CD PLAYER CAMBRIDGE AUDIO AZUR 840C WOJCIECH PACUŁA

Cambridge Audio is a company, that from some time, after a performance slow-down, is in the hands of owners of a chain of audio shops in Great Britain. These are pragmatic people, that do not want to make the same errors as their predecessors, changed the whole philosophy of their product approach. First of all the whole production was transferred to China. A predictable step, made also by other players on that market, among others NAD, that gave the company financial fluency. Design and research is still conducted in England, in London, where Cambridge Audio has their headquarters (in Cambridge the headquarters of Arcama is located...). For long time, the formula for devices "by CA" included only propositions from the lowest price range, the budget shelf. Every generation of stereo products showed, that this was a good move, and furthermore they showed, that the company is developing, as every new electronic "child" was better than its older brother - looking for example at the new versions "v2" much, much better than the original devices. The CA portfolio for long time covered three series - '3xx', '5xx' and '6xx'. For some time however, it became known, that this British company readies a more ambitious project. And here we are: we have got an amplifier and CD player from the '8xx' series with the symbols 840A and 840C respectively.

Because this is a "digital" issue "HIGH Fidelity OnLine", we will take care only of the CD player, although it is worth to mention, that Cambridge does not just cut coupons from their past and in both products proposed refined and expensive technologies - especially in the amplifier. About the XD class, in which the power amplifier part of the 840A works, you will be able to read shortly in "Audio", on the side of the test of the amplifier (probably in confrontation with the Arcam DiVA A90), but in context of this test I have to mention, that some of the listening to the 840C was done together with the dedicated amplifier and that this was an excellent combination - both in the terms of sound quality as well as its character.

While the XD (a variety of the AB) class is quite innovative, its probably most important asset is, that it was implemented in a quite inexpensive, and for hi-end even 'budget' device. A similar value is in the "heart" of the tested player: upsampling. It is in itself nothing exciting, as ICs offering this kind of digital signal processing can be bought from many companies. But an absolute surprise is the fact, that in the 840C the expensive, splendid upsampler Q5 from the company Anagram Technology was used, the same that can be found in the most expensive players from the French manufacturer Audio Aero, or the Swiss Orpheusa. Furthermore, to exploit fully the possibilities of this digital power plant, the player is equipped with digital inputs (4 pieces), what allows the 840C to be the center of a digital system - with a digital TV receiver, DVD player (with stereo sound) or a DAB radio. Really an impressive showing for such a small money.

LISTENING EXPERIENCE

From longer time, from the moment that in dCS-a devices first hi-end upsamplers were implemented (I am talking about Purcell), in the audiophile world a discussion started, if this technology is worth of being dealt with, and if it brings more advantages of disadvantages. Because about the fact if upsampling changes sound there is no dispute - this can be heard clearly. In general, three "schools" of work can be distinguished. The first one, the "main stream" school tells, that you should use all the goodies the digital technology offers, with oversampling, digital filters, and similar, but for no more than necessary. That is the way most devices function. The second school, let us call it "minimalistic" says that the less processing the better, and going back to the classic Philips TDA1541 converters, resigns completely from oversampling and digital filtering, applying only simple analog filters in the output stage. And finally there is the third school, that uses everything possible, to better the sound, regardless if there is something added or removed from the sound path. For this group the upsampling is the "golden path", a key to the absolute sound. I do not know with which group you sympathize, but I must say that I heard as many good as bad examples of every recalled way of thinking. So I am not ideologically bound to any technology, because while I use "mainstream" thinking hidden - at least in the digital processing part - in the Lektor Prime (test HERE), but on my list "best of the best" are also the dCS that uses upsampling as well as the minimalistic Zanden. However I think, that in the price range of the 840C upsampling almost always brings many advantages and little shortcomings.

