When and how all started with the Trinity Audio?
The TRINITY Electronic Design GmbH was established in 2011 as a spin-off of the audio business of the GTE GmbH, which we established in 1995 for our sensor business. The GTE GmbH developed beside their sensor business the unique triangle shape DAC and Power Amps.
Why the name?
In the GTE GmbH we had a sensor system with the registered brand name TRINITY®, so we used the same name for our audio products.
Did you have anyone that inspired your work or still do?
No. I do not consider my work as a burden, since the design of outstanding electronic is my passion.
Do you consider yourself as an audiophile?
I do not know the definition of the word “audiophile”, but since I was 10 years old I have spent all the money I earned on the best audio electronic I could afford or LPs. I was thrilled by the music.
What would you say it is the difference between music lover and audiophile?
Again, what is the definition of a music lover and an audiophile?
Let us imagine we have a 24bit/192kHz recording equipment, but without excellent musicians you can only record fantastic low noise. The bottom line for me is the musician and their skills to write a song, to play an instrument or sing a song are the most important part.
For instances Dire Strait`s – “Brothers In Arms” album is an wonderful album, but it was recorded in 48kHz and 16bit, which was state of the art at that time: link.
The counter question is: Is this album “just” good music or an audiophile recording?
I do not twist my brain to solve the question I enjoy listening this wonderful recorded music.
Beside the technical parameters like sample rates, counts of bits and so on the most important thing is the music and the recording itself or in other words the close collaboration of the recording engineer with the musicians to capture the sound the band had in mind.
Have a look on the new remastered versions of the old recordings a lot come with different takes all were done on the same recording electronic, but the sound is different from take to take and at the end of the day the musicians published the takes the liked most.
For instance The Blu-Ray of YES`s “Relayer” includes numerous additional mixes for the listener to choose
1. A new stereo album mix in DTS-HD Master Audio in 96 kHz / 24-bit resolution mixed from the multichannel master recordings
1. A Flat transfers of the original album mix in 192 kHz / 24-bit resolution DTS-HD Master Audio, made from the original master tape source
1. An exclusive 96 kHz / 24-bit needle-drop of an original UK vinyl A1/B1 pressing
1. The 96 kHz / 24-bit LPCM stereo “Archive Master”
1. And so on
I personally want to listen to the flat master tape sound, which is the sound the band had in mind.
By the way even the old Greeks worked on music theories: link.
Can you tell us a bit more about your impressive background?
I studied semiconductor physics at University in Ilmenau (1979-1984). After the study I designed 2 years digital and analog electronics for needle printers, before I was handpicked by the East German “officials” to design electronic circuits for space applications at Carl Zeiss Jena. In 1989 I escaped from “East Germany” to “West Germany” four months before they opened the wall. With 6 years of experiences in high level design it was easy to get a new job in the research department of THOMSON.
Here I developed the complete electronic for the signal path of a hi-res digital video recorder, which was the only one ever built in Europe. All the electronic circuits between the error-correction were done by me including Channel coding, write amplifier for the magnetic heads, magnetic head read amplifier as chip on board, channel equalizer, decision circuitry, clock recovery and Channel decoding.
After that I designed the read and write electronics for the DVC tape recorder, which were used by THOMSON for the standardization.
After that the focus was on optical disc recording and I developed the write electronic and laser drivers for DVD-R, DVD-RW and Blu-Ray. I was the first who finalized a laser driver design for a blue laser diode with rise times of 350ps.
After finalizing of the optical standards I was asked to support the IC development department, where I designed the electronic for up- and down-converters to distribute WIFI on a regular coax cable in a frequency band of 900…1000Mhz normally used for the distribution of TV signals in a multi-family house.
After that I changed the company and designed power amps in GaAs for mobile phones, before I decided in 2010 to work as a full timer for my own company GTE.
What is the goal of Trinity!?
The design of audio devices which all of have the best measurement results in the world for a reasonable price.
Kindly list all of the current products and product lines…
1. Golden Reference Edition:
1.1. External Power Supply Unit for DAC, preamp, Phono and mediaPC
1.1. Preamp
1.1. Phono
1.1. Vibration damped Rack with touch screen in the top level for the control of the mediaPC & Drive (self-resonance 3Hz)
1.1. mediaPC & Drive with 20TB HDD just for the storage of the music
1.1. Power supply Conditioner with earth line filtering and de-coupling and overvoltage protection
1. Reference Line
1.1. Phono
1.1. Preamp
1.1. DAC with our patented LIANOTEC (Linear Analog Oversampling Technique)
1.1. Power amp
1.1. Power distribution filter with earth line filtering and decoupling
1.1. Vibration damped Rack in stainless steel (self-resonance 3Hz)
1.1. Vibration damped Base for the power amp (self-resonance 3Hz)
1. Professional Line:
1.1. Microphone Preamp with PHONO and NAB tape equalizer
What are the differences between the lines?
