Vintage Quad Reviews
AT A GLANCE: The Quad is a full-range electrostatic loudspeaker, covering from slightly below 50 cps to well above the limits of audibility. Having no heavy cone or voice coil, it possesses very low mass and correspondingly fine transient response.
The Quad has several limitations of a technical, aesthetic, and economic nature that may limit its general acceptance. In our opinion, however, it represents, by a wide margin, the closest approach to truly natural reproduction of sound in the home that we have yet heard.
Unlike other types of speakers, the Quad should not be installed closer than two feet to a wall, or three feet to a corner. The bass perfor mance is degraded by such improper positioning. As may be imagined, this large and uniquely designed speaker assumes considerable prominence when it is standing in the clear, undisguised as a piece of furniture or anything else but what it is - a radiator of sound. For this reason, those interested in decor more than fine sound may regard it as not readily adaptable to the usual living room. In stereo pairs this visual problem may be further accentuated.
The instruction booklet accompanying the Quad is quite specific in limiting its application to rooms of less than 5,000 cubic feet. Our ex perience suggests that it will perform to best ad vantage in rooms considerably smaller than that, and in fact will do a fine job in a room too small to accommodate any conventional speaker. Unlike cone radiators, the electrostatic speaker is literally a window opening on the concert hall, and one can listen to it in comfort at a two-foot distance as well as at twenty feet or more.
In using a Quad some modification of listening habits will probably be necessary for people accustomed to conventional speakers. This speaker should not, and cannot, be subjected to the room shattering levels beloved by some audiophiles.
To do so is to invite breakup and distortion well below the level where the windows rattle. The Quad should be listened to at natural levels. When it is heard somewhat above normal level, the effect is that of being transported towards the orchestra; softer levels move the listener to the rear of the auditorium. This effect is very real, and we have not experienced it to anything like this degree in conventional speaker systems.
The Quad sounds quite different from any other speaker we have heard. It is crisp and taut at all frequencies, including the middles and bass. There is not a trace of the boom or boxiness present to some degree in most conventional speakers. The separation of instruments in the orchestra, even in monophonic reproduction, is strikingly superior to anything we have previously heard Any doubts as to its bass performance were dispelled when records having large bass drum sounds were played. A comparison against the best cone speakers we could muster showed that the cone speakers had much more apparent bass below 50 cps. A thump from the bass drum shook the room in a most satisfying manner. Switching to the Quad eliminated the strong, room-filled bass which we had come to equate with the best in high-fidelity sound. The drum sounded just like a large bass drum - no more, and no less. The absence of bass hangover and excitation of room resonances probably had a lot to do with this naturalness. Others who heard our Quad were equally impressed in its favor.
Having established by listening that this was a superior speaker we were curious to see what our measurements would show. Frequency response, taken out-of-doors, follows the contour of our microphone calibration rather closely. A line drawn through the centers of the many small peaks and dips in the response lies within 5 db of the microphone response from 60 to 15,000 cps. More important is the absence of any of the huge holes or peaks often found on lesser speakers.
The low frequency radiation shows a slow smooth decline starting at a few hundred cycles. But without the sudden change of slope characteristic of box speakers having a system resonance.
Bass frequency distortion is not outstandingly low between 50 and 80 cps, and rises sharply below 50 cps, which we consider to be the effec tive lower limit of the speaker's response. At moderate levels and with some increase in dis tortion, a useful output can be obtained at 45 cps. Obviously, this speaker will not reproduce the lower pedal notes of the pipe organ as well as some conventional systems may, but this is probably the only type of music with which it shows any limitation. On bass transients, such as those produced by drums or keyboard instru ments, the loss of extremely low frequencies is more than compensated for by the lack of re sonance and hangover.
The polar response (not plotted) is a smooth cardioid pattern, down about 10 db at 45 degrees off center axis. The Quad booklet shows a 70-degree effective angle in the horizontal plane, and our measurements confirm this. We measured it at 7 kc, but polar response of the Quad changes relatively little with frequency. This no doubt also contributes to its listening quality.
Tone burst patterns revealed one frequency at which ringing occurred (2.85 kc). At all other fre quencies, however, the tone burst pictures were virtually ideal.
The Quad should be listened to carefully, and preferably in one's own home, before buying. Don't be surprised if the result is a feeling of dis satisfaction with your present speaker system.
H. H. LABS.(From HIGH FIDELITY MAGAZINE - October 1960)
AT A GLANCE: Behind its deceptively simple front panel and compact enclosure, the Quad 22 stereo- control unit houses a host of operating features and conveniences. Being designed to operate principally with the Quad TI power amplifier, rather than with a large diversity of power amplifiers, the Quad 22 has succeeded in eliminating most of the gadgetry and complexity we are accustomed to seeing in stereo preamplifiers.
