.

.

• Anmerkung : Die am Anfang beginnenden Themen und Artikel sind scheinbar trivial und weitgehend bekannt, das stimmt aber nicht. In den Artikeln wird wirklich kompetent recherchiert und berichtet und auch nach unserem Wissen realistisch bewertet. Interessant wird dann die zweite Hälfte dieser Seite, wenn Julian Hirsch aus dem "Nähkästchen der Erfahrung" plaudert !!!

.
The use of published specifications as a basis for judging product performance is probably more practical for a tape recorder than for any other hi-fi component. This fortunate situation occurs because a tape recorder (at least, a good one) has relatively little effect on the quality of programs recorded or played back through it.

The noise level may be increased somewhat and, in most cases, there will be a small (hopefully inaudible) amount of flutter added to the sound. The frequency response and distortion existing in the incoming signal will usually determine the final quality; a properly operated recorder causes remarkably little degradation of sound.
.

This is in sharp contrast to loudspeakers, whose sound has thus far defied all efforts to define it in terms of measured parameters, and to amplifiers, most of which are capable of such high quality that a discussion of their sonic characteristics belongs more in the area of opinion than provable fact.

The tape recorder, on the other hand, is a highly refined but nevertheless imperfect device, many of whose capabilities and imperfections can be inferred from its electrical and mechanical specifications.
.

Not all recorders are specified with equal thoroughness. The best machines, however, carry an impressive array of figures and technical terms, probably more confusing than edifying to a nontechnical reader.

Using the published specifications of a typical high-quality open-reel recorder as an outline, here are definitions and explanations of some of the specifications, with observations on those with greatest significance as well as those of little or no importance.
.

Most of the specifications for an open-reel recorder can also be applied to a cassette recorder. Until recently, many of the operating features of open-reel tape machines were not available in the cassette format, and there were considerable differences in some of their performance characteristics. This situation is changing rapidly and the gap (including price!) between cassette and open-reel machines is closing.
.

Tape-recorder performance ratings, like those of other components, are based on specified test conditions, which unfortunately are rarely specified by the manufacturer in his literature.

Wherever appropriate, we will indicate the standard test conditions or those used in Hirsch-Houck Laboratories' tests of recorders for Stereo Review and Popular Electronics where we deviate from standard practice or where no standard exists.
.

A tape transport may use one, two or three motors. A three-motor machine has a constant-speed motor to drive the capstan, which determines the tape speed.

Each tape reel is driven by its own torque motor which maintains a constant tape tension, within the tensile limits of the tape base material. During high-speed operation (fast-forward or rewind), the capstan does not contact the tape and one reel motor operates at high speed while the other supplies a controlled amount of "drag" to prevent tape spills.
.

Many inexpensive open-reel machines and most cassette recorders use a single motor which drives the capstan at constant speed and operates the reels through a system of belts and slip clutches.

Single-motor transports usually have somewhat more flutter than three-motor types. However, this is also a function of the overall quality of construction and three inexpensive and poorly balanced motors could be inferior to a good single-motor transport.
.

Three-motor tape transports are inherently more reliable, due to the elimination of belts and clutches which require periodic adjustment or replacement. They lend themselves to remote control, via electrically actuated solenoids; many of the better three-motor machines are equipped for optional remote control. A very practical advantage of a three-motor transport is its higher speed in fast-forward or rewind, typically two to three times as fast as a single-motor machine.
.

Some cassette decks (and a few open-reel recorders) use two motors; one for the capstan and one driving the tape hubs through belts. At least one cassette machine has a three-motor transport.
.

To ensure constant tape speed, the capstan of a tape recorder should be driven by a constant-speed motor, such as a synchronous motor, whose speed is a function of the power-line frequency and is independent of line voltage or minor load variations.

Synchronous motors are almost universally used for capstan drive in high-quality recorders, with induction motors to drive the reels. The torque and speed characteristics of induction motors are well suited to this application.

