Tested This Month:
Burwen THE 7000 Noise Eliminator; Shure V15 Type IV Cartridge; Dynaco Stereo 416 Power Amp; Wharfedale E-70 Speaker System; Teac A-103 Stereo Cassette Deck
Burwen TNE 7000 Transient Noise Eliminator
SINCE the name Richard Burwen has been closely associated with audio-signal processing for many years, we were not too surprised a while ago to learn that there is now a Burwen Transient Noise Eliminator.
The Model TNE 7000 from Burwen Research (now a part of KLH) is what is popularly known as a "tick and pop" suppressor, a device intended to remove those annoying transient disturbances from phonograph-record reproduction.
There have been a number of attempts to do this in recent years, and they have met with varying degrees of success. As with any noise-reducing device, the real problems are to distinguish the noise from the program and to remove it unobtrusively, so that the listener is never aware of the sonic surgery that has been performed.
Burwen's TNE 7000 has been designed to detect the differences between transient noises and high-frequency program transients, which is no mean feat since they share a number of similar properties and have broadly overlapping frequency spectra. But there are differences (which we will describe shortly) that make the distinction possible.
Once the unwanted noise transient has been isolated, it must be removed from the pro gram with minimum effect on the desired signal. Normally this is done by somehow blanking the sound for the duration of the noise impulse. ft is then necessary to fill in the "hole" in the sound waveform with something that sounds as though it belongs there. It is possible to do this only because the turn-off interval is very short.
Time delay plays an important role in any transient suppressor. A sensing circuit re quires some period, however small, to determine that a transient noise is present-by which time it is already past! Therefore, the program is passed through a separate channel in which it is delayed for a time equal to the decision time of the sensing system, which then has plenty of opportunity to blank out the offending portion. The preceding comments apply to every successful transient eliminator, although each of the several on the market uses a different circuit approach and makes different compromises (signal processing almost always involves some kind of compromise or trade-off).
In the case of the Burwen TNE 7000, it was decided that the noises to be removed would be those resulting from the marks and scratches that occur on a record in ordinary use, rather than gross defects. Burwen found that musical transients have a typical attack time of about 2 milliseconds, after which they gradually "decay" over a period ranging from a tenth of a second to several seconds. Noise transients, in contrast, are much faster and shorter, with an attack time of from 50 to 200 microseconds and a duration of no more than 2 milliseconds (2,000 microseconds). Because of its rapid rise, a noise transient has considerable energy content in the ultrasonic region (20,000 to 50,000 Hz), where there is little or no musical content.
---------------- The TNE 7000 should be connected in the tape-monitor loop, with the tape machine plugged into it. This permits processing both during playback and before recording.
Working with these facts, Burwen developed an extremely fast "switch" to turn off the program channel. It is controlled by a sensor that monitors the program energy from 30,000 Hz up (and cuts off sharply below that frequency to prevent its being triggered by high-frequency program harmonics). The sensor also monitors the difference between the two signal channels, since a physical scratch will not normally damage a groove in such a way as to produce equal-amplitude, in-phase outputs from both channels. The muting circuit turns off the sound for the duration of the transient, which is usually in the range of 80 to 600 microseconds. Because of its operating speed, the TNE 7000 is able to use a signal de lay of only 40 microseconds, which Burwen claims is less than that of any other currently manufactured transient-noise suppressor.
The Burwen instruction pamphlet, in an effort to put the matter simply, states that a smoothly varying signal is substituted for the regular program during the "off" time to make the "hole" inaudible. Burwen also says that the TNE 7000 filter uses matched and trimmed components to achieve a precise, flat frequency response throughout the audio range. The TNE 7000 has two control knobs that are used to set the SENSITIVITY of the noise-sensing circuits (in accordance with the wide-band noise level of the program) and the THRESHOLD level at which suppression be gins. Each control has a LED indicator to monitor its operation. The TNE 7000 is meant to be placed in the tape-monitoring loop of an amplifier or receiver; tape-recorder input and output jacks are duplicated in its rear. A front-panel pushbutton switch connects the tape recorder into the signal path (with the transient-noise suppression also acting on the signals going to the recording inputs). A second DEFEAT pushbutton bypasses the noise-reduction circuits and converts the TNE 7000 to a unity-gain buffer amplifier with excellent noise and distortion properties but no transient-elimination capabilities.
The operating levels, noise, and distortion of the TNE 7000 are such that they cannot de grade the final sound quality. Although the primary purpose is to remove disc noises, it can be used with some success to process FM broadcasts and tape recordings. The user is reminded that pilot-carrier leakage in the tuner's audio outputs can disturb the operation of the sensing circuits. And, of course, CD-4 records cannot be processed by the Burwen unit, since their 30-kHz carrier frequency lies in the most sensitive part of its operating range. However, a phono cartridge with the CD-4 range of response is highly recommended, since it will provide the information required for most effective operation of the transient-noise sensor.
The Burwen TNE 7000 is a flat-format package approximately 163A inches wide, 3 inches high, and 7 3/4 inches deep. It weighs 7 pounds. No power switch is included, since it is expected that the unit will be switched by the associated amplifier. It consumes only 8 watts from the power line and can be left on continuously if desired.
Price: $299.95.
Comment. We made a number of measurements, but found them of little significance in respect to the audible performance of the TNE 7000. We therefore evaluated the unit entirely through use tests.
We were aware that dealer demonstrations of some other transient-noise suppressors are being made with records that have been deliberately scratched with a radial spoke pattern.
