⌈ The Japanese company C.E.C. was founded in 1954. For many years, it was primarily an OEM manufacturer, supplying turntable components such as motors, bearings and suspensions for external companies such as SME and Oracle. In 1991, however, it presented an invention that would define it for the years to come - a belt drive mechanism for a Compact Disc drive. For this test we got a two-box Compact Disc player with SUPERLINK connection. ⌋

HE THIRTY-YEAR ANNIVERSARY OF THE CEC's BELT DRIVE is not particularly celebrated (the company is called 'C.E.C.', but I will use the simplified notation 'CEC'). I guess the reason are problems related to the pandemic, which are particularly strong in Japan. Let me celebrate this anniversary with the test you will find below - the drive in question is one of the most interesting solution related to the COMPACT DISC playback.

| Belt Drive

COMPACT DISC IS A PLASTIC or glass disc in which spiral recesses are pressed, the so-called "pits". A layer of metal reflecting the laser beam is vaporized on the surface of the disc. The speed of reading the signal from the CD is constant - Constant Linear Velocity; CLV - therefore, the closer we are to its end, the disc must spin slower. Therefore one of the tasks of a CD transport is to establish the precise rotational speed at each point on the disc, which is also different for each of these points.

Apart from C.E.C transports, all other CD transport mechanisms are the so-called Direct Drive ones. The disc is placed in them on a plate, which is mounted directly on the axis of the drive motor. This is a proven, reliable solution, but it has a disadvantage - vibrations from the motor are transferred directly to the CD, and electromagnetic noise is transferred to the optical reading system. Although theoretically it should not matter much, and the errors should be handled by the correction circuits, it is one of the biggest issues of this solution.

C.E.C.’s engineers therefore assumed that reading a digital signal from a CD is in fact a process similar to reading a signal from a vinyl record. And if so, a significant improvement in the sound would be achieved by the use of a belt drive, similar to what we know from turntables.

This is how the Belt Drive CD was developed - a transport mechanism in which the torque is transferred via a drive belt from a motor axis to an axis one places a CD on. The first device of this type, the TL1 model, was presented in 1991. In the top TL0 3.0 transport, as well as in the tested TL2 N, the optical module is also moved with a belt. Since such a system could use a much more powerful motor than in classic players, the axis is part of a heavy shaft that gives it a lot of torque. In order to increase the weight of the system even more and to ensure that the disc always has a flat profile, a heavy clamp is placed on top of it. ♦

THE CASE OF PRODUCTS that we would like to present to you this time is quite special. The device in question consists of two elements: the CD TL2N transport and the DA SL digital-to-analog converter. So we would normally say that we are testing a "CD transport + DAC". Only that the newest CEC DAC is designed ONLY for transports of this Japanese manufacturer, because its only digital input is the SUPERLINK. So to properly describe it, we would be closest to the truth calling it a: "two-box Compact Disc player".

We have already tested one of its elements - we liked the TL2 N transport very much and it left behind a very good impression. It has been on the market for a long time, and its test can be found in the High Fidelity № 184 from August 1, 2019 (more HERE; accessed on 3/08/2021). The above-mentioned test contains all the information about it, which we will now only briefly recapitulate. The D/A converter, on the other hand, is a complete novelty and I should elaborate on its design a bit more. The test will, however, concern a system composed of the transport and the DAC.

▌ TL2 N

TRANSPORT OF COMPACT DISC DESIGNATED TL2 N was introduced to the market in 2017. It uses Belt Drive mechanics, which requires the use of a sliding lid and loading discs from the top - hence we're talking about a classic top-loader. The CEC's lineup is divided into four series, ranging from the top '0' to '5' - the tested model is therefore closer to the CD5 player than the higher series (check the CD TL0 3.0 review HERE and CD5 player’s HERE; accessed: 3.08.2021).

Nevertheless, the TL2 N is a device unique for this company - although it is a model belonging to the '2' series, its drive features two belts, one for moving the disc and the other for moving the optics mechanism, like in more expensive transports from this manufacturer, including the top '0' model. In addition, the device looks classic like any other CD transport. At the front there is a small display, as well as five control buttons and a sixth power button. Next to the display, information related to the device's operation are illuminated in blue. Two of them relate to upsampling. The digital signal available on three outputs - RCA, TOSLink and AES/EBU - can be upsampled to 24 bits and 88.2 or 176.4 kHz.

In the tested system - a two-box CD player - the most important thing is the SUPERLINK output. On this output, the signal is not upsampled, and it is sent directly from the transport. There is another BNC socket next to it, this time an input - to improve the signal timing, we can use an external 44.1kHz reference clock.

