding is in the eating. So building a speaker system as a showcase for this technology seemed the logical next step. Beware!

Specification

Impedance: 4Ω
Max Power Handling: 250W
Dimensions: 210W x 385D x 520H mm
Weight: 35Kg
Finish: Black or Silver anodized aluminum
Sensitivity: 90db
Frequency band: 34~20 000Hz

3607169639b3a41b6e343c990c02dd88
d7e0ba8220875a830d90de371446e7da

Thrax Basus Motional Feedback Bass System

The Basus subwoofer system consists of a bass driver with motional feedback in a sealed metal enclosure, a 1 kilowatt power amplifier, and an active crossover system with switch selectable preset frequencies. The active crossover system allows the Basus system to be used with any speaker/amplifier combination.

The Basus subwoofer project was started as a bass extension unit for the Lyra speaker that evolved in the ultimate universal woofer, for that we provide a matching stand to mount the Lyra on top of the Basus making it one system visually. The Lyra/Basus/Teres combination creates an active speaker system with unmatched synergy and visceral impact. Undistorted bass at high enough levels creates a different perception of the music. When the group delay is very low pulses and percussive instruments become solid and with well defined bodies cleaning up the murkiness in the soundstage creating a see through effect allowing one to hear the reverb and reflections in the back of the sound stage. You don’t hear more bass but more music!

980b3cf9ef1caad9720b06b035f6eae9

Basus and the need for better bass

​In Music school you are taught music has three main elements; rhythm, harmony, and melody. The bass notes lay the foundation for two of these three elements; Rhythm and Harmony. Numerous studies confirm that people react to low frequency beats and can detect timing errors in a low frequency beat more accurately than in mid and high frequency ranges. Bass is not rumbling noise associated with a passing starship, but the foundation of a musical passage. It is the lowest note that defines the harmonic structure in frequencies above it. It can make music sound Sad or Bright, or Mellow. Music contains two very different types of bass; one that is continuous and sustained, and the other pulsed and very short (percussive). Proper phase alignment in sustained bass creates the right waveform allowing you to identify the instrument and to enjoy its sound texture. The pulsed bass gives you the beat and timing of the piece. Any delay there and the sound is ever so slightly off beat making the music less enjoyable. Small ported speakers and home cinema subwoofers exhibits very large group delay giving meaning to the term “murky bass”. Large ported enclosures behave better having this problem only in the very low bass. The best overall performance is achievable with sealed enclosures and especially large sealed enclosures. In large sealed enclosures the group delay is the lowest and less obtrusive. Unfortunately, the efficiency of sealed box systems is very low. For any given cut-off frequency, the box size will have to double for each 3dB higher efficiency. Now you can understand which manufacturers "misinform" you. Below are the maximum theoretically possible efficiencies for sealed box speakers expressed as SPL per watt:

2 cu ft 30hz = 84dB
4 cu ft 30hz = 87dB
8 cu ft 30hz = 90dB
16 cu ft 30hz = 93dB
32 cu ft 30hz = 96dB

The conclusion is that to have deep bass at high levels you need large speakers and a lot of power

Cabinet’s re-radiated energy- think of a drum, vibraphone or an organ pipe. The speaker enclosure is a resonator in its own right. If you create a pulse at its opening in a bit it will respond with its own voice - like an echo in the mountains. Containing this unwanted sonic signature of the box in the box and not letting it escape the enclosure is a difficult task. Heavy boxes with thick walls still leave just a thin diaphragm between you and the inner rumblings of the box. A heavier stiffer membrane and hard surround are used extensively in the more powerful subwoofers. The box resonances, inherently creates delays, and out of time tone bursts making the bass sound murky or boxy.

Harmonic distortion in the bass region seems to be so widely accepted that on the order of 5-10% of people complain about its absence, when it is not present. Putting this in the context of audibility thresholds means that in many systems you hear the “newly created” harmonic structure of a bass note without hearing the note itself. (Bose acoustimass, laptops, tv sets).

Modulation effect of bass on rest of music

The human ear is known to produce a fair bit of harmonic distortion itself leading to the production of intermodulation products. What does it mean? Illustrated for a simple explanation if you listen to a clean tone - an opera singer and you add a low frequency signal with appropriate but still unobtrusive level, meaning you don’t notice it, you will notice in fact a change in the tone of the voice. The voice will be modulated. A clean 3k signal will sound like a few 3k signals. This is some of the Basus of music creation. A bass tone modulates a higher pitch tone giving it its beat and coloring it, so in a solo an off beat note is accentuated by having a different character as not being modulated by the bass burst. Interesting.