The CA player plays with a very pleasant sound, with a large volume. The instruments are given quite close, and do not have deep perspective, but have a kind of substance, density - the element found most commonly with upsampling - elements improving physicality and palpability of the sound. So the saxophone from Rollins from the disc "Way Out West" (Contemporary Records/JVC, VICJ-60088, XRCD) was big, strong, filled and you had the feeling that it is placed in the spot of the speaker (this is a multi-mono recording, where the instruments are placed in utmost positions in the order: left-middle-right). It was surely aided by the outstanding speakers Marcus of the company Harpia Acoustics (test HERE), but without supplying them with such a signal they would not be able to conjure this up by themselves. Anyway, the sound was presented in a congested, slightly warm (especially in the midrange) way, that made all events acquire "body" and mass. The upper treble was slightly withdrawn, or it seemed to be like that in comparison with the strong midrange, however surprisingly well the character of the recordings was reproduced. Warm devices, with a mild sound, sound that way not only due to the withdrawn treble, but also due to the lack of sufficient resolution and high distortion. Almost always they equalize the recordings, showing them in a nice "civilized" way. CA was different - the sound had depth due to the warm, large midrange and fleshy bass (I am not talking about space, but timbre), but the kind of recording is shown immediately by the 840C. When we listen to the CA play the newest disc of Sting - Songs of the Labyrinth (Deutsche Grammophon, 170 3139, CD) with recordings of lute and voice (music of John Dowland and one piece of Robert Johnson), then from the first minutes we know, that there is some kind of error, and we have not heard so badly recorded voice for long time. As if Sting would record the vocal part on an airport, in a toilet, on a Dictaphone, in breaks between the flights. So bad that it is ridiculous. It is not about the musical value, but about the quality (or lack of it) of the sound. CA reproduced the lute a bit closer than the Lektor, closer than the CD-E5 EWAE from E-Sound (test in this edition of HFOL), in a bit more fleshy, more filled way. Every appearance of the sharp, exaggerated voice of Sting was shown without the makeup, in full light. CA tried to show the whole in an attractive way, what was somewhat successful with the lute, but not at the cost of accuracy, but keeping maximum fidelity, not enhancing the shortcomings.

As the sound of the lute showed, somewhere in the frequency range of the CA is a slight "boost" somewhere in the range of the mids and the upper bass, we have the "fleshiness", body, added to every happening. This gives a slightly warmer sound, however it comes with good resolution and discriminating near placed sounds, that do not confluence, but are shown in a noble, resonant way. Euphony is perceived, especially in context of the device's price and elements that will probably surround it, meaning the amplifier and speakers, probably inexpensive, in a unequivocally positive way. We have no objection with discs carrying electronic music, where there is no "model" and we can only compare different renditions (different devices), and where the characteristics, we discussed, always come handy. That was the case with the beautiful disc Geometry of Love issued by J.M.Jarre under the name Project by Jarre (Warner Music 60693-2, CD) - a disc that shows, that this musician still has much to tell, and the last recordings under his own name are subject to a slight downfall of his form, and not a permanent musical amnesia. The disc was reproduced in such a palpable, dense way, that a few times I felt creeps coming down my spine. Although the bass was not well controlled, it is far from the performance of the Bladeliusa Freja MkII player (test in this edition of HFOL), but the two times lower price explains it largely. It must be said, that very good players from the same price level, and even cheaper, can sound with a lower, and more compact bass (an example of such a quick and very true player is the extremely cheap CDX-397 of Yamahy). The point is, that they will not surpass the very good timbre, that is guaranteed by the 840C, similar in this aspect to the E-Sound. If I had to compare the CA player with something else in that aspect then it would be the player Unico from the company Unison Research with tubes in the output stage, fitting this description of timbre, but beaten by the British device in terms of resolution.

After connecting to the digital input (we use the upsampler and converter of the 840C) the Arcam DVD FMJ DV29 (test HERE; there is already its successor - FMJ DV139, that, I hope, will come to my hands shortly...), a very successful device, costing twice the price of the CA, it was "locked" successfully, with both the signal from CD as well as DVD-A (24/96). The only unforeseen was the displaying by the 840C a report that the incoming signal has the parameters 20/96, meaning 20bit words, although the discs were clearly 24bits. However, the CA manual states, that some devices incorrectly stamps the transmitted signal, stating 20 instead of 24bits, and this is probably the case.
Regardless, I am pleased to announce, that CA reproduced the DAD discs of Chesky (discs issued by Chesky Records in the 24/94 format, without the menu and the need of using the screen) and HDAD Classic Records (those discs have the material recorded in 24/96 on one side and 24/192 on the other - the latter is converted on the digital outputs of DVD-Audio players and transmitted in 24/48 format) in a way similar to the Arcam. This is a big thing, as the Arcam is one of the leading DVD-A players in the vicinity of 10000zł. The sound of the CA was presented a bit closer, but without big replacements, in general the differences were subtle - once a bit less clear cymbals from the CA, otherwise a not so precise bass like in the Arcam. In general, the difference was not big. We can expect, that f we hook-up a DVD player to the 840C, with a decent drive section, we get superb sound.

