Exclusive interview with Stavros Danos from Aries Cerat. This is a must read for anyone highly interested in ultra high-end, extreme tube amplifiers, digital, ultimate horn speakers etc. Its a lengthly reading and it took quite some time to finish, but this is the man who have something to say! Enjoy...
How did all started with you and electronics?
I was always fascinated with electronics, as a young kid, always playing with random parts and circuits, trying to put stuff together. I was reading about the Tesla, Faraday, and I was immersed in their world. As an adolescent, I was playing with digital, trying to design controllers and processors from gates and registers. My obsession with digital was complete with a controller that had the opportunity to develop and implement at the robotics lab at the university. However, analog and tube electronics finally won me over. You can say I went the other way around ☺.
My early inspirations regarding electronics was N. Tesla. My audio inspirations in my early steps, was… music to tell you the truth. Late night music sessions were the drive to make components that would make the heart warming music i was listening to, to sound lifelike.
Do you consider yourself an audiophile?
I am passionate about music, and even as an adolescent, always pursued the highest fidelity possible for the music that gave me shivers while enjoying it. Though audiophile is a commonly misunderstood term, you can say I am a audiophile.
When did Aries Cerat formed and why?
Aries Cerat was formed in January 2010. Although,many of my designs, were implemented and developed even before the establishment of the company. My ambition was to get as much people as possible interested in true high fidelity audio. The disappointment about the majority of the audio components being mediocre at best pushed me to develop designs that reproduce music as true to life as possible. Audiophiles and people from the industry really liked our designs even before the company launch. The formation of the company was… something waiting to happen.
Why the name?
Aries is the ram,the mouflon( coincidentally our country’s national animal:). The latin word Cerat is derived from the greek word eras, a word which describes the horn shape.
What do you produce in house and what are your standards when it comes to subcontractors for specific parts?
Many parts needed for our designs are custom made. Fortunately, and after a lot of “trial and error” procedures, we found part manufacturers that can meet our standards and meet our needs with reliability, for parts like film capacitors, transformers etc.
Our three CNC work stations (glass sheet work centre, pub work station) and our pride, a 5 axis CNC with the largest Z-axis working envelope in the CNC market, gives us great possibilities in experimenting, developing and manufacturing horns and speakers designs which would not be possible to be built otherwise. The only limit in our speaker manufacturing capabilities is the imagination.
We use our own proprietary sandblasting technology instead of silk screening for the back and front panels, as well as 3d glass engraving for the product’s facias. All in house. We developed many custom tools for producing and finishing our metal and wood chassis. We also developed a special liquid glass finish for the Symphonia speakers, though my personal attention is the inner workings.
Kindly list the complete Aries Cerat line?
Starting with sources, we have the Cassandra Series DA converters, with two available models, and the Talos phono stage. We have three preamp designs, the Pondera, the Impera II / Impera II Sig, as well as our smallest design yet the Incite.
Our SET amplifiers are the Exsequor,the Concero 25 and Concero 65, the Collato OTL series with 20W stereo and 80W monos, as well as our special hybrid,the Ianus mono blocks. Our one and only integrated amplifier is the Diana SET tube integrated amplifier.
Our speaker line starts with the Stentor and the Gladius, which are direct radiating speakers and our horn speakers, the Symphonia and the Contendo Reference speaker system.
Do you believe in complete system approach by one company and why?
A system with all components designed by the same designer’s to be driven and drive their respecting components, will always have the edge.
As a small example, a system which’s total gain (IV stage of the DAC up to the speaker’s drivers) is designed without no gain being wasted at maximum undistorted acoustic output, will always sound better than another which does not satisfy this criteria. All other parameters being equal of course.
Another example. We state and always demonstrate that low impedance topologies and circuits will always sound better. A tube preamplifier with a 10k input impedance will always sound better than another with 100k input impedance (high value resistive attenuator). All other things being equal. Yet not all sources (phono/dac) can easily drive 10k loads optimally. A designer optimising the whole audio chain will optimise all components having in mind what stage proceeds and what follows.
However, the statement that company X’s amp drives the Company’s X speaker better, because each one complements each other, is wrong all the way.
A proper amplifier will and must drive all quality speakers equally. Same goes for preamps etc.
As a company, we are pursuing of offering a complete system solution, because we strongly believe that the quality of each of our audio chain components is designed not to be a bottleneck, but as the highlight of a system.
Our perfectionist agenda is what drives our complete system solution, not a “”we too”” mentality, as many big audio companies marketing departments push.
It seems every product in your line is special in its own way?
Almost all of our products have it’s own story to tell. Each product was commercially designed and put to the market, as a showcase of some new breakthrough, something we wanted to share with the audio community. So yes, almost every product is very special. The Pondera preamplifier was a showcase of the NOIES power supply breakthrough. same for Impera II series (Inverted triode), the Ianus hybrid(TriodeFet element), speaker designs etc. All have something new to show.
What is your take on exotic electronic parts. Do you implement them?
Audiophile parts is both smoke and mirrors, but also many times exact science. We do not put everything in the same basket.
Example. Resistor quality is both exact science, and do have their sonic signature. Now, if some designer/manufacturer is willing to pay for installing a 100 EUR resistor, while a 5 EUR is way better, that’s another story.
