Manufacturer's Specifications:
Frequency Response: 5 Hz to 20' kHz.
Dynamic Range: Greater than 90 dB.
Channel Separation: 90 dB at 1 kHz
Harmonic Distortion: 0.003% at 1 kHz.
Output Levels: Line, 2.0 V; phones, 7.8 mW at 8 ohms.
Number of Programmable Selections: 16.
Power Consumption: 35 watts.
Dimensions: 13-3/16 in. (33.5 cm) W x 3 1/2 in. (8.8 cm) H x 10 1/8 in. (27 cm) D.
Weight: 13.2 lbs. (6 kg).
Price: $599.00.
Company Address: Sears, Roebuck & Co., Sears Tower, Chicago, Ill. 60684, USA.
This Sears Compact Disc player had, when I tested it, the dubious distinction of being the lowest priced CD player to be marketed in the United States. Indeed, Sears even reduced the already low price by $100.00 for a month in their pre-Christmas catalog. I use the word "dubious" simply because my tests reveal that attempting to come in with a lower priced product than anyone else's doesn't always result in instant success for the product; neither does it necessarily enhance the image of the manufacturer. I fully expected the Sears unit to be missing certain sophisticated programming features, and indeed I was correct in that regard. I did not, however, expect the tracking capability or error-correction capability of the machine to be as poor as they turned out to be, but more about that presently.
The Sears CD player is a front-loading unit which has the now-familiar slide-out drawer for disc loading. It features random programming for up to 16 selections (tracks only, no Index recognition), track-by-track forward and reverse searching, and repeat-play of either the entire disc or of any specific tracks that have been programmed into memory.
Numbered LED indicators show the track being played as well as tracks that have been memorized during programming. If there are more than 16 selections on a disc, the last LED remains illuminated as play progresses to higher numbered tracks. The player does not incorporate any sort of time display.
The Sears unit is extremely compact, standing only 3 1/2 inches tall and measuring less than 13 1/4 inches wide. It appears to be extremely shallow from front to back as well, but this is in part illusory, since the power transformer projects out the back for well over 2 additional inches. Both the front panel and the metal wrap are light silver colored, and nomenclature is highly legible.
Fig. 1-Frequency response, left (top) and right channels, at 0-dB level.
Fig. 2-THD vs. frequency at levels of (from top to bottom)- 30,- 24, and
0 dB, from 41 Hz to 16 kHz (see text).
Fig. 3-Spectrum analysis, 0 Hz to 50 kHz, showing 19-kHz tone (tall spike)
and spurious beat tone at 25.1 kHz (small spike).
Control Layout
The power on/off pushbutton and stereo phone jack are located to the left of the disc-tray slot. That slot is normally dust-protected by a metal flap which swings out of the way when the tray comes forward to accept a disc. Tray motion is initiated by touching an "Open/Close" button to the right of the disc compartment. "Repeat" and "Play" buttons are located just below the "Open/Close" button. "Pause," "Stop" and "Search" touch pads are to the right of the "Play" button, along the lower edge of the player. The "Search" button takes the form of a rocker pad; touching one side of it moves the pickup forward (to the next track of a disc), while touching its other end moves the laser pickup backwards, to the start of the current selection or to previous tracks, if touched repeatedly.
The upper right corner is devoted to the LED track display (16 numbered LEDs) and four separate status indicator lights. These are labeled "Disc In," "Repeat," "Play" and "Pause." Three more touch buttons, below the bank of track LEDs, are used for programming operations. These are labeled "Program," "Memory" and "Clear." The rear panel is equipped with the usual pair of output jacks, the aforementioned protruding power transformer, some metal heat-sink fins, and a rather mysterious switch called a "Pickup Fixing Switch" which must be set to the off position in addition to removing a shipping screw before you can use the player. No doubt the "Pickup Fixing Switch" activates some sort of electromechanical device which locks the laser pickup in place to avoid damage during transport.
Measurements
The player exhibited a roll-off of approximately 1 dB at 20 kHz for both channels, as seen in Fig. 1. (Note that vertical sensitivity in this plot is 2 dB per division.) The cursor was deliberately set at 18.5 kHz, rather than at 20 kHz, so that it would not be obscured by the solid vertical 20-kHz line on the graph.
Harmonic distortion at mid-frequencies, for maximum (0") recorded level, measured 0.004%, rising slightly to 0.0055% at 41 Hz (the lowest test frequency on my Philips test disc). At the high end of the spectrum, I ran into a situation which has come up with a couple of CD players previously tested. A high-level "beat" frequency showed up, causing the distortion meter to read 0.5% or so. Since this "beat" component is not, strictly speaking, harmonic distortion, I have stopped the plot of distortion versus frequency (Fig. 2) at 16 kHz. Figure 3 is a spectrum analyzer sweep from 0 Hz to 50 kHz (linearly swept, with each horizontal division equal to 5 kHz). The desired signal was at 19 kHz (the tall spike near center-screen), while the supersonic beat is seen around 6 kHz above the desired output. If you take the difference between 44.1 kHz (the sampling frequency) and 19 kHz, sure enough you come up with a difference of 25.1 kHz-the frequency of the beat observed in Fig. 3. Needless to say, this beat is not audible, but it is present on some CD players and not on others.
