By HERMAN BURSTEIN
Must you really demagnetize your tape deck's heads (and other metal parts contacting the tape) every so often? The answer which emerged from the various views expressed in my roundup on the subject in the April, 1981 Audio is that the need varies with different decks and situations.
We invited reader comment, and many illuminating letters have come in but still none so illuminating as to reveal one clear, shining, completely unassailable truth. The different viewpoints do increase our understanding of the matter, and here they are: To begin, Phil Sutterlin, of Ampex, Cupertino, Cal., suggests how the problem occurs: "Heads can be magnetized by an applied magnetic field or a current applied to a head's winding if this field or current is asymmetrical, thus containing a d.c. component, or if it is symmetrical but abruptly changes magnitude; Fourier analysis tells us that there is also a d.c. component in the latter case." Sutterlin continues: "Magnetic fields can be applied to heads by recorded tapes, head demagnetizers, or other less common sources. Magnetizing currents applied to a head winding can originate in purposely applied signals, namely audio, bias, and erase; d.c. leakage from the amplifier connected to the head, and transients which occur when turning the deck on and off.
"In my experience, heads are most often magnetized suddenly by mistakes such as: Turning the tape machine off while in the record mode (which can magnetize the erase and record heads), or improper use of a head demagnetizer (the extreme case being that of unplugging the demagnetizer while it is in contact with the head). In these cases, a demagnetized head can change into a magnetized one in a fraction of a second, assuming that current is in mid-cycle and therefore flowing through the demagnetizer or the head.
"Slower magnetization does occur from asymmetrical waveforms, whether audio, bias, or erase. For example, I have found that recording a signal with a 10% second-harmonic component--hence an asymmetrical waveform--causes magnetization." My April 1981 article stated that a magnetized head produces noise and treble loss. Sutterlin, rightly and importantly, points out that it also causes distortion. "Magnetized heads cause an increase in second-harmonic distortion. Since third-harmonic distortion (caused mainly by the tape) is usually dominant, a spectrum analyzer is required to measure the second harmonics. If a tape deck has very low distortion in its electronics, it is very easy to measure any performance degradation caused by magnetized heads by measuring second-harmonic distortion. Noise and treble losses due to a magnetized head are also measurable with a spectrum analyzer." I. W. Borner of Revox ( Regensdorf, Switzerland), is dubious that the asymmetrical waveform encountered in audio produces significant magnetization, and suggests that routine demagnetization, although widely performed, is based on ritual rather than necessity: "The theory of the asymmetrical waveform, which is supposed to be responsible for head magnetization, is difficult to accept. Would this not mean that the electronics in our recorders must be capable of passing the d.c. component resulting there from? Yes, the audio signal may be asymmetrical, but only in that a brief strong positive spike is followed by a longer but weaker negative half; the areas under each are equal, with no d.c. component.
"In querying the younger generation with experience in audio, I find that they have never come across a situation where demagnetization is necessary. Yet, they carry on the practice of regular demagnetization because this is what they have learned from the older generation." (Earlier in his letter, Borner agreed that the tape decks used by the "older generation" did frequently require demagnetization.) "In essence, I don't believe that magnetism in the heads is something that creeps up with time, like head wear. It is something that may happen as the result of careless handling of magnetized tools, or it may be caused by design shortcomings, for example: A d.c. surge through the record head when it is connected to the record electronics, solenoids which create a strong magnetic field near the heads, or suddenness of the oscillator's attack (and possible decay). Since we usually don't know how a recorder behaves in such respects, we have an uneasy feeling that an important recording may not come out as well as it should. The routine of demagnetizing heads is therefore performed with the same regularity as paying insurance premiums." Like Borner, John W. Dana ( Hamden, Conn.), is disinclined to believe that normal operation of a tape deck produces significant head magnetization; rather, he feels that normal operation tends to demagnetize the heads. "Unless d.c. is accidentally fed to the heads, I don't see how they can become more than minutely magnetized, no matter how much the machine is used. Magnetically recorded tape contains alternating north and south magnetized bits of equal and opposite intensity. If a head or guide were magnetized, it would be demagnetized by the alternating north and south magnetized bits as the tape is played.
