CASSETTE TAPE PROGRESS--An expert looks at where technology is taking us.
ONE EXPERT'S VIEW OF PAST, PRESENT, AND FUTURE DEVELOPMENTS
By Robert Donadio
As recently as ten years ago, the making of recording tape was still something of a magical art, but the industry has made tremendous strides in the interim in response to consumer demand for ever-better tapes. This is nowhere more apparent than in cassette technology, in which, in three short years, we have seen the introduction of cobalt-enhanced fern c oxide, ferrichrome, and now a second generation of chromium dioxide. Further, several companies have already made and successfully tested all-metal tapes, and their market introduction awaits only the development of suit able machines to play them on.
The race to find the "perfect" magnetic-tape particle is hotter now than ever before, and the manufacturers of tape software are once again challenging the manufacturers of machine hard ware to catch up. Poised as we are on the brink of even greater things, it is perhaps a good time to ask a few leading questions: How far have we come? How far have we yet to go? And, particularly, in which direction does the future of the cassette lie? Despite the considerable progress of the past ten years, the major deficiency of the cassette format in comparison to open-reel studio tape still lies in the area of high-frequency output. Up to about 5,000 Hz, most cassettes are pretty much equal to open-reel, quite capable of reaching the required amplitude for even the most critical music recording. Beyond that, however, cassette performance begins to drop off.
Ferric Oxide
The history of the development of open-reel ferric-oxide tapes since the late 1940's has been one of increased signal-to-noise ratio (S/N). With the introduction of the Philips cassette, with its much slower tape speed, the emphasis has switched to high-frequency performance above 12,000 Hz, and progress over the past decade or so has been dramatic. At the low frequencies, the S/N of ferric-oxide cassette tapes has been improved (with successful cobalt treatment) from about 45 dB to more than 60 dB, and at the high frequencies the S/N has gone from less than 30 dB to over 40 dB.
Better Oxides. The drive to improve ferric-oxide tape formulations was sparked in 1970 with the introduction of chromium dioxide (CrO2) by Du Pont. In one dramatic step, CrO2 enhanced the potential S/N of cassettes by as much as 5 dB at the high end.
Ferrichrome was introduced several years later as an attempt to combine the virtues of ferric oxide with those of CrO2. On machines properly set up to accept it, ferrichrome too provided certain advantages. But there was more to come. The breakthrough for ferric oxide came in 1975, when the Japanese were successful in perfecting the cobalt treatment process and introduced a number of "chrome-substitute" tapes.
With coercivities high enough to use chromium-dioxide bias and well suited to the 70-microsecond (CrO2) time-constant playback equalization, these tapes equaled and in some ways surpassed the performance of the best chrome then available.
Cobalt enhancement, or "doping," involves chemistry that is simple in theory but proved at first to be very difficult to apply. Each ferric-oxide particle has a certain specific crystal structure made up of atoms of iron and atoms of oxygen. Most of these fit together in predictable ways, but a small percent age of the "sites"-the hookup points-in each crystal are ambivalent in that they can be filled either with oxygen or iron atoms (or any "impurity"). This ambivalence is actually un desirable, but it is possible to take advantage of it nonetheless by filling the ambivalent sites with cobalt (rather than iron or oxygen) atoms.
Cobalt improves the properties of each crystal and therefore the overall magnetic performance of the tape coating. In doping a crystal, however, the chemists were forcing the atoms to do things they didn't want to do, so the early formulations were very unstable.
Eventually, however, the scientists learned the proper stage at which to introduce the cobalt atoms and the proper conditions of temperature and pres sure; the result was the so-called "ferri-cobalt" tapes.
Better Binders. None of these tapes have yet achieved the theoretical ultimate performance we can expect from ferric oxide, however, for it can-in theory, at least-be improved about 7 dB at the high end by smoothing the tape surface and thus improving tape-
to-head contact (2 dB) and by enhancing magnetic properties through better particle orientation and faster, more controlled processing of the coating materials (5 dB).
To improve tape-to-head contact, the average roughness of the tape has to be reduced about 30 percent. Right now, the average tape-to-head separation is roughly twice the peak-to-peak variation in the surface roughness-about 20 microinches. This head-to-tape gap can be brought down to less than 8 microinches (it has already been done in video-tape formulations), making possible 2 to 2.5 dB more output at 15,000 Hz.
The 5-dB gain will come from improving the magnetic properties of the particles through physical means. In the history of magnetic tape, we have usually depended on the oxide manufacturer to do the work of improvement--and, indeed, we can still expect some contributions from that source.