Paying 300 EUR for single piece of an known electrolytic super audiophile quality gives you marketing hype. We choose otherwise, we come up and install a space tech cap, with 1000x current surge, 100x less ESR and ESL, but no audiophile pedigree. We leave hype to others and do not use exotic-named parts, but do use parts with exotic specs.
What importance do you give to the built in elements? Are they very important to the way you construct your amplifiers or is it the matter of synergy?
Every component in each of our designs, is chosen after we are sure that it is the best possible choice for the application intended. We give a lot of time and effort in AB testing in every possible part selected and topology/circuit. We always AB test components, topologies, PSU configuration, everything that is responsible for the final result. A lot of time and effort is invested in training, to correlate these component/topology/circuits and the actual sonic result. After years of testing, ear training, measurements and listening experience, you can know, even before you assemble a prototype, what you will hear, when first plugged in. Synergy is usually claimed to explain why something of unknown parameters sounds the way it sounds. Our designs sound the way they sound because of the exact combination of circuits and materials.
Please tell us more about your innovative Inverted Triode stage?
Our Inverted triode stage is something completely new in the tube (and not only)audio world. The performance and operation parameters of the inverted triode is completely above any triode, and sets new sonic benchmarks.
To understand the inverted triode you must first understand the triode principles.
A triode is simply a device with three elements. A cathode, an anode and a control element in between.
The cathode is heated, to a temperature high enough that electrons have enough energy to escape the cathode’s atom “grid”. The electrons are attracted and accelerated towards the anode, as the anode has a higher potential than the cathode. A control element between the
cathode-anode, the grid, a thin wire structure, usually a wire wrapped around the cathode, controls the flow of the electrons by having enough negative potential as to partially “shadow” the electrical field of the anode, restricting the flow of electrons toward the anode.
An inverted triode works differently. Again a cathode is heated as to emit electrons, accelerated towards an anode which now has a wire grid-like structure, as to be electrically “semi-transparent”. There is no control element structure between the anode and cathode. Instead, solid plate surrounding the cathode-anode structure is now the control grid. This works by “pushing” the electrons back from the anode, using the very strong negative field of the outside plate instead of restricting the electron flow between anode and cathode.
This new active element, the inverted triode, presents completely new set of working parameters, operating envelope and some exciting
The inverted triode has power gain, but no voltage gain. In fact the voltage gain is much lower than unity. It is an electrical equivalent of a step down transformer, but it has power gain and many times the usable bandwidth.
Lower internal resistance: A 845 tube has internal resistance of 1600ohm, a 6SN7 has 8000 ohm, our inverted triode has around 10(no typo:)!
Near zero input capacitance: It has near zero input capacitance, unlike any triode, thus having many times the usable bandwidth of any signal tube.
Low voltage/high current: It can deliver 50mA of current from as low of 15V of anode potential .That is the current capability of a power tube in over twenty times lower anode voltage.
These specifications state that we can design some really exotic circuits.
An example of our initial research, was the comparison of our inverted triode stage and a power tube, both driving the same bifilar inter-stage transformer:
The power tube as a driver, had a 10Vrms -3db bandwidth of 15Hz-120KHz, which is a fantastic performance. The inverted triode stage had 2Hz-500Khz!! All other parts being equal of course(transformer, input stage etc). This brings completely another level of performance.
We strongly believe that this innovation opens a new world of possibilities in the audio industry, and with our Impera II series, which uses the Inverted triode stage as output stage, we can share this exciting new innovation, with it’s out-of-this-world sound, to the audio community.
You put a lot of time and efforts into transformers. Do you built them in house?
We have three transformer vendors, which wind transformers to our exact specifications and the same goes for the core selection, winding configuration etc. Winding technique share the same importance as core material, core size etc. Everything is built to our specs, tested and measured to meet our special requirements.
How important are the output transformers?
An audio design is a chain of subsystems, each to it’s own significance and value. A poorly designed SET amplifier with the best output transformer, will sound just as bad as it’s worst sub-system, circuit.
I can see a lot of badly designed tube amplifiers and the last thing they need is a top notch output transformer. However, a carefully designed amplifier will need a top quality OPT to match.
Unfortunately, quality SE OPT, need to be big. You cannot bypass physical properties. And 15kg OTP is just too big for the majority of the stylish amplifiers of the market.
We give our best effort to design the best possible transformer for all our designs, DAC’s, pre-amps ,amps and make our best effort to have the best tube for the purpose driving it. After all a SOTA transformer can sound only as good as the circuit driving it.
What is your view on volume control? What’s the ultimate way to do it?
There are two methods in volume control. First method, is the use of a variable voltage divider; by the use of two variable, or one variable one fixed resistor (series +shunt resistor). This is done either by potentiometer, discrete resistors, LDR and of whatever topology, ladder, shunt etc. The basic circuit is the same. You attenuate the signal by the use of a voltage divider. This is the method used in the overwhelming majority of equipment regardless of price. Second method is the use of a multi-secondary or multi tap transformers, where a separate winding or tap is selected, each tap with different step down ratio from the primary.