▌ DA SL

FROM THE OUTSIDE CEC'S NEWEST DAC is similar to the transport it accompanies. You can see that the manufacturer aimed for a design coherence and lower production costs by using basic housing modules for both devices. It was not a difficult task, because a "design coherence" is a far-fetched concept for this set. Suffice it to say that these are black or silver devices with housings made of thick, stiff, bent steel sheets with aluminum fronts.

The only element that breaks us out of this monochromatic utilitarianism is a „window” in the center, under which there is an acrylic glass covered with black paint, in which several holes are cut through which blue light shines through. There we will read the type of the selected Superlink input, the selected digital filter and the presence of an external reference clock.

I am talking about different Superlink inputs because historically this has been done in two ways. One was to use a 9-pin D-SUB socket, and the other was to use four BNC sockets. The former was used in devices where there was no room for such an extensive link, and the latter in all "full-size" devices, since it is a much better solution.

| Superlink

"CEC SUPERLINK Digital Signal Transmission System Connection", or Superlink for short, is a 16 bit, 44.1kHz digital audio transmission protocol developed by CEC and used exclusively in CEC products. In the DL SL version it is based on four 75 Ω cables with BNC plugs:

1. MCK (master clock): the main clock sent from D/A converter to the CD transport,
2. BCK (bit clock): sends a clock for identifying and extracting data such as sampling rate from the CD transport to the DAC,
3. LRCK (L/R clock): clock signal for left and right channels,
4. DATA: audio signal sent from the CD transport to the DAC.

In this way, the clock of the left and right audio signal channels, the clock of a signal returning from the DAC to the transport and the audio signal are separated. It resembles another method of transmission, IIS, which I described recently in the DENAFRIPS TERMINATOR PLUS converter review. Let me remind you that dCS (Dual-AES/EBU), Esoteric (ES-Link) and Accuphase (S-Link) also have their own transmission protocol. In Asia, the ZeroLink link, used by SOULNOTE, SFORZATTO and others, has recently become more and more popular. It uses a DVI digital video socket.

The CEC with Superlink belongs to an elite "club" which resigns from the classic S/PDIF link (RCA, BNC, TOSLink and AES/EBU), in which all signals are sent via the same cable, in series. They need to be "packed" on the transmitter side, and unpacked on the receiver side, which generates jitter. The downside of this solution is the lack of compatibility with third-party devices. ♦

The DL SL DIGITAL-ANALOGUE CONVERTER is therefore a device designed to work only with CEC CD transports. When we look inside it, it turns out that is an extremely simple circuit, based on a single ESS ES9028PRO DAC. It is an eight-channel D/A chip, in which four channels are combined into one, thus improving the signal-to-noise ratio and minimizing distortion. The chip has a theoretical resolution of 32 bits and is based on the company's HyperStream architecture. After that, there are only two integrated circuits per channel and that's it.

So I assume that the DL SL designers wanted to eliminate from the signal’s path as many circuits and elements as possible. Among those are digital inputs with switches and their power supply. The analog output is also simple as it only has two integrated circuits per channel. The device offers a balanced XLR and unbalanced RCA outputs, the former’s output is not twice the voltage of the latter, which suggests two separate output paths.

Performance can be improved by using an external 44.1kHz reference clock. You can also choose one of the two digital filters, branded as Flat and Pulse, offered by the ESS chip. The first is a classic brick wall filter with a steep slope behind the Nyquist frequency. It offers good results because it suppresses mirror images (the signal from the frequency range is duplicated outside the passband of the system), i.e. ALIASING DISTORTION.

Its problem, however, is its not very cool behavior in the time domain - each impulse is accompanied by oscillations before (!) and after the impulse. They cause sharp and "digital" sound. Over time, manufacturers have learned to deal with them, but these are only half measures. This is where the Pulse filter comes into play. It is a filter with a slow decline in the frequency response, so that before and after an impulse there is only a single “wobble”. The downside is that the aliasing distortion is greater with it and the frequency response at 20kHz is lowered by 2dB. As always, the choice is ours.

Just shortly about remote control. The CEC uses the ugliest and least handy remotes I know. Although the one supplied with the DAC, the RU-217 model, is slightly better than the one accompanying the CD transport, it does not make much of a difference.

▌ SOUND

⸤ HOW WE LISTENED The CEC was tested as a system, ie as a two-box Compact Disc player. The manufacturer supplies the converter with BNC cables. However, I would suggest that you think of them same as about loudspeaker jumpers - they should be replaced with higher quality ones as soon as possible. In the test, I used four FURUTECH ALPHA SERIES FX-αAg cables prepared by the company's Polish distributor.

