Hirsch-Houck Laboratory test results on: the Mitsubishi DA-P10 preamplifier and DA-A15 power amplifier, Stanton 8815 stereo phono cartridge, Fisher MT-6225 turntable, ESS amt lb speaker system, and Pioneer RT-707 open-reel tape deck.
Mitsubishi DA-P10 Preamplifier and DA-A15 Power Amplifier
Fo, anyone who has had difficulty choosing between an integrated amplifier and a separate preamplifier and power amplifier, it appears that it is now possible to have the best of both worlds. An ingenious mechanical and electrical design in Mitsubishi's deluxe line makes it possible either to join a preamplifier with one of two power amplifiers (100 or 150 watts per channel) to form a single integrated amplifier or to use them as "separates." The DA-P10 preamplifier is actually what Mitsubishi calls a "Dual Monaural" unit. It consists of two distinct mono preamplifiers in a single cabinet with separate power supplies but a common power transformer. Since it is, after all, intended as a stereo control unit, the DA-P10 has a master volume control (ATTENUATOR) and input selector, with positions for two magnetic phono cartridges and two high-level sources (TUNER and Aux). Other switches affecting both channels are the power switch, MODE (stereo/mono), and tape monitor.
Each channel has its own subsonic filter, with a 12-dB-per-octave slope below 18 Hz controlled by a lever switch. The bass and treble tone controls, separate for the two channels, are eleven-position types with the center marked DEFEAT (the tone-control circuits are completely bypassed in that position). Finally, each channel has its own level control, which is normally used to set signal balance between channels.
The principal reason for the thorough separation of the two channels is Mitsubishi's conviction that crosstalk between channels, even at very low levels, can degrade listening quality. By separating all possible signal circuits as well as the power supplies, they achieve a claimed separation of 80 to 100 dB, depending on the frequency.
The same philosophy carries over into the DA-A15 power amplifier, which we tested together with the DA-P10. The two power-amplifier channels are completely separate from power transformer to output terminals, sharing only the chassis. A similar degree of interchannel audio-signal isolation is claimed for the DA-A15.
The Mitsubishi DA-A15 is rated to deliver 150 watts per channel to 8-ohm loads from 20 to 20,000 Hz with less than 0.1 percent total harmonic or intermodulation distortion. A 1-volt input signal is required for full output (the sensitivity can be reduced by screwdriver adjustments next to the input jacks), and the nominal rated output of the DA-P10 preamplifier is also 1-volt (although it is capable of about 9 volts output before overload).
The speaker outputs of the power amplifier employ insulated binding posts that make a very positive contact. There are outputs for two pairs of speakers, only one of which can be driven at a time. A three-setting slide switch in the rear of the amplifier energizes either set of output terminals (through internal relays that also provide a turn-on time de lay and protect the loads against amplifier failure), and the third setting silences all speaker outputs.
Both the DA-A15 amplifier and the DA-P10 preamplifier are finished in semi-gloss black, with functional block diagrams and specifications printed on their top surfaces. No dust cover hides the major components of the power amplifier, the two power transformers and large power-supply capacitors occupying the center of the chassis and the heat-sink fins being mounted along the chassis sides.
Handles are supplied with the power amplifier for optional installation convenience. To join the two units into a single integrated amplifier, the preamplifier is slipped over the power-amplifier handles and fastened in place with heavy screws. The signal outputs of the preamplifier are connected to the inputs of the power amplifier, the power-amplifier line cord is plugged into one of the switched outlets on the preamplifier, and (if desired) the remote-control sockets for the speaker selector are linked with a supplied cable. When this is done, a SPEAKERS Switch on the panel of the preamplifier operates the out put relays in the DA-A15. The various cables are folded along the sides of the power amplifier into channels that are styled to match the heat-sink fins above them.
-------------- The DA-A15 power amplifier is shown at left. This unit
docks physically and connects electrically with the rear of the DA-P10 preamplifier
(right). The two are joined by bolting the handles of the DA-A15 to the frame
of the DA-P10.
The basically black amplifier units contrast with the satin gold panel of the preamplifier and its matching machined metal knobs. Externally, there is no indication that this is not a conventional integrated amplifier. Since the power-amplifier outputs are not routed through the preamplifier, the phone jack on the panel of the latter is driven from a separate amplifier stage in the DA-P10 that is designed to drive 8-ohm headphones. (It will not drive high-impedance dynamic phones or electrostatic phones to an adequate listening level.) The signal-input and output jacks are recessed into the right side of the preamplifier, and four a.c. outlets (two of them switched) are in its left side. The front dimensions of the Mitsubishi units are 16 3/4 inches wide by 6 3/4 inches high. The preamplifier is 8 inches deep and the power amplifier is 11 3/4 inches deep.