Nevertheless, the important operating requirements of a good control unit - those which contribute to listening enjoyment - are well represented in the Quad. Practically all the requirements of the most critical music lover (as contrasted to the gadgetry) will be met in a completely satisfactory manner by the Quad 22.
The Quad II power amplifier, which powers the preamplifier, is also a remarkably simple design, with extremely high performance. Rated at 15 watts, it will deliver 50% more than this over most of the audio range, and at most output levels has unmeasurably low distortion. Construction is jewel like, in the best 13ritish tradition, and all tubes and components are operated most conservatively. Control unit: $150; power amplifier: $125.
IN DETAIL: To begin with, the Quad 22 is one of the smallest stereo control units available, measuring 10 _ in. wide by 3 _ in. high by 7 in. deep over-all. Most of its control knobs are recessed flush with the surface of its panel. Their functions are clearly marked. Input channel selection is by a group of push buttons below the knobs. Much of the operation of these buttons seems obvious from their markings, but there is far more to this preamplifier than meets the eye.
For example, the two left-hand buttons are marked stereo and mon. Not much doubt about their use, except that under the buttons is a line linking them, marked 2 mon. This signifies that pressing both buttons plays a single-channel mono input through both power amplifiers and speakers.
The next four buttons are marked radio, mic, disc, and tape. Also self-explanatory, but once more there are further markings under the buttons. Joining radio and mic is a line marked radio 2. When both buttons are depressed, a second high level input, for another radio, TV, etc., is switched in. Joining the mic, disc, and tape buttons is a line marked discs. The disc button must be pressed for any phono operation, and places the RIAA equalization into effect. When both mic and disc are pressed, somewhat more high frequency rolloff is applied, as would be used with older Columbia LP records. With both disc and tape in use, the low frequency turnover frequency is reduced from 500 cps to approximately 350 cps, and there is less high frequency rolloff than in the RIAA curve. Operating all three buttons gives still less high frequency rolloff, plus extended low frequency boost.
When a stereo pickup is used, its channels are automatically paralleled when the mono position is used. A third pickup jack, for mono pickups, is provided, and is only selected when the mon button is used. Although all basic equalization is performed within the preamplifier, the various loading and matching requirements of the many types of pickup cartridges are handled by an external plug-in adapter. A socket is provided for this in the rear of the preamplifier, as well as one for a tape adapter. The pickup adapter terminates the cartridge correctly and attenuates its output to a level which will not overload the phono preamplifier stage. A table in the Quad instruction manual lists the adapter units for all commonly used cartridges.
Also on the rear of the preamplifier are three other sockets. These can supply power to one or all three external tuners, a multiplex adapter, etc. The plate power to these outlets is switched by the corresponding input selector button. The practice of powering tuners from the amplifier is common in England (Quad makes an FM tuner which operates in this manner) but is rarely found in this country.
The volume control of the Quad 22, operated by a large dial, is very silky and smooth in its operation. Underneath this dial is a small button which moves in an arc of 90 degrees. This operates the balance control, which has a range of plus or minus 5 db on each channel.
The two tone controls (ganged for both channels) are clearly marked for their level, or flat, position. The tone controls are of the feedback type, with a sliding inflection point. Mild operation of the controls gives a satisfying control of frequency extremes without being heavy handed in the middle region.
One of the outstanding features of the Quad 22 is the variable slope high frequency cutoff filter. Three cutoff frequencies (10 kc, 7 kc, and 5 kc) are provided. A second knob controls the slope of the cutoff characteristic from zero to 25 db/octave. The filter (which is used for both channels) is not a tone or equalizing control, but solely for removing distortion products in the higher frequency regions with a minimum effect on the musical content of the program. It does this with remarkable effectiveness. The need for this sort of control depends on the quality of oneÕs speakers, with the finest speakers needing it the most. On a truly wide-range speaker slight amounts of high frequency distortion can become extremely annoying. We found that practically all FM broadcasts could benefit greatly from proper application of this filter.
Another position on this filter is marked cancel. This position removes all filters and tone controls from the circuit. It serves both as an extremely flat , wide-range position, and as a means of determining the effectiveness of the tone control and filters, by an A-B comparison.
The Quad II power amplifier is not a newcomer to the high-fidelity scene. It too is very compact, measuring only 13 by 4_ by 6_ in. This tiny package, weighing only 18_ pounds, contains a superb amplifier and its power supply, capable of powering a preamplifier plus a tuner. Its circuit has no controls for balancing or other adjustments. The push-pull KT66 output tubes are operated in a unique Quad circuit, which has windings for the cathodes of the output tubes as well as the plates. This increases their efficiency, as well as reduces their distortion. The components of the Quad II amplifier are operated with utmost conservatism (tubes at 70% of rated voltage, etc.).