Today, however, we're also exploring tape binders as a means of accomplishing at least part of this goal. Hereto fore, a binder system was only some thing to hold the oxide needles together on the substrate (the base film). In theory, however, the binder can and should not only hold them there, but hold them in a certain orientation. Powerful magnets are used to orient the oxide particles physically while the binder is still wet (unfortunately, not every particle winds up being properly orient ed). The oxide particles have two directions of movement within the wet coating: up-and-down and side-to-side. Be cause of the coating thickness, a particle has a great deal of difficulty moving in the vertical plane. Under the influence of the magnetic field it naturally wants to do that; just as naturally, we do not want it to. The best kind of binder system is therefore one that will re strict vertical motion of the oxide particles, yet allow them to move easily in "The race to find the ‘perfect' magnetic-tape particle is hotter now than ever before...." the horizontal plane. Using such a binder, if we move very quickly from the coating point to the orientation point the particles can be "frozen" more or less in the direction of tape travel. This longitudinal orientation directly affects the output signal because it increases low-frequency flux and reduces interference between the magnetized particles at high frequencies We could, theoretically, achieve total longitudinal orientation, but the best result achieved so far under laboratory conditions is 7 on a scale of 10. The best tapes now available have orientation factors of 3 on the same scale. By using new binder technology and better magnetic particles, the figure could be improved to 4.5.
Faster Milling. Milling is another critical stage in the process of preparing the oxide dispersion, and the more time the magnetic particles spend in the process the greater the chances are that they will be broken. Most of the effort in this part of the manufacturing process goes into developing milling methods that are faster and (consequently) less damaging to them.
This three-pronged approach-better oxides, better binders, and faster milling-will give us about 5 dB additional dynamic range for the ferric-oxide tapes at the high end, plus the 2 dB that comes from having a smoother surface.
That, theoretically, is all we can expect from the ferric tapes-cobalt-doped, epitaxial, or whatever-until we get into pure-metal powders. Still, this is very respectable performance when considered in the light of the other ad vantages of ferric oxide, and if the problems of bias consistency can be solved worldwide, there is a bright fu ture for the cobalt-enhanced ferrics, especially at the 120-microsecond play back equalization.
Chromium Dioxide
The other major category of cassette tapes, chromium dioxide, presents a far more encouraging picture even though it has had only a fraction of the developmental attention paid to it that the ferric tapes have enjoyed. The problem from the beginning has been that chrome was too good. At its introduction it was so far ahead of the other tapes then in use that its manufacturers were not motivated to invest in any re search and development efforts to improve it. Though research on ferric oxide goes back more than twenty years, CrO2 has seen only one improvement in its whole history.
The development direction for chrome from this point on is very likely similar to that for ferric tapes-creating a more uniform particle with a higher "crystal energy," which means more coercivity (high-frequency response) and more remanence (overall output). Fortunately for chromium di oxide, this is easy to accomplish-the particle is a lot more controllable than that of iron oxide. So far, it has been possible to increase chrome's coercivity without changing the bias from the "classic" chrome bias. This provides an increase in the short-wavelength response, thus adding 4 to 5 dB to the output at 15,000 Hz.
It is impossible to predict the ultimate practical performance of the chrome formulation, but the next theoretical goal is an additional 3 dB at low and middle frequencies and another 6 dB in the range between 14,000 and 20,000 Hz. And this is on top of the achievements of second-generation chrome, which has already demonstrated greater dynamic range and more high-frequency headroom than even the best chrome substitutes on the market today.
I do not mean to be critical of cobalt-doped iron-oxide tapes, but you cannot break the laws of nature. When we look at the theoretical limitations of ferric and chrome tapes, we find that the potential for chrome is way ahead.
This is true partly because CrO2 is the precisely controllable result of technology (CrO2 does not occur in nature), while ferric oxide has to be grown or "cultured" in much the same way as yoghurt or bread dough.
Perhaps this idea of an "unnatural" oxide gave some credence to the "wear scare" that has hindered the sale of chrome cassettes over the past several years. The rumor that CrO2 wears heads at a faster rate than iron oxide started shortly after the introduction of the chrome-substitute cassettes and was widely circulated among audiophiles, retailers, and even tape-deck manufacturers. According to all avail able current research, just the reverse is true: cobalt-treated ferric-oxide tapes actually wear heads faster than any current chrome formulation.