The results of such a demonstration can be most impressive, and we have used the method ourselves in evaluating some earlier click-suppressing devices. Somewhat to our surprise, the TNE 7000 did little to remove the effects of such gross record defects. We contacted Burwen and were informed that the TNE 7000 had been designed to remove common "real-world" record scratches, ticks, and pops rather than artificial "demonstration" blemishes.
From that point onward, we began searching for the most tick-laden of our older records, and we found enough of them to give the TNE 7000 a good workout. We were pleased to find that it did its difficult job with almost total success. To Use it, the SENSITIVITY knob is turned clockwise until the adjacent LED dims noticeably, indicating that the program noise threshold has been reached.
Next, the THRESHOLD control is advanced until the clicks and pops are audibly reduced.
Each time the suppressor operates, a red LED next to the THRESHOLD control flashes for 17 milliseconds (even if the actual suppression period is a small fraction of that du ration). One soon learns to adjust the controls until the noises disappear (which, amazingly, they do) while the program quality is unaffected. An incorrect (excessive) setting of the suppressor circuit will produce an unmistakable harsh distortion (and the suppression light will be on full time). The simple solution is to back off a bit on the THRESHOLD knob setting. We found negligible dulling of the transients in guitar music or the sounds of flamenco dancers' shoes and castanets, even though they caused the LED to flash.
We were surprised at how well the TNE 7000 coped with occasional pops and crackles in FM programs. True, we were using a good tuner, with adequate pilot-carrier suppression, but we found that many of the transient noises in FM broadcasts originating on their records could be largely or even totally eliminated. The suppressor worked fairly well even on the occasional case of automobile-ignition interference we experienced.
We wondered what "weaknesses" the TNE 7000 might have other than its inability to eliminate the effects of knife or razor gouges on a record. After much listening, we found none. That does not mean that the TNE 7000 is perfect-just that we found no weak nesses in its performance in normal use. We would consider it highly effective against transient noises on any record that one might seriously wish to listen to (in other words, if a record has been deliberately gouged or trod upon, it is hardly a candidate for hi-fi reproduction). If it has any operating faults, they are more than outweighed by its "pluses." Does this mean that the TNE 7000 belongs in everyone's hi-fi system, or even in every deluxe system? Not necessarily. What it really comes down to is whether it is worth $300 to you to eliminate-in most cases really eliminate-ticks and pops from your records. If that is a problem that plagues you, the TNE 7000 is worth careful consideration.
++++++++++++++++++
Shure V-15 Type IV phono-cartridge
FOR about five years, the V15 Type III has headed the Shure phono-cartridge line. Since progress is inevitable, in cartridges as in all other things, the venerable Type III (which will still remain in the Shure line) has been edged out in the top spot by the company's new V15 Type IV.
Despite a distinct family resemblance be tween the two cartridges (and their common use of the moving-magnet principle), the V15 Type IV is an all-new unit, and its performance improvements over the Type III are by no means slight. The trackability data sup plied by Shure show a moderate increase in trackability at both high and low frequencies (amounting to about 3 dB in the 5,000- to 10,000-Hz range) compared with the Type III, which was already one of the best tracking cartridges on the market. At the infrasonic frequencies, however, where warp excitation can cause tracking problems, the improvement offered by the Type IV is a dramatic 10 dB or more.
Much of the credit for this goes to a newly designed stylus assembly, lighter in effective mass than the former one, but with no sacrifice of rigidity or strength. A new stylus shape, which Shure calls "hyperelliptical," is claimed to reduce high-frequency tracing distortion by as much as 25 percent as compared with the performance of conventional elliptical styli. It is Shure's version of the extended-line contact styli introduced by several manufacturers in the last year or so. The Type IV, however, is not intended specifically for use with CD-4 discs.
Although it often happens that a reduction in the stylus' effective mass results in a lower signal-output voltage from a cartridge, Shure has redesigned the magnetic system of the V15 Type IV for greater efficiency, so that its output voltage is, if anything, slightly higher than that of the Type III. They have also made a long-overdue change in the loading requirements for the cartridge. For years, Shure's top cartridges have given their flattest frequency response when terminated in a rather high capacitance of 400 to 500 picofarads (pF) instead of the usual 200 to 300 pF for which other cartridges are designed. As a result, because of the low-capacitance cables used in most record players since the introduction of CD-4 records, it has often been necessary to add external capacitance to the phono circuits when using a Shure cartridge to get the flattest possible frequency response. The V15 Type IV is designed to work into a capacitance of 200 to 300 pF in parallel with the usual 47,000 ohms, and even that load is not critical, so that it is compatible with virtually every record player and amplifier in current use.
--- FREQUENCY IN HZ (CYCLES PER SECOND); PEAK VELOCITY IN CM/SEC OF TEST DISC
In the graph at left, the upper curve represents the smoothed, aver aged 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. ------------
Shure has been concerned for some time with the problems of tracking warped records, and the Shure cartridges have been de signed to resonate in typical tone arms in the 8- to 10-Hz range, where the warp velocities are relatively low. To reduce the problem even further, they have now taken a rather novel and highly effective approach in the Type IV. The swing-away stylus guard, al though it looks much like those found on other Shure replaceable stylus assemblies, is much more than it appears to be at first glance. It carries a small brush, about V4 inch wide, that rides on the record just ahead of the stylus. The brush is made of some 10,000 graphite fibers only 0.3 mil in diameter, so that about ten fibers span the width of a record groove. This permits the fibers to fit into the groove and remove dust from there as well as from the surface of the record. How ever, dust removal is not the primary purpose of the assembly.