Many single-motor machines also use them for capstan drive. The speed constancy of an induction motor (it can vary slightly with load changes or large line-voltage changes) is adequate for most home recording purposes, but not for professional applications, where precise timing is important.
.

Two-speed synchronous motors are widely used, since they do not require mechanical shifting of belts or drive wheels when changing tape speed (except in a three-speed machine, where this is usually done to select the lowest speed).

A few very fine tape transports use an electronically servo-controlled d.c. motor for their capstan drive. This system, although expensive, provides complete electrical control of tape speed, including such possibilities as vernier adjustment or even continuous adjustment over a wide range. It is also entirely unaffected by line voltage or frequency variations. Many cassette recorders also use servo-controlled d.c. motors, and at least one has a more sophisticated direct-drive motor operating at capstan speed.
.

Most recorders can be classified as two-head or three-head machines. As a minimum, an erase head and a combined record/playback head are required. The electronic portions of two-head recorders are switched between the record and playback functions, which accounts for this being the most popular system in low-priced recorders.

For best performance, different gap widths are required on the recording and playback heads. A combination head is usually designed to favor the playback function, which compromises some of its recording characteristics. The ideal solution is to have separate heads for the two functions and this is now the rule in all but the least expensive open-reel recorders.
.

The separate playback head makes it possible to monitor the program from the tape an instant after it was recorded. Separate recording and playback amplifiers are also required for this feature, which is now found on most tape decks with any pretensions of high-fidelity performance. Off-the-tape monitoring simplifies setting optimum program levels. Distortion or hiss from excessive or insufficient level shows up immediately, as well as hum and unexpected background noises.
Almost all cassette recorders have two heads, required by the Philips patented cassette design. A few have been built with separate record and playback heads, with a third head for non-critical monitoring or with a single head structure containing separate record and playback gaps.

Most home tape recorders, either open reel or cassette, are four-track (quarter-track) machines. A stereo program is recorded on two tracks in one direction, and when the reels are interchanged the second pair of tracks is recorded in the opposite direction. A number of open-reel recorders, and a few cassette machines, are designed to play in both directions without physically interchanging the reels or turning over the cassette. Usually, the reversal can be accomplished automatically with the aid of a piece of conducting tape attached to the recorded tape or by a sub-sonic tone added to the tape recording. In the case of cassettes, the stalling of the transport at the end of play initiates the reversal instead of simply shutting off the recorder.

In order for the playback head to contact the second pair of tracks in the reverse direction, it must be shifted mechanically or a second head can be used. A few recorders can also record in the reverse direction. In this case, as many as six heads may be needed.

For more demanding applications where ease of editing, maximum dynamic range, and lowest distortion are required, the two-track (half-track) format is used. The two tracks occupy the entire tape width and are recorded in one direction only. Many semi-professional machines are optionally available with two-track heads.
.

Quadraphonic open-reel recorders employ the standard four-track format, using only one direction of tape movement. Except for the additional recording and playback amplifiers, they are essentially similar to stereo machines.

An alternate arrangement used in a number of open-reel recorders is a conventional three-head, four-track stereo configuration, plus a separate four-channel playback head with two additional amplifiers for playback only of four-channel tapes.

At this time, no four-channel cassette recorders have reached the market, but there have been public demonstrations of both four-track and eight-track quadraphonic cassette recorders.
.

Most home tape recorders operate at 7V2 ips (19 cm/sec) and 33/4 ips (9.5 cm/sec). The higher speed provides slightly extended high-frequency response, but many of today's open-reel recorders can cover the audible frequency range with equal effectiveness at either speed.

Nevertheless, a 7 1/2-ips recording is easier to edit and will almost always have a better S/N ratio, lower distortion, and lower flutter than one made at 3 3/4 ips on the same machine. Balancing this is the economy of the slower speed, which requires half as much tape for the same recording time. Some recorders also have a 1 7/8 ips (4.75 cm/sec) speed, principally for recording voice or non-critical musical material.
.

