ALMOST COINCIDENT with the introduction of the 33 1/3 long-playing record, magnetic tape became the standard medium for producing a master recording. Then as now, it was valued for its ease of handling and editing, wide frequency and dynamic range, and low distortion. On the other hand, there were also some problems like tape hiss, print-through, and crosstalk. That was in 1949, and although there were advances in magnetic head structure, tape oxides, and standardization of equalization, these noise problems still plagued the industry for 17 long years, until in 1966 a young physicist named Ray Dolby introduced a revolutionary tape-noise-reduction system.
The rest is history. Although a few die-hard skeptics still view the Dolby System with a jaundiced eye, most of the recording industry has enthusiastically endorsed it, and it is in use throughout the world ... even behind the Iron Curtain! Most record companies here and abroad use the Dolby A301 noise reduction units in multiples. To name just two, London/Decca has over 50 units, Columbia something of the order of 80. Dolby has become a common word in the lexicon of every knowledgeable audiophile. Most of them have some idea of how the system works and they own and enjoy disk recordings free of tape hiss, cut from Dolby System tape masters.
While these audiophiles rightly regarded the Dolby System as the "tool" of the professional recording engineer, they were not unaware of the consumer-oriented potentialities of the system. The advantages of a noise-reduction system for home recording and for pre-recorded tapes were obvious. The question was whether a much simpler and far less expensive system could be derived from the highly sophisticated Dolby A301 unit. Dr. Dolby designed just such a system, and called his home-type noise-reduction system the "B" Type. A little later on I'll explain the differences between the two systems.
"B" Type in KLH Recorder
At any rate, the first result of Dr. Dolby's venture into the consumer field was a licensing agreement with KLH to incorporate the "B" Type system into a tape machine of their manufacture. The $650 KLH Model 40 tape deck was introduced in 1969, as was the Model 41, a simpler and less expensive ( $229) tape deck produced in Japan. Both tape machines were well received by the hi-fi press at demonstrations which effectively proved the worth of "B" Type noise reduction. When sufficient recorders were in the hands of the public to get a meaningful reaction, it was of enthusiastic acceptance of the system. Thus was consumer-type tape noise reduction launched.
However, it was obvious that not everyone was willing to buy a tape recorder in order to obtain the Dolby "B" system--especially those who already owned machines. What was needed was a "black box," an externalized version of what was in the KLH recorders, so the "B" Type noise reduction could be used with the kind of tape machines owned by most audiophiles.
On February 10th, 1970 the Advent Corp. of Cambridge, Mass., a licensee of Dolby Laboratories, introduced their version of the "black box" and demonstrated it to the hi-fi press and to some record company executives. They call their unit a " Tape Noise Reduction Center." Reaction to the device and to the idea was enthusiastic, especially when it was realized that this independent unit made tape noise reduction possible with any format-open reel, 8-track cartridge, or cassette and paved the way for "Dolby-ized" pre-recorded tapes in these formats.
At this point I think it would be pertinent to review briefly the operation of the Dolby A301, the professional system, and how the "B" Type works.
Essentially the Dolby A301 is a highly sophisticated type of compression/expansion system, with some elegant solutions to the problems that have plagued this type of device for many years. For one thing, previous compression/expansion systems operated over the whole frequency band at all signal levels and under dynamic conditions one could hear the "swishing" and "breathing" sounds characteristic of that kind of circuitry.
The A301 is set up to work on low-level signal components over four independent frequency bands. The bands are set for compression thresholds of 40 dB below peak operating level. The bands are divided as follows; Band One, 80-Hz lowpass; Band Two, 80-Hz-3-kHz band-pass; Band Three, 3-kHz high-pass; Band Four, 9-kHz high-pass. Band One provides noise reduction in the hum and rumble frequency range; Band Two in the mid audio range (broadband noise, crosstalk, print-through); Bands Three and Four, in the hiss range. All the bands work together, in varying degrees of momentary noise reduction in their respective frequency ranges. The A301 is in two sections. One is the record processor, the other the playback processor. It is important to remember that high-level signals pass through the system unaltered, the masking effect of the ear making inaudible any background noise at those amplitudes. The noise that bothers us is that during low-level passages. In the record processor, all low-level signals are boosted 10 dB up to about 5 kHz. Above 5 kHz, the boost rises smoothly to 15 dB at 15 kHz and then levels out. This boosting (compression) occurs before the recording.
The processed signals are recorded on a typical professional recorder, which does not require any special modification for use with the Dolby System. In the playback processor of the A301, the recording is attenuated in a complementary way-in exact proportion to the boosted signals. At the same time, noise acquired in the audio channel (tape machine, etc.) is reduced. The amount of noise reduction is the same as the boost in the compressor ... 10 dB up to 5 kHz, rising to 15 dB at 15 kHz. So the A301 reduces the tape hiss, hum, print-through, and crosstalk to levels which in general recording practice are inaudible. This is accomplished with no degradation of frequency response or added distortion.
