Archive of the former Yahoo!Groups mailing list: The Mellotron Group

 

 

 

"First, what I'm about to write will be viewed by many as controversial, by others as wrong, by still others as hopefully insightful. But I have no doubt that, because what I'll say flies in the face of "common wisdom", "standard practices" and even many maintenance manuals, there will be those that just ignore all of this. That's fine, do what makes you feel good, but there is at least some science behind my statements below.

Several decades ago I did some in-depth research into the effectiveness, and necessity of demagnetizing recorder parts as part of routine maintenance. I won't bore with details, but part of the research involved renting a lab-grade hall-effect probe gaussmeter in order to measure DC magnetic fields, analyzing their effect on the recording process, and their effect on pre-recorded test signals. In addition to the gaussmeter, a swept spectrum analyzer was used, as was a low distortion sine generator, and several strong permanent magnets as DC field sources.

In short, what I found is this:

1. Operating a tape recorder does not cause anything in the tape path to become magnetized. Not even a tiny bit.
2. All parts including heads have made from materials with a natural neutral magnetic state, to which they tend to return even if deliberately magnetized.
3. The magnetic flux required to affect a recorded tape is much higher than anything anywhere around the tape recorder, except in the erase and record head gaps. The effect of a steady-state magnetic field (a DC field) on recorded tape is first noticed as partial erasure of high frequencies, and that occurred with a field somewhere above 200 Gauss. Tape has very high coercivity by design, meaning you need a high field strength to cross the threshold above which the tape will become magnetized. That's what the bias oscillator does. By being many times hotter than the actual audio signal on the head, the bias forces the signal over the coercivity threshold into the more "linear" range of the tape. But that means it's harder to erase tape than you might think. Typical guide and head residuals were in the zero to 20 gauss range, far below tape coercivity levels.

4. Erase and record heads are self-demagnetized by the bias oscillator, which provides an AC field in the gap that is thousands of times higher than what a de-magnetizer can induce.

5. The field found on recorded tape is incapable of magnetizing anything. It is minuscule, and when the tape moves, it becomes an AC field, which would demagnetize if it had any effect.

6. The effect of a DC field near or in a record head produces two measurable results: First, a significant elevation of even-order harmonic distortion, and second, an elevation in low frequency noise (a sort of gravelly sounding noise). But again, the field has to be fairly high for this to happen. You need a spectrum analyzer to differentiate even harmonics from the normal odd harmonics created in the recording process.

7. If tape machine parts become magnetized, there is a defect in manufacture (the material doesn't have a natural, neutral magnetic state), or a defect in the tape machine, either due to a failure or design flaw. Asymmetrical bias waveforms will present a DC component, any DC on a head due to a circuit defect (bad blocking cap, for example). Demagnetizing may temporarily remove the residual, but if it creeps back, something is "broken", and needs repair or replacement.

8. Lastly, casual use of a demagnetizing device, even if the classic proper techniques are observed (slow removal, switching of when several feet away), sometimes resulted in an increased magnetic state rather than a decrease. There's no way to tell if you are demagnetizing, or magnetizing, without further testing for the effects of magnetism. Once again, these are HF erasure of a recorded tape, or increase in noise and even-order harmonic distortion during record. Yes, you need instrumentation to do this.

During the research, I also found that attempts to deliberately magnetize a tape head with a strong permanent magnet were successful, but over time, the head returned to a neutral state by itself. It took 24 to 48 hours, but then all the effects of magnetism vanished. The same was true of guide parts, though some could not be magnetized being non-ferrous.

In summary, routine demagnetization is unnecessary, and could actually make things worse. If you plan to demagnetize, you also need to be able to measure the effects of residual magnetism to see if you've made it better or worse, or if it was needed at all.

Have fun,

Jim