Thrax Audio Tube Technology


Has over 100 years of history. It started with the simplest possible amplifier, which is shown below (fig “A”).  
This is a `single ended’ common cathode amplifier, referred to in the industry as Single ended triode or SET. 
This type of gain stage is the basis of our amplifiers and is recognized for its superior sonic signature.




Audiophiles have been led to believe through published measurements at maximum output power and uneducated reviewers that single-ended triode amplifiers produce vast amounts of harmonic distortion. 
As a matter of fact triode vacuum tubes are by far the most linear amplifying devices in existence today. They produce the least amount of distortion, and that distortion is predominately second harmonic, which is the least obtrusive type for the sound. By contrast, pentodes produce greater distortion, and the third harmonic tends to dominate. A transistor looks at best like a very bad pentode.


To state the obvious, a single-ended circuit must operate in Class A1 or A2. A push-pull amplifier may be Class A1, A2, AB1, AB2, B1, or B2. Class A indicates that each output tube handles the full cycle of the audio signal, while AB and B allow some of the devices to cut-off during a portion of the cycle. 

Subscript “1” indicates that no grid current is drawn by the output tube, while subscript “2” indicates that the output stage enters the grid current region of operation. In the grid current region, the impedance presented to the driver stage is abruptly lower, and real drive power is required, not just drive voltage. 

The grid current region tends to be a rather non-linear load for the driver and most designers will avoid it. 

Single-ended and push-pull circuits may be built with triodes, beam power tubes, pentodes, or the latter two in ultra-linear mode. We use exclusively Class A push-pull circuits for our output stages, there is a natural cancellation of even-order harmonic distortion products in this topology. 

The cancellation is not complete, but it would be unusual to see large amounts of second harmonic distortion from a push-pull circuit. Note that a push-pull circuit has no significant ability to cancel odd-order distortion products. If low distortion performance is required, one must avoid the generation of odd-order harmonics in the first place. A good triode tube meets this requirement. 


In single ended operation there is no mechanism to naturally cancel harmonic distortions and available power output is greatly limited. The full DC current for the output tube(s) flows through the transformer primary and strongly magnetizes the core of the transformer. Thus, much of the core’s ability to couple the audio signal is used up by the non-audio DC current, and causes the core to saturate asymmetrically with audio signals. Adding parallel output tubes for more power directly increases the DC magnetization current, thus exacerbates the distortion problem. 

To deal with this an “air gap” may be introduced into the transformer core. 
In most cases also a greater amount of core material is used, which in turn makes the whole unit larger. By increasing the size of the coils we soon become limited by parasitic capacitance and leakage inductance affecting the bandwidth of the transformer. The final result is either a higher degree of distortion (all harmonics with the second dominating, increasing with decreasing frequency), a measurably peaked frequency response, or both. As observed the best sounding single ended designs rarely reach above 20W and it is practically impossible to manufacture a transformer of utmost quality for more than 80W in single ended triode operation. 

Since the distortion in the single-ended transformer is asymmetrical, a system based around this type of amplifier might be more sensitive to absolute polarity. 

The same problems are true for the push-pull transformers but in that case the power limit for the same size transformer is 4 times higher.


Traditional theory gives negative feedback high marks. Consider that when the “error” signal is fed back into a non-linear amplifier, it multiplies the distortion order. For example, if an amplifier naturally produces second harmonic, negative feedback will create a second harmonic of that second harmonic, which is the fourth harmonic. If the basic amplifier has second and third, the fed-back amplifier will contain second, fourth, sixth, and ninth. As is well known, the higher orders of distortion are far more objectionable to the ear than lower orders, and odd orders more offensive than even orders. 

Thus it may be possible to lower the level of distortion products and still have the distortion be more audible? 
The application of negative voltage feedback also reduces an amplifier’s measured output resistance, i.e., it raises the “damping factor.” Here again, the measurement fails to capture the essence of things. 

A feedback amplifier has better control of speaker motion because that motion creates a voltage (the back e.m.f.) which enters the feedback loop through the amp’s output terminals. The amplifier then cancels out the error signal. 

A negative feedback amplifier maintains better control over speaker motion because the erroneous motion creates a voltage (the back e.m.f.) which enters the feedback loop through the amp’s output terminals. The amplifier then cancels out this error signal, which corrects the motion of the speaker. 
However, like many theories, this is an oversimplification and, in practice, the opposite result may be obtained. Quite often the motion of a speaker’s voice coil former may not match the acoustical output due to cone break-up and the fact that the motion of the coil former is being sensed by the voice coil, which is a reactive element with phase shifts and delays. The back e.m.f. passes through a cross-over network, which will again alter phase and frequency relations. By the time the error signal reaches the power amplifier it is arguably an erroneous error signal. As the power amplifier attempts to correct for this signal, it may actually do the exact opposite with respect to the speaker’s acoustic output. 
Small-Signal Distortion in Feedback Amplifiers for Audio


In general a signal passing through an amplification stage will have some distortion added to it. (In our case this will be almost only second harmonic.) And when that signal passes through the next stage it would be adding distortion to the distortion generating a minimum amount of fourth order distortion and so on. (Pretty much like the effect of feedback described above)

Consider now the following: the usual preamplifier (tube or solid state) has 3 stages – input buffer, gain stage, output buffer; then the simplest power amplifiers have 4 stages – input, phase splitter, driver, output buffer. And all of this is dependent on the signal amplitude via an extremely non-linear function. 

