Nakamichi Dragon Cassette Deck
When Nakamichi introduced the first three-head cassette deck, the original Model 1000, it included a system that enabled the user to align the azimuth of the recording head to that of the playback head, so that every cassette made on the machine would play back on it with the highest possible accuracy. (The current version of that model, the 1000ZXL, performs this task automatically before you start recording.) But alignment of the recording head does nothing to eliminate the disparities that can occur when you try to play a cassette that wasn't recorded on that particular deck.
Herein lies the Dragon's breakthrough: Instead of aligning its recording head to match the playback head's azimuth, it adjusts the playback head to match the azimuth of whatever tape is being played. Its right-channel playback gaps (of which there are two-one for the forward and one for the reverse direction) are split, so that the inner and outer edges of the track are read separately. The Dragon compares the signals coming from the two sections of the gap and automatically adjusts the playback head's azimuth until their phases exactly match. (The direction indicator in use- forward or reverse-flashes until any ' 'perceived" error is corrected.) So the azimuth of any tape, no matter what its origin, can be matched by the adjustment process as long as the error falls within the Dragon's adjustment range (±12 minutes of arc, or one-fifth of a degree). Its inventors call this feature NAAC, for Nakamichi Auto Azimuth Correction, and its range is broad enough to encompass azimuth disparities that, left uncorrected, would produce serious high-frequency rolloff, with response theoretically down by about 3 dB at 10 kHz and 10 dB at 17 kHz. We found many tapes in our collection that required azimuth adjustment away from the playback head's median position, but none whose azimuth was beyond NAAC's range.
The Dragon's closed-loop dual-capstan drive follows the design of Nakamichi's ZX-9, except that in Dragon, each capstan is individually driven by its own motor. Furthermore, its direct-drive capstans must be used in two directions, whereas all previous Nakamichis have been unidirectional. The use of different capstan diameters (and, therefore, rotation speeds), to prevent cumulative resonances in the two drive systems, complicates the issue. Nakamichi solves the problem with a quartz-lock system that keeps the takeup capstan running 0.2% faster than the supply capstan (for optimum tape tension and head contact) no matter which direction is in use.
There are three heads: erase, recording (for the forward direction only), and playback (a four-track design, with a bifurcated right-channel gap for each direction). Superficially, this is not too different from the head arrangement in some other decks we have tested with bidirectional playback; actually, the automatic azimuth adjustment, added to the inherently flat, wide-band response that is a Nakamichi hallmark, makes the Dragon the first automatic-reverse model we have tested for which no performance apology need be made. Normally, some tradeoff is required to get auto-reverse convenience-though, in fairness, most reversing decks we've tested sell for less than half of this one's price.
The Dragon, however, is simply a superb machine. With Nakamichi's tapes (SX Type 2 ferricobalt, ZX Type 4 metal, and EX-II Type 1 ferric), response is extremely flat and extended, noise is very low, headroom is very high, and flutter is the lowest Diversified Science Laboratories has ever measured. Erasure (thanks to Nakamichi's dual-gap erase head, no doubt) is also exceptional: almost 70 dB with metal tape and more than 80 dB with the ferricobalt. Where the lab measured behavior in the reverse direction, the figures invariably were in the same ballpark as those shown here, which represent the forward direction. Such consistency of behavior, which Nakamichi says was a design goal, is frequently lacking in bidirectional decks.
The metering is calibrated from -40 to +10 dB with respect to the conservatively placed zero indication, 4 dB below DIN 0 dB, with steps of 1 to 2 dB between the 0-dB and +7-dB marks. When you use the meters to "fine-tune" the deck for the tape you're using, the 0-dB indication becomes the calibration point and the scale "stretches" to show finer gradations than it does with input signals. The 15-kHz bias-adjustment test tone (which would create compression at 0 dB) is recorded at -20 dB, with the metering sensitivity increased accordingly, so that the calibration point remains the same.
There are six small knobs for adjusting sensitivity (Dolby tracking): one for the left channel and one for the right for each of the three tape types. Six more identical knobs, similarly disposed, provide for bias calibration. Adjustments are made by selecting the appropriate test oscillator (for sensitivity or bias) and recording its tone with the noise reduction off, monitoring from the tape. The manual (excellent by ordinary standards, though perhaps a little shy of the "superb" rating we usually give Nakamichi's efforts) suggests checking sensitivity first, then bias. (Pushing one selector button cancels the other.) We'd also urge that you go back and forth between the two a couple of times, particularly if your new tape is significantly different from that for which the deck was set previously, since each adjustment can influence the other. In any event, the process is quite simple.
There are no reference settings to fall back on when in doubt, though the uncalibrated midpoints of the knobs' rotations will give you at least an approximation of the settings for the Nakamichi tapes. Standardizing on a particular formulation for each tape type (always a good idea so you know exactly what to expect from your tape/deck combination) should help to minimize the need for tweaking-or the degree of tweaking needed-before you begin to record. But since the optimization scheme enables the Dragon to accommodate a wide variety of formulations, you need feel no compunction about trying any good tape of the recommended types (1,2, or 4).
Playback equalization (and a reciprocal part of the recording equalization) is controlled by a separate switch that is not influenced by the tape-type selectors, so you can also use Type 3 ferrichromes. (Set the tape selector to "EX" for Type 1 bias and the EQ to 70 microseconds.) This totally manual bias and EQ selection also forestalls the difficulties that can occur when an oddball cassette is inserted into a deck that uses the standard shell keyways to determine the tape type-and hence the bias and EQ.
The flexibility is welcome, but the EQ switch that makes it possible is less so. It is one of fourteen small, black, short-throw pushbuttons on the front panel, the seven at the right end being particularly important. Most decks use equally small pushbuttons, some in even greater profusion; but most such decks have illuminating indicators (or buttons) to show which functions are engaged, or they at least contrive to make the "in" and "out" positions more readily distinguishable. As a result, the Dragon requires somewhat more care than usual in routinizing your recording practices. Even once we became reasonably familiar with the deck, we found it easy to miss incorrect settings of the EQ or noise reduction buttons until a recording was already in progress.
That the Dragon is worth a little extra care in use-as well as its hefty price tag- is beyond question: Nakamichi has again set a pace that others will find hard to match. The automatic azimuth adjustment is an important advance in cassette technology for recordists who want the best the medium has to offer. As usual, there is back-panel provision for a remote control and a DC power-supply connection for Nakamichi's BlackBox accessories. And there are the usual conveniences of an autoreverse deck (unidirectional/bidirectional/ repeat operation, timer modes, and memory modes), plus a little bit of extra smarts: After forty seconds of silent tape, the deck will automatically fast forward to the end of the tape, reverse, fast wind till it finds music again, back up slightly to the start of the selection, and then return to play mode at just the right point.
Going further beyond the conventional, Nakamichi has also included an excellent cueing mode that provides reduced output at two relatively slow speeds of fast wind in either direction, so you can find your place on the tape by ear, and an automatic recording pause, which activates if the deck has been recording for thirty seconds with no input signal. But nicest of all the "extras," perhaps, is the automatic fader, which in six or two seconds (depending on your touch) will bring the recording signal up to the levels you have preset (on the master, left, and right rotary controls) or fade out to silence at the end of your recording.
All considered, the Dragon's features are very well chosen and integrated for the user who wants top quality together with convenience where it counts without a lot of fussy little afterthoughts. Moreover, this is achieved without compromising the deck's basic performance, which is outstanding by any standard, even without benefit of its unique NAAC azimuth-correction system. But it is that final touch that gives the Dragon its fire and Nakamichi another feather in its cap.