When they are joined, the combined unit is only 16 3/4 inches deep owing to overlap be tween the sections. The preamplifier weighs 13 pounds and the power amplifier weighs 40 pounds, so that the combination weighs a massive 53 pounds. Price: DA-P10 $290, DA A15 $590.
Laboratory Measurements. The two units were measured separately but were joined for use tests. The preamplifier tone controls had good characteristics, with a maximum range of ±12 dB. The bass control took effect be tween 100 and 400 Hz as it was varied. The treble characteristic was hinged at about 2,000 Hz. The response in the center (DEFEAT) position of the controls was ±0.5 dB from 20 to 20,000 Hz. The phono response, including the complete preamplifier circuitry, was equally flat, measuring ±0.5 dB from 20 to 20,000 Hz, and it did not change detectably when measured through the inductance of some typical phono cartridges. The subsonic filter reduced the output by 2.3 dB at 20 Hz and by 14 dB at 5 Hz.
To produce the rated 1-volt output, a high level input of 0.145 volt or a low-level input of 1.95 millivolts was needed. The phono-over load capability was an excellent 310 millivolts. The unweighted noise output of the DA-P10 was too low to measure through its high-level inputs, being less than the 100-microvolt minimum indication of our meter (which is 80 dB below 1-volt). Through the phono inputs, the noise was a barely measureable -79.2 dB, one of the lowest phono noise levels we have encountered in such a unit.
The preamplifier output clipped at 10.3 volts. At any usable signal level, the distortion was well under 0.01 percent except at 20 Hz, where it reached 0.05 percent at 3 volts output (at 1 volt, the distortion was typically between 0.002 and 0.006 percent).
The power amplifier was not fazed by the one-hour preconditioning period at one-third rated power. Its heat sinks have temperature sensors that change color from red to black at 140 degrees F (60 degrees C), but they did not change color during any of our tests. The out puts clipped at 178 watts per channel when driven at 1,000 Hz into 8-ohm loads. The 4-ohm and 16-ohm clipping levels were 267 and 112 watts.
The distortion of the DA-A15 was comparable to that of our test instruments (about 0.002 percent) at most power levels and frequencies. From 0.0022 percent at 1 watt it rose smoothly to 0.0036 percent at 100 watts and 0.0056 percent at 180 watts, just as clipping began. The IM distortion was between 0.004 and 0.008 percent up to 100 watts and reached 0.013 percent at 180 watts. Unlike just about every other amplifier we have tested, the distortion of the Mitsubishi DA-A15 did not increase even slightly at very low power levels. The IM analyzer reading in creased from 0.004 percent at 1 watt to 0.032 percent at 3 milliwatts, but this proved to be entirely random noise in the analyzer.
At its rated 150 watts output, the distortion of the DA-A15 was about 0.02 percent in the range of 20 to 30 Hz and 0.005 percent from 100 to 1,000 Hz, rising to 0.04 percent at 20,000 Hz. At lower power levels the distortion was slightly less. A 0.27-volt input drives the amplifier to a reference 10-watt output, and the unweighted noise level was 87 dB be low that power. The square-wave rise time of 0.7 microsecond was exactly as rated, and the measured slew rate of 20 volts per microsecond somewhat surpassed the rated 15 volts per microsecond.
We also measured the interchannel isolation of the DA-A15 power amplifier. This was possible only with the use of our H-P 3580A spectrum analyzer, since the crosstalk was-120 dB at 20 Hz,-110 dB at 1,000 Hz, and-80 dB at 20,000 Hz.
Comment. In its basic electrical performance, the Mitsubishi amplifier (we will treat the two units as a single integrated amplifier, since that is the way we used them) is as near "state-of-the-art" performance as could be desired. Although a few other fine amplifiers rival it or marginally surpass it in selected individual characteristics, we have never seen one with the overall excellence exhibited by this superb combination.
It should surprise no one, therefore, to learn that we found no special audible qualities in the sound as reproduced through these components, since every other link in the audio chain, from microphone to speaker, is far below their quality level.
Certain features omitted from the DA-P10 preamplifier, such as loudness compensation and a high-cut audio filter, will hardly be missed. On the other hand, a control unit of this caliber should have made provision for the handling of at least two tape decks. And with most headphones presently manufactured being in the 200-ohm impedance range (although rated for 8-ohm outputs), it is a pity that they cannot be driven to really high levels (because of clipping in the internal headphone amplifier stage of the DA-P10).