Nor can the action of the system be heard--there is no "breathing" or "swishing." The Dolby "B" Type noise-reduction system works in the same fashion as the A301, except that instead of four independent frequency bands, the "B" Type has one band designed to reduce tape hiss. From the earliest days of home recording and the earliest pre-recorded tapes, tape hiss has been the bane of the tape enthusiast's life. It has been the greatest barrier to the growth of tape recording and has restricted the sales of pre-recorded tapes. The problems of print-through and crosstalk in home tapes has never been of the magnitude of the hiss, and in recent years their effect has been negligible. The "B" Type system gives 8 dB of noise reduction at 2 kHz, rising to 10 dB at 5 khz. As in the A301, the "B" Type works only on low-level signals, boosts them before recording, attenuates them in complementary fashion on playback, and reduces tape hiss at the same time. The "B" Type system employs a record and a playback processor just like the A301, and there is no degradation of the audio signals in any parameter. In short, the system is a "kissin' cousin" of the professional A301, except it works on a single-band basis for the reduction of tape hiss.
Prototype Systems
When Dr. Dolby decided to expand the "B" Type system beyond its use in the KLH recorders he had his laboratory make up some prototype "B" system boxes for various interested parties. I was fortunate enough to be given two units ( need two for four-channel recording) and I have had a chance to use them for some months now. I also have a prototype Advent " Tape Noise Reduction Center," and it is a handsome unit and it works flawlessly. The Dolby boxes are sleekly professional .in appearance, and using them has taught me what to expect in the way of performance from the "B" Type system. The Advent unit was not found wanting in any respect, and as a unit designed for the consumer, it has some convenience facilities lacking on the Dolby boxes.
Quoting Advent on their "B" Type unit, "it is intended to be inserted in a component stereo system between an amplifier (or receiver) and a tape recorder. It becomes a unified tape recording control center with the following features: Separate input-level controls on both stereo channels for both microphone and line inputs. These maintain input-mixing capabilities for any recorder, and add these to any recorders presently lacking them; a master recording-level control that governs both stereo channels and allows recording level to be set without disturbing the balance of stereo channels or individual inputs; output level controls for each channel that permit matching the requirements of any pre amp, amplifier, or receiver; a multiplex filter switch that prevents recording interference from inadequate suppression of multiplex-carrier or pilot-tone frequencies by a tuner; complete calibration facilities for optimum use of the Dolby System with any recorder. These include calibration meters, an internal test-tone oscillator, and two "Dolby Level" tapes ( open reel and cassette) that enable the unit to be set to a standard characteristic for all "Dolby-ized" tapes, including prerecorded commercial releases. There are headphone output and source-tape monitor switches."
What Advent left out was a playback tone level control on each channel, and a switch for each channel that in one position takes the Dolby noise reduction out of the circuit, and makes it operative in the other position at which point a tiny signal light goes on next to each switch. Setting up to use the Advent unit is fairly easy. Here is a typical situation: you feed the tape outputs of a pre-amp into the line inputs on the back panel of the Advent (all connections are standard RCA, except phone jacks for the microphones); from the Advent you plug into the line inputs on a tape recorder. From the outputs of the tape machine you plug into the "from tape" receptacles on the Advent. The "amp-out" on the Advent is connected to "tape-monitor in" on the pre-amp. Now you are all set to calibrate the system. For this you need special tapes ( which will be supplied by Advent with each unit). The important thing is that the tapes must have a certain level of magnetic flux at a certain frequency.
For reel-to-reel recorders if you want to, operate at 7 1/2 ips, a standard Ampex reproduce alignment tape will serve nicely. The reference flux level is 185 nano-webers per meter at 700 Hz. You play this tape back on your machine and by means of the playback-tone level controls of the Advent you set the needle of the calibration meters to the inscribed mark.
Having done this, you do not again touch these controls, nor the output-gain controls on your tape recorder. For playback of "B" Type tapes this is all you need to do. If you want to record a Dolby tape with your Advent unit, you place a blank reel of tape on your recorder, switch on the internal oscillator on the Advent, place the input-gain controls on your recorder to some arbitrary position, say at the half way point of rotation, turn the record calibration level pots on the read of. the Advent about half way. Now record 30 or 40 seconds of the signal.
Switch to the monitor positions on the Advent and play back the signal you have just recorded. The calibration meters should deflect to the inscribed mark. If they do not, you must adjust either the tape recorder input-gain controls or the Advent's record-calibration pots for either more or less signal as the case may be.
Record at the new setting and check the meters on playback. Once the meters are calibrated, the calibration pots on the Advent and the input gain controls on your recorder should not be moved. Control of recordings is by the master record control on the Advent, and you still set your overall recording levels by the VU meters on your recorder. Sounds a little complicated but it isn't once you have done it a few times. If you intend to use a cassette machine the calibration procedure is the same, except that you must use the correct cassette calibration tape with a reference flux level of 200 nano webers per meter at 440 Hz.