A few high-priced machines offer 15 ips (38 cm/sec), either in addition to the two normal speeds or replacing the slower one. This feature is usually found in recorders designed to handle 10 1/2-inch reels, which provide the same playing time as a 7-inch reel operating at 7 1/2 ips. Compatibility with tapes that have been made on professional machines, or which must be played on them, is the principal reason for using the 15-ips tape speed.
.

The cassette recorder operates only at 1 7/8 ips. However, by virtue of special head designs and tape formulations, many of them are capable of true high-fidelity performance. The best cassette machines compare favorably with a good open-reel recorder in listening quality.
.

Many manufacturers specify the time required to move a given length of tape in fast-forward or rewind. Some single-motor transports require as much as 3 minutes or more to handle a 1200-foot (370 m), 7-inch reel of standard 1.5-mil tape. Three-motor machines typically operate about twice as fast. With thinner tape (1 mil or 0.5 mil), either 1800 feet or 2400 feet can be wound on a 7-inch reel and proportionally longer times are required.
.

For a given playing time, most cassette recorders are comparable to three-motor open-reel machines in their fast speed operation. Times of 75 to 90 seconds are typical for a C-60 cassette (equivalent to 1200 feet of tape at 7V2 ips). A few cassette transports can move a C-60 tape in 60 seconds and a couple are as fast as 40 to 45 seconds.

Wow and flutter are the audible effects of frequency modulation of the program material, caused by uneven motion of the tape across the heads. When the speed fluctuation occurs at a low rate (under 10 Hz), the characteristic "wow" sound can be heard. This is especially apparent on tones of extended duration, such as the organ and the delay of piano notes. Faster rates, up to 200 Hz or more, are heard as "flutter"  - a slight muddying of the sound and in extreme cases, a "gargling" quality.

Wow and flutter measurements are frequently combined into a single flutter rating, expressed as a percentage of frequency modulation. For example, a 0.2% flutter will cause a 1000-Hz tone to vary between 998 Hz and 1002 Hz. Higher and lower frequencies will be affected proportionately. The audibility of flutter depends on several factors, including: magnitude, rate, and program material (duration and frequency range of tones).
.
The human ear is most sensitive to flutter affecting frequencies in the vicinity of 3000 Hz, which is why a frequency of 3150 Hz is now generally used for flutter measurement. Standard test tapes are recorded with a 3150-Hz tone that has a very low intrinsic flutter (typically less than 0.02%). The tape is played on the recorder and its output measured with a flutter meter. This is essentially a calibrated FM receiver, fix-tuned to 3150 Hz, with a meter indicating the percentage of frequency modulation.
.
Unweighted rms flutter measurements respond equally to flutter rates over a wide range (such as 0.5 Hz to 10 Hz for flutter, 10 Hz to 200 Hz for flutter, or 0.5 Hz to 200 Hz for a combined measurement). Since the most audibly objectionable flutter rates occur between 1 and 10 Hz, current IEEE standards call for a weighted peak flutter measurement, emphasizing that frequency range and reducing the contribution of higher and lower frequencies to the final measurement. Some recorder manufacturers use a similar weighting curve applied to rms rather than peak measurements. These are usually identified as "Wrms" flutter measurements.

Weighted readings are always less than unweighted readings, usually by about 20 to 30%. A peak measurement will always be greater than an rms measurement. A comparison among published flutter ratings for different recorders is only valid if the same technique was used in all cases.

Cassette recorders are tested in the same manner, except that presently available test cassettes have a residual flutter level between 0.1 % and 0.2%, which is more than that claimed for the latest recorder designs. To test these machines, a standard test tone is recorded and played back into the flutter meter. Some flutter is introduced when recording and some during playback, (these may add or cancel each other at different times). By taking several readings and averaging them, it is possible to establish an approximate flutter rating.

In multi-speed recorders, flutter is usually less at the higher speeds. Bi-directional recorders may show slight differences in flutter when running in forward and reverse directions due to variations in the tape tensioning and guidance system. As a rule these effects are minor.