If cross-coupling between channels, either steady-state or transient, is as much of a problem as Mitsubishi and several other manufacturers believe it to be, then the Mitsubishi amplifier is the one least likely to suffer from this "fault." However, since it is highly un likely that the stereo separation in the pro gram material itself will ever exceed 30 dB, and that only over a limited portion of the spectrum, the 80 to 100 dB of interchannel isolation afforded by the Mitsubishi design would seem to fall into the category of over kill. But, in any case, the extra 50 dB or more of audio-signal separation between the two channels certainly does no harm.
Taking it strictly on its own demonstrable performance, the Mitsubishi amplifier is a superbly constructed, state-of-the-art amplifier with exceptional flexibility of installation (and as an integrated amplifier it is one of the most powerful on the market despite its relatively compact size). Our experience with the Mitsubishi amplifier convinces us that its appearance of elegance and superior electrical quality is completely consistent with its actual performance. It is a beautiful product.
Stanton 881S Stereo Phono Cartridge
THE new Stanton 881S is now this company's finest phono cartridge, inheriting that distinction from the 681 series of cartridges, which will continue to be represented by the 681EEE. The new 881S, though it resembles the 681 cartridges in external appearance, is totally different in its internal design details. The 681 cartridges use the moving-iron principle, while the 881S is a moving-magnet cartridge. The effective stylus mass is kept low by the use of a "rare-earth" magnetic material that is about ten times as powerful as other metallic magnet materials used in previous cartridge designs. The more powerful magnetic field has also made it possible to reduce the cartridge's coil inductance, there by minimizing its sensitivity to external hum fields and to load conditions presented by the preamplifier and connecting cables.
The stylus jewel of the 881S has Stanton's "Stereohedron" shape, a variant of the CD-4 stylus shape with a vertically elongated con tact area said to improve high-frequency tracking ability and reduce record wear. In combination with the rare-earth magnetic material, the Stereohedron stylus is credited with making possible the very low tip mass of 0.2 milligram.
Like most other Stanton cartridges, the 881S has a hinged brush as part of its stylus assembly; this rides on the record surface and removes surface dust. The brush is easily detached if desired. The cartridge is designed to track at forces from 0.75 to 1.25 grams. (The additional gram applied to compensate for the brush does not become part of the stylus tracking force.) Each cartridge is supplied with individual calibration data, a box for storing extra styli, and a small screwdriver, as well as mounting hardware. Price: $150.
Laboratory Measurements. To test the Stanton 881S, we installed it in the tone arm of a Fisher MT-6225 record player. The cartridge load was 47,000 ohms in parallel with 290 picofarads (the nominal rated load is 275 picofarads).
From the 3.54-cm/sec (centimeters per second), 1,000-Hz bands of the CBS STR 100 test record, the 881S produced an output of 4.5 millivolts, the levels from the two channels matching within 0.8 dB. The measured vertical angle of the stylus was 22 degrees. The extraordinary tracking ability of the 881S was demonstrated by the fact that it played the 30 cm/sec, 1,000-Hz tones of one disc at a mere 0.5-gram tracking force, and the very high level 32-Hz tones of another at 0.4 gram. The exceptionally demanding German Hi-Fi Institute record, with its progressively higher-amplitude 300-Hz tones, could be played to the 80-micron level at 0.5 gram and to its maximum level of 100 microns at 0.75 gram (very few other cartridges can do as well at any tracking force). For our other tests, we used the nominal rated force of 1 gram.
The frequency response, measured with the STR 100 record, was flat within ±1 dB over the 40- to 20,000-Hz range of the record. Reducing the load capacitance to 150 picofarads had a negligible effect, giving a slight rise in output above 15,000 Hz but the same overall variation. However, a very high capacitance of 520 picofarads (which is required by some other cartridges for flattest response) produced a slight peak at 10,000 Hz and a drop in output above that frequency.
The measured channel separation was different on the two channels (a property of the test record) but averaged 20 to 35 dB over the entire 40- to 20,000-Hz range of the record. In the Fisher tone arm, whose mass of 18 grams is typical of many recent arms, the cartridge compliance resonated at 8 Hz with an amplitude of 10 dB. Its flat frequency response made it possible for the 881S to reproduce an excellent 1,000-Hz square wave from the CBS STR 112 record with only a slight overshoot.
The intermodulation distortion, measured with the Shure TTR-IO2 record, increased smoothly from 2 percent at 7 cm/sec to 5 percent at 27 cm/sec. Even at the low 1-gram operating force, the cartridge never mistracked on this record, though its highest levels are of ten unplayable by other cartridges even at their maximum rated forces. High-frequency tracking was measured with the 10.8-kHz tone bursts of the Shure TTR-103 record. The repetition-rate distortion of the 881S was comparable to what we have observed on other fine cartridges.