The graphite fibers are electrically conducting, and through the metal supporting bracket there is an electrical connection between the tip of each fiber and the ground terminal of the cartridge (and from there to the system ground). They are intended to drain off electrostatic charges from the record as it is being played, and they do this with considerable effectiveness. Several benefits accrue from reducing the electrostatic charge on the record surface. For one thing, the vertical tracking force will then be determined almost entirely by the arm adjustment instead of being significantly increased by electrostatic attraction between the cartridge and the record (which can be a genuine problem when the in tended tracking force is a gram or less). Also, a lower charge makes the record less likely to attract dust, and it is therefore easier for the graphite fibers to remove any dust that may be present.
Probably the most important purpose of the brush assembly is its function as a damper (or "Dynamic Stabilizer," as Shure calls it): by introducing a mechanical resistance at the low-frequency arm/cartridge resonance, it reduces the amplitude of the resonance and improves the cartridge trackability at very low frequencies. This is accomplished by viscous-damped pivots which give the assembly some of the properties of the accessory dampers that are meant to be attached to the headshell of the tone arm. In the V15 Type IV, the integral damper can be designed for optimum performance with this specific cartridge.
The effect of the assembly on the normal arm-balancing procedure is compensated for by adding 0.5 gram to the stylus tracking force (intended to be effectively between 0.75 and 1.25 grams). Thus the arm force is set in the range of 1.25 to 1.75 grams. If one does not wish to use the damper (though it is hard to imagine why one would not want to), it locks in an upward position when fully lifted (in which case the force should be set 0.5 gram lower). And, as mentioned earlier, when the damper is swung down all the way it serves as a stylus guard.
In addition to the hyperelliptical stylus furnished with the V15 Type IV, two other styli are available. They are the VN4G, with a 0.6-mil spherical stylus, and the VN478E elliptical (biradial) stylus, whose 0.5 mil x 2.5 mil dimensions make it possible to play 78-rpm records at forces between 0.75 and 1.25 grams.
The price of the Shure V15 Type IV is $150.
Each purchaser of a Type W receives a coupon redeemable for a free copy of Shure's new ‘Audio Obstacle Course-Era IV’ (1TR-115) test record.
Laboratory Measurements. The Shure V15 Type IV was tested in the tone arm of a Dual 701 turntable. The cartridge was loaded with 240 pF in parallel with 47,000 ohms. For most tests, the total downward tracking force was set to 1.5 grams with the damper in use, equivalent to a stylus force of 1 gram. Certain tests were repeated with different tracking force and cartridge loads.
Our initial tracking tests showed that the V15 Type IV could play the highest levels on our low- and middle-frequency test records at stylus forces between 0.5 and 0.75 gram. On the German Hi Fi Institute test record, the 70-
micron level of the 300-Hz test band could be played at 0.75 gram, 80 microns at 1 gram, and the maximum level of 100 microns was playable at the cartridge's rated maximum of 1.25 grams.
The outputs of the two channels matched within 0.5 dB, and the average output was 3.85 millivolts at a stylus velocity of 3.54 centimeters per second (cm/sec). The vertical stylus angle was 20 degrees, as rated.
The frequency response, playing the CBS STR 100 record, was well within Shure's tight specification of ± 1 dB up to 8,000 Hz and ±2 dB at 20,000 Hz. Our test sample, which gave the same response on both channels, measured flat within ±0.8 dB from 40 to 20,000 Hz.
The stereo separation was about 30 dB or better in the mid-range, 17 dB at 10,000 Hz, and 8 to 15 dB at 20,000 Hz. A reduction in load capacitance to 150 pF made virtually no change in the frequency response, confirming the compatibility of the cartridge with any modern record player. An increase to 375 pF boosted the output about 1 dB between 6,000 and 12,000 Hz and dropped it slightly above 17,000 Hz. This change was not significant from a listening standpoint.
We were especially interested in the effect of the damper. With the damper lifted and not in use, the low-frequency resonance had a low-amplitude, "double-hump" shape characteristic of the Dual 701 tonearm, whose double-suspended counterweight is designed to distribute and reduce the effects of bass resonance. There was a peak of about 3 dB at 9 Hz and a broad rise of about 1 dB between 12 and 18 Hz. Then we repeated the measurement with the damper in use, changing the vertical force as required. This time there was absolutely no resonant rise. In fact, the out put began to drop off gradually below 20 to 30 Hz, and it was down about 3 dB at 11 Hz and 5 dB at 7 Hz, a substantial improvement.
Since Shure has made high trackability and low distortion two of the major design goals in creating the Type IV, it was not too surprising to find that the distortion when playing the Shure 1TR-102 IM test record was very low, measuring less than 1 per cent from 7 cm/sec to 25 cm/sec and only 2 per cent at the record's maximum of 27 cm/sec. For all we know, these may be the residual levels of distortion in the record; we have never measured lower. Similarly, the 10.8-kHz tone-burst distortion from the Shure TTR-103 record was as low as we have seen--between 0.7 and 0.9 per cent from 15 to 30 cm/sec. All distortion measurements were made with a 1-gram stylus force.
Listening to the older Shure "Audio Obstacle Course-Era III" (TTR-110) record, we found that the Type IV played all levels on the record without a trace of strain at a 1-gram force. A few other cartridges we have used have done nearly as well, but always at a substantially higher tracking force. Evaluating performance with the new TTR-115 "Era IV" record was not easy, and we would prefer to wait until we have had a chance to use it with other cartridges before commenting in any detail on it. It does not appear to be as obvious as the 'FTR-110 in revealing mistracking, but that may simply mean that the Type IV does not mistrack on it! We played some of our collection of warped records, with and without the damper in use. The damper Usually made it possible to play records that were otherwise unplayable except with very low-mass tone arms. There was no audible loss of bass with the damper in use, since its effects are confined to the frequencies below 20 Hz, but its effect on warp tracking was certainly dramatic.