The player was placed on the FINITE ELEMENTE MASTER REFERENCE PAGODE EDITION MkII rack - transport on carbon fiber shelves, and the DAC on a classic wooden one. The CEC was compared to the AYON AUDIO CD-35 HF EDITION SACD player.

Power was supplied by the Harmonix X-DC350M2R Improved-Version (transport) and Acrolink 8N-PC8100 PERFORMANTE NERO EDITION cables. The signal to the AYON AUDIO SPHERIS III preamplifier was led through RCA SILTECH TRIPLE CROWN interconnects. I put passive EMI/RFI VERICTUM X BLOCK filters on top of the DAC and the transport.

Recordings used in the test | a selection

⸜ AUDIOPHILE ANALOG COLLECTION Vol. 1 + Vol. 2, 2xHD Fusion 2HDFT-C1143/1167 | 2 x CD (2020).
⸜ ISTANBUL, wyk. Hespèrion XXI, Jordi Savall, Alia Vox, AVSA 9870, „Raices & Memoria, vol. IX", SACD/CD (2009).
⸜ JESTEŚ BOGIEM, soundtrack, Magic Records 3719533, CD (2012).
⸜ PAGANINI FOR TWO, Gil Shaham, Göran Söllscher, Deutsche Grammophon/Universal Music Ltd, Taiwan 480 246-5, XRCD24 (1993/2009).
⸜ BECK, Sea Change, Geffin Records/Mobile Fidelity, UDCD 780, „Special Limited Edition | No. 01837”, Gold-CD (2002/2009).
⸜ DEAD CAN DANCE, Into The Labyrinth, 4AD/Beggars Japan WPCB-10076, „Audiophile Edition”, SACD/CD (1994/2008).
⸜ LEONARD COHEN, Popular Problems, Sony Music Labels SICP-4329, CD (2014);
⸜ SANTANA, Supernatural, Arista | BMG ‎07822 19080 2, CD (1999).

INTERESTINGLY, WHILE IN MANY ASPECTS the two-box CEC player is different from the reference player, if I had to briefly summarize its sound, I would use similar, or even identical terms, which I would use when describing Ayon Audio CD-35 High Fidelity Edition (№ 1/50).

So I would talk about fluidity and density, I would recall comparisons with "analogue" sound, at least when it comes to its stereotypical distinguishing features, but I would also point to something more, the inner richness and credibility of this presentation, rather characteristic of a good analog tape recorder than of a turntable. Of course, it would only be an approximation to what this device can do and how it really relates to the aforementioned Ayon, but an approximation surprisingly, in my opinion, internally coherent and true.

Take, for example, the track opening the Audiophile Analog Collection Vol. 2 sampler - a set prepared by our friend René Laflamme (2xHD, Fidelio, Nagra). The piece was recorded at CBS Studio Quebec using a pair of Fidelio microphones in the M-S arrangement directly on the Nagra IV-S stereo tape recorder, so one intended as a reporter tool and for field recordings. Which means a good, but only good, dynamics and a good, yet not the best signal-to-noise ratio compared to better devices.

The M-S technique uses two microphones, one with a "figure-eight" directional characteristic and the other with a supercardioid directivity, placed coaxially one above the other. A special circuit mixes their sound, allowing for the recreation of unique stereophony. Its advantage is excellent imaging, due to the fact that microphones can be placed close to the performers, and the disadvantage is the necessity to use them with small ensembles. Krzysztof Sztekmiler in the Podstawy nagłośnienia i realizacji nagrań (PL) textbook even says that it is a "reporter" method.

The proximity of the microphones means that the sound image is close to us, it is at our fingertips. CEC showed this recording exactly this, and in addition in a very dynamic and powerful way. The spatial elements, like the dense air surrounding the guitar in the left channel, as well as the cymbals placed quite far in the back, also on the left side, were perfectly captured and re-created. Right from the start, one could also hear a certain feature of this sound, which will repeat itself with each subsequent disc, namely the momentum - the tested player sounds “full”, sounds strong, and finally - deep.

It does this by bringing the foreground slightly in and blowing it up while deepening the mid-bass range. I know from experience that these are the features of a belt drive and the higher the model of the device, the more expensive it is - because that's how it works - the more sophisticated it is, and yet it is still about the same set of features. It seems that the DA SL converter is able to catch it very well and tries to "keep up" with this sound, because the TL2 N transport itself connected to the Ayon digital input did not have these elements that are so strong, so unambiguous.