Comment. The subjective tracking test with the Shure "Audio Obstacle Course-Era III " record confirmed our measurements of the tracking ability of the Stanton 881S. At its 3/4-gram minimum rated force, the cartridge played the highest levels of all selections on this record except for the bass drum (which required 1 gram for its maximum level). In view of the enormous tracking difficulties imposed by this record, we would judge that the 881S can be used with complete success at 44 gram and that its tracking ability will be taxed by few, if any, records. Of course, operation at 3/4 gram requires a good tone arm, and many otherwise satisfactory arms become marginal in performance below I gram. It is also a good idea to check the correctness of the arm's tracking force and other adjustments by external means, since a slight inaccuracy in the arm's calibration could easily carry it outside the cartridge's narrow range of recommended tracking forces.
Incidentally, we found that the integral brush, though easily removed, is not so easily replaced. Rather than risk damage to the stylus, we left the brush off for most of our testing and listening (it has no detectable effect on the performance of the cartridge).
In general, the sound of the Stanton 881S is completely neutral. It injects no coloration, emphasis, or de-emphasis into any part of the frequency spectrum, and it has a notable freedom from audible tracking distortions of any kind. Since it is so easy to forget that there is a cartridge in the reproducing chain, this is the kind of cartridge we prefer to use when listening to records for musical enjoyment, rather than as a means to uncover flaws in cartridge performance.
The 881S is undeniably expensive and, as with any premium-price product, many of its performance refinements are of a subtle nature (the law of diminishing returns takes over rapidly when cartridge prices exceed about $70 or $80). Nevertheless, the special qualities of the Stanton 881S are real, and anyone who teams this cartridge with other components of comparable quality will be able to appreciate them as much as we did.
------- In the graph at left, the upper curve represents the smoothed, averaged frequency response of the cartridge's right and left channels; the distance (calibrated in decibels) between it and the lower curve represents the separation between the two channels. The inset oscilloscope photo shows the cartridge's response to a recorded 1,000-Hz square wave (see text), which indicates resonances and overall frequency response. At right is the cartridge's response to the intermodulation-distortion (IM) and 10.8-kHz tone-burst test bands of the TTR-102 and TTR-103 test records. These high velocities provide a severe test of a phono cartridge's performance. The intermodulation-distortion (IM) readings for any given cartridge can vary widely, depending on the particular IM test record used. The actual distortion figure measured is not as important as the maximum velocity the cartridge is able to track before a sudden and radical increase in distortion takes place. There are very few commercial phonograph discs that embody musical audio signals with recorded velocities much higher than about 15 cm/sec.
+++++++++++++++++
Fisher MT-6225 Turntable
THE Fisher MT-6225 "Studio Standard" turntable is a direct-drive record player driven by a novel 120-pole, three-phase mo tor. The 120 permanent-magnet poles are in a flexible strip of magnetic material cemented to an 8 1/4-inch ring on the underside of the cast platter. The stator is a 90-degree arc of closely spaced pole pieces on top of the motorboard and immediately adjacent to the rotor strip when the platter is placed on its spindle. Fisher describes this as a "linear drive" motor. Its very large number of poles (most direct-drive motors use eight to twenty poles) are intended to minimize flutter and rumble from the torque impulses exerted on each pole as the platter turns.
The nonferrous platter (together with its heavy rubber mat) weighs 2 3/4 pounds. It operates at 33 1/3 or 45 rpm, with speeds selected by pushbuttons on the control panel to the right of the platter. The outer rim of the platter has four rings of stroboscope marks (for operation with 50- and 60-Hz current at both speeds) lit from below through a prism. There am separate vernier-adjustment knobs for the two speeds.
The Fisher MT-6225 is a semi-automatic turntable whose tone arm must be positioned manually (the motor turns on when the arm is lifted from its rest). At the end of a record, or whenever the reject button is pressed, the arm returns to its rest automatically and the motor shuts off.
The tonearm is an S-shaped tube on gimbal pivots. Its threaded counterweight has a tracking-force scale calibrated at 0.1-gram intervals from 0 to 3 grams. The cartridge shell, cut away to minimize its mass, uses the popular four-pin bayonet mount and has a well-designed finger lift. Next to the arm base is an antiskating dial, calibrated from 0 to 4 grams, and a cueing lift lever. The lift is undamped, but the descent is slow and damped.
The MT-6225 comes on a solid walnut-finish wooden base with softly sprung feet. It has a removable, hinged plastic dust cover.
The record player is 17 3/4 inches wide, 14 4/2 inches deep, and 5% inches high. It weighs 17 1/4 pounds. Suggested price: $200.