We started with the Reference Line strictly speaking we started with the PHONO of the Reference Line, since it was clear this will be a “first time done” development. The time constants were well known since 1957 and all what we needed was an outstanding gain stage, which we had already developed for the voltage gain stage for the triangle shaped power amp inside the GTE.
The following Preamp was also a no brainer, we combined the same gain stages with the volume control we used in the triangle shaped DAC and the design was done.
That means these first two devices had no special design cost at all.
The triangle shaped DAC was already based on the PCM1704, but for the new Reference Line we had to implement a new asynchrony USB interface, that means we had to design two outstanding OCVCXO one for the 44.1kHz and one for the 48kHz recordings. The proprietary LIANOTEC (Linear Analog Oversampling Technique), which we had already patented and implemented in our triangle shaped DAC, was the simplest implementation.
After the DAC I wanted to design the power amp, but then I realized that it made more sense to first design a matching source for the DAC to give TRINITY DAC owners the possibility to play high-res files over the USB interface. Therefore I designed the media-PC&Drive.
Here we have to say playing a CD with such a mediaPC outperforms without any doubts any high-end CD transport, since the mediaPC sends the audio data over the asynchrony USB interface with its two OCVCXOs.
“Common” CD transports use AES/EBU, SPDIF or Toslink interfaces, which all need a clock recovery on the DAC side, whereby such SPDIF receiver ICs have system related significantly higher jitter than an OCVCXO.
After the PC I finalized the Power Amp of the Reference Line, whereby the electronics was based on the electronics we used in our triangle shaped power amp. At this point the Reference line was almost done.
Only the introduction of the rack requested an additional power distribution filter to make the power supply wiring and the earth line filtering and decoupling in the rack much easier.
At this time I reached the limits of my UPL audio analyzer, this means I could not improve the electronics further.
What inspired you for the introduction of the Golden Reference Edition?
In fall 2014 Audio Precision introduced the new flag ship audio analyzer APx 555.
Now we had the first time test signals with no distortions up to -140dB. Of course I was the first client in Europe who bought this analyzer.
The first measurements with the brand new analyzer have shown that the devices of the Reference Line are better that I had specified with the UPL analyzer, which had “only” -115dB spurious free dynamic range. On the other hand these measurements have shown also that there was some room for improvements above 1kHz.
It was clear that the architecture I used in the Reference Line was exhausted, therefore I had to start from scratch to design a new universal gain stage.
The target was to get again on the limit of the best analyzer in the world and we reached these limits with the Golden Reference Edition a year later.
Please tell us about your DACs and what’s unique in your approach?
The really unique is the proprietary and patented LIANOTED architecture. This architecture improves all parameters of a DA converter and reduces the sensitivity to clock jitter. The TRINITY DAC is the only DAC, which does not sound digital! Beside the technical parameters, this is the DAC, which can reproduce a wide and extreme deep image even at very low sound pressure levels. The best way to check this is to listen the BOLERO from Percussion Museum.
By the way “Bolero” and “Pictures at an Exhibition” are my favorite classic.
By the way TRINITY does **not** support MQA, we support only real high-res music.
“_“Connecting the dots, we see that_ [_Spencer_ _Chrislu_ _(MQA Director of Content Services) acknowledged in August 2016: link. “If a studio does their archive at 24-bit/192kHz and then uses that same file as something to sell on a hi-rez_ _site, that is basically giving away the crown jewels upon which their entire business is based” link. What this basically implies is that MQA is a way to defer release of a full resolution “studio master”._ _An opportunity to sell music lovers a version that must not by definition hold the full value of said “jewels”._ _And so it goes, perceived opportunities to sell the same music yet again because the precious, awesome-sounding, crown jewels are safe_ _in_ _those concealed music vaults…”_
Analog vs Digital!? You offer products from both realms. What is your take on subject matter?
There are a lot of people who still have huge counts of LPs and not all LPs are available in a digital format or sometime in a very bad compressed digital format. If you analyze some LPs you will see that there are LPs, which have much more dynamic then the matching CD, therefore we should also please these analog music lovers.
How about your phono preamplifiers? What make them so different?
Actually I designed the PHONO Preamp without having a turntable. I did not see a need for it.
The time constants were described by the RIAA standard in 1957 and the friction noise of the needle in an empty groove of a copper master is also well known. All you have to do is to design a gain stage, which has lower noise than the friction noise of the needle and you have to design an exact equalizer network for the time constants. Not really rocket science.
I just run in a small problem as I tested the first assembled PCB with UPL audio analyzer. I saw a small gain peak around 150Hz. My simulation did not show this artefact and there was no technical explanation, how this could happen. Therefore I analyzed the look-up table of the UPL audio analyzer and saw that the calculated values (the inverse RIAA response) of the UPL audio analyzer from Rohde&Schwarz were not really accurate, therefore I replace this lock-up table by the one I calculated (with 8digits after the dot) and that is the end of the story. I forgot to mention of course the gain stage should have even at high gains a bandwidth in the range of 500kHz.
