WORLD PREMIERE

⌈ AURORASOUND is a Japanese company established in 2010, founded by Mr. SHINOBU KARAKI. Its lineup includes line and phono preamplifiers as well as amplifiers, including headphone ones. We are the first magazine in the world to test the company's newest linear preamplifier, the PREDA-III. ⌋

RUPERT NEVE, WHO PASSED IN FEBRUARY EARLIER THIS YEAR, founder of NEVE and FOCUSRITE, a designer of the Amtek, one of the giants of the professional audio world, whose mixing consoles and devices are synonymous with top-class sound, in an interview for the online magazine "Audio Technology" shared his thoughts on tube technology, which he started his adventure in audio with, and solid-state technology, which he devoted most of his professional life to:

I had to re-educate myself to start over on the semiconductors. I found that you could actually do more with them than you could with valves, and I got quite excited about it. But there was a lot of folklore going around about the fact that these devices were unreliable and noisy. So it all started with making sure the semiconductor designs produced a sound quality at least as good as the valves. That meant a lot of listening, and a lot of measuring. The more I got into it, the more of a perfectionist I became. I think the same goes for not only myself, but a lot of people who design equipment – you find that you can get a bit of an improvement, so you put it in.

⸜ GREG SIMMONS, Rupert Neve Interview. Part 1, www.AUDIOTECHNOLOGY.com Issue 69, January 15, 1998, accessed: 26.04.2021.

Further on, he points to the advantages of tubes resulting from their high supply voltage, and thus huge dynamics, but the conversation is not directed towards confrontation, but instead he tries to explain what is so interesting about transistors, which is not found in tubes, without crossing out the latter. This short fragment already shows that transistor circuits, if well designed, can be an equivalent alternative to tubes. The point is not to replace one technique with another, but to have a choice.

That is why the products of the Japanese company Aurorasound are so interesting - they seem to be an emanation of what Rupert Neve talked about, offering a sound based on both listening and measurements, which is not to pretend anything, but rather to bring the best out of a given technique, or what is most important in it with as little compromise as possible.

▌ PREDA-III

THE FOUNDER OF AURORASOUND, Mr. SHINOBU KARAKI, seems to share Neve's approach as almost all - except the PADA-300B power amplifier - are transistor devices. Same goes for the tested line preamplifier, the PREDA-III model. It is a dual-mono design, from the power supply, through the input, to the output, and its circuit is balanced, again - from the input to the output. I asked the designer himself about the changes compared to its predecessor, the PREDA-II model.

A few simple words…

⸜ SHINOBU KARAKI
Founder, owner, chief designer

WOJCIECH PACUŁA When did you start thinking about PREDA-III?
SHINOBU KARAKI It all started maybe one and a half year ago.

WP Are all Aurorasound’s new project so time-consuming? How long does it take to tune a new device?
SK The R&D (research and development) stage took us around 1.5 years - that’s time needed to get to the prototype. Most important element of the process is, what we call, the Product Concept - product positioning, figuring who our target customer is, coming up with an exterior design, knob and SW placement, functions, total signal path and material cost. I take 80% of total R&D stage for those studies. 

We have lot ideas for electronics circuits and components we could use but the very important element of the whole concept is figuring out how to integrate them and build into a good product. Other important factors that translate into sound quality are chassis construction, power supply section and PCB design. The ultimate sound and sonic character are already being determined in this stage. For final tuning, I do try to find proper resistors, capacitors, and optimal voltage/current distribution for every section always remembering also about thermal stability.

WP Can you tell me what makes PREDA-III unique, different from other preamplifiers available on the market?
SK PREDA-III is very basic and versatile pre-amplifier. No gimmick, no exotic functions. But PREDA-III can satisfy all requirement of what a pre-amplifier is supposed to do in any audio system, whether combined with solid-state or tube amplifier. It offers balanced and unbalanced connectivity, a large number of inputs and outputs, and it simply a pleasure to use from the very first to the very last second.

The NEC/Toshiba transistors may be unique in these days, because they don’t make them today anymore. Also the very ridge chassis may be unique.