In most cases, a flutter level of 0.1 % or less will not be audible. Most of the better open-reel recorders (and a few cassette decks) can meet this requirement. Typical good cassette recorders have 0.15% to 0.25% flutter, while low-priced open-reel machines fall in the same range. With some types of music this can be audible but, in general, it would be apparent only to a critical listener. Flutter levels exceeding 0.3% are not consistent with high-fidelity reproduction.
.

The dynamic range of a tape recorder is limited by the maximum level that can be recorded and played back with acceptably low distortion and by the residual noise level in the playback output. The ratio of these two levels, expressed in decibels, is the signal-to-noise ratio (S/N). Strictly speaking, it is the signal plus noise-to-noise ratio, (S + N)/N, but the difference is minor in this case.

Usually, a single figure (e.g., 55 dB) is given as the S/N rating. Implicit in such a rating is a specific (but often unstated) distortion level at maximum signal input. A total harmonic distortion (THD)of 3% is generally used as a reference for S/N rating of home tape machines. Like all hi-fi components, the distortion of a tape recorder increases with program level, especially near its maximum capability. However, one cannot assume that 3% THD will coincide with a "0 dB" or other indicated maximum recommended recording level on the machine's meters.
.

As a rule, the reference distortion will be reached with an input of +3 dB to +6 dB, allowing some reserve recording "headroom" for brief peaks that might not register on the meters. In the case of cassette recorders, the headroom is usually not more than 2 or 3 dB at middle frequencies (and sometimes considerably less) and reduces greatly at higher frequencies due to the greater recording equalization necessary to achieve a wide frequency response.

The audible noise level in the playback output consists mostly of hiss or wideband random noise. Usually there will be some low-frequency noise as well (such as power-line hum), but this is much less audible due to the characteristics of human hearing. An unweighted noise measurement responds equally to hum and hiss and may give an unduly pessimistic result in terms of the subjective character of the noise. Therefore, it is customary to "weight" the noise measurement to discriminate against the less audible low and high frequencies. Sometimes the weighting curve is specified (e.g., ANSI "A" weighting, etc.) but often it is not. As with flutter, S/N ratings can only be compared when they are based on the same reference distortion and noise weighting characteristics.
.

Tape-recorder distortions are difficult to summarize in a single specification, since they vary widely with level, frequency, and the nature of the test signals. At middle frequencies, such as 1000 Hz, the THD is easy to measure. At frequencies exceeding one half to one third of the recorder's maximum response frequency, the harmonic levels in the playback will be reduced or eliminated by the machine's inherently limited response so that THD cannot be used to measure nonlinearity at the higher audio frequencies where it is most serious. Two-tone intermodulation distortion (IM) measurements are needed for this but there is no universal standard for such tests at present.

The THD distortion ratings published for tape recorders (e.g., "less than 1%") can be assumed in the absence of other information to be measured with a 1000-Hz signal recorded at an indicated level of 0 dB on the recorder's meters. This is the procedure followed at Hirsch-Houck Laboratories and we believe it to be typical of industry practice. Since distortion is also affected by tape speed, it should be specified at each operating speed.

Like many other tape-recorder specifications, distortion can also be affected by the type of tape used. Fortunately, in the case of open-reel recorders, the differences are minor, within any one classification of tape ("standard," "low noise," etc.). Most manufacturers do not specify the tape used in their own tests, but our test reports do.

In the case of the cassette recorder, the tape is a critical factor and must be known for any meaningful interpretation of the ratings.
.

A frequency-response specification, as a minimum, should include the limits (e.g., 40-20,000 Hz), the variation over that range (±3 dB), and the tape speed (7 1/2 ips, 19 cm/sec). The frequency response also depends on the tape used, but with most open-reel recorders this is not very critical.

Most tape recorders, especially at the lower tape speeds, cannot achieve their rated frequency response at maximum recording level. This is due to saturation of the magnetic tape coating at the higher frequencies which are boosted by the recording equalization and appear as a roll-off of high-frequency response. It is customary to measure frequency response at a lower level, not more than -10 to -15 dB, to avoid high-frequency saturation problems.