Which is not to say that either type presents a serious wear problem. If you operate your machine an average of 150 hours a year for three years, you may notice a 2- to 3-dB deterioration of the signal at 8,000 Hz, depending on the composition of your tape head-and that "s with the most abrasive ferric-oxide tape. And so it turns out that the "issue"' of recorder head wear was never really an issue at all.
Corning Up
What performance levels can we expect in tomorrow's tapes? Certainly there will be improvements in chrome-substitute ferric-oxide formulations, though perhaps not to the point of reaching the theoretical limit of 7 dB more "head room" at high frequencies immediately. However, they should soon be capable of rivaling second-generation "super-chrome" formulations.
Furthermore, ferrichrome. perhaps with some improvements, should continue to expand its popularity as more and more cassette decks add bias/ equalization switching especially for it.
And, of course, a steady-and rapid improvement in chromium dioxide seems to be guaranteed now that the industry is finally committed to realizing its full potential.
But the biggest tape news of the near future will not be made by either ferric or chrome formulations but by pure-metal tapes or digital technology. A bit of a race is shaping up here, with the decision going to whoever is first to develop successful-best and cheapest, that is-hardware. Which one triumphs is not really all that important. Al though the technologies are vastly different, the goal remains the same: an additional 10 dB of S/N more or less "across the board."
Pure Metal. The figure of 10 dB is, at least, what many future manufacturers of pure-metal tapes are aiming for. It may. prove to be a trifle optimistic, but it is necessary to identify some target so that equipment manufacturers can plan and design accordingly. Reports 7 "In tape development, we've come farther in the Seventies than in any comparable previous period." of early work indicate that a 5-dB improvement from 20 to 20,000 Hz has al ready been achieved. Furthermore, 10 dB has evidently been squeezed out at high frequencies in the laboratories, though with a modest sacrifice of improvement at lower frequencies.
As far as we know, no one has yet managed to demonstrate the full 10-dB, all-frequencies increase with the consistency mass production requires, but the tape industry as a whole is probably 50 to 80 percent of the way to the goal.
Once it gets there it will be up to the machine manufacturers to produce affordable equipment capable of applying several times the present-day bias levels to the tape. This will entail development of entirely new head technologies and characteristics.
Digital. It is safe to assume that the hardware manufacturers would prefer a system in which the tape is not so critical a factor-a digital system, in other words-whereas most tape com panies would be willing z to go either way. The demands made on the tape would be strikingly different for the two systems, however. Initially, at least, a tape for digital audio recording would be an easier product to design and manufacture than even a cobalt-
treated ferric-oxide tape-initially be cause the ultimate product will depend on what decisions the system designers make. In the matter of sampling rate, for example, it will make a major difference in tape requirements if the signal waveform is sampled and quantized 60,000 times a second rather than, say, 40,000 times a second [see this month's "Tape Talk"]. The packing density available from the tape and the permissible occurrence of signal dropouts will be critical.
Actually, the tape industry need expect no major problems in creating a workable digital audio tape. Historically, analog recording tapes have maintained-and had to maintain-a higher standard of performance than tapes in tended for digital applications. Digital technique essentially involves working at a fixed frequency and either saturating the oxide particles or not recording on them at all. This is easier to cope with than the analog situation, in which the frequency bandwidth is fairly wide and the tape must meet higher criteria for noise and distortion.
Our considered opinion is that chromium dioxide is, once again, the best presently available magnetic material for digital applications. Far more bits of information per inch can be record ed onto and retrieved from chromium dioxide than can be handled by other materials. (When I joined BASF some years ago, computers worked with a packing density of 556 bits per inch of tape track. Today we're up to 6,250 bits per inch, and this figure is determined by the digital electronics rather than the tape. which is capable of more.) Chrome has all the characteristics of packing density, physical smoothness, and especially consistency that we'll require from a digital medium.
THIS year, magnetic recording tape will be forty-six years old; the cassette concept will barely have turned twenty. In tape development, we've come farther in the Seventies than in any comparable previous period. There's no doubt that the rate of improvement in this recording medium is accelerating. We have seen the design of tape change from near-alchemy to a science, and this is only the beginning. The future for the cassette-and for the ultimate satisfaction of the consumer-looks very bright indeed.
Robert Donadio is Technical Development Manager at BASF Systems, where he has worked for the past twelve years, primarily on the development of new tape formulations.
Also see:
Source: Stereo Review (USA magazine)