That is the reason why the TRINITY Phono Preamps and of course also phono path in the Mic preamp as well are so dynamic.
I am often confronted with statements like “nobody needs audio electronic device with a bandwidth above 20kHz”, even from engineers and they are so wrong:
Please remember we have more than one device in the signal chain and the resuming bandwidth can be calculated roughly with this rule of thumb:
After this rough estimation let us look to a correct simulation.
3 cascaded identical gain stages and each has a 3dB bandwidth of 20kHz It simulates a chain like: phono – preamp – power amp
After the 1st Stage we have the -3dB at 20kHz, but after the 3rd stage we have the -3dB at **10kHz!!!!** _Not really high-end_
I think a resulting bandwidth of 10kHz speaks for itself and do not need any further explanations. This short and simple illustration shows you, why all TRINITY devices have an ultra-wide bandwidth, we consider of course the whole chain and not just one device.
How about balanced circuits and topology? A must?
From my engineering point of view a balanced architecture will always outperform a single-ended architecture.
This statement is based on my experiences as I worked for digital magnetic storage in the 90`s. The magnetic head amplifiers we used were modified head amplifiers designed for HDD and all these amplifiers and signal paths were balanced with a noise level of 0.5nV/ at that time. The fully balanced signal path was essential to get the best BER (Bit Error Rates).
The only disadvantage of a balanced design is it doubles the cost. Therefore from a marketing point of view a single ended is the better solution.
Please tell us more about your amplifiers?
I assume you mean the power amplifiers. The difference between a preamp and power amps is not only that the power amp has to deliver higher currents and voltages, but also that the PA has to drive a complex load, whereby the preamp drives “only” a more or less resistive load plus the cable capacitance.
My target was not to copy one of the major audio power amplifier designs I wanted to realize my own ideas.
There are countless good designs described over the last decades. For instances Motorola had published decades ago an app note AN1308, where they described a design of a cascode output stage, which can easily be trimmed to 1kW by adding more transistor and use higher supply voltages. All these designs had one in common: good performance to a reasonable price.
My interest was it again to design an amplifier, which can be called an “ideal” amplifier. Please read also my manuals they are full with additional explanations.
In our case the power amplifier is parted in a voltage gain stage and a current gain stage. Actually this makes the design expansive, but gives you the best technical parameters.
The output stage is based on 144 pairs of very high frequency transistors to get an outstanding bandwidth. In general we do not claim things we measure it and we measure all parameters under worst case conditions. The current gain stage has a small signal bandwidth above 10MHz and under full power it can drive 20A into 2Ohm width a power bandwidth of 500kHz. Yes, we specify our power amp even at 2Ohms, since a lot of loudspeakers have impedance drops down to 1.4Ohm. The group delay, the time a signal needs from the input to the output, is only 20ns!!!!! Please remember the group delay in a double terminated 1m 50Ohm coax line is in a range of 4.8ns.
We did not followed the approach of an error corrected design to reduce the THD, since we have the opinion, if we have no distortion on the output we do not have to compensate something, which would needs an additional compensation loop.
How about the negative feedback?
There was a time as the transistor amplifiers sounded bad even if they had lower harmonic distortions than the valve amps. One who had a closer look on it was Matti Otala and Eero Leinonen, who published a paper “Possible Method for The Measurement of Transient Intermodulation Distortions”. As a consequence, from there on all audio designs were based on low feedback and wide bandwidth, but this was 1976!!!
We are now in a new century and should use newer design architectures and we have access to much better electronic devices, which were not available at that time.
So the TRINITY Power AMP has a very high bandwidth regardless of the implemented very high open loop gain above 130dB.
There are some ultra-high power amplifiers on the market with no negative feedback, but I would guess the reason is not to improve the sound, but to make the amp stable, since the wiring of so many transistors over a huge heatsink generates a hell of reactive elements capacitive elements, inductive element and a coupling between these complex elements. If you apply a negative feedback around such amp all these complex elements are inside of the closed loop and have a direct impact on the phase margin or in simpler words it is very likely that this design is not stable and starts to oscillate.
Please note: the negative feedback improves the THD and it reduces the output impedance and increases therefore the damping factor, this means in my eyes, it is the only correct way to build up precision amplifiers. By the way the output impedance of the TRINITY Power Amp is so low, that you cannot measure it.
What I can say for sure is that every loudspeaker sounds much better on our power amp and even horn loudspeakers sound fantastic. There are some rumors that a horn has to be driven by valve amps, but in our case it is completely wrong.
Even if you run the TRINITY line at full volume on a high efficiency horn and you place your ear in the horn you will hear no noise or hum!!!!