In order to minimize this effect the preceding stage of any amplification stage should have at least 2-3 times lower distortion than the latter. (Would that be possible when we have 7 stages in the signal path?)
Minimizing the number of stages reduces drastically the order of distortion and its inter modulation products. 
The sonic result is vastly improved transparency and speed. Our products feature the minimum sensible number of stages implemented with the most linear devices available operating as close to theoretically perfect operating conditions as possible. 

Result? You be the judge!


Amplification stages are only half of the story. As seen on fig 1 the power supply is represented by a single capacitor. This assumption alone has ruined many beautiful designs. 
In theory a large enough capacitor is as close as you get to the theoretically perfect power supply. Well, not so in practice! You have a rectifier and the mains supply connected to it. For every cycle of the mains you have two things happening. You charge the capacitor through the diodes and the mains transformer for a limited amount of time by connecting it to the mains supply with all its noise and garbage and then you discharge it through the amplifier and load until the next charge cycle. So your power supply is constantly varying its value and for a portion of the time is connected directly to the polluted mains line. (don’t forget the fact that this capacitor is part of the signal loop!!!) 
And this seems to be acceptable for all electronic design gurus?! Here is an example of a good amplifier design and its power supply.
By looking at the whole picture you see a coil (choke) separates the reservoir capacitor from the signal capacitor, thus preventing any noise from the mains reaching the signal capacitor and as a side effect keeping a constant voltage across it. 
It is very unfortunate that no currently available commercial products feature similar topology. On first glimpse it looks simple yet it does all that is required by the PSU to approach the theoretically perfect with minimum component count.



A further improvement to the above example is the use of a choke-loaded rectifier. This reduces the current fluctuations (pulses) needed to charge the reservoir capacitors and stops the diodes’ switching noise from entering the signal loop. Although high frequency interference noise is not audible its intermodulation products from the interaction with the audio signal are very noticeable. 
So avoid it at all costs. Now leap forward a hundred years and implement this technology with the most advanced components and circuits possible and you are pretty close to perfection.


In all directly heated tubes the signal current passes through the cathode of the tube with its specific resistance generating AC voltage across it. Its amount is dependent on the tubes used and operating conditions but it is ALWAYS there. 
In almost all implementations today there is a large size smoothing capacitor across it producing a practical short circuit for the AC signal. The designers justify this in order to have a stable DC and no hum in the filament, but it ruins the operation of the cathode in AC. The simplest solution was implemented again many years ago, it is a simple choke in series with the filament. 
This is the simplest possible solution providing separation of the AC path and the DC path. Simple yet almost always overlooked. It is practically impossible to make a good enough inductor with reasonable size and cost to tackle the task, but a modern circuit called a gyrator can achieve the same effect.


Now that we have designed a simple, yet sophisticated amplifier, building it in practice to perform to design specifications presents a number of issues on its own. One of the most overlooked aspects of design is the mechanical construction and component vibration control. 

As you probably know sound is recorded by sensing air pressure with microphones. There are quite a few types of microphones but they fall into two main categories: electromagnetic and electrostatic. 
Electromagnetic types work by sensing the variations in a magnetic field and logically the electrostatic work by sensing the variations in a static electric field. 

Well guess what, every wire in a magnetic field is an electromagnetic sensor. Does not matter what the cause of magnetic field variation is (the wire vibrates in a constant field or the field changes) you have a parasitic signal entering the signal path. Each capacitor is in practice also a microphone superimposing the environmental vibrations on the voltage across it. This effect is only dependent on the mechanical construction of the device itself. 

This applies to vacuum tubes as well as to solid-state devices like transistors. The only way to reduce those effects beyond the threshold of audibility is by proper mechanical design of the equipment layout and chassis and the selection of components with a correct construction and not only the correct electrical parameters.


Another source of impurities that might enter the signal path comes from the air. Known as EMI/RFI (Electromagnetic interference and Radio Frequency interference). This is the effect of every piece of wire working as an antenna for airborne electromagnetic waves. 
Those get rectified and produce signals that interact with the audio signal and the intermodulation that occurs is very audible. Some of you may be old enough to remember AM radio and its specific background noise – this is what you get in your amplifiers. 
Quite often the source of this interference is within the chassis of the device itself (bad PSU design, transformer and rectifiers) and more often than not enters through the power supply or other attached wires (noisy power line).