0ba746688a4028b95f01d36faccb69d8

Noise and signal to noise
Up to now we discuss about how good it is to have bass modulating the upper tones and how it creates music, but when this is is done ad hoc without being related to the music we could classify it as noise and distortion. the level of these artifacts relative to the level of the reproduced signal will define the speaker signal to noise ratio. It varies with amplitude and spectral content and this is one of the reasons for not being able to quantify, but once experienced everyone can distinguish between various speaker systems.

One means of reducing those artifacts is to make sure they are not created in the first place. The use of feedback has been widely discussed and is the principle behind 99% of the amplifiers in use today. Why is it not applied to speakers? Not as easy and well understood or maybe too complex and with limited commercial value to be worth the effort, we won’t know for sure. Speaker drivers are linear electric motors, much like the one spinning a fan or lately the wheels of your Tesla. Those same motors are used in CNC machines and everywhere in our daily life. To make them do what we want we need to know what they do. The PC on which you are reading this has a fan mounted on the CPU that adjusts its speed according to the temperature of the CPU. This is just one example of a feedback system. You start playing a computer game and your CPU load increases dramatically, a bit later the fan stats spinning much faster trying to cool the CPU. The time it takes for the CPU to heat up so the fan speeds up and then restores the CPU temperature is called latency. Every feedback system has such a thing.​

1d3f1786422ca8544c1e2b81feb45644

​Now apply this to speakers. You have a bass driver and you apply a signal to it. The current running through the coil creates a driving Force accelerating the speaker cone - (voltage is not present in formulas for force driving the cone - only current) the acceleration creates a sound wave, but when you stop the signal the cone is still moving. It does not accelerate anymore so it does not produce sound anymore right? Not quite, now you have the cone mass, surround and spider forming a resonant system that damps the excitation. This system has its own “deceleration force acting on the cone (producing sound not related to our signal) and last but not least the magnetic motor generates EMF that when run trough a load (the crossover and amplifier output ) again creates current accelerating the cone in some unpredictable direction. It is a bit like beating a drum and trying to control the sound with the sticks.

Introducing a system that measures the acceleration and velocity of the cone and comparing it to the input signal gives us the ability to create a control signal forcing the cone to start and stop as does the input signal and not as the resonant system it comprises.

As described earlier all feedback systems have latency and a number of other limitations allowing us to apply only a certain amount of correction in a limited frequency band. In other terms, the sound can never be made perfect but over a limited frequency range it can improve performance more than 10 times!!! With the implemented feedback system we are able to control (lower) the resonance of the mechanical system in the box and the total Q of the system. Taking advantage of the fact that speaker drivers operating above their resonant frequency have much lower phase shift. 

To handle the 1kW of power and the high pressures and mechanical stress we built it in a all metal enclosure capable of withstanding the high sound pressures. 1kW? What for? At resonance the impedance of a bass driver can rise more than 10 times. To put this in perspective means that the power going to it is 10 times less. So your favorite 100W class A  amp that can pump 30 Amps in a speaker in reality can’t deliver more than 10 watts to the bass driver!! Yet once we go to a higher frequency with very reactive part of the impedance it can pump a few hundred watts in it for no reason what so ever charging and discharging some capacitor. 

Mass and its effect

A myth that large heavy cones make bass is just not true. The heavier the cone the more force and time is needed to accelerate it and to stop it. Heavier units are more robust and can take more power to compensate for the lower efficiency, so with a powerful amp you can make them sound just as loud as the lighter ones. It makes no sense but sells more speakers 😉 Anyone that studied acoustics will know that the formula for sealed speaker efficiency and extension there is a corresponding volume of the enclosure, but there is no mention about driver size, weight and shape. It is irrelevant. Only practical manufacturing reasons will define the size and weight of the driver.

Matching an active bass unit to an existing system requires removing the frequency band from the main speakers and level matching the two. We have built a crossover system in the Basus allowing for multiple interface types and level adjustments to suit every system. You always wanted that 20W SET amp to power the mid/highs and have thunderous bass at the same time, now is your chance. With it’s i/o flexibility Basus allows for balanced and unbalanced signals as well as 6 predefined crossover frequencies. amplifier damping illustrated  -  monkey phase shift around resonance (imaginary component illustrated).