Laboratory Measurements. We used the Fisher MT-6225 with several different cartridges, but most performance tests were made with a Stanton 881S installed. The cartridge installation itself was relatively easy, but the adjustment for minimum tracking error requires that the stylus overhang the turn table center by 19/32 inch. However, since no reference mark or gauge was supplied, we do not see how that setting could be made with acceptable accuracy by the average user. We used an external stylus protractor to set the stylus overhang. The tracking error was then very low-less than 0.2 degree per inch over most of the record and 0.5 degree per inch at a 6-inch radius.
When the arm had been balanced, we found that the section of the counterweight that carries the tracking-force scale had little or no friction to retain its position relative to the rest of the weight. As a result, great care was necessary in adjusting the force to be sure that the two sections did not shift relative to each other. Fisher explains that some of the earliest units, of which our test sample is one, do have this fault, which has since been corrected. By our estimate, the calibration of the scale was off by no more than 10 or 15 percent when care was used in the setup. In any case, we would suggest that any force settings within 0.5 gram of the minimum recommended for a cartridge be made using an external gauge. At the 3/4-gram to 1-gram forces we used with the Stanton 881S, an error of 1/4 gram could have been serious in respect to adequacy of applied tracking force.
The unweighted turntable rumble was -35 dB, largely in the horizontal plane, and with ARLL weighting it was -59 dB. The rumble was mostly below 30 Hz, with no specific frequencies characteristic of the rumble visible on a spectrum analyzer. The wow and flutter were 0.07 and 0.04 percent, for a combined reading of 0.08 percent. Flutter components were found at 30, 50, and 70 Hz in addition to the major concentration at frequencies below about 10 Hz. The speed verniers had a range of +5.1 to -5.9 percent at 45 rpm and from +1.5 to-4 percent at 33 1/3 rpm. The speed changed only 0.2 percent as the a.c.-line volt age was shifted for test purposes from approximately 95 to 135 volts.
The tone arm, including the supplied low-capacitance signal cable, had a capacitance to ground of 86 picofarads per channel, which is compatible with CD-4 cartridges. The measured arm mass was 18.3 grams-fairly low for arms of this type. The antiskating calibration was quite good, since setting it to match the tracking force always gave acceptable antiskating correction. The cueing device lowered the pickup with only a slight outward drift that caused less than 2 seconds of the record to repeat.
The isolation against external base-conducted vibration was about average for direct-drive turntables. The major feedback sensitivity was at 90 Hz, with lesser transmission peaks at 35, 190, 260, and 450 Hz.
Comment. Our criticisms, as should have been evident from the preceding comments, apply mostly to the inadequate setup procedures of the MT-6225. Overall, the MT-6225 is easy to handle, attractive to the eye, and in respect to its measured specifications delivers the kind of overall performance we would expect from a direct-drive record player in its price class.
++++++++++++++++
ESS amt 1b Speaker System
SINCE its introduction a few years ago by ESS, the Heil "Air-Motion Transformer" (AMT) has undergone the inevitable improvement process that follows the appearance of any radically new device. The AMT is a tweeter of unconventional design with exceptional performance characteristics. Its "voice coil" is in the form of flat, conducting aluminum-foil strips bonded to a Teflon diaphragm. The diaphragm is pleated, accordion-like, and operates in a powerful magnetic field. The audio-signal current passing through the conductors causes the alternate sets of pleats to either draw closer together or move apart, thus "squeezing out" the air be tween them and generating a sound-pressure wave. In its most recent version (as used in the amt lb system), the AMT diaphragm is bonded to a gauzelike nylon material which serves the dual purpose of preventing adjacent "turns" of the aluminum strips from shorting out under extreme drive conditions and of smoothing the frequency response in the upper-middle and high-frequency ranges.
Simultaneously with the tweeter improvements, the woofer section of the amt lb has also undergone a complete redesign. It now features a 12-inch cone made of Bextrene, a rubberized polystyrene material also used by a few British manufacturers for their woofer cones. Bextrene is said to provide superior damping of undesired cone-vibration modes, thus smoothing out the woofer response.
(ESS manufactures all its own drivers, including the new woofer and the Heil AMT.) In the amt lb the woofer output is augmented at very low frequencies by a passive radiator of the same diameter located in the rear of the enclosure. The crossover network departs from current practice in having 18-dB-per-octave slopes at its crossover frequency of 1,000 Hz.
The amt lb, like its predecessors, is in the shape of a truncated pyramid containing the forward-facing woofer and the rearward-facing passive cone. The Heil driver is mounted on the top of the woofer enclosure together with two frequency-balance adjustments.