The resistor method, is very simple to analyse, regardless of the choice of resistors, topology and method of varying the resistance. The main problem is the output impedance. I’m leaving aside the problems associated with sub optimal solutions like pots etc, tracking problems, problems with resistor linearity and noise etc.
The output impedance is a function of the total impedance of the resistors (series+shunt),and a function of the voltage ratio or attenuation rate.
This resistance, which depends on the volume position if you like, forms a filter with the active stage following the attenuator. This is either it a solid state or tube. This low pass filter presents many problems in the audio chain, in limiting the active stage bandwidth. A common practice is the use of high value resistance attenuators in the region of 50-100k, of even more. The filter created between the stage following the active stage input capacitance and the output resistance of the attenuator, makes a low pass filter, and this is audible. This filter is always present, either the attenuator is at the input, middle or the output of the device, usually a preamp or integrated. The input resistance gets higher the higher attenuation ratio is chosen.
The TVC method is in my opinion the best method possible. A multi-tap transformer is the mechanical equivalent of a gear box with make a large number of gear ratios available. The output impedance is lower than the output impedance of the active stage proceeding(example the source) and the more attenuation rate is selected, the output impedance gets lower by the voltage ratio squared. This gives supreme resolution bandwidth and transparency.
Of course, the TVC quality is detrimental. In my experience, almost all commercial TVC available do have their own sonic signature. It is very difficult to make a near perfect TVC, but even a standard quality TVC is much better than the best resistor attenuator.
The TVC core is very important, but even more important is the winding technique used and this is where most of the TVC get their sonic
A small example, of the output impedance importance, is a test using a E280F super tube, which has high enough input capacitance, driven by three attenuators; a 100K shunt, a 10K shunt and a TVC.
The 100K,at -20db position, the tube stage has a -1db upper bandwidth limit at 20KHz! Here a 10K pot,same -20db position has a -1db upper limit at 200KHz! Of course, not all sources can drive a 10K attenuator, even though they should. Same stage with TVC is an easy 250KHz regardless of position.
In my opinion,a TVC is a way to go, and if not, a VERY low total impedance discrete resistor attenuator is second best.
Is OTL the ultimate principle in high-end or can this be achieved with output transformers also?
A simpleminded opinion might be, that the tubes are high voltage-low current devices, thus incompatible with the task of driving low impedance loads, as speaker drivers for example. Another simpleminded opinion would be, that as far as tube audio goes, OTL have the benefit of removing the output transformer and all it’s traits along, so they are superior than say SET amplifiers. Things are a bit more complicated than this. I strongly believe, that an OTL has it’s place in the audio industry, but is it not for everyday speakers or all loads.
Indeed tubes are high voltage low current devices, and most manufacturers work with high current tubes such as the 6C33 or likes, working on Class AB and each tube conducting heavily on half the cycle. They use high local and global feedback ratios to correct driving unbalances of the output tubes and global feedback to lower the output impedance.
An OTL can perform excellent, only when designed right. And then again, only if the speakers allows it. No OTL on the market can drive all speakers, where many SET amplifiers can drive loads down to 1ohm. A properly designed tube amplifier with the proper OPT can drive absolutely anything, within it’s power range, where a perfect OTL cannot.
Another bad reputation OTL have, is the habit of destroying speakers when a tube goes wrong, a runaway tube occurrence or even a rail fuse blow. We beg to differ….
You’re working on your own new OTL amplifiers. Tell us more please…
Having studied almost all that there is out there in OTL technology, having studied the traits and weaknesses, we decided to skip everything and reinvent everything from scratch. Our Collatio OTL Series is something unique. The Collatio OTL amplifiers are current output amplifiers, not voltage amplifiers that is the industry norm, either tube, transistor.
A typical direct radiating speaker, with the usual Best regards, of sealed box bass loading, do not like to be driven by current amplifiers. Many drivers used in horn speakers designs however, perform excellent when current-driven. In a way, our OTL amplifiers are a special purpose amplifiers, combined with a properly designed horn speaker makes an absolute world class performer.
All OTL amplifiers are push pull amplifiers, Working in Class AB. The main difference among the various designs on the market, is the topology and feedback loops between the drivers and output tubes.
Our output stage is a unique Single ended stage, not a push pull, where it’s output tubes are loaded with 4 chokes in parallel; the amplifier is a Single Ended parallel feed stage.
Single ended means Class A only. Where other OTL can get away with 200mA of bias for each tube bank, the Collatio 20W amplifier is biased at 2.6A per channel. This high current demand from the output stage means that a super sized power supply must be implemented, to provide 2.6A per channel steady state, with lowest noise possible.
That is why the output stage power supplies, come with double choke filtration and 6 high current high speed capacitors and it is massive.
The driving stage is a Directly Heated triode, the 814 tube, a big bottle tube that can deliver 10W of driving power to the output stage. Driving 20 output tubes per channel at high frequency needs a lot of current and the 814 tube has a lot to spare. The driver tube is loaded with a choke and is DC coupled to the output stage.
Sonic performance aside, the Collatio amplifiers are bullet proof in terms of stability and speaker protection. Their output stage is AC coupled, so that no DC can be present at any times at the speaker terminal. You can break all tubes while in operation with a hammer, still no damage can occur to the speaker. It is really bullet proof. The capacitor used is a 5kg cap each, high performance film capacitor, with not so typical specs.