WP With which power amplifiers combined with PREDA-III did you achieve the best sonic results?
SK The best partner for PREDA-III is the PADA-300B amp. Yet, the PREDA-III is a versatile device that will work very well with other amplifiers.

WP Why did you choose the Alps potentiometer for the new PREDA?
SK The volume control choice took me the longest time and a lot of effort for PREDA-III.  Because the original PREDA uses autoformer type I had to find a solution for the PREDA-III that would offer a superior sound. I have evaluated many types of volume control parts, shunt register, ladder resistor, semiconductor type, ......etc. Finally I found out that this ALPS offered the best sound and the best feeling while using the knob. This part is very expensive, and there is no remote control, but I decided to use this ALPS for the sound quality it offered.

WP Recently you introduced two new interesting products, AFE-10 and AFE-11. Could you tell us something about then?
SK Yes, AFE-10 and AFE11 are very unique products. AFE means Analog Front End. They are accessories for enjoying analog audio. The former is an MM cartridges Expander, that allow users to more precisely match impedance to a specific cartridge. The latter is a very unique device that allows you to send a signal from a CD Player to a phonostage and use its RIAA equalization.

I will release the AFE-12 next. This is a multi EQ curve equalizer device for 78 r.p.m. SPs and  mono LPs.

WP If you don’t mind me asking - are you currently working on any „full-size” project for?
SK Many people inquired about an integrated amplifier and D/A Converter and I do have some ideas for such products. Yet, an implementation of those ideas will take some time. In a short-time perspective I will focus on devices for analogue records playback. ♦

THE PREDA-III IS A solid-state line preamplifier. It features six line INPUTS, three of which are balanced (XLR) and three are unbalanced (RCA) ones. There are also three line OUTPUTS - two balanced and one unbalanced. There is also a switch for changing the absolute phase. Unlike previous designs by this company, the PREDA-III does not feature a remote control.

As its designer mentioned, for this design he used NOS type transistors, i.e. no longer produced ones, from NEC / Toshiba, MOSFET (field) type. The design of the device, apart from obtaining the best possible sound, was also about the lowest possible distortions and noise.

Mr. Karaki lists eight basic elements, he used to achieve his goals: channel separation into two monophonic circuits in one chassis, the use of a balanced signal path, the use of new AMP-2 Rev.4 amplification modules, the use of high-class sockets, including those from Neutrik XLR and rhodium-plated, made of turned brass RCA sockets, reaching for an optical encoder used to switch inputs, enabling the absolute phase change and a high-class, rigid, vibration-resistant housing.

▌ SOUND

⸜ HOW WE LISTENED The Aurorasound PREDA-III preamplifier was tested in the HIGH FIDELITY reference system, where it was compared to the Ayon Audio Spheris III tube preamplifier, in which we have only one tube, Siemens C3g and a transformer with taps controlled by relays.

The system is connected with unbalanced interconnects and this is also what I tested the Japanese preamplifier with: the signal from the SACD player was led using Siltech Triple Crown cables, and from the preamplifier to the Crystal Cable Absolute Sound power amplifier. The PREDA-III was powered by the Harmonix X-DC350M2R Improved-Version cable and stood on the top shelf of the Finite Elemente Master Reference Pagode Edition rack, on its own feet.

It was an A/B/A comparison test with A and B known.

Recordings used for the test | a selection

⸜ ANDRZEJ KURYLEWICZ QUINTET, Go Right, Polskie Nagrania „Muza”/Warner Music Poland 4648809, „Polish Jazz | vol. 0”, Master CD-R (1963/2016);
⸜ FRANK SINATRA, Sinatra’s Sinatra: A Collection Of Frank's Favourites, Reprise Records/Stereo Sound SSVS-017, Stereo Sound Reference Record | Signature Sinatra, SACD/CD (1963/2021)
⸜ JOHNY CASH My Mother’s Hymn Book, American Recordings ‎B0002362-02 | Compact Disc (2004);
⸜ MAYO NAKANO PIANO TRIO, MIWAKU, Briphonic BRPN-7007GL, Extreme Hard Glass CD-R (2017);
⸜ SONNY ROLLINS, A Night At The “Village Vanguard”, Blue Note/Universal Music LLC UCGQ-9011, SHM-SACD (1957/2017)
⸜ THE OSCAR PETERSON TRIO, We Get Request, Verve/Lasting Impression Music LIM K2HD 032, K2HD Mastering, „24 Gold Direct-from-Master Edition UDM”, Master CD-R (1964/2009)
⸜ Tomasz Pauszek, LO-FI LO-VE, Audio Anatomy AA-006-17-CD, 2 x Master CD-R (2017);
⸜ YES, Fragile, Atlantic Records/Mobile Fidelity Sound Lab UDCD 766, Gold-CD (1972/2006)
⸜ THE BEATLES, The Capitol Albums Vol. 2: Sampler, Apple Records | Capitol Records DPRO-009463-59566-2-7, REFERENCE CD-R (2004)