(The "presence" control varies the output at all frequencies above 1,000 Hz with a shelved characteristic. The "brilliance" control affects only frequencies above 5,000 Hz, with slope changes of ±3 dB per octave above that frequency.) The upper portion of the cabinet contains the bass drivers and Heil tweeter, both concealed by a removable cloth-covered frame.
The ESS amt lb has a nominal 6-ohm impedance and is rated to handle up to 375 watts of clean program material (the "clean" qualifier is significant, since relatively low power levels of clipped program material can damage any speaker). The dimensions of the amt lb are 16 1/4 inches square (maximum) by 351/4 inches high, and it weighs about 75 pounds. Price: $450. A bookshelf version of the amt lb, with essentially the same specifications, is also available. Price: $398.
Laboratory Measurements. The ESS amt lb was tested with its level controls set to the centers of their indicated "normal" ranges.
The smoothed frequency response in the reverberant field of our test room (10 to 15 feet from the speakers) was joined to a curve de rived from closely miked measurements of the woofer and the passive cone to form a single composite frequency-response curve. Additional measurements were made to verify the effects of the level controls.
The bass distortion of the driven cone was measured with nominal inputs of 1 watt and 10 watts (into 8 ohms), and the output sound-pressure level was measured at a 1-meter distance with 1 watt of random noise in the octave centered at 1,000 Hz. The speaker's impedance and tone-burst response were measured over the full audio range.
The combined frequency response was flat through the mid-range, with a slight bass rise at 7th Hz and with the output strongly sustained down to 20 Hz. The output of the passive radiator was dominant below 50 Hz. The low-frequency response would, of course, be affected appreciably by the room characteristics and the placement of the speaker within the room. With our preferred settings, the AMT high-frequency driver response rose smoothly above 5,000 Hz to a maximum of +9 dB in the vicinity of 13,000 Hz. The response could have been considerably flattened with the speaker's controls, but we found that the best subjective balance was obtained with the "normal" control settings.
The overall response in our live-room measurements was approximately ±5 dB from 30 to 15,000 Hz.
The ESS amt lb was moderately efficient, delivering a sound-pressure level of 89.5 dB at I meter with a 1-watt input. The bass distortion at that level was very low, less than 1 percent down to about 40 Hz and only 7 percent at 25 Hz. When the power level was raised to 10 watts, the distortion increased only moderately above 40 Hz, to between 1.5 and 2.5 percent; it rose much more rapidly at lower frequencies, to 14 percent at 30 Hz. The system impedance was just under 5 ohms (which we take as the rated impedance) at its minimum point of 120 Hz, and about 16 ohms at the maximum, in the vicinity of 1,000 Hz. The tone-burst response was excellent at all frequencies, especially in the vicinity of the crossover, where many speakers exhibit severe ringing or other anomalies.
------- The excellent tone-burst response of the ESS amt lb speaker
system is shown (left to right) at 100, 1,000, and 6,000 Hz. The input signal
appears above the speaker's output in each case.
Comment. Recalling the difficulties experienced by some people (though not by us) with the original ESS amt 1, whose Heil driver could sometimes be blown out by overzealous application of a super-power amplifier, we deliberately drove the amt lb, with program material having strong high-frequency content, to levels of at least 400 watts per channel (as monitored on peak-level indicators). The sound was stupendous, as might be imagined--at times it was quite overpowering in our listening room--but we never heard distortion or strain and, perhaps needless to say, the speakers suffered no damage.
Just as the "proof of the pudding is in the eating," there is no substitute for listening when evaluating a speaker. The simulated "live-vs.-recorded" test, which covers the frequency range above 200 Hz, confirmed the superior accuracy of the amt lb at those frequencies. The duplication of the 'live" sound source, on a side-by-side basis, was essentially perfect.
Listening to a wider variety of program material, we noted two characteristics of the amt lb, both of which were consistent with our measured response curves. On some pro gram material the bass seemed at times slightly heavy, though never muddy or overbearing. The top end was airy, open, and crisp. At first hearing, the speaker sometimes seemed slightly bright (but never strident). However, after listening to the amt lb for a time, almost any other speaker we switched to seemed to be slightly closed-in and dull. Since we did have a reasonably objective reference standard in our recorded "live" music source, we knew that our reactions were not totally subjective and that the amt lb was really a highly accurate reproducer.
Probably more significant than any of these reactions is that during the several months we had the amt lb for testing we found ourselves almost invariably choosing it to listen to in preference to some other very fine speakers we had on hand, probably because of its utter lack of harshness, coloration, or other un pleasant distortions. Without attempting to distribute the credit between the new Heil driver elements, the Bextrene cone of the woofer, and the l8-dB-per-octave crossover network, we can say that the amt lb is obviously an excellent performer judged both by ear and by test, and one of the speakers we have most enjoyed using during the time we had it for testing.