The output impedance is fixed at 16 ohms (4- and 8-ohm speakers can be handled by changing jumper wires inside the amplifier). The power output is rated at 15 watts from 20 to 20,000 cps.
In our tests, the Quad 22 and Quad II were treated as a unit amplifier, since they are designed to be used together rather than with other amplifiers and preamplifiers.
The frequency response was just about as flat as our measuring equipment, except for a rise of slightly over 0.5 db at 20 cps. The tone controls, for the first half of their rotation, have a very mild effect, particularly on the high frequency end.
The filter curves show the response for the 10 kc, 7 kc, and 5 kc positions, at slopes of approximately 12 db/octave and at the maximum of 25 db/octave. Unfortunately these curves do not convey an adequate impression of the improvement in listening quality resulting from intelligent use of the filter controls.
The RIAA phono equalization is within plus or minus 1 db from 20 to 20,000 cps, which is about as good as we have seen anywhere.
The power amplifier proved to be as conservatively rated as one could wish. At the rated 15 watts output (at 1,000 cps) the distortion was unmeasurable by our equipment (it was the residual level of somewhat less than 0.1%). The harmonic distortion reached 2% at 23 watts. At 20 cps the Quad II did not fare quite so well. Its output transformer is comparatively small, and we were surprised to find it as good as it proved to be. The 20 cps distortion was under 0.5% up to 4 watts output, and did not become severe until nearly 10 watts was reached. The intermodulation distortion, which is a rough measure of the amplifierÕs performance at 60 cps, was under 0.4% up to 15 watts, and reached 2% at 23 watts.
According to the IHFM standards for amplifier power measurements, the Quad II has a power bandwidth of 24 to 16,000 cps at 11.5 watts and 1% distortion.
Incidentally, all the preceding response and distortion figures apply to the combination of preamplifier and power amplifier.
The hum and noise on high level inputs are totally inaudible. On the phone input, one can hear a slight hiss and hum at maximum gain by placing oneÕs ear against the speaker, but under normal conditions it is completely silent. The gain is moderate on phono, with about 3 millivolts needed for 10 watts output. On radio input, only 54 millivolts is required (British tuners usually have lower audio output levels than ours).
The 10 kc square wave response on the Quad amplifiers showed a slight rounding of the leading edge, which changed to a single overshoot in the cancel position of the filter control. Adding capacitive loads of up to 0.22 mf in parallel with a 16-ohm resistive load produced a slight ringing, which became accentuated when the resistive load was removed. We were curious about the behavior of this amplifier with capacitive loads, since it is designed to drive the Quad full-range electrostatic speaker. We therefore examined its square wave response with this speaker as a load. It proved to be quite stable, with no ringing, but with considerable rounding of the waveform. The important thing, of course, is that is does drive this unusual speaker, and does a superb job of it.
When one weighs the virtues of the Quad amplifiers against the many features they do not have, the balance is heavily in favor of the Quad. Instead of blend, reversing, phasing, loudness compensator, and other controls (which certainly have their place in the scheme of things but may not be used too often) the Quad has a pleasing simplicity of appearance, design, and operation. Its tone controls are really usable, and its filters are worth more than any other feature we can think of when it comes to getting the best sound from less than the best program sources. The over-all cleanness of the power amplifier is as good as its curves suggest. In this day of super-power amplifiers, it is sobering to listen to a really good 15 watt amplifier, even on a moderately low efficiency speaker system. When a pair of these are used in a stereo system, they are more than adequate for most listening situations.
The purpose of this review is to compare the Quad ELS with other high quality speakers now available and with original sound sources. The ELS is basically a large capacitor with outer per forated plates and a central charged diaphragm. The signal is applied in push-pull to the outer plates which deflect the diaphragm.
The electrostatic system has several advan tages over moving coil in that the diaphragm is driven over its complete surface, making its movement predictable. An electrostatic dia phragm can also he made light enough to follow the signal even at very high frequencies.
The designer of such a speaker has three main problems to overcome. First. a basic capacitor transducer is non linear as the force on the dia phragm is not proportional to the applied signal voltage. Second, the load presented to an amplifier is essentially capacitive and therefore difficult to match. Thirdly. the manufacture of a practical full range unit at a realistic price is not easy. In the Quad ELS the first problem is overcome by a simple but effective device. Instead of applying a constant voltage to the diaphragm, once it is charged the polarising potential is disconnected. The diaphragm now carries a constant charge and experiences a force proportional to the applied signal voltage. In this way it in turn applies a force which acts directlv on the air and is a linear function of the applied signal voltage. The second and third problems are solved together bv constructing the speaker of strip units progressively increasing in plate spacing and area from the centre line, together with suitable crossover net-works. In conventional jargon, the centre strip is the super tweeter, the two on either side of this the tweeters, and the two outside strips the woofers. Since the centre strip is vertical and narrow, its horizontal dispersion is excellent. Unless some measure were taken to improve it, however, the vertical dispersion would be poor. In the Quad ELS the plates are curved to assist the vertical dispersion.