AS I SAID IN THE INTRODUCTION, the division into tube and transistor products is largely an ideological, not a logical one. Because the biggest surprise for all those looking for special distinguishing features of "tube" or "transistor" sound will be the incredible smoothness of the preamplifier’s sound that was developed by Mr. Karaki-san. I listened to the opening guitar on YES's Fragile album and had shivers going down my spine, literally, it sounded so dense and so smooth at the same time. The album I am using was released by Mobile Fidelity on a golden disc in 2006, so it is not the newest remix of Steve Wilson, and yet the sound was excellent.

The smoothness I am talking about manifests itself in the beautiful fluidity with which the device lets the music flow through itself, shaping it in such a way that it is whole, thickening it rather than showing shallows. And yet the PREDA-III turned out to be a preamplifier with a very detailed imaging. This is a complex issue, because we usually ascribe a clear detailed, sharp imaging to semiconductor devices, and at the same time we indicate tube devices as those that show the bodies of instruments more clearly, in a more resolving way.

These two views are not mutually exclusive - the first is about detail, and the second about information and both are wrongly attributed only to one or to the second technique. The Japanese preamplifier could serve as a kind of "icebreaker" breaking the bottlenecks in thinking, because it shows large and saturated shapes, but in a clear, accurate way.

| Our albums

⸜ THE BEATLES The Capitol Albums Vol. 2: Sampler
Apple Records | Capitol Records DPRO-009463-59566-2-7

REFERENCE CD-R (2004)

AT THE END OF 2004, the AMERICAN LABEL Capitol Records released a compilation The Capitol Albums, Volume 1 with The Beatles' albums originally released in the USA in 1964. These recordings are sourced from tapes prepared by Dave Dexter Jr., then director of A&R. He used master tapes intended for US releases and added reverb to them, and some simulated stereo effect. Its second part was released in 2006. Both boxes contained several CDs, so special samplers with selected songs were also prepared.

Their sound is different from the British versions, it is not so intimate and tangible. On the other hand, the remaster is very good and the sound is surprisingly resolved and full. The version of the Reference CD-R sampler that I am talking about was prepared by the Columbia mastering department as an internal reference against which the final CD versions were compared. Its sound is more resolved and more precise than ready-made Compact Discs, it also has clearly higher energy.


Once again it is confirmed that the audio signal is modified between the mastering studio and the CD, mainly during the preparation of the glass master and then during pressing. ♦

ALL I HAD TO DO WAS TO PLAY THE BEATLES, but just from the aforementioned reference version burned in the Capitol studio mastering on the CD-R, to hear in the album opening track Baby It's You the harmony and the rawness of the sound. On the right side there are vocals recorded with a close microphone, without any „beautification”, and on the left side we have drums, in the center, in the back, the guitars. Lennon's vocals were shown very close by the tested preamplifier, almost with a "visible" slight hoarseness, while the drums had quite a long reverberation, sounded as if they had been picked up by microphones from a distance.

It was a very good reproduction of this recording, without emphasizing the roughness in Lennon's voice and without deepening the reverb on the percussion. And yet both of these elements were clear, maybe even more than with the reference preamplifier, i.e. the tube preamplifier. It seems that this is because the PREDA-III shows information in the 1-5 kHz range more clearly.