THE MAKE

The Cambridge Audio Azur 840C is the first try, from many years, to devices more expensive than the budget "start". The device has significant dimensions, and is packed with intriguing technologies and solutions. The main characteristic is announced already in the full name of the player: Upsampling Compact Disc. Fortunately the wrong, but often used, sometimes by very renowned companies, nomenclature like 24/192 CD Player, or dumber, was not used. CD is CD and has fixed parameters of the signal 44.1kHz/16bits, and everything above that is an addition, as signaled in the Cambridge Audio name. For the self-restraint - an A. The mentioned name was screen printed on the thick aluminum front panel. We will find on it only a few buttons to access drive functions and two more, marked 'menu' and 'select', not found on CD players, that are used to navigate the menu. The 840C has, in addition of a typical set of outputs, unbalanced analog RCA, digital RCA and TOSLINK, balanced XLR output and two sets (RCA+TOSLINK) digital INPUTS. It can be treated a drive with an elaborated converter section, and playback stereo sound from DVD players, satellite tuner or DAB radio. Every input can be assigned a name, and determined if the input signal is to be upsampled or not. In addition, we can chose the format of the signal on the digital outputs, and decide if we want to add dither or not. If you use a recorder, or a separate converter, then let the dither always be "on". All this operations need a readable display, and in the 840C we have the best one found below 10000zł and above. It is snowy white, exceptionally large LCD display, with black numbers and letters. The track number and time is being displayed on the lower, much larger line of the display - the digits have the same size as the LED displays used in the Lektor players by Ancient Audio, but have worse contrast, and the upper row, with information on upsampling, menu and chosen output is smaller. The enclosure, as usual in CA, is very solid, composed of a few rigid (the sides are also profiled) elements, so the top cover is screwed from the top and not a top with bent sides.

The inside reveals a few interesting things. First of all - the drive. It is a modified Sony drive, with damping elements on the bridge connecting the sides and on the disc clamp ring, and written in Cambridge (or actually in London, where the company is located, in Cambridge resides Arcam...) servo controller program. So I think, that similar like in the case of the modified Philips drive used in devices from QUAD and Audiolab, talking about "own" drive is justified. The circuitry controlling the drive is hidden below a metal screen (interestingly in the cheaper CA players it is covered with copper...) and here is also the power supply for this section. The drive is located on a very solid, specially shaped "spider" from cast aluminum, looking similar to Finite Elemente tables, in the right scale, o course. The power is supplied by a large toroidal transformer, larger than the one in the tested in this issue E-Sound CD-5E EWAE player, but smaller than the one found in the universal player from Bladelius, the Freja MkII (also tested in this issue). From the transformer three separate secondary windings are supplying independently the upsampling circuitry, converters and analog section. On a large circuit board in the back of the enclosure expanded filtering and rectifying circuits are found, with, among others, a bank of ten nice looking capacitors with Cambridge Audio logo.

The most important IC is the upsampling chip. This is a DSP chip, with software in the Q5 version, prepared by the French company Anagram Technologies, changing the input words to 24/384kHz and output that signal to the converters. Those, hidden behind the power rectifiers heat sinks, were provided by Analog Devices and this are two ICs (one for each channel) AD1955 - multi bit converters delta-sigma 24/192, with a very high, real resolution of 20 bits. The circuit is internally balanced and has such an output. In the current/voltage conversion Burr-Browna OPA842 chips are working, and in the filters and output the ICs OPA2134 of the same company. The output is coupled by a DC-Servo circuit. The whole assembly is SMD, except for the power supply, and the WIMA capacitors in the filters near the converters. The outputs are keyed with relays. The whole looks really impressive.
The only problem is that I cannot see how the signal is led from the drive to the converter, because this path is bridged by only one, shielded cable (two wires and shield), which provides the signal (in a balanced form) to the upsampling chip. This is not the I2S, protocol, so if this is the only way, then the signal is transferred in that balanced form of S/PDIF.

HOME PAGE