Comment. The sound of the V15 Type IV can be described in much the same way as that of a good amplifier; there is really no particular sound at all that can be attributed to the cartridge. It is, after all, essentially flat, with distortions that seem to be below those inherent in even the best test records, and with far greater tracking ability over the entire audio band than any other cartridge we know of. If such a cartridge were to have a sound of its own, it could only be as the result of a lack of the distortions present in the outputs of other cartridges. Frankly, we did not discern any essential difference in the sound of this cartridge compared, with that of the V15 Type III or even the M95ED when each cartridge was properly loaded and operated within the limits of its tracking ability.
Aside from sound quality, the forte of the Type IV is its all-around tracking ability.
When recorded groove modulation or one of the many kinds of warps imposes a severe task on the stylus system, this cartridge is al most certain to cope with it better than any other we have seen. The Type IV is able to play records that other cartridges cannot.
The price of $150 is considerable for a pho no cartridge, even in today's inflated economy, and the V15 Type IV is certainly not for everyone. For those who can afford it and can appreciate what it does for them and does not do to them, there is nothing else quite like it.
++++++++++++++++++++++
Dynaco Stereo 416 Power Amplifier
THE Dynaco Stereo 416 power amplifier resembles the earlier Stereo 400 in size and in many of its external features. However, it differs from the 400 in so many ways that it certainly deserves its status as a new product.
The Stereo 416 is a powerful amplifier, available either factory-wired or as a kit, rated to deliver at least 200 watts per channel to 8-ohm loads from 20 to 20,000 Hz with less than 0.25 per cent total harmonic distortion. It shares that particular specification with the Stereo 400, but beyond that point the differences become apparent. For example, the Stereo 416 carries FTC-type power ratings for operation with 4-ohm and 16-ohm loads (300 watts and 100 watts per channel, respectively) and can even be used safely with 2-ohm loads, into which its mid-range power rating (at clip ping) is 450 watts per channel! The Stereo 416 has one of the most extensive protective systems we have seen on a home-entertainment power amplifier, and it has been designed to be as rugged and fool proof as possible. The amplifier's power sup plies are protected by an a.c.-line circuit breaker and four d.c. fuses. Electronic volt ampere limiting protects the output transistors (eight for each channel), and heat sensors mounted directly on two of the output transistors shut down the amplifier if their tempera ture rises to an unsafe value. The heat sink, which occupies the entire rear of the amplifier, has an area of over 1,000 square inches and is cooled by a fan that normally operates very quietly at a low speed. If the heat-sink temperature rises to 55 degrees C, the fan switches to a high-speed mode. And if, for any reason, the heat sink reaches 85 degrees C, the entire amplifier shuts down and a red front-panel HI TEMP light glows. When power is first applied, a relay delays connection of the speakers to eliminate any transient thumps. The same relay cuts off the output instantly if a significant d.c.-voltage component appears at the output.
Other protective features include individual fuses in each speaker output (located on the front panel) and the Dynaguard power-limiting circuit that was originally featured on the Stereo 400. This can be set to go into operation at nominal power outputs (into 8-ohm loads) of 20, 40, 80, or 120 watts, or it can be shut off entirely. It integrates the amplifier outputs (separately for each channel) so that brief transients are passed unmodified up to the full power capability of the amplifier.
With signals of longer duration that exceed the threshold of the circuit, the output is smoothly limited to the selected value, with a response time roughly inversely proportional to the amount by which the uncontrolled signal exceeds the threshold. Lights on the panel of the amplifier glow when the Dynaguard circuit operates.
The power switch is a three-position rotary control, the "off" position is in the center, and rotation either to left or right turns the amplifier on. (The two "on" positions serve to reverse the polarity of the power-line connection to eliminate any possible hum from system ground loops.) Other front-panel controls of the Stereo 416 include individual level controls, low- and high-cut filters, and a switch that bypasses all signal controls as well as the Dynaguard. In line with current trends in amplifier design, the Stereo 416 uses LED indicators instead of meters to monitor its output-power levels. Two horizontal rows of LED's display the instantaneous output of each channel in 3-dB steps from rated power (0 dB = 200 watts into 8 ohms) down to -21 dB.
The lights are also calibrated in percentages of rated power, but these turn out to be rather unwieldy values, such as 1.56 percent for-18 dB, so that on the whole the decibel calibrations appear to be more useful. The LED's are virtually instantaneous in their action and respond to peak signal amplitudes (although they are calibrated in terms of continuous, or average, power output, which the manual erroneously refers to as "rms power"). A knob below the indicators increases their sensitivity by 6 or 12 dB or shuts them off entirely. In the most sensitive setting, the -21-dB light glows with an output of about 0.1 watt.
A unique accessory available for use with the Stereo 416 is the Dynaco C-100 Energy Storage System, (see heading illustration), a unit that needs a little explanation. One of the principal limitations on the low-frequency output of an amplifier is the energy available from its power supply. The full supply volt age, whatever it may be, is present in the absence of a signal and remains essentially unchanged for the duration of a short transient signal, even of high power. However, if heavy currents are drawn from the power supply, as happens during high-power continuous operation or during high-level low-frequency transients (especially when driving low-impedance loads), the voltage across the power-supply capacitors drops, usually faster than the power supply can re charge them. Thus the voltage under heavy drain can be appreciably lower than the no load voltage, which is why the continuous output of an amplifier is less than its short term transient output (this is the rationale for using amplifier dynamic "music-power" ratings). Economic and physical considerations usually make it impractical to design a high-current power supply with sufficient regulation to maintain constant voltage under varying load conditions.