The Collatio 20W stereo amplifiers can be used as mono blocks, delivering 80W of OTL SE power with no modification from the user.
What do you think about silver wiring and use of silver generally in tube amplifiers? How about copper?
Silver use in transformers would have some value if and only if every other aspect of the amplifier is perfect. I am baffled to see silver wire winded transformers used in amplifiers where the last thing they need is silver wire. You cannot make a poorly designed amplifier sound great by throwing a pound of silver.
Regarding the everlasting argument between copper and silver, i found more sonic differences between type of extrusion method, wire geometry and braiding than the material itself. Silver wire itself could sound way different if you change the type of forging/extrusion. In my opinion both copper and silver can sound perfect.
Did materials and their quality changed over the pact decades?
Some materials and components, have changed for the better. Some have not. The tech industry is not driven by audiophiles and audio designers. So, the direction and actual target that drives a specific industry does not always help the audio industry.
Some examples. The Delta Sigma converted chips, have paper specs un-reached by early converters with doubled theoretical resolution and extremely low noise floor levels. However, a Delta Sigma DAC was not developed for it’s sound quality, but because it was cheaper and easier to built. A r2r doc will always outperform it sonic wise, but the industry building these just do not care, if the built costs 100X more. So they dropped them.
This is not absolute. In areas like capacitors for example, we see some amazing achievements, which a designer can benefit. It all depends on the specific components, drivers, materials and techniques used.
N.O.S. tubes and parts. Worthy or hyped?
NOS tubes is a tricky matter. Speaking of tubes, the hype driving their prices is another story, which deserves a book. But in terms of quality, letting aside the risk of getting a dud, usually, but not always, manufacturers like RCA, WE, GE and Mullard. Those are hard to beat.
We always provide NOS tested and matched tubes with our designs, so we make the testing and choosing for you.
NOS components, again it always depends on the part.
You seems to prefer tubes. Why?
Because you can make better sounding circuits using tubes. Their linearity is exceptional and when tube circuits are designed right, it is the way to go.
Where is the line between overusing old tube concepts and designs and implementing the contemporary solutions? Where does Aries Cerat stand in this?
Some old tube concepts were well ahead of their time and have
something to say even today. Others are just the best they could do at the time. As a designer you must be able to fully understand some old designs and concepts, before you being able to either use some aspects that are timeless and adjust/modify them for today’s designs or just leave them in peace. Unfortunately, this is not always the case, as many designers use older concepts, circuits or materials without even knowing why.
Tubes vs solid state?
Tubes and especially triodes are the most linear active components ever created. Any semiconductor, vintage or today’s state of the art, does not even come close in linearity. Given this, you can design circuits that sound better in their simplicity and especially without feedback, where with transistors, you cannot.
Of course tubes are not panacea , and especially today, tubes are overused and misused. A badly designed tube circuit has all the traits that many wrongfully name as “tube sound”, where true tube sound is nothing like.
What is your approach and specific to your topology?
We have no specific topology. Our approach is to design the best possible audio component, then price it. Not the other way around. Then throw AB comparisons at it of what the competition has to present at more than twice the price range. Only when it outperforms them in every aspect, only then it is put on the market.
To achieve this, we break all norms and always start from scratch in every step of design.
We try many tubes types for the specific job, not what the mainstream use. We try old topologies, but also try to innovate, so we know that each stage and each part of the design is the best it can be.
AB testing is repeated in each step of the development, either for electronics of speaker design.
Every component is repeatedly AB tested, every topology, PSU design, speaker drivers, crossover design, everything is vigorously tested to make sure that this is the best way to do it.
I designed a rig, where you can test everything on the fly, and make AB testing fast and reliable, keeping always control of what you test and hear. In time you can correlate many technical aspects of any design, either electronics, speakers, crossover to the actual acoustic result, which accelerates our development procedure.
It seems analogue refuse to die. Do you still strongly believe in analogue advantages?
Analogue has it’s benefits,and rightfully so, not only has it’s place in top systems, but many times is the main source of choice. The digital vs analog is a never ending debate, but for me it was over long ago. Analog enthusiasts, reject rebook for high performance systems. I tend to believe that they never listened to a system where the system really brought rebook to it’s technical limits.
Same goes for digital enthusiast that reject analog for the technical problems of analog reproduction. Again i believe not many listened to a system which could really bring analog to it’s limits. Even the argument that the claimed dynamic range limitation of the vinyl, is out of place, when 99% of the speakers have lower DR.
Your gear is monumental. Why such a big proportions?
Our designs are as big enough as to to accommodate the circuit components. No bigger, no smaller☺
Unfortunately, when you choose to make no compromises in circuits, components, things get big really fast, not because of over engineering, but of proper engineering. We invest a lot of time and efforts in trying to make the layout as small as possible, at the same time satisfying all layout principles we strictly follow in terms of heat management, magnetic field isolation, power and signal optimal roughing and at the same time keeping volume to minimum. Many times something has to give up, and we always choose size instead of compromising the design in any way. Several layout forms are designed in 2D, then 3D to reveal any mistakes, then comes the prototype for verification. Sometimes you can pack everything as close as possible, some times you cannot. Unfortunately form follows function and big proportions and big weight is something you learn to live by :)
Please tell us more about Symphonia speakers...
