TAPE GUIDE (Jan. 1986)

Analog Versus Digital

Q. If I make a cassette or open-reel recording of a Compact Disc, the music sounds more like a phono record--softer, less harsh. Does audio tape round off the transients? Is it possible that digital faithfully records the hard edges that exist in all live music? If I sit close to an orchestra, their instruments can sound very harsh at times, especially when they play loud.

None of my audiophile friends like the sound of digital. Could it be that they like the way magnetic tape affects the music in an analog recording? Isn't the tape that is used for a digital recording magnetic? If so, why would digital sound different? How does digital eliminate tape hiss?

-Anthony Mouldin; Lewisville, Tex.

A. Analog cassette decks produce a fairly sharp cutoff above a frequency which is usually under 20 kHz; often this cutoff point is as low as 14 or 15 kHz, depending on the deck, on whether it has a separate head for playback, and on the tape used.

Open-reel decks perform similarly, except that the cutoff frequency tends to be higher, depending on tape speed; the lower the speed, the lower the cutoff frequency. These cutoffs may be eliminating the frequencies that produce harshness to your ears. In the case of CD, cutoff does not begin until beyond 20 kHz. Hence CD may be presenting, to a substantial extent, frequencies that are subdued or eliminated when copied on analog tape by your cassette or open-reel deck. These very high frequencies may represent the original music, or they may be distortion products that occur somewhere in the recording chain. It should be noted that many CDs are based on analog tapes which may have been recorded without the maximum possible attention to high fidelity.

Yes, magnetic tape is used for digital recording. The standards set for digital recording preserve frequency response to at least 20 kHz (although not past 22 kHz), so that high treble response is typically more extended than on analog tapes or phono records. And digital recording has a much higher signal-to-noise ratio, about 98 dB unweighted, compared with S/N ratios usually in the 60s or 70s for the other media.

Analog recordings attempt to reproduce the infinite number of gradations of sound by an equally infinite number of gradations in the recording medium-in the case of tape, minute variations in magnetic flux. Such systems are unable to distinguish between very small real signals and the accidental "signals" that come from irregularity in the tape and its magnetization; these irregularities are then heard as hiss.

Digital recordings first encode the sound by a finite (but very large) number of amplitude values, and then encode those values as binary ones and zeroes. The playback system must only distinguish between the two signal values representing ones and zeroes, and these two values can be made very different from each other. (A "one," for example, might be a pulse that saturates the tape, and a "zero" might be another saturated pulse of opposite polarity; each pulse would last for a specified period of time.) The minute irregularities that produce hiss in analog systems are simply ignored by digital systems. In digital recordings, noise depends on the number of bits used to quantize the recorded signal. Sixteen bits are used in CD recordings, which signifies that the audio signal can be represented by 65,536 (or 216) different levels. The more bits, the less noise.

Mail Problem

Q. I exchange cassette tapes by airmail with friends in Switzerland and France. When I play a tape sent by them, I am astonished at how dull and weary the tape sounds. Is it possible that the tapes have been demagnetized by the Postal Service's equipment or by the X-ray devices used by the U.S. Customs Service?

-Kevin P. Moylan; Long Beach, Cal.

A. Your question is a new one to me, and I'm not sure of the answer; perhaps other readers can comment. I suppose it is possible for X-ray inspection to have a deleterious effect, although I have been led to believe otherwise. Possibly the foreign tapes were recorded at slightly too high a speed, so that they play back slow on your equipment and therefore sound "dull and weary." Or perhaps your deck is running slow.

To find out whether cassettes are being harmed in the mail, send one of your own cassettes to a friend abroad, and ask him to mail it back. If it has been harmed, consult the postal authorities about precautions you and your friends might take in the future.

Erase-Head Retraction

Q. Why is it that almost all cassette decks (with the exception of a couple of old portable mono units) permit the erase head to be engaged in the playback mode? Common sense indicates that this causes unnecessary head and tape wear. Or is there a reason that I'm unaware of?

-Mitch Bradford; Gypsum, Colo.

A. One probable reason is that of cost; it would be more expensive to design and make a mechanism for disengaging the erase head in playback but not in recording. Second, it may well be that tape alignment would suffer with the erase head disengaged; the tape might skew more and/or differently. Third, I think the problem of head and tape wear is negligible, so far as the erase head is concerned.

Out, Damned Spot

Q. A few months after I bought a three-head tape deck, I noticed a rust colored spot on the record head. Although the deck's performance hasn't been affected substantially, sometimes it seems I can't record at the same levels I used to without getting distortion. I have tried many kinds of head cleaner, but the spot remains. Does it affect my deck's performance? How can I get rid of it?

-Albert Saldarriaga; Mesa, Ariz.

A. Normally a rust-colored spot on a tape head would be produced by tape oxide that has shedded on the head.

This would be easily removed by isopropyl alcohol, special tape-head cleaner, etc. But since you say the spot isn't removable, it can't be due to tape oxide. It could be an imperfection in the head, which might account for your inability to record at adequately high levels. Your best course, if the deck is under warranty, is to take it to your dealer or to an authorized service shop for repair.

Recording at Low Levels

Q. I own a Nakamichi 1000ZXL cassette deck, which I use with a dbx 224 noise-reduction unit. Because the dbx unit virtually eliminates tape hiss, I have been recording pop and rock music from discs at relatively low levels-with average VU readings of 10 dB and peaks of 0 dB. I have, however, noticed a print-through problem during quiet passages of the music on tapes that were recorded about one year ago. How do I eliminate or reduce this problem? I don't think I can record at a lower level. I use C-90 cassettes. Would a C-60 reduce the print-through? Also, I'm in doubt as to whether it's better to record at a relatively high level to avoid the tape noise floor, or to record at a relatively low level to avoid tape saturation, print-through, etc. If I record at a low level, will the tape eventually self-erase itself? I am under the impression that, since tape-deck frequency response specifications are usually given at -20 dB, the lower the level used during recording the better will be the frequency response of the recording Is this true?

-Robert R. Maigetter; Kewaunee, Wisc.

A. There isn't very much you can do to get rid of the print-through that you have already incurred. It may help somewhat to put such a tape through fast-wind and rewind once or twice before playing it, particularly after a long period of storage. For the future, it may help to use somewhat thicker tapes, such as C-60, which provide more protection against print-through. It may also help to try recording at slightly lower levels, perhaps 3 dB or so lower After all, with a very high-quality deck such as yours, and with dbx, you probably have S/N in the neighborhood of 90 dB, and it seems quite likely you could give away a few dB of signal-to-noise ratio without an appreciable in-crease in audible noise. Once you have S/N of about 70 dB, you have very quiet tape reproduction unless you play at thunderous levels. So you may have as much as 15 to 20 dB of S/N to play around with.

Recording at a lower level will not cause self-erasure, which increases with frequency, not level. Recording at lower levels does tend to improve high-frequency response, because it avoids tape saturation. However, it is not necessary or advisable to record in the vicinity of-20 dB, because most pro gram material has a natural drop in the high-frequency region-down 10 dB or more by the time 10 kHz is reached.

This offsets the effect of treble boost in recording and thereby helps to avoid tape saturation. On the other hand, some program material may have an unusual amount of high-end energy, and in such cases it is advisable to reduce recording level a few dB; how much to reduce it is a matter of judgment guided by what one hears.

(Source: Audio magazine, Jan. 1986, HERMAN BURSTEIN)

= = = =