Handbook for Sound Engineers

Magnetic Recording and Playback 1095

part of the filter. For a 50: oscillator, multiply the
capacitor values by 2.57.) The resulting high-frequency
playback response of an equalized flux loop will be flat
except for any residual high-frequency discrepancies
due to eddy current losses or self-resonance of the play-
back head and cabling.

The flux loop can also be used in reverse as a pickup
device to probe the magnetic fields generated at the
gaps of the record and erase heads. If the driving
network is disconnected and the loop connected directly
to the inputs of an oscilloscope and meter, the relative
magnitude of the bias and audio fields can be examined.
Care must be exercised to correct for the 6 dB/octave
rise in flux loop output voltage due to the inductive
nature of the flux loop. (A resistor in series with the
input and a capacitor shunted across the input can be
used to create an integrating low-pass filter that will
flatten out this 6 dB/octave rise.)

Details regarding the construction and use of a flux
loop, along with detailed mechanical alignment proce-
dures for azimuth, height, and tape wrap, are available
from the various tape recorder manufacturers.

28.8.1 Head Relapping

The performance characteristics gradually change as the
abrasive action of the tape wears away the faces of the
heads. The resulting decreases in gap depth will reduce
shunting effects, leading to an increase in efficiency for
both the record and playback heads. Bias and audio
levels must be gradually reduced to offset the rising
efficiency. A critical point is reached, however, when
the useful face of the head has been completely
removed and the length of the gap begins to increase
quickly with wear. The top end of the playback response
will drop abruptly within a matter of only a few hours of

use, rendering the recorder unusable. At this point, the

head must be replaced to restore normal performance.

The heads on most recorders require attention long

before this point of ultimate failure is reached. On most

machines, the tape wears away the rounded apex at the

gap of the head, leading to a drop in contact pressure

with the tape at the gap. The tape begins to lift off the

head slightly, creating erratic short-wavelength perfor-

mance due to the spacing loss effect.

The common solution is to recontour the face of the

head to restore the contact pressure. This process,

known as head relapping, can be utilized two or three

times during the useful life of a head to restore original

performance. Although the average technician can be

trained in the relapping process, the high cost of a

mistake with a 2 inch (50 mm) multitrack head

assembly suggests that the more exotic relapping tasks

should be handled by relapping specialists.

28.9 Routine Signal Alignment Procedure

A common problem arises with conventional recorders

and alignment procedures—namely, that the procedures

require a change in each adjustment to verify that the

optimum point has been reached. This typically leads to

not only the premature demise of many trimmer potenti-

ometers (which are typically rated by the manufacturer

for a life of 200 adjustment cycles) and head azimuth

hardware, but also many operator errors due to the

tedious nature of adjusting a multitrack machine that

may have as many as 1000 adjustments.

If the operator is willing to adopt a philosophy that

most of the adjustments are probably adequately close to

optimum and that they need not be readjusted, then the

alignment task shifts to looking for the exceptions to the

norm rather than arbitrarily resetting everything. This

strategy promotes better results since each iteration of

the alignment procedure serves to fine-tune the results

rather than to erase all past efforts and start afresh for

each alignment with a high probability of error.

A few exceptions to the need for tweaking to verify

proper performance are worthy of note. For example,

head azimuth can be verified with a differential method

that uses alternating test segments that have equal but

opposite amounts of deliberate azimuth error. If the

drop in level is equal for both directions of tilt, then the

head must be correctly aligned to the correct vertical

reference. No head adjustments are required if the test

results are satisfactory.

A similar noninvasive test procedure for optimizing

the bias level can be achieved if the bias system

contains a master bias level trimmer that varies the level

Figure 28-62. Flux loop.

7Oscillator

100 7 Multiple turns of fine wire

C

Loop Head

EQ standard C =

EQ time constant

Rsource Rloop

30 in/s

1.5 in/s IEC

7.5 & 15 m/s NAB

7.5 in/s IEC

0.204 MF

0.408 MF

0.583 MF

0.816 MF

Equalized flux loop