Output linearity was accurate to within 0.2 dB from 0-dB record level down to- 60 dB. Stereo separation is plotted for the left and right channels in Fig. 4. I measured between 82.5 and 85 dB of separation at mid-frequencies. Separation decreased very slightly to between 77 and 78 dB at the high end of the spectrum.
SMPTE-IM distortion measured 0.007% at 0-dB record level, increasing to 0.03% at a-20 dB output level of the test disc. Signal-to-noise analysis, without and with A-weighting, is shown in the plots of Fig. 5. Unweighted S/N measured 95.4 dB, while weighted S/N was exactly 100 dB below reference (0 dB) output level.
Reproduction of a 1-kHz, digitally generated square-wave signal, shown in the 'scope photo of Fig. 6, is typical of the shape encountered with those CD players using multi-pole, steep-analog filters. The same also applies to the digitally generated unit-pulse test signal shown in the oscilloscope photo of Fig. 7.
I have received at least three letters from readers of Audio in recent weeks, all of them pointing out an error which I have been making in some of my reports on CD players.
The readers point out that my analyses of phase linearity, using the test signals available on my Philips test disc, have been wrong. As one of these readers points out, an 11-pole filter, such as that used in many CD players, will introduce a phase shift at 20 kHz which is well in excess of 360° of error.
Thus, my reporting of angular errors amounting to 25° or 30° or 60° in the past should more than likely have been 385° (360° + 25°), 390°, or 420°. In truth, once the error exceeds 360°, it is impossible to accurately pinpoint the angular error using the particular pair of signals provided on the Philips test disc. With those CD players using digital filters ahead of D/A conversion, plus gentle one- or two-pole filters with higher frequency cutoffs, the test signals remain valid, and I shall continue to use them and to show the resulting 'scope photo. In the case of this Sears player, however, I have elected to use another pair of test signals available on the same test disc, 200 Hz and 2 kHz (see Fig. 8).
Fig. 4-Separation vs. frequency.
Fig. 5-S/N analysis, both unweighted (A) and A-weighted (B).
At least at these lower frequencies, we can see that both signals have their positive-going crossings of the zero axis at the same point, indicating little or no phase error. What happens at 20 kHz with respect to actual phase error is anyone's guess-at least until someone comes up with a test signal that is valid for accurately determining such phase errors. I thank the readers who wrote for setting me straight on this point, though in all honesty, I doubt whether it would have made much difference in terms of my subjective evaluations of the various CD players in question.
Up to this point, I had been mentally equating the Sears CD player's limited programming features and random access features with its very low retail price. I was certainly willing to accept the former in return for the latter-until I used my special tracking and error-correction test disc.
Readers of earlier reports will recall that this disc contains a radial wedge of increasing width (to simulate opaque scratches on the surface of a disc), several black dots of increasing diameter (to simulate dust particles on the surface of a disc), and a simulated fingerprint smudge. Most of the CD players I have tested were able to play right through wedge thicknesses of 800 microns and through dust specks having maximum diameters of 600 microns. A very few machines were even able to play right through all of the defects (up to 900 microns on the wedge and 800 microns for the black dust dots). The Sears CD player was unable to track the wedge when it was only 500-microns wide, and it failed to play through a dust speck 500 microns in diameter. This is the poorest error-correction or tracking capability I have encountered in any CD player tested to date. What it means is that, unless your CDs are absolutely perfect, free of even minute scratches and free of any manufacturing defects or "bubbles" in the surface of the plastic, you may well experience tracking difficulties. To make matters worse, susceptibility to mistracking caused by mechanical shocks applied to the body of the player was not very good either.
Fig. 6-Response to 1-kHz square wave.
Fig. 7-Single-pulse reproduction.
Fig. 8-Two-tone phase check (200 Hz left-channel, 2 kHz right-channel: see
text).
Use and Listening Tests
Several Compact Discs in my collection already have minor surface scratches. These scratches don't seem to affect playback on my reference CD player, but they immediately presented difficulties to the Sears unit-difficulties that resulted in audible mistracking, muting, and skipping.
Since the unit performed perfectly when playing my test discs and when reproducing music from discs known to be free of any minor scratches, I cannot conclude that I was given an atypical sample to test. On the other hand, neither can I conclude that other units of the same model will do as poorly as this one when tracking Compact Discs, since some production variations in tracking ability must surely exist from unit to unit.
As matters stand, it really doesn't matter how good the reproduced sound was when I played mint-condition CDs.
The fact is that CDs are not indestructible and, with repeated playing and handling, are likely to become scratched, however superficially. The standards set for CDs by Philips and Sony allow for several sophisticated levels of error correction to take care of such minor defects in discs.
Whoever made this unit for Sears, whether by design (in an effort to keep costs down) or for other reasons, has either failed to take full advantage of the error-correction capabilities inherent in the CD format or has come up with an inferior servo-tracking system. Whatever the reason, if I were looking for the right CD player to buy, I would want to think carefully (and, preferably, check several more Sears units) before I opted for this model.
-Leonard Feldman
Also see: Sanyo DAD 8 CD player (April 1984)
(Source: Audio magazine, April 1984)
= = = =