How it Performs
What is it like using a "B" Type noise reduction system? It is a fascinating experience and at first a bit eerie. Mr. Marc Aubort of Dolby Laboratories and Elite Records was kind enough to furnish me with some "B" Type recordings, made from Dolby A301 masters. In a solo piano recording that was superbly clean, when you got to the quiet sections-the extreme pianissimos, and long rests in the music--the absence of hiss was almost unreal! As you get used to it you notice the much wider dynamic range, the extreme clarity of the recording. It is like exploring a whole new world. In the last movement of the Brahms 3rd symphony, it is startling to hear the big orchestra in a full forte, and then a moment later a pianissimo and some rests-without a smidgin of hiss to mar your enjoyment. Needless to say, you can record FM "off-the-air" and from records with noise reduction, but you must remember that the signal-to-noise ratio will be only as good as the medium from which you are recording.
The Dolby System cannot remove noise retroactively! However, you may still like the results for at least your tape machine will not be adding any noise of its own.
If you are fortunate enough to make some live recordings, then you will get the full benefit of the noise reduction. In this respect, I think Advent might have placed the microphone input jacks on the front panel, rather than the rear. I have made some live recordings with the Advent with superbly quiet results. However, my most exciting live recording thus far, was one in which I used the Dolby prototype box. I was fortunate enough to record the Suffolk Symphonic Orchestra, an estimable group of 75 musicians. The concert was held in a school auditorium with pretty fair acoustics, although the stage treatment dampened some instruments too much. The program was ideal for testing the Dolby system-a Mozart "Figaro" overture, a cello concerto by an old Russian Romantic, Davidov, with a brilliant young soloist, Jeffrey Solow, who made his Carnegie Hall debut shortly after this concert. And, ambitious wonder, the complete 1947 version of Stravinsky's "Petrouchka"! Now the Dolby boxes have no mike inputs ... just line only. I solved this by using the great Ampex AM-10 mixer, which among other things has a master gain control over both channels, two big VU meters, with a range switch to work at plus 4, 8, or 12 dBm, and best of all the input channels can be switched to either "A" or "B" channel or both. I used my ReVox two track machine so I could edit, and was working at 7 1/2 ips since the Dolby was calibrated for that speed. I used the fine Schoeps condenser mikes for left and right, and that superbly tight cardioid, the Electro-Voice RE-15 in the middle, split left and right so that on playback I would have a nice phantom center channel. Of course it also allowed better control on the cello in the concerto. The results were just terrific. There were many sections in the concerto that were quiet and Petrouchka was quite unbelievable in the great dynamic range I achieved, and the clarity of everything.
There are many rests in Petrouchka and it was just great not to hear illusion destroying hiss. When I had a chance a little later, I played the concert back through the Advent, and having calibrated it properly, the noise reduction worked like a charm. One of the things you can do with a Dolby tape is make a one-to-one copy to another tape machine without going through a noise reduction unit. I did this and then played the copy through the Advent unit to achieve noise reduction. Nice to know this if you want to send a Dolby copy to a friend who has the means to restore it via a "B" Type box.
A Deluxe Accessory Of course, it is very obvious that this Advent box in a deluxe unit and it will sell for about $250. It is equally obvious that a simple playback-only box, suitable for use with pre-recorded open-reel or cassette tapes would be a desirable product. This is especially true with cassettes, whose terrible hiss has kept them from consideration as a medium for high quality sound. I took the Petrouchka I had recorded and transferred it to cassette, using TDK "SD" and a Wollensak 4700 cassette deck, a fine unit with good motion and very low hum. The playback of the cassette was an eye-opener-not quite as quiet as the open-reel 7 1/2-ips master, but incredible for a cassette.
The quality was amazing too, in terms of dynamics, frequency response, and transient response. Many of my friends who have heard it can hardly believe this is coming from a cassette. Getting back to the playback-only "B" Type box, Advent has this in mind as a product too, and before long. Dr. Dolby has come up with some new developments in this respect and it is reportedly possible that such a unit can be made to sell to the consumer for between $50 and $100. If this is so, this will lend great impetus to the possibilities for Dolby-ized prerecorded tapes in every format. In fact, I am very pleased to report to you that the ball is already rolling. Jac Holzman of Elektra Records, who was first on the market with a Dolby disk, is going to issue a Dolby-ized "B" Type cassette of Sibelius' "Legends of Lemmenkainen." Seymour Solomon, president of Vanguard Records and another Dolby pioneer, has told me that within two months he will issue a limited number of Dolbyized "B" Type open-reel and cassette recordings.
I know of one other record company which is tottering on the brink, and I'm pretty sure will have agreed to put out Dolby tapes by the time you read this.
It is the old story of the chicken and the egg that we went through in the early days of stereo. As soon as a few of the Advents reach the market, and the public hears how great an improvement noise reduction makes in their music, things ought to open up, and the major record companies will hop on the bandwagon.
It is an idea whose time has definitely come. While it is true that one of the advantages of the Dolby system is that you can get top-quality results by recording at slower speeds and thus save on tape, I think that's fine; but the main thrust is that at last we will be able to buy pre-recorded tapes unsullied by that old demon hiss!
(Audio magazine, May 1970; Bert Whyte)
= = = =