What is the importance of proper gain distribution across the complete chain?
To be honest I have never heard about “proper Gain distribution” before and I do not see any technical need to think about it, at least not for our gain stages, since they are distortion free between 0dB gain and 46dB gain.
In common closed loop designs (negative feedback) the THD are depending on the gain if the amplifier itself is not ultra linear, but the TRINITY MIC preamp has shown that we have no distortions at all independent of the gain settings.
And connected to the gain, how does Trinity Audio deals with the volume gain (preamplifier)?
Actually in the Reference Line and the Golden Reference Edition the volume control is not based on changing the gain. The gain is fixed and the volume is attenuated. The architecture is in both cases the same, but in the Golden Reference Edition we use two relay contacts for each point instead of one and we have an improved control electronics. The 8dB output attenuator in the MIC-Preamp you have reviewed is based on the volume control of the Reference Line. If you increase the volume in the Golden Reference Edition you will hear no mechanical noise or switching noise at all. Just the music gets louder.
The architecture is not based on a simple lossy pad architecture, which is common even in high end electronics, we use a more complex approach normally found in high frequency technique. The advantage of this architecture is the “sound” does not change with the volume setting.
How about power supplies? You’re using some of the non-conventional approaches…
It depends on the point of view, actually for me it is a conventional design even if I use wideband SMPS (Switched mode Power Supply Modules). Beside the much higher cost they have only technical advantages, if the design is done correct. Even the power supplies of the Phono preamps or MIC preamps are based on SMPS and you know you hear no noise at all, obviously a very good choice.
Of course I have countless passive filters and linear voltage regulators behind the SMPS to complete remove the switching noise and at the end of the line I have POL low noise linear voltage regulators for the final voltage regulation for each gain stage.
Often you can find transformer based solutions even in an external enclosure, where the manufacturer offers “mu-metal options” at the owner’s expense as a workaround to reduce the radiated hum in the signal electronic.
The TRINITY power supplies are designed in that way that such options with costs are not necessary, therefore I think the carefully use of SMPS modules is the better way.
Does form follows the function with Trinity?
Yes of course, if I start the electronic design I also start with the enclosure and look for a matching flight case for the enclosure as well. By the way I do both electronic and mechanic designs by myself. I only order a PCB if the mechanical design is finalized as well. If you look inside you will see that all XLR connectors for analog inputs and outputs are soldered on the same PCB, this means the location of these connectors determines the location of the cut-outs in the enclosure. To solder wires between an audio connector and a PCB is a no go and has nothing to do with real high-end.
Would you say Trinity DNA is clearly evident across the complete product portfolio?
Without any doubt, I would never ever sell a device, which is not worthy of carrying the TRINITY name.
Exotic electronics parts and materials. What’s your take?
Let us define exotics first, if we equate exotics with unique than I would say yes, since I use ICs you have never seen before in other audio devices, therefore they are exotics.
For instance the non-re-clocking data recovery electronics in the SPDIF, TOSLINK and XLR inputs is based on such parts.
The SPDIF signal (Sony Philips Digital Interface) is not simply passed on, it is first regenerated.
Figure 1 shows the output signal at channel **A1** as supplied by a UPL Audio Analyzer from R&S. Channel **A2** shows the regenerated signal leaving the TRINITY®-DAC controller.
The audio analyzer features an option for simulating a 100-m long cable. Figure 2 shows the same test setup, but with the cable simulator activated.
Figure 1
Figure 2
The additional digital non-re-clocking data recovery has an Input Sensitivity of 500µV (BER = ).
The goal of this circuit is it to reduces the jitter in the clock recovery of the input receiver IC. I think this is something we can describe as a unique or exotic solution.
On the other hand I would assume I am the only one who uses two OCVCXO in an audio DAC.
These full customized oscillators have not only the lowest phase noise characteristic in the audio market, they have additionally the highest accuracy as well. The accuracy is better than 1ppb (parts per billion).
Typical very good oscillators are in the range around 10ppm (parts per million).
The accuracy of TCXO (temperature compensated Xrystal Oscillators) often used in the best DAC converter can be around 0.1ppm.
Compared to the best TCXO used in audio the TRINITY OCVCXOs have a 100 times higher accuracy. For all who are too lazy to calculate the 1ppb deviation by hand: Link.
1ppb from a 22.5792MHz is 0.02257Hz.
TRINITY OCVCXO
Standard deviation (Jitter) is 30.3mHz
Accuracy 22.579200,0083MHz (better than 1ppb)
Based on my experiences in countless fields of electronics, I use parts, which are exotics in the audio field, but they are common in space or other high-tech industrial applications.
Of course I know that there are other oscillators on the market, where the most exotic part is the price, but that is a different story.
As an engineer I do not use “exotic” parts, which need a fairy tale afflicted marketing story behind them.
One example for instance is the use of Ohno copper cables in the audio business.