The Symphonia speakers began as project even before the establishment of the company.
Many ideas and concepts were made possible with the highly specialised equipment the company acquired, such is our 5-axis work station with the highest Z-axis envelope possible, ideal to carve anything from our material or preference,that is low density plywood.
The Symphonia went into many stages of development,and along with the parallel development of our Contendo Reference speaker, gave us two speaker development platforms to exchange concepts and innovation between the two development platforms. For example,the horn loaded ribbon tweeter, with it’s innovative 360deg flare, was developed for the Contendo and utilised in the final Symphonia design. Same goes for the solid wood midrange horn, with it’s special drop shaped enclosure, to eliminate all diffraction, (same principle used for the tweeter horns) was developed for Symphonia and used in the new Contendo.
You can say that the Symphonia was not an isolated project, but a project that grew along with the company. It adopted many breakthroughs from other speaker projects.
Our proprietary variable slope filters, with the constant delay characteristics, were being developed for many years ,though our smaller speaker got to use them first, but the Symponia was the developing platform.
A log of work and effort went into eliminating the problems that come with horn physics, that if let unattended result in mediocre result. Problems that if not solved, the horns speakers get the nasty sonic traits that many, wrongfully, attribute to horns in general. I my opinion, a badly designed horn is worst sounding than a badly designed conventional speaker.
So if a horn speaker design, is compromised from the beginning, either from budget limitations, built tech resource limitations of simply lack of horn physics understanding, then horns are better skipped and built a conventional speaker. Horns done correctly is the ultimate electro-acoustic transducer, but only if specific aspects of horn design are respected and done correctly. This is what we tried to do with the Symphonia and the Contendo speakers, and strongly believe we succeed so.
How would you compare them with other hour manufacturers like Avantgarde Acoustics for example?
Horn design, requires addressing some issues that horn physics dictate, and if not addressed, in my opinion, horns are better left to peace☺.
Some of the problems are improper flare for the bandwidth/too wide bandwidth for a specific horn. Many designs get this very wrong. A specific horn has a specific bandwidth that is working optimally. Usually, designers “stretch” the horn’s passband to get away with less horns/drivers for a particular design, making the design smaller or aesthetically more appealing. A horn operating in the wrong bandwidth presents high order distortions as well as non linear distortions.
Improper driver. Not all drivers are suited for horn loading. Proper drivers are usually expensive. Also, finding the best driver for the intended bandwidth, is very tricky and never straight forward.
Horn construction/material. A horn presents sound pressure many times the pressure present in conventional speakers. A plastic horn, however thick, it will ring and resonate thus colouring the sound. A thin plastic one, is a complete nightmare.
Horns are very sensitive to their termination, that is what geometry follows after the flare ends(after the mouth). Discontinuities of the geometry, generates a secondary emission. This delayed emission returns back to the horn throat and horn driver,as well as radiated towards the listener. This is called horn diffraction, and is very very audible.
Proper filtering/time aligning.
Flushing the horns might result in aesthetically pleasing speaker, but the sonic result is incoherent sound, which many horns are accused for. Same goes for improper filtering of the horns. Crossover design and physical alignment must address all issues that derive from physical position of the horns as well and the actual acoustic filtering inherit of a horn-loaded driver. This is rarely addressed.
Please tell us more about you Contendo Reference speaker project?
The Condento Reference speaker, is our effort to design and built the best speaker, if the perfect speakers does exists.
Our Contendo speaker is a development platform, to push the envelope of what it is though possible in speaker performance. It is an ongoing project, which went through a lot of stages and each development stage seems to push the limit of our design and manufacturing abilities.
With this project we decided to address the main problems that are associated with horns and speakers in general and implement radical solutions.
Horn diffraction problem, is the major factor in creating the honking sound that most horn are associated to. We solved this problems with the drop shaped horn enclosures, for the midrange and tweeter. These bullet shaped 360deg horn flares are a complete nightmare to design and built and could not be possible to be manufactured without extensive use of our workstations, but in my opinion it is worth it, as it solves the ongoing problem that seems to exist in other designs.
Solid wood construction of the horns,results in very robust and sonically dead horn flares. Our horns have 15cm minimum thickness, no comparison with composite horns.
Another problem is the crossover and alignment problems. The development of our constant group delay filter was a major breakthrough.
Along with our custom measuring rig we designed, we create a true time aligned and perfectly coherent speaker.
What does state of the art amplification actually means for Aries Cerat?
SOTA amplification in my opinion happens when an amplifier is not the bottleneck of the overall performance of the system. A top system must give the illusion that the sound is not amplified in any way and that an acoustical instrument is playing un-amplified in front of you. Most amplifiers with SOTA measuring and technical parameters almost always give this “electric” signature. What you are listening to is signal passed through a chain of audio components. A true SOTA system must give this un-amplified illusion and an amplifier is a big part of this.
Do you think balanced topology is a must for best sound?
I am not in favour of balanced topologies, except in very specific cases.