Recorders differ considerably in their susceptibility to high-frequency tape saturation. The recording equalization, which is a key factor, is rarely (if ever) specified and cannot be measured from outside the machine. The easiest way to judge a recorder's high-frequency "headroom" is to measure its frequency response at 0 dB and at a lower level such as -20 dB. The smaller the difference in high-frequency response between the two measurements, the better the recorder in this respect.

In the Hirsch-Houck Laboratories' tests, we record at -20 dB using a sweeping oscillator and play back into a chart recorder synchronized to the frequency sweep. Similar results, with less resolution, can be obtained by recording spot frequencies and readings the playback levels on a meter.

Cassette recorders present special problems. The high-frequency recording equalization is greater than that used in open-reel machines and the test level should not exceed -20 dB. One manufacturer even recommends a -30 dB level for testing his machine. For best results, the recording bias should be matched to the tape characteristics. This is important even with open-reel recorders but is absolutely vital with cassette decks. Not only must the bias be set for the type of tape, but for the specific brand. A few recorder manufacturers specify the tapes for which their machines are set up at the factory, but most do not. Except for the two- or three-position switches used to set the bias for a general category of tape, external bias adjustments are rare in open-reel recorders and almost non-existent in cassette recorders. For the consumer, this means that he must restrict himself to the specific tape used by the manufacturer in his final adjustment of the machine, unless he is competent to re-adjust the bias himself. Most recorder manufacturers will be glad to suggest suitable tapes, if they are not listed in the instruction manual. Alternatively, the consumer can try several brands and standardize on the one that sounds best.

When using chromium-dioxide (Cr02) tape, it is only necessary to have a machine with a switch that selects the appropriate bias (and generally, equalization) for this tape. Although Cr02 tape can only be used on recorders equipped for it, it has the advantage of being magnetically interchangeable among manufacturers. Chrome cassettes, of any manufacture, are so nearly alike in their bias requirements that the user is not restricted to any particular brand.

A frequency-response rating of "40 to 20,000 Hz, ±3 dB" means that the output can vary a total of 6 dB from its maximum to its minimum, within that frequency range. Without an actual response curve (rarely supplied by the manufacturer, but included in most equipment test reports), one cannot assume that two recorders with identical ratings will sound alike. One may have a rising response at the high end and will sound bright. Another, still within the ±3 dB rating, could have a falling high-frequency response and sound subdued or even dull. Between these extremes there are an infinite number of response variations which can only be defined graphically. If the variation were small, such as ±1 dB, the ratings could be compared with greater validity, but most recorder manufacturers use a broader tolerance to encompass the wider range of variations found within an extended range of frequencies.

Another aspect of recorder frequency response, not always specified separately, is its playback response. This indicates the accuracy of its playback equalization and its suitability for playing commercially recorded tapes or tapes made on another recorder. Playback response is measured with a standard test tape, which usually has a relatively limited range compared to the coverage of most of today's recorders. Typically, these are 50-15,000 Hz at 7 1/2 ips, 50-7500 Hz at 3 3/4 ips, and 31.5-10,000 Hz for cassettes.
.

These are different manifestations of the same effect - leakage of a signal from one track of the tape to another. This leakage is largely a function of head design but can also occur in the wiring to the heads and in the recorder's electronics.

Both effects are usually specified at a middle frequency, such as 1000 Hz, although they vary with frequency. Crosstalk is the more serious, from a listening standpoint. It is a transfer of signal from one pair of tracks to the other. When playing the tape in the forward direction, the second pair of tracks are being played backwards, and the crosstalk will be in the form of noise or garbled sounds, with no relationship to the desired program.

When the leakage occurs between the two stereo channels in the same direction of tape travel, its only effect is to slightly dilute the audible separation of the program. Since any tape recorder is likely to have much better channel separation than the program being recorded actually requires, this "problem" is trivial. Typical specifications for a high-quality recorder might be: Crosstalk more than 60 dB at 1 kHz; separation more than 50 dB at 1 kHz.
.