The process itself is from the engineering point of a view really something special. It is well described in the paper: “The Ohno Continuous Casting Process”.
At the end of this article you can read:
_“But with OFHC and OCC, the nomenclature “six nines” or “eight nines” has almost no meaning. All else being equal, higher purity is a straight forward benefit. However, grain structure, softness and surface finish can each make more difference than a “nine” or two. Then there is the matter of measurable purity. Due to contamination caused by the measuring process, there is a serious question as to whether any metal can be verified as having greater than six nines purity._ **_Also, since “nines” became a selling point, some quite absurd and dubious claims have been made.”_**
I would expect that each of these cables come at least with a metallurgical attestation like you get, if you buy a gold ring at Tiffanies.
What’s your take on tubes vs solid state?
During my time at the university we had valves on the curriculum and it was very interesting, but I am a solid state designer.
How about high-end audio cables?
In my case the word “high-end audio cable” is misleading, since we would assume that my cables are designed especially for the very limited 20kHz audio band. Our analog cables are designed for high frequency signal up to 1GHz.
A simple single 50Ohm or 75 cable is not a brainer.
You can easily order one here with the connectors and the matching cable you need: link.
For the distribution of my Rubidium clock to all of my measurement instruments I use the excellent Multiflex 86 Low Loss. You see an outstanding good cable does not cost the earth, as long as it is a “standard” cable and runs in high volume.
For balanced cables it is a different story. Here the selection is very limited and shielded silver coated copper cables with a silver coated shield do not run in high volumes, therefore they are much, much more expensive and they are only built to order.
The TRINITY tonearm cable has shown you, that we have our own opinion about cables.
As I started the design, I was looking for a well-balanced high-frequency cable, since the bandwidth of my devices goes beyond 1MHz. Of course it should have a controlled impedance of 100Ohm.
Controlled impedance is important for signal integrity: it is the propagation of signals without distortion.
Impedance is an AC characteristic, meaning that it is related to frequency. The controlled impedance is of course more or less only needed in double terminated high-frequency transmission lines, but the tight tolerances needed for such cable are also extremely helpful in a simple audio application.
There are a few very good cable manufacturers specialized in designing and manufacturing of such balanced cables for space applications. For the cable insulation I had a special form of PTFE in mind. This kind of cable construction reduces time delay and skew of signal transmission, while minimizing electromagnetic interference. On the other hand such “space” cables have to work even at high acceleration during the rocket launch, therefore these cables must be resistant against at vibration and shock. The latter is very important if you want to use such cable as a tone arm cable.
Now let me explain skew. Skew is the time difference between the two wires, this means the time delay on both wires is not equal. The main factors contributing to skew are cable length, cable type and quality of cable. Skew for a given cable is typically specified by the manufacturer as skew (in picoseconds) per unit length. The balanced TRINITY cable has a worst case skew of only 100ps/m.
There are countless parameters you can measure on a cable, but honestly I have never ever seen that a dedicated high-end audio cable manufacturer has ever published such parameters.
Therefore I am very reserved, when I hear another fairy tale about a new “high-end audio cable”.
Beside all the serious audio cable manufacturers, there are countless quacksalver on this market as well.
To summarize it, the TRINITY audio cables are based on the best high performance high frequency cables and they are not manufactured or assembled at special moon or sun phases.
At the end of the day we decided to manufacture our own tone arm cable with the following spec.
Of course I use the same cable for my XLR cable as well.
How much effort do you put in dealing with micro vibrations, unwanted resonances etc…
Micro vibrations are really a serious problem. I remember 20 years ago I was measuring an amplifier and a tool dropped from my hand and landed on my desk. In the same moment I saw that the noise level on my measurement instrument went up. I reenacted the same procedure and yes, if the tool hits the desk it created additional noise. The vibration of the desk created a charge injection in the PCB.
The high of the charge injection depends on the material of the PCB and the impedance of the circuit.
Therefore my audio designs are based on very low impedances, which request higher output currents and driver capability of the gain stages, but as I mentioned above the gain stages can drive 100Ohms distortion free.
On the mechanical side I use aluminum enclosures, which are milled from one piece and the bottom plate is made from 5mm stainless steel and the devices stand on a small contact surface of 10mm ceramic balls.
The devices of the Golden Reference Edition stand on 10mm ruby balls.
The TRINITY Racks and power amp bases are mounted on special spring loaded feet, which have an extreme low self-resonance of only 3Hz. These feet reduce 85% of the vibration normally found in a house. The used special springs in the feet are carefully calculated depending on the weight of each audio device.
Please tell us more about your highly unique amplifiers?
An ideal audio amplifier should have the following characteristics:
1. Infinite Power Bandwidth
1. Infinite Output Current
1. Zero Output Impedance
1. no Distortion
1. no Noise
1. no hum neither mechanical from a transformer nor electronic hum in the signal
1. cost nothing
Everybody will agree to design an ideal amp is impossible, but how close a design comes to all of these requirements determine the real quality of the amp.