---- FREQUENCY IN Hz (CYCLES PER SECOND)
---- CONTINUOUS AND EQUIVALENT SINE-WAVE WATTS/CHANNEL
One improvement that can be made is to use larger (higher capacitance) filter capacitors in the power supply. Since they can hold more energy, they maintain their voltage better under conditions of heavy current drain.
The use of very high power-supply capacitance is one of the distinctions between ex pensive amplifiers and those selling at lower prices. Their inclusion inevitably exacts a cost in dollars, space, and weight.
Dynaco has taken a novel approach to the energy-storage problem by providing an out board "energy storage system," the C-100.
This is nothing more than a huge bank containing 100,000 uF of capacitance, or 50,000 for each power-supply polarity. (By contrast, the Stereo 416 itself has 10,000 uF for each power-supply polarity.) The C-100 plugs into a socket on the side of the Stereo 416, which charges its capacitors slowly when the unit is turned on. After about one minute there is a click from a relay in the C-100, signifying that charging is complete, and a red CHARGE light on its panel is extinguished and replaced by an ON light.
The principal benefit to the user of the add ed energy-storage capacity is a higher clip ping-power level for transient signals, especially at very low frequencies. There is also an increase in the continuous-power capability at lower frequencies, such as 20 Hz, especially when driving 4-ohm or 2-ohm loads.
The C-100 matches the panel width and appearance of the Stereo 416, and it can be placed on top of it in normal operation. This does not impede the cooling of the amplifier, and in fact the C-100 helps form a tunnel around the heat sinks, through which the cooling air is blown by the fan.
The Dynaco Stereo 416 and C-100 are de signed to be mounted in standard 19-inch equipment racks. The front panel of the Stereo 416 is 7 inches high and 19 inches wide;
the amplifier is 14 inches deep. It weighs about 53 pounds. The C-100 has a 31/2 x 19 inch front panel and is the same depth as the amplifier. Price of the Stereo 416: kit, $649; wired, $949. The C-100 is available only as a factory-wired unit and costs $229.
Laboratory Measurements. The amplifier tested for this report was constructed from a kit. However, a second unit, factory wired, was also tested and the two were found to be essentially identical in their performance.
The FTC-mandated preconditioning period which preceded all measurements gave the first clue that the Stereo 416 was no ordinary amplifier. Not only did it not overheat at any time during the one hour at one-third power and five minutes at full power, but its heat sinks never became more than moderately warm. In fact, the perforated metal grille over the power supply and low-level circuit boards was by far the warmest part of the amplifier's exterior (and it was completely comfortable to the touch). The cooling fan never switched into high-speed operation, either during the tests or in actual use.
With both channels driving 8-ohm loads at 1,000 Hz, the outputs clipped at about 260 watts per channel. The 4-ohm and 16-ohm clipping powers were 400 watts and 144 watts per channel. The harmonic distortion at 1,000 Hz was well under 0.01 percent at all power outputs from 0.1 watt to more than 200 watts, and the intermodulation distortion was typically less than 0.03 per cent for outputs up to 200 watts or more. At the rated 200 watts out put into 8 ohms, the distortion was well under 0.01 per cent from 50 to 1,000 Hz, rising smoothly at higher frequencies to about 0.1 percent at 20,000 Hz. It also increased at low frequencies, to 0.02 percent at 20 Hz, but this represents the residual distortion of our signal generator.
The distortion was essentially the same at half and one-tenth power as at full power. It is noteworthy that the distortion in the output of the Stereo 416 was almost pure second harmonic. There was virtually no third harmonic, and no higher-order harmonics. It is generally considered that low-order even harmonics, and especially the second harmonic, are the least offensive types of distortion.
An input of about 0.32 volt produced a reference output of 10 watts, and the wide-band hum and noise level was-81 dB referred to 10 watts (-94 dB referred to 200 watts). The amplifier rise time was 1.2 microseconds and the slew rate (rated at a rather low 8 volts per microsecond) was 18 volts per microsecond.
The Low and HIGH filters reduced the response by 3 dB at 42 and 14,700 Hz. Whereas most LED power indicators we have seen are so approximate as to have little more than cosmetic value, in the Stereo 416 the error of the LED calibrations at 1,000 Hz was usually well under 1 dB and reached a maximum of 2 dB only at the lowest level of-21 dB. The 6-dB and 12-dB attenuation switch settings were also very accurate.
On a steady-state basis, the Dynaguard circuit came into operation somewhat below the indicated power levels. At the 20-, 40-, 80-, and 120-watt settings of the Dynaguard switch, a continuous power output of about 15, 30, 70, or 105 watts caused the Dynaguard light to come on after a number of seconds of operation. But, as Dynaco points out, even the 20-watt setting will not affect the normal operation of the amplifier up to its full clip ping output with most program material, which typically has high-level peaks of very brief duration. Consequently, it is recommended for maximum protection with most speaker systems.
Comment. Our kit builder reports that construction of the Stereo 416 took 40 hours.
Construction was straightforward, with the most tedious operation being the mounting and wiring of the sixteen output transistors.
The instructions were very clear. Maneuvering space was limited toward the end of assembly, but manageable nonetheless. The Stereo 416 is probably not a kit for the absolute novice, although the instructions are simple enough to be followed by someone with little experience (nothing is left to guess work). Afterward, the instruction book/ operating manual serves as an excellent technical manual for the amplifier.
In use tests, we found the Dynaco Stereo 416 to be an absolutely first-rate amplifier.