Single ended always sound better and the benefits of noise rejection in balanced topologies does not apply in home audio. Single ended components now days have very low noise floors in home environment. They always have the edge in sound quality, all other parameters being equal of course.
Balanced topology has it’s applications in our designs in the Cassandra series of Dacs. We use two banks of paralleled days in complementary mode, (differential mode).This improves the linearity of the days near zero crossing and also increases dynamic range and lowers noise floor of the converter.
We believe that this is one of the very few “chapters” in the audio chain where symmetrical not only makes sense but has an edge.
What would you say sets apart Aries Cerat products and designs above other manufacturers?
No compromises in design and built.
How do you see the traditional Japanese tube products. For example like Kondo, Shindo etc.?
I have a lot of respect in Japanese tube manufacturers. They kept tube audio alive in times where transistors were thought to be the way forward.
Most of the Japanese tube products have similar sonic traits, and have a lot to like about. I personally find them more soft and round than my taste, but that is just me ☺.
Are physical mediums obsolete?
I hope not. At least analog will continue to be the king of physical mediums. Digital is another story. I always used PC as a player even when it was considered heresy. In Digital physical mediums are obsolete.
What about analog front ends and phono preamplifiers. What’s your view and what makes your phono unique?
Our Talos phono stage takes a different approach of the three sub circuits of a phono pre preamp and tries to set itself apart from the norm.
We use balanced input stage, which cancels all noise picked up from the cartridge to the phono input. This lowers the noise floor and gives true lower level detail which is often lost in noise. The input stage is the most important part of a phono pre and we tried to make a stage that retrieves all information coming from the cartridge generator.
The RIAA stage is not implemented with the usual RC filters, but is a low impedance LRC filter. LRC which although very rare, is not something unique, but the way it is used, is. We use a constant impedance driver for the LRC filter, using a transformer loaded SE triode stage to drive the LRC circuit.
The output stage is another small SE amplifier, a transformer loaded SE triode stage.
I designed Talos to be a world reference source ,and I believe we succeeded.
Would you say, that one love for music reflects in his creations?
Speaking for myself I can say that the love for music is the driving force in designing. I always try to create something that will make the music I love, move me even more.
There is the high-end and ultra high-end. What is the difference in your view?
In my opinion if you listen to a system which does not allow you to commend on the sound, bass, stage, timbre; all the attributes we audiophiles care to comment, but instead you just listen to the music. This is the ultimate level. A system which gives you the illusion of natural un-amplified sound and just do not care about the system. Everything else is just HiFi.
Single Ended tube concepts. Utmost way to go or?
Sonic performance says so.
Pure Class A. Is it a must and still a holly grail?
Class A is again a misused term.Class A operation term is stretched and bend by marketing agendas, as to name any device delivering couple of watts in “Class A” region as Class A.
Class A gets it’s sonic benefits in two ways. First and well known, is that the active component(tube or transistor) is never entering cut off or even near cutoff, it always conducts current on any condition and any load.
(By these term alone, almost 9 out of 10 transistor amplifiers fall short of this criteria. A tell-tell sign is the heatsink size, which gives away that the actual bias level is adequate for a few watts only and usually in 8 ohms (half of that in 4 ohm loads)).
The most important aspect in strict Class A operation is that the active element is biased well into high current region. Usually in a very nice spot in the triode/transistor characteristic curve chart. This is the main reason of the sonic advantages. The active component is biased into the most linear region of the envelope of operation and not the most non-linear, which is near the cutoff region.
Your products are actually reasonable priced compared to most stuff out there. How so?
As a company,we want to present our designs to as many audiophiles and musiclover possible. We want to show them how true High End audio reproduction sounds like, and expose as many enthusiasts to our designs and sound. I always try to give the absolute best. Not only at the product’s price range.but well above. I would never put a product in the market, if i knew that it would be just another product in this price range. I always approve a design to be on the market, when i am absolutely sure that it completely outperforms everything else in the market well above it’s price
How important is the power supply in your view?
Power supply is as important as the circuit it feeds. No less and no more. On top of that some topologies are more sensitive in power supply design and characteristics. Like for example SE topologies.
Of course a PSU design is designed accordingly to what’s feeding. A bias circuit has different design keys that a digital circuit or a tube filament PSU. Even if they share the same noise and current profile characteristics.
What is specific to you approach regarding power supply?
In analog circuits we avoid the use of actively regulated PSUs. Active regulation always uses a feedback loop. A feedback loop in the PSU never makes a PSU respond perfect in sharp transients. It is never fast enough. A transient response snapshot of any actively regulated PSU tells the whole story. A PSU is part of the signal chain and feedback loop in the PSU IMHO is a feed back loop in the audio chain.
With our tube circuits, we always use choke filtration. A dynamic analogy of a choke, is a moving mass. For example a flywheel, where a capacitor’s mechanical equivalent is a spring. Both can store energy, but the way and rate that the energy is delivered is very different. No active PSU can store energy like a choke does. We fine tune our CLC and usually CLCLC filters, to provide ultra clean power with excellent transient response characteristics. A CLC circuit in parallel with a load, is a highly sensitive tuned system. Each load requires a specific filter to operate optimally. We tune our filters to be as fast as possible with minimum sagging and critical dampening, as no ringing is present in whatever load conditions.