As constructed, the ELS is a doublet source -i.e. the diaphragm radiates on both faces (at least at low frequencies). Having no upward or side ways radiation, it cannot directly excite room modes in two out of three room dimensions. In addition, its polar diagram is such that the mean spherical radiation is reduced by a factor of three at all frequencies, further reducing colour due to the listening room by the same factor.
For optimum results this construction requires that the speaker be free standing and placed well into the room. Two ELS were used in this way for the listening tests. They were also tried close to a wall and found to give excellent results provided they were not placed parallel to it. My usual test tape was played and the opinions of the listening panel are given below.
|Choir:||Clear natural sound, very pleasing.|
|Bell and percussion:||Excellent transients, very clean and bright.|
|Qrgan:||Full pleasant and natural tone. Having heard the popular fiction that the ELS is lacking in bass, we were surprised by the amount of bass produced.|
|Folk singer:||Voice and guitar both very natural.|
|Dance hand:||Natural pleasant sound. The leader claimed that the sound was exactly what he heard when conducting.|
|Piano concerto:||The strings had the right sort of ' sheen' and the piano a pleasant singing tone.|
|Wind quartet:||Excellent balance with natural sound from all instruments.|
|Speech:||Opinion was divided here. Some listeners thought male spech a little nasal, others that it was the most natural they had heard. Faults in the recordings were clearly heard.|
|Full orchestra:||Climaxes handled well. A good sound generally with firm bass.|
|Military band:||Listeners claimed that the sound was exactly what they heard when listening to bands in the park The sound certainly bad an 'open air' quality though the recordings were made in a concert hall.|
Comparison with other monitor speakers generally showed up the coloration in the other speakers. It was only when compared with the Spendor BC1 that the 'nasal' quality on speech became apparent.
In all these speakers reviews, any apparent fault in reproduction is checked by a live versus recorded comparison so a male voice was recorded first balancing on the Spendor, listening to the playback on the Spendor and then on the Quad. Then balancing on the Quad and again listening to playback on the two systems. In the first case, the Spendor sounded right and the Quad slightly nasal. In the second case, the Quad sounded right and the Spendor slightly bass heavy. Both these effects were marginal and needed careful listening to detect. On these tests alone it would he impossible to state that one of the speakers was right and the other wrong.
Frequency response curves were taken in free air conditions and show the speaker to have a figure of eight polar characteristic at low frequencies and a cardioid at high frequencies. The stereo image was good over quite a wide listening area but within a smaller area the image was even better This effect has probably given rise to the other popular fallacy that the ELS permits only one stereo seat. Over the wider listening area, the image was as good as many other speakers tested and considerably better than most As can be ex pected from the excellent response curves, the speaker is one of the least coloured ever tested.
This, coupled with its property of exciting room resonances less than more conventional speakers, makes it very suitable for use where acoustic treatment of the listening room is not possible. The ELS gives a particularly clear and clean sound and even listeners who preferred the 'warmer' sound of the Spendor or Rogers speakers were impressed by this.
The transient response of the ELS is quite ex ceptional and, as all material tried on the speaker gave such excellent results we decided to try to stretch it to the limit A recording was made con sisting largely of deep pedal notes on a large organ together with triangle gong, cymbals, side drum, bass drum tympani and tubular bells. It thus consisted of frequencies largely below 50 Hz and above 8 kHz with sustained deep bass together with fierce transients, providing a searching test for any set of replay equipment. It was difficult to find a combination of amplifier and speaker that would cope adequately with this tape as it stretched most systems to their limit, and cracked some, but the ELS fed by the Quad 303 and 33 functioned well. Listeners to this last recording were struck by the amount of bass the ELS would handle and also by the surprisingly high volume of sound they produced. The stereo image was rock steady, all the percussion instruments being accurately pin-pointed.
The flat axial frequency response shows one reason for the pleasant uncoloured sound and the excellent polar diagrams account for the very good stereo image. In sound quality, the ELS stands up to the very best monitor speakers available today. The ELS handles sufficient power for monitoring classical music at moderate levels and is well suited to this purpose. Comparing the performance and measured curves of older ELS units with those submitted for re view showed a consistency over five to ten years. Only a careful check of serial numbers enabled the units to be accurately distinguished It seems that every ELS is like every other almost regard less of age. Barring excessive overloading, it seems that the units are outstandingly reliable. Many of them have given trouble free per formance over a large number of years.
A device like this which is reliable and consis tent from sample to sample is too rarely met. It is still the standard by which others can be judged and is highly recommended to all who want to hear clean, uncoloured and undistorted sound.