Why no one else does and says this? Well that's not quite true. The principle we use were developed by Philips in the 60’s and 70’s and thousands of studio monitoring speakers have used it over the last 35 years. Unfortunately, most of the development effort was directed to making a small box sound like a large one and no one ever considered building commercially a big system. You can say that we have scaled, hot-rodded and turbocharged the concept, by developing a custom drive unit, drive electronics and bomb proof enclosure.. Is it that good?  And, then some!

TECHNICAL SPECIFICATION: BUILT IN AMPLIFIER

POWER: 1000W

DIMENSIONS: 450W x 450D x 800H mm

FREQUENCY BAND: 25~250Hz

WEIGHT: 130Kg

Thrax Modular Systems

ENYO Modular Audiophile System

51fcbf57f667c634010b80105a0998d0
a91f0de5f41e3a079d7bf25b72ba20cf

Enclosed in a single chassis is a complete audiophile system with the sound you want and the features you need. The ensemble features two fully balanced vacuum tube power amplifiers. The amplifiers are based on a widely available long life military transmission tube . We use a ultra linear push pull arrangement with cathode feedback for the output stage driven by a fully differential input and driver. The absence of global feedback and choke filtered power supplies contribute to the the exemplary transparency and speed. The amplification modules are fed trough a 4 input fully balanced relay based selector and volume control board.

On the input selector board you can add our new RIAA module accommodating both Moving Magnet and Moving Coil cartridges. This module is based around an exotic transconductance amplifier having properties uncommon in the price range. Gain and loading are selectable in the configuration menu.

For the digital age we have a second input board accommodating all modern digital interfaces. AES, Spdif, Toslink, USB, Network and Bluethoot inputs. This input module apart from switching the inputs has the ability to apply different filters and upsampling to supply the DACs always with the optimal data rate and format making it the truly universal digital input module. All data is re-clocked to the high precision internal clock generator. The DAC modules sit on a separate card allowing upgrades as new technologies become available. The integrated network player board supports DLNA, Roon, TIDAL (including MQA) and DSD.

The DAC modules are simplified versions of our multi-bit R2R module developed for the Maximinus DAC offering similar tonal rendition and lifelike dynamics.

Of course there are separate power supplies to power the digital and analog circuits.

So you get a multibit DAC, an Upsampler, a Network player, a Phono preamplifier and two 50 watt vacuum tube power amplifiers in a single box. Each designed to work with the other in the most optimal way.

The modularity of the system let’s you grow as you need and to keep it current for long time to come.

AVAILABLE FINISHES: SILVER ANODIZED ALUMINUM and BLACK ANODIZED ALUMINUM.

ARES Modular Audiophile System

2ce8f5a149f142dc173fce8c199edd09

Enclosed in a single chassis is a complete audiophile system with the sound you want and the features you need. The ensemble features two true mono power amplifiers with independent power supplies based on a new concept in solid state amplification. We use a non switching output stage arrangement isolated by a single ended buffer delivering Class A liquidity and texture uncommon to solid state designs without the associated heat. The absence of global feedback and regulated supplies of the drivers contributes to the the exemplary transparency and speed. The amplification modules are fed trough a 4 input fully balanced relay based selector and volume control board.

On the input selector board you can add our new RIAA module accommodating both Moving Magnet and Moving Coil cartridges. This module is based around an exotic transconductance amplifier having properties uncommon in the price range. Gain and loading are selectable in the configuration menu.

For the digital age we have a second input board accommodating all modern digital interfaces. AES, Spdif, Toslink, USB, Network and Bluethoot inputs. This input module apart from switching the inputs has the ability to apply different filters and upsampling to supply the DACs always with the optimal data rate and format making it the truly universal digital input module. All data is re-clocked to the high precision internal clock generator. The DAC modules sit on a separate card allowing upgrades as new technologies become available. The integrated network player board supports DLNA, Roon, TIDAL (including MQA) and DSD.

The DAC modules are simplified versions of our multi-bit R2R module developed for the Maximinus DAC offering similar tonal rendition and lifelike dynamics.

Of course there are separate power supplies to power the digital and analog circuits independent from each other and the power amplifiers.

So you get a multibit DAC, an Upsampler, a Network player, a Phono preamplifier and two 120 watt mono power amplifiers in a single box. Each designed to work with the other in the most optimal way.

The modularity of the system let’s you grow as you need and to keep it current for long time to come.

AVAILABLE FINISHES: SILVER ANODIZED ALUMINUM and BLACK ANODIZED ALUMINUM.