On digital circuits, we use local stabilisation with LC decoupling in every local circuit, every IC or power circuit.
Digital circuits, have a lot of noise in their power rails, which contributes to jitter in DAC’s driving circuits. We use extensive LC decoupling in all areas possible, so we do have very clean rails and ground planes.
Classical music or studio recorded music. What do you see as ultimate test for ultra high-end?
Big symphonic plays, concerts ,symphonies are simply the ultimate test in any audio system. This is what separates the men from the boys. Many top echelon systems collapse in large orchestra crescendos, in organ passages etc.
There is no comparison between studio recordings and big orchestra recordings.
It always brings many top systems to it’s knees, as many performance aspects of such plays, you cannot get around it.
Especially if someone has live concerts first hand experience he can get a full assessment of system with just one big orchestra crescendo replay.
How do you see the rise of digital audio?
Digital audio in a true step ahead, as it gives the average music lover completely new tools for expanding, sharing and managing his music library. To the average HIFI enthusiast it gives a complete new level of easy of integration and a new all in one mentality rise in audio systems. This doesn’t mean much in no compromise High End systems, but is a means to expose the music lovers and audiophile community to more people.
Is Redbook 44khz 16bit CD format obsolete?
By no means the Redbook is obsolete. A recording making the best of the available dynamic range, with as less studio intervention possible, converted by a true SOTA DAC can bring almost all systems to their limits. Very few systems worldwide can bring the Redbook format to it’s limits, if the recording and converter are top notch.
How about vinyl? We heard how its limited compared to digital, but…
But is not. As I mentioned, a top analog source will never be the bottleneck of performance even in top systems.
Cassandra DAC again makes quite a statement. Tell us more about your view regarding digital and DAC’s in general.
Cassandra was developed at the start of the company. While aiming to design and built the best speakers and electronics possible we needed a top reference source. Something that was the absolute best and holding nothing back in developing of our line-up.
The Cassandra’s prototype was developed as a cost no object lab tool and it was decided at a later stage to make it commercially available.
We found new converters launching everyday with impressive specs, but they fell short of what we call analog sound. They always present this specific electronic signature that gives digital reproduction it’s bad name. From the beginning we had in mind to create a true analog source, which happens to take Ones and Zeroes as input :)
Ladder DAC. Why you see it as only real deal for utmost performance?
There is a lot of information regarding the superiority of ladder days over delta sigma modulators. DS converters were developed because it is many times cheaper and easier for the IC company to make a DS modulator than to laser trim resistors to an accuracy of 1/128000 it’s value☺.
Many modern DS instead of one bit modulators are turning into multiunit modulators. 2,4 or 6 bit. Hey some day they will return to the full ladder converters☺.
High resolution audio. Is it hype or evolutional step into the right direction?
Higher sampling rate is a big evolutional step as long as the extra samples have true information and are not just extrapolated by a fancy up sampling filter. Higher sampling frequency means more information. Higher bit depth, however is another story.
A DA converter, regardless of the bit depth marketed due to the noise picked up from it’s analog filters up to the preamp input has a much higher noise floor. This degenerates the true SNR and actual bit depth. The typical noise floor of any audio system degenerates the bit depth of ANY doc down to 15-17 bit regardless of the theoretical n-bit depth.
I leave alone the actual amplifier-speaker dynamic range, and actual recordings dynamic range. The additional bits after 16-17 higher bits are just noise.
What bothers me in the race of new music distribution formats is the need to re-buy your music collection every now and then. This is not right.
PCM vs DSD?
DSD is one bit streaming of data,roughly the native sort of speech “language” of Delta sigma AD and DA converters. It has it’s purpose as long as the ADC-ADC is delta sigma. When speaking for R2R days,it is of no true value.
Where is the fine line between, resolution, transparency and musicality?
Resolution, musicality and transparency are terms that for almost every audiophile has a slight different meaning of the words. In my opinion, natural sound is the absolute benchmark that these terms must be referenced to. Natural sound has all the resolution in the world, yet no details stick out like needles At present many systems have a pushed treble; where one might think is analytical, where it is not.
Same with transparency. For example a thin sounded speakers usually gives the impression that they are highly transparent and of high resolution, where again they are not. In my opinion there should be only one benchmark and this is where a system stands in comparison to natural sound.
How do you tune your products or what is your specific goal in creating Aries Cerat way of sound?
My approach when tuning a product, is that the product under test by no means restricts dynamics, tone, energy and life out of the audio chain.
A product cannot be tested stand alone in sonic terms. So testing a design among a long chain, needs time, knowledge and experience. A design is approved when it appear to be no bottleneck of performance in any sonic aspect considered in High Fidelity world.
What are your musical references for your when designing the products?
When making listening tests, AB comparison between, components, topologies,tubes, PSUs and secondary circuits or even final fine tuning we follow specific steps. There is a long process which dials in after many listening tests, measurements and AB testing, where in time you train the ear/brain, to correlate certain sonic parameters with specific technical parameters/measurements. It other words you train the ear/brain network to interpret what you are listening in reference to actual technical parameters and the opposite,designing and measuring something,while having in mind how it will sound. This takes a LOT of time and effort.