You probably don't remember, there is no reason for you to, so let me remind you that when testing the excellent, two-box MARK LEVINSON №52 preamplifier I wrote that the reference preamplifier, this is the Ayon Audio Spheris III, sounded a bit dark compared to it and that I thought that Levinson, however, this one element, i.e. reaching information from the mentioned range, does more effectively, better. I would say that the Aurorasound, although it is a much cheaper device, does it in a very similar way.

I did not mention it earlier, so I will say it now that when listening to the Yes album with the PREDA-III, the noise of the tape was better audible, it was slight, but present. Ayon modified it a bit, it brought out stronger bodies, but it was just the importance of this information that was diminished. What's really interesting is that the tested preamplifier does not sound aggressive or bright, absolutely not! As I said, the sound is smooth, nice, you could even say: sweet. And yet it shows a lot of information.

As on the SONNY ROLLINS mono album, a live recording from Village Vanguarde A Night At The "Village Vanguard", where in the opening seconds, still without music, you can hear the laughing lady, with a Japanese preamplifier it was more real, more "present". The saxophone also had clearer contours, it was stronger, and yet there was not even a hint of brightness or harshness.

The more expensive preamplifiers, on the other hand, are more energetic and show a slightly larger volume of instruments. The important information, however, is that these are not big differences. Without an exact 1:1 comparison, with a calibrated volume level, so that the playback on the PREDA-III and Spheris III would have the same loudness, it would not be immediately clear, and perhaps not at all.

After all, the voice of FRANK SINATRA from the Sinatra's Sinatra: A Collection Of Frank's Favorites, listened to in the SACD version, published by the Japanese magazine "Stereo Sound", was large, full-bodied and dense. You could hear a lot of reverb that was put on it, but it was not extended, the reverb on the strings sounded stronger, as in the track In The Wee Small Hours. It was on this album that I heard an important feature of the Japanese device's sound, namely low, dense, but at the same time slightly rounded bass. An extremely interesting situation - the transistor preamplifier, and thus stereotypically more precise, sounded much more "tube-like" than the tube preamplifier.

Please do not get me wrong, the PREDA-III has one of the best basses I have heard with a preamp, but expensive devices, I mean: much more expensive, they do it a bit better, differentiating the tension within the sound more clearly, its "mobility". And yet I couldn't stop listening to the Japanese device with a smile on my face, it all sounded so good.

On the Sinatra's disc it was about the sound of the double bass, while on the disc of TOMASZ PAUSZEK LO-FI LO-VE it was about recreating electronically generated sound. It was audible with it, that at the very bottom the energy is not as high as with the - twice as expensive - reference preamplifier. Both comparisons also showed that the tested device is fantastically musical. It focuses our attention on the listening axis, which is why monophonic discs sounded incredibly addictive and real, but they did not close the edges of the sound stage, which was large and full-bodied during the auditions - and these are derivatives of a saturated low range.

▌ SUMMARY

I have to admit to you that while listening to the Aurorasound PREDA-III preamplifier, I did not miss my preamplifier at all. Yes, when I got back to it I had a feeling of coming home, I knew it was - ultimately - a better device. Its "betterness" was, however, put in quotation marks, and the Japanese device remained in my memory as something special.

It shows large phantom images, a bit further away than the Ayon, but closer than - say - ACCUPHASE C-3900 or the aforementioned Mark Levinson. It was most reminiscent of what I had heard with Naim's Statement system (test HERE).

The treble and midrange are flawlessly smooth and pleasant with it, yet extremely accurate and carry a lot of information. The reverbs are reproduced in a very good way and only the micro-information about the immediate surroundings of the instruments is better with the reference preamplifier, which makes their sound a bit less “tense”. The low bass is strong, dense, but it does not extend as low as in the most expensive preamplifiers.

The most important thing, however, is the pleasure of communing with the tested device. One, that is well-made and two, that it simply sounds great. It does not sound like a tube or a solid-state device - it just sounds like the Aurorasound.

▌ DESIGN

AURORASOUND PREDA-III IS A BEAUTIFUL, solidly built device. Its chassis is made of perfectly matched aluminum plates - silver on the front (10 mm) and back, and black on the other sides: Sides: 8 mm, top and bottom: 5 mm thick). The device rests on cast iron feet with felt inserts.