Pioneer RT-707 Open-reel Tape Deck
MOST of today's open-reel tape recorders are as notable for their large size, weight, and price as they are for their generally outstanding performance. This might lead one to conclude that there is little market potential for more modestly endowed open-reel machines costing from $500 to $1,000, since that is a price range where the buyer also can choose among a great many deluxe cassette and Elcaset decks.
Evidently Pioneer feels otherwise, judging by the appearance of the Model RT-707 recorder in their new product line. The RT-707 is an open-reel, four-track stereo machine operating at 3 3/4 and 7 1/2 ips. It has the usual three-head format, plus a second playback head for use in the reverse direction of tape movement. It is equipped for automatic tape reversal, initiated by a piece of conducting-foil tape attached to the coated side of the magnetic tape, and it also can be reversed at any time by touching a button on the panel. It records only in the normal forward direction.
The Pioneer RT-707 has a three-motor tape transport with six-pole induction motors for each of the tape hubs and a direct-drive a.c. servomotor for the capstan. The use of a direct-drive capstan motor eliminates the belts and pulleys required to couple a high-speed motor to a slowly turning capstan, and with them go the periodic maintenance procedures and potential failures associated with such mechanical systems.
Most of the panel space of the RT-707 is devoted to the two 7-inch tape reels (a metal take-up reel is supplied with the recorder).
Between them are two large illuminated level meters with vertically oriented scales. Be tween the meters, red and green LED's indicate when the machine is in the RECORD or PAUSE mode.
Above the meters are seven pushbutton switches. Two are REC MODE selectors for the two channels which must be engaged in order to make a recording (they thus serve as a safety device to prevent accidental erasure of a recorded tape). Since it is possible to record on one channel while playing the other, special-effects recordings such as sound-with-sound can be made with external jumper connections between the recorder's input and output jacks.
There are separate recording Bias and equalization (EQ) buttons. Each has STD (standard) and LH (low-noise/high-output) positions; a table in the instruction manual lists recommended settings for most popular tapes. The MONITOR button channels either the SOURCE signal or the TAPE playback pro gram to the line outputs. The remaining but tons are the SPEED selector and the POWER switch.
Below the meters are the four-digit index counter, its reset button, a REPEAT button, and a PITCH control knob. The REPEAT function allows a tape, or any portion of it, to be repeated indefinitely by switching from for ward to reverse playback when the metal foil is contacted and from reverse to forward when the index counter has returned to its "0000" setting. The PITCH knob is a speed vernier, operating only during playback, with a nominal ±6 percent range. It is detented at its center, which establishes the correct playing speeds.
The head assembly is flanked by two rubber rollers and tension arms. When the tension arms are raised to their uppermost (latching) positions, the tape can be loaded in a straight line across the heads. The single capstan is to the right of the heads and nearest the take-up reel in the forward direction of tape motion.
Screwdriver access holes in the head cover permit easy azimuth adjustment of the re cording and playback heads if required.
At the lower left corner of the panel are the microphone jacks and the headphone jacks plus separate recording-level controls for the microphone and line sources, which can be mixed. Each control is a concentric pair coupled by a slip clutch for individual adjustment of channel levels.
The tape-transport controls are grouped at the lower right of the panel. They are mechanical pushbuttons that actuate electrical solenoids. Although there is no remote-control facility in the RT-707, the controls can be preset so that unattended recording or play back can be initiated with an external timer switch in the a.c. power circuit. There are fast-speed buttons for both directions, a STOP button, and the PLAY and REC buttons that must be engaged simultaneously to make a re cording. (By pressing them both while playing a tape, it is also possible to make a "flying start" recording.) Next to the PAUSE button are small playback-direction selectors on which illuminated arrows show the direction of tape travel.
The line inputs and outputs are in the rear of the recorder, with separate playback-level controls for each channel. These are detented at their mid-points. There is also a single un switched a.c. outlet.
The Pioneer RT-707 has a distinctive size and shape, considerably more compact than the typical open-reel recorder. Its satin-finish aluminum panel is 19 by 9 inches and is slot ted for mounting in a standard EIA equipment rack. The recorder is 14 inches deep and weighs about 43 1/2 pounds. Price: $575 (the RT-701, identical except that it lacks the reverse-play feature, is $525).