The bias frequency is often specified, but has little direct bearing on the suitability of the recorder for home service. As long as it exceeds about five times the highest frequency to be recorded, there is no need to be unduly concerned about the actual bias frequency. Most recorders have a bias frequency between 75 kHz and 125 kHz.
.

.

There are a number of tape-recorder characteristics which are never mentioned in published specifications. Most of them are probably of little interest to the average user, who can generally obtain the information from the manufacturer if he wishes. A few items which may be meaningful to the home recordist deserve mention, however.
.

The level meters of many recorders are marked as "VU" meters, usually incorrectly. The abbreviation "VU" means "volume unit", and the characteristics of a "VU" meter are detailed in an IEEE/ANSI specification.

These include a frequency response within ±0.5 dB from 25 to 16,000 Hz, and a 0 VU reading corresponding to 1 milliwatt at a specified impedance (usually 600 ohms). When a signal which will result in a 0 VU reading is applied, the meter pointer should read 99% of that reading in 0.3 second, with an overshoot of 1 to 1.5%, and should return to its rest in approximately the same time when the signal is removed.

Typical home-recorder meters rarely have the ballistic characteristics of a VU meter and are more correctly described as "dB meters" (assuming that their dB calibrations are correct, which is not always the case). In our recorder tests, we apply toneburst signals to check the response of the meters. Some overshoot considerably but most are too slow and read substantially less than a true VU meter on transient signals.

It is important to know the general response characteristics of a recorder's meters to be able to allow sufficient reserve range for brief program peaks. Most meters, including true VU meters, are average-responding devices and will not indicate the peak level of the signal.

Some recorders have peak-reading meters, able to respond in a few milliseconds. In some units, the meters read the outputs of the recording amplifiers thereby including the effect of the equalization and reducing the likelihood of over recording a signal having strong high-frequency content.
.

Many recorders use electronic circuits to establish correct recording levels. In a number of cassette recorders, and a few open-reel machines, these are automatic-gain-control (a.g.c.) circuits, which may entirely replace manual recording level controls.

An a.g.c. circuit operates the recording inputs at maximum gain in the absence of a signal and quickly reduces the gain to accommodate any signal level. Although the attack is fast (a few milliseconds), the release time, (return to higher gain) is much slower, typically a number of seconds. Such a.g.c. systems are convenient for recording conferences where voices may originate at different distances, but are unsuitable for recording music. During pauses in the program, the upward surge of background noise is noticeable and frequently objectionable.

Limiters, found in some of the better cassette and open-reel recorders, are quite different in their effect. Although a limiter, as used in home recorders, may be very similar in its attack and release times to an a.g.c. system, it does not come into operation at normal program levels. The recording-level controls are set in the usual manner and, under most conditions, the limiter circuit has no effect. However, if the level exceeds the recorder's rated maximum (usually +1 to +2 dB on its meters), the limiter reduces the gain almost instantly. If the recording level is set too high, the result may be quite similar to an a.g.c. system, with "pumping" and audible variation of background noise. With correct levels, the action of a good limiter will not be detectable yet it can provide effective insurance against distortion from unexpectedly high signal levels.
.

The noise level of a good open-reel recorder, or a Dolby-equipped cassette recorder, is negligible for most practical purposes. When taping from records or FM radio, the noise in the input signal will usually exceed that introduced by the recorder.

A notable exception is the case where microphones are used for a "live" recording. Assuming that ambient background noise is low and that precautions have been taken to eliminate hum pickup, the noise (hiss) in the recorder's microphone preamplifiers will often be unpleasantly audible, especially when operating near maximum gain.

Although no recorder manufacturer we know of specifies the noise contribution of his microphone amplifiers, our laboratory measurements include a check of the increase in noise at maximum microphone gain as well as at lower settings. Some recorders will increase the noise by only 3 dB or less through the microphone inputs (as compared to the line inputs), while others add 20 dB or more.