"Along this line of reasoning, I believe that two-head decks, which use the same head for record and playback, would have no problem with head magnetization if the machine is used for recording at least once in a while. The alternating signal going through the head would create a strong alternating magnetic field, which would remove any residual magnetism, so long as the recording signal is strong enough to drive the VU meter into the red. One could occasionally record FM interstation noise, turn up the recording level so that the VU meter goes into the red, and gradually reduce the record level to zero." Assuming that a strong a.c. signal through the record-playback head is an effective demagnetizer, Dana's expedient strikes me as unnecessary.
Bias current is such a signal; its magnitude is roughly 10 times that of the peak audio signal ordinarily fed to the head.
Dan Dugan of Dan Dugan Sound Design ( San Francisco, Cal.), points to a prominent cause of head magnetization and stresses the need for an efficient demagnetizer. "In my experience, first as an amateur and then as a professional recordist, the most serious cause of head magnetization is the transient which occurs when the deck is switched into the record mode.
I had one machine, which didn't last long on the market, that would sometimes magnetize its head with just one actuation of the record button. But since 1970, many machines have included time-delay circuits in their bias oscillators to soften this effect.
"No discussion of head demagnetization is complete without mentioning the need for a really effective demagnetizer. I've found that most of the demagnetizers sold for consumer use are too weak to do any good, and the persons using them usually have no way of knowing whether demagnetization has been accomplished or not.
The R. B. Annis Company of Indianapolis, Ind. sells heavy-duty demagnetizers that really do the job, and it also sells magnetometers to measure head magnetization." Some types of heads appear to be more susceptible to magnetization than others. Nakamichi, in its "Technical Bulletin 4" published a few years ago, states: "Perhaps one of the most disturbing properties of ferrite heads is the spontaneous build-up of magnetism.... There are many stresses imposed on ferrite during the cutting processes because of its great hardness.
The high temperature process of melting the crystal glass onto ferrite, furthermore, compounds the stress. Even after cooling, the ferrite head is under constant stress because of the differing coefficients of expansion of ferrite and crystal glass. ... Whereas ... it was once believed that physical shock was required to induce magnetism in ferrite heads, it is now known that the constant stresses caused by changing ambient temperature have the same effect as shock." On the other hand, there is reason to believe that the disadvantage of ferrite heads, with respect to magnetization, may have been overcome to a significant extent, as demonstrated by Nakamichi which does use ferrite heads in its decks, although not for playback.
Still, the point remains: Depending on materials used and type construction, some heads may be more subject to magnetization than others.
We are reminded that not only heads but also other metallic components may become magnetized and endanger our recordings: The Standard Tape Manual, published by Standard Tape Laboratory, Inc. (26120 Eden Landing Rd. #5, Hayward, Cal.), states that magnetization is possible in "the guides, the capstan, and any metallic surface which contacts the tape.
Even 'stainless' materials called 'nonmagnetic' are capable of retaining some field and are therefore suspect."
The Manual goes on to warn of stray a.c. fields which may cause erasure for which the user may blame the heads. "Many top selling machines have been designed without any consideration by the engineering departments as to a.c. magnetic field contamination." As a horrible example, "one professional machine actually showed an a.c. field, caused by a misplaced solenoid, of 34 Gauss (in nontechnical language, a helluva strong field-H.B.). ... The field was concentrated at the supply reel side of the transport at the flutter idler guide." Further, "a poorly filtered d.c. supply may carry enough a.c. ripple to do a lot of damage to a tape passing near a solenoid or relay coil." We are indebted to Mike Hardwick of Westronix Hi-Fi ( Salem, Ore.), for calling our attention to the above information in the Standard Tape Manual.
In closing, John J. Swetko (Glen Oaks, N.Y.), writes: "During 18 years of tape deck ownership, which has ranged from tape manglers to top-notch decks, I have experienced only one episode of tape degradation. When I first opened my TEAC I immediately played a tape, and in one pass I immediately destroyed it. After that event, I have never failed to demagnetize after moving a deck. Other than that, my demagnetizing has been haphazard yet without ill effect." The correspondence indicates that what was once a definite threat is now less of a problem. Nevertheless, head magnetization can occur and caution should be exercised when conditions put recordings at risk.
(Source: Audio magazine, Sept. 1982)
Also see:
Cassettes: Focus On Shell Mechanics (Sept. 1981)
Open Reel Recorders (The Mechanism Of Magnetic Tape Erasure; Focus On Head Demagnetization) (April 1981)
Performance of High Energy in Magnetic Materials in Audio Cassette Recording Tapes (Sept. 1978)
All That Data: Tape Deck Frequency Response and Headroom (Jan. 1981)