Not only was its sound at least the equal of that provided by any amplifier we have used, but it was totally free of the extraneous noises that can be annoying in a lower-power amplifier and positively hazardous to one's speakers (and the amplifier itself) in a high-power unit.
The turn-on time delay keeps all transients from the speakers until the amplifier (and other system components) are fully stabilized. If any large d.c. component is present in the signal from another signal source, the relay instantly cuts off the outputs until the dangerous condition is removed. No matter how loudly one plays the system, the Dynaguard gives one peace of mind (and intact speakers!). Yet, even when the circuit operates, there is no sense of distortion or even of com pressed dynamics (for a moderate overload; sustained overload is not recommended for this or any other protective device). The fan, which could be heard at close range in a quiet room, was completely inaudible in use, even with background-music levels.
In reviewing our test results with the Stereo 416, we were struck by how its "philosophy" differs from the current trends in amplifier de sign, and by how those differences have not in any way hampered its performance. At a time when slew rate and transient intermodulation considerations are occupying the attention of many amplifier designers, the Stereo 416 has what may be the slowest slew-rate specification of any amplifier in its power class. If this fact should disturb anyone, the amplifier has a built-in prevention for TIM (as opposed to a cure) in the form of the high-cut filter that slows down input transients to keep them well within the transient capabilities of the following power stages. Although it does not impair the amplifier's listening qualities, we never felt any need to use it (since the slew rates demanded by any program material at our disposal will not tax the Stereo 416 in the slightest). Another interesting aspect of the amplifier is that its high-frequency harmonic distortion is substantially higher than that of some amplifiers designed for very high slew rates, although 0.1 per cent at 20,000 Hz cannot be considered excessive by any rational per son-and, being purely second-harmonic, it is about as innocuous as any nonlinear distortion can be.
The Dynaco Stereo 416 is far from cheap, either in kit or wired versions; nevertheless, it is one of the best "200-watt" amplifiers one can buy at any price, and almost certainly the one least likely to damage either itself or one's speakers if stressed or used carelessly.
++++++++++++++++++
Wharfedale E-70 Speaker System
FOR many years, Wharfedale was a household word among audiophiles. In the formative years of the high-fidelity industry this British loudspeaker manufacturer was noted for neutral-sounding speakers relatively free of the deliberate frequency-response exaggerations and aberrations that characterized many of the hi-fi speakers of the time.
Now, after some years' absence from this market, the Wharfedale parent organization (Rank Hi-Fi) has begun to import a new generation of Wharfedale speakers into the Unit ed States. The top-of-the-line speaker is the E-70, which was tested for this report.
The E-70 is a large, floor-standing system with an open-mesh grille that is virtually as transparent to light as it is to sound. The four drivers and the port opening are framed in bright metal rings, giving a visual effect more reminiscent of some recent Japanese speakers than of the Wharfedale speakers of earlier years. The E-70 is a three-way unit, with a single 10-inch woofer in a ported enclosure described as a "maximally flat fourth-order Butterworth system" that is said to be responsible for high efficiency and flat bass response. The crossover to the mid-range drivers (a pair of 4-inch cone speakers) is at 800 Hz, and there is another crossover at 7,000 Hz to a horn-loaded compression tweeter.
Two front-panel knobs adjust the mid-range and tweeter levels from a nominally flat response (at maximum) downward in five steps of 1 dB each.
The Wharfedale E-70 is an 8-ohm system rated for use with amplifiers delivering up to 100 watts output. The "E" apparently stands for efficiency, one of the principal design aims for the system. It is rated to deliver a sound-pressure level (SPL) of 94 dB at 1 meter when driven by 1 watt, yet its rated frequency response is 50 to 18,000 Hz ±3 dB.
Obviously, this speaker would appear to re quire much less amplifier power than the usual acoustic-suspension design (or even most ported enclosures) for a given sound output.
The veneered wooden cabinet is handsomely finished in oiled walnut, and matching grains are used on speakers bought in pairs. The en closure is 32 inches high, 13 1/2 inches wide, and 14 inches deep. Weight is about 70 pounds. Price: $475.
Laboratory Measurements. With both level controls set to maximum, the smoothed semi-reverberant frequency-response curve was flat within ±2 dB over most of the range covered by the mid-range and high-frequency speakers. A slight rise above 10,000 Hz was the most significant departure from flatness.
The bass response was measured separately at the woofer cone and the port, using close mike spacing, and the two curves were combined to generate a total bass-output curve.
This proved to be exceptionally flat, with none of the bass rise in the 70- to 100-Hz range that mars the sound of so many speakers. Combining the bass and mid/high-frequency curves, we obtained a composite frequency response that was flat within ±2.5 dB from 35 to 10,000 Hz, and nearly within ±3.5 dB from 35 to 15,000 Hz.
The mid-range control (which is actually marked Low) affected the response between 120 and 1,500 Hz, with a maximum cut in out put of about 5 dB. The high-frequency control had a range of 5 dB also, affecting frequencies above 1,500 Hz. The flattest overall response was obtained when both controls were set to maximum.
Our live-room measurement gave a slightly higher efficiency reading than Wharfedale's already exceptional anechoic rating, and the 98-dB SPL we measured in front of the E-70 (it was being driven by 1 watt of random noise in the 1,000-Hz octave) represents the highest speaker efficiency we have measured to date.