There is a way to reduce induced noises in the interface between components. 
Transformers stop many of the polluters from entering the signal path by the nature of their operation and construction. 
A good description of how and why is available in the white papers at:


Using transformers to separate each gain stage of the system from the outside world is a good way to secure clean signal handling. Resulting in high s/n and much more expressive contrasts in the music nuances and timbre. This is also one of the reasons why our amps have such unprecedented clarity.


This could be argued a lot. But here are our observations: At normal listening level (approaching and slightly exceeding the natural level of sounds) the peak SPL reaches 110-112db. So a high fidelity system has to achieve this in the listener’s room. Average listening volume is around 86-93db, the higher levels being peaks in the program. Consider also that the average noise level in a listening room rarely goes below 24-26db SPL. 
Now add speaker efficiency to the equation and you will get something like this: 
If your speakers are “modern” high performance super duper design with a 82-86db/watt sensitivity look for a power amp capable of 400+Watts. (most probably class D types) 
If you have been more sensible and opted for a speaker with a 88-93db sensitivity you don’t need more than 100w to enjoy the full scale of orchestral pieces and the selection of fine solid state and tube amplifiers the market has to offer.
If you have your house built around a pair of full range horns with 108-112db/w you might consider a “spud” tube (single tube) amplifier with 3-6W to rock the neighborhood. 

Any other combination will not lead to the desired result.


There are three different aspects to be discussed in DA conversion. 

First is the Conversion process. 
Second are the processes before conversion takes place (DSP), third is the timing of it all or in other terms – clocks. There are multiple books on each of the above subjects, so don’t expect an in-depth analysis but more a guide for what to look for in a DAC. 

Multibit vs. Single bit (bitstream) converters. Google it. There are 1000s of pages, but it comes down to this: Multibit when correctly implemented is better, but far more difficult to make at high resolution. Hence cost is prohibitive. There is no single chip solution as of today! There were some 20-bit converter chips (burr brown) in production and they have legendary status among users. Even today there are new designs coming out built around the first Philips chips (20 year old tech) claiming audiophile performance and unbeatable sound. We have taken the road of multibit converters and have requested MSB to develop a version of their discrete DAC chips to meet our requirements. You can read about their principle of operation on the MSB site. 

The hype about high sampling rates and the argument for non-oversampling DACs. Well in order to avoid this argument we decided to support both. 
We support high sampling rates up to 384kHz accommodating all existing formats, yet we are not bats – it is the intermodulation and filter characteristics that are audible, not ultrasonic signals. 
Again to overcome all arguments and to accommodate user preferences we support multiple options for upsampling, re-clocking and digital filtering. Any signal processing on a 24-bit word of data will generate extra data. Almost all other DACs and processors will truncate the data back to 24-bit. We let it flow at 32-bits keeping all information, as insignificant as it might be. Otherwise those errors accumulate and after 3-4 processes (re clocking, upsampling, dithering, energy filters) they become significant and are clearly audible.


Apart from data fed to the converter, the other major influence is the time it takes for the conversion to happen. You can google “jitter” and will be swamped with lots of meaningless pages. In essence you are interested in the cleanliness of your clock and short-term stability, also known as phase noise. You will see companies offering rubidium, atomic, GPS locked, tourbillon or hourglass clocks out there but a properly designed and manufactured quartz oscillator is pretty much unbeatable for audio. Having said that please note that less than 1% of the ones we have seen were up to standard. . 

The most interesting property for us is low phase noise at the very low frequencies and high s/n of the clock signal with no distortion. 
All conventional DACs require buffers and filters. Those are always implemented with opamps. (even discrete opamps) They isolate the DAC from the output and sum the outputs of multiple DAC chips, convert I to V and other tasks, but it does not matter how good they are, they always severely limit the sonic performance of the DACs. Whatever you have done up to that point performs at the quality level of the opamp used. (isn’t this sad?). Fortunately on multibit DACs with low impedance resistor ladder (it has to be low resistance to keep the noise down) you can take the output directly from the conversion resistors. This way we can sum the multiple DACs with a transformer with multiple primaries and by taking the output from the secondary of the transformer we have a complete solution. 
Isolation of the output from the DAC, stopping any high frequency content passing trough the band limited transformer avoiding IM distortion and summing out of phase, removes all common mode noise and artifacts. You can call it „the magic of coils”.


Transformers don’t measure that well. They have limited bandwidth and rising distortion at high levels and low frequencies. All those shortcomings are true, but would seem to be substantially better performing in dynamic conditions than ANY op amp. The effect is jawdropping not subtle.


As explained above our products are very complex and use very special parts at substantial procurement cost. The price of our products reflects directly the cost of materials involved in its manufacturing and the time required to assemble them. All technologies developed are reused in our other solutions and products. 
Constant R&D is undertaken to further the performance of the existing products and to optimize manufacturing process, which should bring better products in the future.