Therefore, the bass line on Nevermind from LEONARD COHEN's Popular Problems played on synthesizer had a deep, low "vibe". It was a really deep sound, which is rare in digital players, regardless of the type, and if anything, I would associate them with SACDs and DSD files. The CEC plays this element stronger than the Ayon, so the whole presentation is placed closer to a listener, and its foreground is bigger.

The transition between the bass and midrange is a bit withdrawn - where the Ayon shows information about sustain, the CEC focuses more on attack and decay. Cohen in this piece recites, does not sing, and in a characteristic way lengthens the vowels in the words ending the phrase, like this (triple vowels are my underscore):

I had to leeeave
My live behiiind
I dug some greeeaves
You’ll never fiiind

Nevermind, Leonard Cohen/Patrick Leonard

The CEC player covered this manner somewhat because it emphasized the consonant attack more. And yet the last thing that can be said about this device is that it sounds "bright", because it doesn’t - many not fully calibrated turntables sound much brighter - but the device emphasizes the attack. Which gives amazing rhythmicity, while maintaining fluidity and natural softness. Which allows tracks recorded in less-then-perfect way to show themselves from a new, better side.

One such album that I love musically and that I have had since its premiere, when I got it for my birthday from my relatives, is SANTANA’s Supernatural. This album, released in 1999, brought a renewed interest to the artist and brought him back to the attention of the media. Thanks to the excellent compositions and cooperation with other famous musicians, it reached the first position on the charts in eleven countries, including the USA, where it was awarded fifteen-times (!) Platinum record status. The first single, Smooth, held stayed number one on the Billboard Hot 100 for twelve weeks.

Still, the album is poorly recorded. It seems that it was created with the charts and car radio playback in mind, so it was made as loud as possible, reducing its dynamics to almost zero. The material was recorded in various studios, even as part of one song, by various sound engineers. This resulted in a spatially blurred sound with low resolution and even lower selectivity. And yet - and this is the triumph of art over technology - one can truly enjoy listening to it.

The tested player did not deepen these problems, although it did not bring out the details that Ayon can add to enrich the presentation. It did something else - because it has a perfect rhythm and can "pick up" the attack of the sound better than most digital devices I know, it gave this album a "flow", fluidity, he let the next pieces to develop their story. I’ve talked about the strong bass we get from this player and I heard it also from the Supernatural. I was afraid that it would weigh it down too much, but no - the sound was deepened, but not slowed down.

Although the reference player is more resolving, the selectivity of the CEC is so good, and at the same time so refined that at first it seems that it shows more details, that they are part of the whole, never separate from it, so they do not attract our attention, but there seem to be a lot of them. They are placed in a large space where the foreground is promoted. However, unlike in "warm" devices (although you could call this sound that way), the space itself can also be heard perfectly.

Let's listen to the The Ubiquitous mr Lovergrove track from the DEAD CAN DANCE album, Into The Labyrinth, in the 2008 Mobile Fidelity remaster. The series in which it was released is not particularly sophisticated in terms of sound - it was created from PCM digital files - but this album was a great success for them. The perfection of the tested device in creating a large, really large space, and at the same time changing space depending on the context, was heard in the way the CEC was able to differentiate the reverberation accompanying Brendan Perry's voice and the "bagpipes" and the reverb superimposed on the synthesizer.

The first two extend behind the line connecting the speakers - far, far into the soundstage. In turn, the reverberation of the synthesizer comes out in our direction, surrounds us, goes out behind our head. The CEC showed it brilliantly, maybe even more clearly than the Ayon, which focused more on differentiating the instruments in their bodies. So with the Japanese player we will never run out of large, spacious, well-organized space.

So we have a strong, dense and low sound with a clear attack, which, however, is so refined that there is no impression of hardness or brightening. The device is also above average rhythmic, but it can also differentiate this element, because it will show acoustic and electric recordings differently, also reference and so-so discs will be nicely differentiated. And it offers an excellent space, no worse than a good turntable does.

⸤ CD vs SACD These are all the elements that usually distinguish Super Audio CDs. While listening to the Dead Can Dance disc, I also paid attention to the differences between the CD layer played by CEC and what we lose by giving up the SACD layer (I also listened to a few CDs and the corresponding single-layer SACDs).

And it is like that - it's not a very big difference. Saying "negligible" I would exaggerate a bit, but I would also never say that it is a "decisive" difference. With SACD, the attack is a bit softer and the timbre more velvety. And yet, due to the fact that the Japanese player has a similar character in itself, it was not something that would limit me a lot. Without direct comparison, the differences disappeared and I quickly forgot about them.