A moderation draws attention in this design, but its maker winked at us, cutting cooling holes in the upper panel, covered with a silver mesh, placing orange LEDs underneath, that operate as current sources in audio and power circuits. The point is that this type of device does not need cooling holes - so they have a decorative function.

FRONT AND REAR The PREDA-III is a classic, analog preamplifier, with two manipulators on the front panel - an input selector and volume control pots. Between them there is a row of light emitting diodes (LEDs).

The first three, blue ones, tell us about the choice of one of the balanced (XLR) inputs, and the next three, purple ones, indicate a choice of unbalanced (RCA) ones. In the previous version of the preamplifier, the volume was shown on the display, but the choice of the attenuator in the new version caused changes - now it is a classic knob with a red dot on the collar, and adjustment cannot be made using the remote control (which is not available).

And there is one more LED that informs about the change of the absolute phase. There is no switch to be associated with it, because it is placed underneath, at the front edge. There is also a power switch next to it.

All connection sockets - both signal and power - are rhodium plated. It is a metal used to protect contacts interchangeably with gold. It is important to know that these two metals produce significantly different sonic results. Sockets look very solid, fair although there are no markings on them.

INSIDE There are two important inscriptions on the back of the preamplifier: "True balanced Signal Path" and "LR Dual Monaural Construction". They are important because they tell about the structure of the tested device: it is a balanced and dual mono design.

Which I saw right after I unscrewed the top cover. The audio circuit is mounted on two PCBs, separately for the left and right channels, just like the power supplies, also on separate PCBs. Each channel also got its own power transformer (toroidal). Screwed one above the other, they are placed in the corner, as far as possible from the audio circuits.

At the input there is a circuit called AMP-2 Rev.4: "High Performance Precision Analog Amplifier Module", covered with a blue aluminum screen. They are mounted on additional, gold-plated boards - at the input we have low-noise MOSFET transistors from NEC / Toshiba, and at the output an integrated circuit that lowers the impedance and ensures high current efficiency; is the so-called "Line driver". As we read in the company materials, this module is 1.5 times faster and has lower noise than its predecessor in the PREDA-II model. This module also uses precise Takman resistors, made of three materials: Bi-Cr-Al, brass plugs and OFC copper terminals.

After this module, the signal goes to the attenuator - it is one of the most beautiful potentiometers I have ever seen, next to the TKD (Tokyo Ko-On Denko) potentiometer, used by the Octave company in the JUBILEE PRE. They are manufactured by the Alps company. It has a heavy, stiff, brass body with a diameter of 50 mm and four separate segments with a stop path. The track has been polished to a mirror surface, and the signal is fed through several silver "brushes". The potentiometer has no motor, therefore the preamplifier is controlled manually. I would like to add that the inputs are changed using Omron relays controlled by the Burns optical encoder.

After attenuating the signal, it goes back to the main board, where we have the output buffer, again with transistors and a "driver" at the output. High-class passive components such as precision resistors, Wima, Vishay and Nichicon capacitors draw attention. Between them and the output - the signal is led through shielded interconnects - you can see an Omron relay, which (I think) changes the absolute phase of the signal.

Equally good components are also mounted in the differential power supply - the rectifiers are screwed to the bottom of the housing. Ripple filtering of the supply voltage is handled by very good Nichicon Muse series capacitors. Let us add that the IEC socket is integrated with the "Pi" type filter and that the ground wire is not connected to anything.

PREDA-III features a relatively simple circuit, but with excellently executed using beautiful components. ■

Technical specifications (according to the manufacturer)

• Frequency range: 10-100 000 Hz (-1.5 dB)
• Total Harmonic Distortion:
XLR: 0.0008% | RCA: 0.001%
• Input impedance: 47 kΩ (RCA and XLR)
• Output impedance: 47 Ω (RCA and XLR)
• Max input signal: 8 V (XLR/RCA)
• S/N (2 V, A-weighted):
XLR: 113 dB | RCA: 108 dB
• Gain: 9.5 dB
• Power consumption: 60 W
• Dimensions: 420 x 360 x 112 mm
• Weight: 10.2 kg