--------- FREQUENCY IN HZ (CYCLES PER SECOND)
Laboratory Measurements. Although the instruction manual lists Scotch 206 as the tape used for deriving the recorder's performance specifications, our test sample had been adjusted for TDK Audua tape. Both of these, plus several other comparable tapes, were used in our tests. The differences between them were slight, with Memorex Quantum giving the widest frequency response and Scotch 206 the best signal-to-noise (S/N) measurements. Since the TDK tape fell be tween these two in all respects and none of the differences were audibly significant, the following test data are based on the use of TDK Audua with LH bias and equalization settings. (Maxell gave essentially the same fine results.) The playback frequency response, measured with Ampex test tapes, was within ±1.5 dB from 50 to 7,500 Hz at 3 3/4 ips (the limits of the tape). At 7 1/2 ips, it was slightly different in the two directions of tape movement. In the forward direction, the response was within ±0.5 dB from 50 to 5,000 Hz, rising to +2.5 dB at 15,000 Hz. In reverse, the response was ±0.5 dB from 50 to 15,000 Hz.
A line input of 35 millivolts (mV) or a microphone input of 0.18 mV produced a 0-dB recording level at maximum gain settings. The resulting playback level was 580 mV with the controls centered and 800 mV with the maximum playback level setting.
At 3 3/4 ips and a-20-dB recording level, the record-playback frequency response was down 4 dB at 30 and 14,000 Hz. At a 0-dB recording level, the high-frequency response was down 4 dB at 10,000 Hz due to tape saturation. At 7 1/2 ips the response was ±2 dB from 20 to 24,500 Hz at a-20-dB level and from 20 to 18,000 Hz at a 0-dB level.
The playback distortion from a 0-dB re corded signal at 1,000 Hz was 0.23 percent, and the level had to be increased to +12 dB (far off the meter scales) before the 3 percent distortion-point reference was reached. The S/N referred to that level at 7 1/2 ips was 63.3 dB in an unweighted rms measurement, 68.5 dB with IEC "A" weighting, and 65 dB with CCIR weighting. The S/N at 3 3/4 ips was about 3 dB worse, and the noise level increased by 8 dB through the microphone inputs at maximum gain. At more normal gain settings there was little added noise.
The wow was less than 0.01 percent under all conditions. Unweighted rms flutter was 0.08 percent at 3 3/4 ips and 0.065 percent at 7 1/2 ips in a combined record-playback measurement. With the Ampex flutter test tapes, we measured 0.09 percent flutter at both speeds in the forward direction. In reverse play, which places the capstan between the supply reel and the heads, the flutter was 0.17 percent at 3 3/4 ips and 0.12 percent at 7 1/2 ips.
The playback speed was exactly the same as the recording speed with the PITCH control set to its detented position. The playback speed could be varied over a +9.2 to-7.6 percent range. In the fast speeds, 1,800 feet of tape were moved forward in 89 seconds and rewound in 102 seconds. The 0-dB level of the meters corresponded to a 180-nW/m flux level, and the meters responded a little slower than a standard VU meter. Tone bursts of 0.3-second duration indicated about 10 percent less than their steady-state levels. The headphone level is fixed and is adequate for general listening via 200-ohm phones.
Comment. The Pioneer RT-707 sounded every bit as good as its excellent measurements suggest, and it was also an easy-to-use, smooth-handling machine. At 7 1/2 ips it appeared to be essentially the equivalent of many other high-quality (and far more expensive) open-reel tape recorders designed for home use. And as with most of them, its performance at 3 3/4 ips is more comparable to that of a medium-price cassette deck, so that its advantages over a cassette machine at the lower speed are principally those of tape-editing convenience rather than basic recording quality. Nonetheless, it is interesting that this combination of versatility and performance is now available at a price hardly more than that of a good cassette recorder and actually far less than the cost of one of the new top-of-the-line deluxe three-head cassette or Elcaset decks.
In using the RT-707 with a number of tape formulations, we discovered that the recommended settings of the BIAS and EQ switches were not always optimum. The Pioneer manual suggests that if one is dissatisfied with the sound, other settings should be tried. We found that recording interstation FM tuner hiss at about a-10-dB level and comparing the incoming and playback signals with the MONITOR switch was the best way to establish the optimum tape bias and equalization. This should be done at 7 1/2 ips, since there will al ways be a distinct dulling of the highs in such a comparison at 3 3/4 ips. At 7 1/2 ips, the RT-707 is capable of virtually flawless reproduction of a random-noise signal-which is about as tough a test as can be made.
Since there is little difference in price or size among many good regular cassette decks, the lower-price Elcaset decks, and the RT-707, the tape hobbyist is now free to make a choice of format solely on the basis of performance or convenience. Certainly one can no longer generally characterize open-reel recorders as large and expensive and cassette recorders as compact and inexpensive. We suspect that "low-profile" open-reel decks such as this one will become more popular in the future.
---
TECHNICAL TALK; JULIAN D. HIRSCH