The bass harmonic distortion at a 1-watt in put was between 1 and 2 per cent from 100 Hz down to 36 Hz, rising to 7 per cent at 30 Hz. It must be remembered that the SPL developed by this speaker with a 1-watt input is greater than that of an acoustic-suspension speaker driven by 10 watts. We also measured the distortion with a constant-output SPL of 90 dB at 1 meter. The curve was similar in shape to that obtained at 1 watt, but higher in level. It was about 2 to 3.5 per cent down to 40 Hz and 8 per cent at 35 Hz.
The impedance of the E-70 was a minimum of 6 ohms in the 150- to 200-Hz range. Typically it was 8 to 10 ohms at most audio frequencies, rising to 25 ohms at the upper bass resonance of 70 Hz and to 35 ohms at the other bass resonance of 25 Hz. The tone-burst response of the system showed no significant signs of resonances or spurious outputs and was consistent with the flat, smooth measured frequency response.
The simulated live-vs.-recorded listening test confirmed the impression we had formed from general listening to the E-70. It produced a somewhat sharp and crisp sound, apparently a result of the strong high-frequency output above 10,000 Hz. The overall sound balance was good, and there were no obvious low-frequency or mid-range colorations. Nevertheless, the E-70 had a tendency to sound a little "hotter" than most speakers when reproducing programs having considerable energy in the highest audible octave.
To our surprise, reducing the setting of the system's high-frequency level control did not change the essential character of the sound.
In fact, we usually preferred to leave both controls at their highest settings. This can be explained (we think) by the fact that the over all frequency response of the speaker is exceptionally flat with both controls at maximum, except for the rise above 10,000 Hz.
The HIGH control introduces a downward slope in the response that negates much of the benefit of its inherently flat character. Per haps an octave equalizer or a tone control capable of modifying the response only above 10,000 Hz would make a worthwhile improvement. We didn't try to equalize the speaker, preferring its sound "as is."
Comment. We would have to agree that, in the E-70, the Wharfedale engineers have achieved their goal of very flat, wide-range, low-distortion performance in a speaker having unusually high efficiency. Certainly the efficiency is at least as high as that of any non-horn-loaded system for home use that we have used. Although we drove the E-70 from high-power amplifiers (200 watts per channel or more) much of the time, we also obtained excellent results with a car-radio amplifier delivering about 2 watts per channel! The low-frequency response of this speaker, no matter what technical terms are used to describe its design principle, is as nearly flat as any we have seen. In fact, its overall measured frequency response and excellent high-frequency dispersion make this system one of the more outstanding ones we have seen in terms of meeting its design goals and specifications.
Our listening (and test) room is relatively well damped-not a "dead" room by any means, but far from "live" or bright-sounding. Speakers whose high-frequency output is at all lacking tend to sound dull and lifeless in such an environment. The Wharfedale E-70, on the other hand, is very close to ideal for our listening conditions. The sharp, almost analytical quality imparted by the high-frequency emphasis can be heard, but we did not find it unnatural or objectionable in any way.
The quality of the cabinet's walnut finish also deserves mention. It doesn't help the sound, but the manner in which the grains are matched shows the care that went into the manufacture of the E-70 and explains some of its rather high price. The sound quality ex plains the rest!
-------- Tone-burst response of the E-70 at 100, 1,000, and 6,000
Hz. Lower trace is the speaker output. Measurements were in a reverberant
room.
+++++++++++++++
Teac A-103 Stereo Cassette Deck
THE Teac A-103 is a moderately priced front-loading cassette deck with the dark-gray and black appearance currently popular in audio-component styling. The tape trans port, which occupies most of the left side of the unit, uses a single d.c. servo-controlled motor. The cassette itself loads into a slot in the door.
Almost the entire cassette can be viewed through a large, clear window in the door, and an orange backlight behind the cassette makes it easy to see how much tape is on each of its hubs. The control keys can be operated in any sequence without going through STOP (except for the REC and EJECT functions). The cassette compartment is flanked by the pushbutton power switch on the left and the index counter on the right. Below the counter are two mic jacks for medium-impedance (600 ohms or higher) microphones.
The right half of the panel carries a functional group of recording and playback controls and indicators. At the top are two large, well-illuminated level meters whose green-lit markings are clearly visible from a consider able distance. To their right are rectangular indicator lights (red and orange) that show when the record function or the Dolby system have been activated.
Below the meters are four bar-shaped push buttons. One selects either the line or the mic / DIN inputs (the two cannot be mixed). To its right are the separate atm and EQ switches, each marked for CrO2 and "normal" tapes.
The fourth button turns on the Dolby system. Below them are the two recording-level control knobs, and concentric with each is a clear plastic ring with a red index line that can be set to mark a preset recording-level adjustment. The headphone jack (for 8-ohm phones) is at the lower right of the panel. In the rear of the recorder are the line inputs and outputs and a DIN connector.
The specifications of the A-103 are not markedly different from those of many similar machines. The frequency response with a chromium-dioxide tape is stated to be 30 to 16,000 Hz, with no tolerance given. The signal-to-noise ratio (S/N) is rated at 55 dB with out Dolby, and wow and flutter is rated at 0.1 per cent in a NAB-weighted measurement.
The Teac A-103 is about 16 inches wide, 6 1/4 inches high, and 11 1/4 inches deep. It weighs 15.4 pounds. Approximate nationally advertised value: $250.
Laboratory Measurements. No specific recommendations are made for tapes compatible with the machine's adjustments. The instructions consist of a single folded sheet, but they are nevertheless adequate in most respects, listing many popular tapes and the corresponding settings for the bias and equalization switches. Lacking further information, we made frequency-response measurements with a number of presumably suitable tapes.