Another important consideration, frequently omitted from recorder specifications, is the maximum signal capability of the microphone amplifiers. It is entirely possible on many recorders, when recording high-level performances such as rock groups, to overload the microphone amplifiers. The distortion this produces is independent of the gain-control settings or the meter readings. Obviously, the more input signal a recorder can handle without distortion, the less likely this is to occur. An external microphone attenuator can be used to prevent this and some recorders have built-in microphone attenuators.
.

Many recorders have outputs for low-impedance (4 to 16 ohms) stereo headphones. The output level is sometimes specified, either in millivolts or milliwatts. Without knowledge of the sensitivity of the specific headphones to be used, this is useful only for comparing recorders.

As a rule, headphone volume is fairly low and might not be suitable for monitoring a recording in the presence of the live sound, even if it is adequate for normal listening. Very few recorders will deliver satisfactory volume with high-impedance phones (200 to 600 ohms), so be sure to check the compatibility of the recorder and the phones you intend to use if this is of importance.
.

Many of the specifications for most tape recorders are devoted to what might be termed features rather than actual performance specifications. Some items, such as input and output levels and impedances, fit into both categories. With the possible exception of microphone sensitivity and impedance, there is little likelihood of difficulty with the interface between recorder and other components. There is a large overlap between recorder-input sensitivities and amplifier tape-output levels as well as between amplifier-input sensitivities and recorder-output levels.
.

Most recorders made for the European market have DIN connectors, usually in addition to the standard U.S. phono connectors. The DIN is a German standard system in which the input and output signals are carried by a single connector.

Compatible receivers and amplifiers have similar connectors and a single cable between recorder and amplifier makes all signal connections. The DIN input level to a recorder is much lower than standard line-input levels so that special adapters and attenuators are needed if an American-type amplifier or receiver is connected to a DIN input socket.
.

Reel size is standardized at 7 inches for home recorders, but 10 1/2-inch professional-size reels can be used on some models. Usually such recorders have a switch to adjust the reel braking for the two sizes. Automatic end-of-play shut off, including mechanical disengagement of the tape drive, is now found on almost all cassette and open-reel recorders.

In the more expensive units, the "Mic" and "Line" inputs have separate recording-level controls and can be mixed. Lower priced models, including most cassette recorders, transfer the recording inputs from "Line" to "Mic" when microphones are plugged in.

Better grade cassette recorders almost universally incorporate the Dolby "B" or some other noise reducing system, such as the ANRS used in JVC recorders. A few expensive open-reel recorders are also "Dolby-ized" the price differential for this feature is considerable in this case since the tape monitoring capability requires four Dolby circuits instead of the two used in cassette recorders.

A number of open-reel recorders, as well as most of the better cassette machines, have switchable bias for different tape formulations. In the case of open-reel machines, this allows for "standard" or "low noise" tapes. With cassette recorders, the choice is between ferric-oxide and chromium-dioxide tapes, although some provide for two grades of ferric oxide as well as Cr02. To obtain optimum results with Cr02, it is desirable to change the recording and playback equalization as well as the bias. The best cassette recorders do this; others change only bias and playback equalization, bias and recording equalization, or simply bias alone.

Most three-head open-reel recorders permit special effects such as sound-on-sound and echo recording, usually in mono only. Except for such applications as language instruction, these are probably of minor importance. In general, they are accomplished by playing back the output of one channel into the input of the other while the second is being recorded, together with new program material. Sometimes this calls for the use of external patch cables but many machines have front-panel switches to make the necessary interconnections.

An important variation on this is the synchronized recording capability of a few deluxe open-reel recorders. These can be operated so that the recording head of one channel serves as a playback head. A performer, listening on headphones, to previously recorded material can record a second track in synchronism with the first. With a four-channel recorder, this makes possible the creation of many musical effects which are typical of commercial recordings, but have in the past been unavailable to the home recordist.
.