▌ SUMMARY

CEC TWO-BOX PLAYER sounds very good. Its key component is a sophisticated transport, and the DAC section is extremely simple - unlike in my reference player. Nevertheless, their sound together has similar features, but on a different level - the Ayon is a better device. It is scary to think what would happen if we combined the Ayon DAC and the CEC transport, particularly if we could use the excellent Superlink.

Anyway, the Japanese player sounds in a very "analog" way, that is, one in which there is no room for irritation, but only for liquidity. You can listen to this sound for hours without getting tired. It has a low bass with an emphasized mid-range, which brings the soundstage closer to us and the foreground is the most important element of the soundstage. The high treble is warm, but mainly when using the Pulse filter. But it is this filter that gives the sound refinement.

If someone says that a CD is "passe", he is only talking about the mainstream - in high-end it is still one of the best ways to reproduce a digital signal - only SACDs offer even better sound. A good file player - see Aurender N20 - can do a lot, but in direct comparison the physical disc is still better, more internally complex sound medium. Therefore, I recommend this two-box player to anyone looking for a reliable, beautiful sounding music source.

▌ DESIGN

CEC DEVICES ARE extremely solid - they are also heavy. This is due to both, the rigid chassis and the weight of its drive. The housing is made of thick, bent metal sheets, and the device stands on albumin feet.

▌ TL2 N

From the outside, the TL2 N looks just like many other CD players and transports. It is distinguished by a sliding top-loader transport lid, which easily gathers dust. The front of the device is made of a 5 mm thick aluminum plate with a cut out window covered with a transparent plate. Underneath there is a small, not very legible display known from devices from the 90s.

Similarly to turntables, here the shaft on which the disc is placed is decoupled. Together with the optics system, it is bolted to a thick, aluminum plate decoupled from the base by four elements resembling silicone rubber - two at the front are different in color than the two at the back, so I assume that they also have different mechanical properties. The disc clamp weighting 380 g is made of brass. The controls and outputs are on the PCB behind the drive.

The power supply is based on a medium-sized Bando toroidal transformer. It has two secondary windings, and thus we have two separate stabilized power supplies. You can see Rubycon filtering capacitors in them.

The remote control is newer than the display, but still quite old. In addition to controlling the drive and upsampling, functions also available from the front of the device, the remote control can also dim or turn off the display and backlight, as well as loop the track and the disc and program its playback.

▌ DA SL

The ELECTRONIC CIRCUIT OF THE DA SL TRANSMITTER looks quite simple. It is mounted on one printed circuit board, occupying less than half of the inside of the housing and divided into two sections: input and output.

In the input section, there are single receivers, but also the entire power supply, consisting of two separate power lines. They are based on a small toroidal transformer from the Japanese company Bando, similar to the one used in transport, mounted on a small vibration damping pedestal. Integrated voltage stabilizing circuits and voltage ripple filtering capacitors are on the audio circuit board; all electrolytic capacitors come from the Japanese company Rubycon.

As I said, the audio circuit is simple: its input has BNC sockets, but without a digital receiver - this is a signal that is directly "understandable" for a D/A converter. The DAC is an ECC technology chip, model ESS ES9028PRO. The output features four integrated circuits - two Burr Brown OPA2604 and two Texas Instruments OPA1688A. The latter are powered by a relatively high voltage (36V), which is to ensure high dynamics of the audio signal. What's more, they also have a high output current, so they can even be used as a headphone amplifier. Before the output sockets, gold-plated RCA and XLR (Neutrik), there are also small relays that "mute" the output.

Nice, good layout, though - let's say it - a very simple one. ■

Technical specifications (acc to the manufacturer)

⸤ TL2 N
Supported formats: CD-DA, finalized CD-R/RW
Disc stabilizer: ø 120 mm, 380 g
Digital outputs: Superlink (4 x BNC), RCA, AES/EBU, TOSLink
Clock input: RCA 44,1 kHz
Upsampling: 24 bits/88.2 kHz, 176.4 kHz
Power consumption: 21 W
Dimensions (W x H x D): 435 x 335 x 103 mm
Weight: 12 kg
Color: silver, black

⸤ DL SL
Digital output: Superlink -16/44.1
Frequency range: 20 Hz-20 kHz (± 0 dB, Flat filter)
S/N: 105 dB (1kHz, 0 dB)
Crosstalk: 105 dB (1kHz, 0 dB)
THD: 0.003% (1kHz, 0 dB)
Output voltage: 1.5 V/XLR | 1.2 V/RCA
Power consumption: 10 W
Dimensions (W x H x D): 435 x 335 x 103 mm
Weight: 8.2 kg