With NORMAL bias and equalization, very similar results were obtained with Maxell UD XL I, Scotch Dyna-range, Memorex MRX2, and BASF Professional I. The Maxell was used for our subsequent measurements with NORMAL tape settings, and most of the tests were also made with Dyna-range, since it was very nearly like the UD-XL I. We also tried TDK AD and Scotch Master I, but the ma chine was evidently under-biased for them and the high-end response rose (though not so much as to make their use impractical).
With the CrO2 control settings, the Scotch Master II was the flattest by a small margin, although TDK SA and Maxell UD-XL II were almost identical to it. BASF Professional II, a true chrome tape, had a strongly rising high-end response.
The overall record-playback response at a-20-dB level was practically the same with any of the tapes we used. In general, the variation was about ±2 dB from about 40 to 15,000 Hz, with the usual low-frequency ripples in the response curve beginning at about 200 Hz. The frequency response at a 0-dB re cording level revealed some interesting facts about the recorder as well as the tapes. With the "normal" tapes, the 0-dB curve intersect ed the-20 dB curve at about 13,000 to 14,000 Hz (in contrast to the 10,000 or 11,000 Hz ob served on many low-price cassette decks).
This indicates relatively good high-frequency tape-saturation characteristics and, by inference, a relatively efficient head design. With the chrome-type tapes, whose high-frequency performance is inherently better than that of a standard ferric tape, the two curves never intersected, the 0-dB curve remaining well above the-20-dB curve at all times. The rapidity with which the high-frequency response drops above 15,000 Hz suggests that the recorder has a built-in low-pass filter to re move any 19-kHz pilot carrier from a stereo FM program that is being taped.
The playback frequency response was measured with a TDK AC-337 test tape for NORMAL (120 u-sec) equalization. It was flat within ± 1 dB over the 40- to 12,500-Hz range of the tape. The CrO2 (70 u-sec) response was measured with a Teac 116SP test tape, and it measured flat within ± 1.5 dB over the 40- to 10,000-Hz range of the tape.
The tracking of the Dolby system was checked at recording levels of-20 and-30 dB. There was less than 1 dB of change in the frequency response with the Dolby system in or out of the circuit at any frequency, indicating very close matching of the Dolby recording and playback frequency-response curves.
This would not have been possible without the recorders' very flat response.
A line input of 55 millivolts (mV) or a mic input of 0.215 mV was sufficient for a 0-dB re cording level. The mic input overloaded at 30 mV, which suggests the need for caution when making live recordings unless a micro phone with very low output (or an attenuator) is used. The playback output (which is at a fixed level) from a 0-dB recording was 325 mV with "normal" tape and 290 mV with CrO2.
The playback distortion (third harmonic) from a 0-dB recording at 1,000 Hz was about 0.8 per cent with Dyna-range tape, 0.5 percent with UD-XL I, and 1.8 percent with Master II. The reference-distortion level of 3 percent was reached with recording inputs of +6, +7, and +3 dB, respectively. The unweighted signal-to-noise ratio (S/N) referred to those re cording levels was 50 dB with Dyna-range, 52.7 dB with UD-XL I, and 48.5 dB with Master II. With CCIR/ARM weighting these readings improved to about 54, 54, and 55 dB, respectively. Finally, with the Dolby system in use, the weighted S/N was 64 dB with each of the tapes. The noise increased by 10 dB through the mic input at maximum gain and proportionately less at lower settings.
The combined wow and flutter, measured with a TDK AC-342 test tape, was 0.05 percent with NAB weighting (average reading), and the same with a JIS reading (rms). A CCIR- or DIN-weighted peak-flutter measurement read ±0.1 percent, as did an unweighted average measurement (0.1 per cent). The tape speed was about 1 per cent fast.
The meter carried a Dolby-level calibration at its +3-dB marking, which supposedly corresponds to the standard (200 nW/m) Dolby level. However, playing a Dolby-level tape (TDK AC-337) gave meter readings of +5 dB.
Since the Dolby system seemed to work with unusual effectiveness, we assume that the error was in the meter calibration and not in the actual adjustment of the Dolby circuits. The meters came very close to meeting true VU meter ballistic specifications (a rarity in home recorders), with a 5 per cent overshoot on 0.3-second tone bursts of 1,000 Hz. In fast for ward and rewind, a C-60 cassette was moved from end to end in about 80 seconds. The 1,000-Hz crosstalk between channels was-54 dB, as measured with a TDK AC-352 tape. Headphone volume with 200-ohm phones was on the low side.
Comment. Reviewing the test data and comparing it with what we have measured on a number of other cassette decks, it becomes clear that the Teac A-103 performs well above expectations for a machine of its price. The ultimate test must be in the listening, and here the A-103 acquitted itself beautifully. When we recorded interstation FM tuner hiss and compared the playback to the original (at a -10-dB level) the fidelity of the recording was virtually perfect. Needless to say, recording music off the air resulted in a playback that was indistinguishable from the original. The same applies to playing high-quality recorded cassettes such as the Advent CR/70 series.
They sounded every bit as good as on machines costing two or three times the price of the A-103.
Evaluating the Teac A-103 left us once more with that feeling of satisfaction that comes from finding a product that not only lives up to its promise, but exceeds it handily.
About all that is missing is "memory re wind," provision for ferrichrome tape, input source mixing, and a playback-level control.
The last is the only one we really missed. It all goes to show that while you can't have every thing for a bargain price, you certainly can come close to it in the Teac A-103.
This is without doubt a very good cassette recorder, one that sounds as good as it measures and handles with complete freedom from "bugs."
------ FREQUENCY IN HZ (CYCLES PER SECOND)
+++++++++++++++++
Also see:
Source: Stereo Review (USA magazine)