The goal during the design of the TRINITY AMP was to get a close as possible to these objectives.
How do you feel about the Class D technology? Any application in high-end?
It has nothing to do with high-end. The design goal for Class D was high efficiency and not HIFI. You have to consider that even the audio analyzer needs a special brick wall filter to measure these very noisy output signal, otherwise it would only measure rubbish.
_“The switching process adds fast rising edges at the switching frequency to the audio output signal. These fast edges are of no consequence to the typical load (a loudspeaker), but present a difficult signal for measurement instruments. The fast switching edges present high energy content and will introduce slew rate limiting when presented to the input stage of most measurement instruments. When stressed by these fast edges, the analyzer input amplifier will usually slew rate limit and will not be able to function effectively in its normal mode.”_ _AP_
_“These fast edges are of no consequence to the typical load (a loudspeaker)” **_I definitely can`t agree_** **_to_** **_this!!!!
Let us consider the whole line. In the most cases we have huge power supply conditioner at the beginning of the audio line to remove all the high frequency noise from the 100Vac…230Vac lines and at the end we add with a Class D amplifier an electronic circuit, which generates a hell of high frequency noise on the way to the loudspeaker.
This is in my eyes not a coherent concept.
Measurements and listening. Are they correlated or to be followed separately?
First of all measure means you compare things. Measurements are essential for the development of products and to compare competitive designs, not only for audio devices.
For instance if you have developed the fastest car in the world, you would have to measure the speed.
In audio it is completely different. In most cases there are only claims with typical values and even these claims reflect only the best case scenarios for a good marketing.
Nothing is guaranteed therefore the buyer has no warranty claims, if his device is far away from the typical value. In my eyes this has nothing to do with high end engineering, it is high-end marketing.
Only with measurements can you improve something, since you need a reference for the comparison.
In audio we compare the input signal from the audio analyzer with the output signal of the electronic and of course the ideal case would be that only the amplitude changes and no noise, distortions and so on are added and no parts are removed. Therefore the limitation is the accuracy of the audio analyzer.
The devices of the Golden Reference Edition are such ideal devices and as long as we have no better audio analyzer, we are unable to improve these devices further. Whereby I do not believe that there will be better audio analyzer available in the future, since we are limited in the spurious free dynamic range by the physical noise level.
Nevertheless after all this technical achievements only listening brings the ultimate decision.
For instances as I designed the first triangle shaped DAC I used different resistor types in the feedback of the current to voltage circuit and you could really hear a difference, whereby I could not measure a difference. In other applications there was no difference audible, if I changed the resistor. It shows me beside the THD measurements and so on there must be something hidden we have not yet addressed.
I recommended you also a listening test, where you to drive your power amp direct with the reviewed “MIC pre” in PHONO mode and use the gain and attenuator settings to adjust the volume instead of using an additional preamplifier between.
Only in this simple configuration can you hear the full performance of the Mic preamp in phono mode.
Can you tell us more about intermodulation distortions?
The harmonic distortions are simple to understand since all harmonic distortions are multiples of odd and even frequencies of the fundamental tone and are therefore always placed above the fundamental frequencies.
Intermodulation Distortions are completely different since they can be mixed down, for instance the famous 19kHz & 20kHz test tone generates 1st order tone at 1kHz in the lower sideband. That means even very high frequency tones can generate audible low frequency tones. I know a case were the non-linearity of a valve amplifier mixed the noise shaping noise of a SACD from above 20kHz down in the audio band below 20kHz and the designer had no idea why his system is so noisy if he plays SACDs and not so noisy if he plays PCM files.
By the way this “mixing” technique is also intentionally used in so-called up and down mixer.
For instance as I mentioned earlier, I had to design a down and up mixer to mix the 100kHz band of a WIFI network down to a 900..1000MHz band used for cable TV band and back.
Back to the roots, the amount of IMs gives you an overview of the linearity of your system and these IMs were measured right from the beginning. The following picture is from the article “Intermodulation Distortion” – published in February 1960 “Electronics World”.
For all who have a deeper interest in this thematic have a look here: link.
What music references you use for testing?
This question implies that I have titles for testing and titles for pleasure. This is not the case.
Nevertheless I have titles, which I can use for testing and for fun as well.
A very good entry level is the “Ultimate Demonstration Disc 1”
I started with the CD and looked to replace the tracks with its high res counter parts like Rebecca Pidgeon`s “Spanish Harlem” to track Mr. Chesky`s introducing comments of each tracks.
To test the depth and width of a sound stage of a DAC I like the “Bolero” from Percussion Museum, which I have as DVD-Audio in 24bits/192kHz. The problem with common DACs is that they have almost no deep stage at very silent parts of the music. That is one reason why a lot of CDs are compressed in the dynamic range. We all know the “Bolero”. After a short intro the music starts very silent and gets louder and louder. During the listening you can monitor if the instruments change their virtual positions over the volume nor not.