Having said that in our tests we use recordings of all sort of musical genres and wide variety of acoustic instrument recordings etc. For example, when testing for dynamics, some tracks with uncompressed drum set will do the job or some orchestra crescendos. You isolate specific traits in each part of each recording in order to evaluate the component/design/circuit in full.
What were your explorations on of Class D?
I am not a stranger regarding switching technologies. Working at the robotics lab at the University I’ve developed a new switching topology, as a motor driver for a robotic/cnc platform I’ve designed. In few words, it was a switching technology driver, which worked as a Class A bridge driver controlled by a DAC, which turned over to switching mode under specific angle/load and could vary the motor vibration characteristics, as you could also vary the current threshold where the transition between Class A/switching mode occurred. Long story short I’m very familiar with the technology and I can see the
potential. However audio is a completely different league. I strongly believe that Class D will never perform at the standards of a SOTA.
Class A amplifier for a number of reasons. As a company we only use Class D modules on the Sub modules of the Condento Reference speaker, (driven by tube crossovers)and for me this is a far we go with this technology.
How close can one get with digital reproduction in comparing to analog in your opinion?
Digital technology is making very big advancements. Yet the aspects of the DA technology that always lacked are going the wrong way. The industry is driven by who can develop the most exotic filter, highest up sampling rate and highest bit length, at the same time using Delta sigma converters, sub optimal IV conversion circuits and poor analog stages.
R2R converters is the way to go in high performance DA conversion. DS technology need higher and higher sampling rates to make filter design easier. At the same time everyone neglects that a typical audio chain has 14bit true resolution,and a hand full of systems around the world have the dynamic range to make full use of the dynamic range available in 15-16bits.
A true perfect DAC can perform very close to a top analog set up in terms where analog is ahead; finesse treble and music flow,while be better in some other ones. I believe that all things being equal, both mediums are pretty close although finding a perfect doc is more difficult than it seems and you cannot find it by looking in spec tables.
Both mediums have their place in an absolute top level audio system.
Does form follows function with Aries Cerat products?
This is a rule that our designs strictly follow.
Many say the first watt is most important. Is it?
Actually in horn speakers the first few milliwatt are most important. Low passages,ambience, sense of space, timbre are in most cases all hidden many db below fundamental frequencies of the instruments. Many sensitive speakers are driven with just a few watts in peak signal. So you can imagine that the integrity of the signal in very low power output,is essential for the listener. This is where SS and most push pull designs fail; the first milliwatt.
Of course, heavy and inefficient speakers mask the first watt amplifier artefacts.
How important are the room acoustics from your experience in making state of the art sounding system?
Room acoustics is a misconceived matter by most audiophiles. Many perceive room treatment as something aesthetically invasive or expensive. Two well designed diffusers will augment the sonic result well beyond any cable lifter or exotic paraphernalia. Yet one chooses blindly one over the other if he only has budget for one or the other.
Nowadays many acoustic treatment companies offer solutions that are also aesthetically pleasing and have excellent integration with the average listening room. There is really no excuse for not treating a acoustically problematic room. Yet no all rooms have inherit acoustic problems. The problems usually starts with the user not being able to properly set up a system in listening space. Again aesthetics win over acoustics… This is just a matter of mentality,nothing much to do
When is it simple enough (design and topology) to say stop, this is where it all ends and product is concluded?
Personally this is one of the hardest “thing” to do when we design a product. A lot of projects are forever kept in pre production stage and development stage. Our constant urge to innovate and improve a design delays our products in reaching reaching the market and production stage. I can give you a lot of examples of designs forever being revised and improved. As a designer, sometimes you have to set a firm personal limit and conclude the design when you feel it is perfect…
How do you see the state of the present high-end society?
I believe High End society is an transition, where not enough people are interested in High End audio and new audio companies make an entry each day. At the same time big established companies try to move to the stratospheric level of High End, not always with products to support their move. High End audio society does not live in a bubble and as long the middle class is annihilated, the middle class of High End will follow. We live in …let’s say interesting times.
Would you say that high quality is more affordable today or you have to pay premium price for best components and sound?
Today’s number of audio companies does not compare to any time in the past. Too many “me too” companies out there and the average audiophile who searches for a decent audio system is lost in the overabundance of choices. I would say that a decent audio system is more or less affordable, but its a way more difficult to be discovered in the endless catalog of audio companies.
If we speak about the top notch systems, I would definitely state that the top grade systems are getting more and more stratospherically priced.
What is the difference between audiophile and music lover?
Some audiophiles are all about their equipment. I believe a true audiophile is indeed a music lover.
Who would you say typical Aries Cerat audio customers are?
I do not feel our designs are destined for a specific type of audiophile. I believe audiophiles who have extended experience of how true natural sound is like are “sound” believers of our designs and company.
Where are Aries Cerat plans for future?
Get our sound known.
What would you say is the goal of Aries Cerat?
We constantly innovate and try to push the limit of what is possible is music reproduction and our perpetual goal is showing to audio community what is possible.
Any last thoughts for our readers?
Just keep the music going and expand your listening experiences.