In an outstandingly good system you can pin point to the instrument in a razor like deep and wide image independent from the volume.
Do you consider classical music to be the last word for evaluation?
No, we would wrong all the other bands, songwriters recording engineers and so on.
On the other had we should differ. A classical “string quartet” has 4 musicians on the stage and at “Friday Night in San Francisco” we have 3 guitar players on the stage. There is not too much difference in the count of musicians, but in the music.
The real difference comes with orchestral works and if we go further orchestral works with an organ like “Symphony No 3 in C minor ‘Organ’, Op 78”
Such orchestral works request from the playback devices to play the full audio bandwidth and to create a wide, deep and razor like image. From this point of view I would say, yes classical orchestral music can be the most challenging music for a playback chain.
By the way as I mentioned above I like “Pictures at an exhibition” not only the piano version also the concert version from Ravel and of course the newer interpretation form ELP and Isao Tomita. My insider tip is the DVD-Audio from Exton (2001 Octavia Records Inc) 192kHz/24bit and Ales Barta on the organ. Unfortunately this DVD-A is no longer available only the lower resolution versions are available at amazon and qobuz
What does the state of the art mean in the world of Trinity and in Dietmar Bräuer’s mind?
State of the art is nothing else than the best available technology used by all serious companies, a TRINITY device has to be significantly better than “State of the art”.
Is there something that could be called Trinity voicing of products or its completely contrary?
The TRINITY devices have no own sound, since they were designed to reproduce the music without adding or removing anything. Only if the electronics add or remove something can it have its own sound. If you use a complete TRINITY line you will hear the recording and not the electronics.
How can a true musicality be reached with high-end audio products?
Avoid “patch work” set-ups, like amp from this company, pre from that company and so on, but to reach true musicality you have to do some room treatments. I would speak about true musicality, if the music touches my soul. If the TRINITY line is set-up in a perfectly treated room than there are songs were I get goose bumps and the amazing thing is I get these goose bumps again, if I play the same title again.
When the hi-fi stops and high-end experience kick in?
In the old days we had the HIFI standard DIN45-500. With the introduction of this standard a lot of manufacturers were very disappointed, since now they had to fulfill some exact described measurements to call their own devices HIFI products.
Like today a lot of manufacturers invented technical neologisms of nonsense parameters to push the marketing, instead to measuring the real characteristics of a device. For instance someone has asked me, if I have also a “mechanical ground”!!
I had no idea what this guy wanted and I tried to explain him that he request something like a black Lipizzaner.
Actually “High-End” audio says nothing about the performance of a device, since we have no standard specification or at least a minimum of requirements you have to fulfill to call a device high-end.
This has led to the fact that phrase “high-end” is overstressed.
I am waiting for neologisms like “Super High-End”, “Highest-End” and so on.
On the other hand, if we look at the High-End show in Munich, we see that the much bigger parts of the show area is occupied by let`s say it that very reasonably priced audio devices and audio accessories.
Shows like this do not care about real high-end anymore.
By the way, I tried to get an atrium room for many years and I always got the answer from the general manager: “It was my personally decision to reject your application!” The visitors have to know, it is not an open show, it is an event of the “High-End Society” and the founding members determine who gets a room and who doesn’t!
So I gave up to apply for this show, since my place is only 12km from the show and I can easily pick-up interested people and drive them to my place and back to the show or hotel.
What sets apart Trinity from the competition?
The most important thing is, we do not claim parameters we measure all relevant electronic parameters under worst case conditions. Every device comes with a measurement protocol.
Since we push the devices to the utmost, we do not introduce a new version of the devices every year. The Reference Line is 5-6 years old and still has the best measurement results in the market and the best price/performance ratio.
A lot of manufacturers bring new versions every year, whereby “new” need not mean better.
Our philosophy is to give our clients a very long lasting product.
You know your clients well. How would you describe the Trinity Audio customers?
I would call them “educated” music aficionado, since they have their own opinion on how a system should sound and they have a lot of experiences. One told me “Dietmar we are so happy with your products since these products give us the possibility to enjoy our music hobby on a much, much higher level.”
What is your view about the current state of the high-end audio?
On one hand there are real good devices on the market, on the other hand there are devices on market, which would be outperformed by 20 years old HIFI device, but the most bothersome are the incredible discounts. Manufacturers who offer a several thousand Euro update for free and so on. This has nothing to do with serious business conduct.
Have you ever heard that Rolls Royce gives 85% discount on a Wraith?
I would guess, if you ask RR for a discount the counter question would be: “What should we cut out?”
Even a learning-disabled amoeba would understand that a new 100.000Euro device offered for 15.000 is not worth even the 15.000Euro.
If we look a little bit deep a lot of these high-end designs are only based on the application circuit in the data sheet. This is solid engineering, but not more. For a high-end design I would expect some own sophisticated electronic circuit design.
What’s your take about the proper room acoustics?
Beside the instrument acoustic we have the room acoustic as well. The acoustic process is to “shape the tonal development beginning with the tonal perception of the performer down to the aural impression of the listener.” _Acoustics and the Performance of Music
Concert halls, operas and so on have a sophisticated acoustic to get the music from the musicians to the ear of the concertgoer.
To get a similar tonal result of a concert hall performance in your listening room, beside the perfect electronics you need a serious effort in the room acoustics.
By the way I have more than 50 books only related to room acoustics.
Is the high price a must!
In a seriously outstanding design a higher price comes alone from the developing cost and the use of the best and therefore most expensive electronic parts. Have a look on the main board of the Golden Reference Edition: link.
This picture shows only one channel and this one channel contains significant more ICs than any other device on the market and all these ICs are mounted on an 8-layer PCB instead of a usual 2-layer PCB. Such a design has of course a much higher bill of material than a conventional design.
Unfortunately there are also devices on the market like Class-D amplifier for 12.000Euro which are based on 120Euro re-branded Class-D OEM modules. Here the high price has no relation to the internal value.
“Price is what you pay. Value is what you get. “_Warren Buffett_”
By the way we have a complete high-tech SMD assembly line for roughly 80.000Euro in-house with a pneumatic solder paste printer, a full automatic “pick & place” machine, which is able to assemble even the smallest SMD parts and we have of course a vapor phase soldering machine. For the visual inspection we have an USB microscope with an optical zoom factor of 10.
For the thermal inspection we use a professional infra-red camera.
For the development of the Golden Reference Edition we had to buy additional measurement instruments also in the range of 80.000Euro. I almost forgot about all the engineering software we use.
You see to play in this league you have to do some very serious investments.
What does the slogan “Be the first or be better” apply to?
It is a simple business rule I read in a book decades ago as we started our GTE GmbH.
To be successful in a new business you have to be the first or better, otherwise there is no reason why someone should buy your product. Whereby “to be better” is not related to better performance, it can be better marketing as well.
An example was the introduction of the video cassette recorder, the VHS standard with the best marketing strategy was chosen and not the ones with the best technical performance like Betamax or Video2000.
How about the “amplifying wire”?
The “amplifying wire” is the Holy Grail of the electronic engineering. Every ambitious engineer, who has to design amplifiers, is on the search for it.
A wire is the simplest electronic part and it can be considered as ideal linear, which means a wire has no distortion, no noise and so on.
The only problem is a wire has of course no gain. Every electronic engineer is on the search for an amplifier, which has a gain and the linearity of a simple wire.
Our search is over we found the Holy Grail in the TRINITY devices of the Golden Reference Edition.
The next screen shot of the APx555 audio analyzer shows the performance of an XLR cable in conjunction with the gain stage used in the Golden Reference Edition and confirms the gain stage is as linear as a simple wire and has 6dB gain.
THD Ratio vs frequency
Left channel **CH1** single gain stage of the Golden Reference Edition
Gain 6dB
Output Voltage 6.3Vrms into 600Ohm
Right channel **CH2** XLR cable
Gain 0dB
Please note: The devices of the Golden Reference Edition are based on 6 parallel working single stages
Is there a possibility to push things even further from your current flagship products?
The PHONO and the Preamp of the Golden Reference Edition can`t be improved further, neither from me nor from any other company, since here we are on the technical limit. The best a competitive device can reach is to be equal with our devices.
There is no Golden Reference amplifier…
It is in the pipe.
What can we expect in the future from the Trinity Audio?
The completion of the Golden Reference Edition with a matching Power Amp and of course a
DAC, based on our proprietary LIANOTEC architecture.
Since two weeks ago I work in my spare time on the new DAC, which will be the last DAC you have to buy. I have a really outstanding concept in mind with up to 32 x PCM1704 per channel!!!
I also finalized the design of a headphone amplifier for the Professional line, which can also be used in parallel as Line amp, but at the moment I am not sure if there is a market for it.
It is said that music can be a window of the soul… How much of such transcendental experience we are bound to experience with high-end audio?
As I already mentioned for me the music matters. The audio electronics allows us to enjoy our beloved music at home, the performance of the electronic determines on which level we can enjoy our hobby. During our first show in Milan many years ago our friend, who did the room tuning, told me after the first day, this is his first show where he has no headache after listening almost 10hours music on a relatively higher sound pressure level.
In other words the soul can only be touched by the music, if the ears and the brain are not irritated by harmonic distortions, noise or hum and so on.
Any last thought for our readers?
I got in the past phone calls, which started with the first sentence or better question, ”What is my best discount!”
If the performance of the music reproduction is no longer important we carry the music to grave.