Handbook for Sound Engineers

1092 Chapter 28

reproducing on the same machine is appropriate. The
method of testing should be noted as part of the perfor-
mance report.
Although test and diagnostic work is commonly
conducted with simultaneous record/reproduce, the final
testing should always be conducted in the repro-
duce-only mode. The tape should be started and stopped
several times, with the various transport elements reori-
ented by hand between runs, to achieve a sampling of
random combinations of the various record and play-
back flutter components. The arithmetic average of the
maximum values of each sample throughout the reel,
excluding any infrequent short-duration bursts, is the
reported value.
If the flutter readings are excessive, the next step is
to analyze the flutter waveform for information to help
pinpoint which tape path component is defective. The
following techniques are helpful in isolating the culprit:

1. The human ear and brain form a very versatile spec-
   trum analyzer that frequently can immediately iden-
   tify the defective component from the characteristics
   of the flutter signal being reproduced in a monitor
   loudspeaker. Take advantage of this free portable
   instrument that is always at your disposal by
   listening to the demodulated output from the flutter
   meter.


2. The various selectable filters of the flutter meter can
   be used to isolate the general portion of the flutter
   spectrum in which the offending component is
   generating flutter.


3. The expected rotational flutter rate from a rotating
   component can be calculated from the diameter of
   the component and the tape speed using the
   expression

(28-13)

where,

ST is the tape speed,

d is the diameter of the component.

These frequencies can range from approximately

0.5 Hz for the once-around of full reel of tape to

60 Hz for a small-diameter capstan shaft. Some

manufacturers include a table of these flutter

frequencies in their maintenance manuals. The small

balls and retainer clips inside the ball bearings used

in many rotating components generate additional

not-so-obvious flutter components at frequencies

higher than the once-around rate of the bearing.

4. If the flutter is very regular, the flutter pattern
   displayed on the oscilloscope can be utilized to

calculate the frequency of the dominant flutter

component. Any flutter components caused by ac

motors or power supply ripple will remain

stationary on the oscilloscope screen if the sweep

triggering mode is set to line.

5. A common search technique is to deliberately create
   flutter by attaching a small piece of masking tape to
   the surface of a rotating component. The rate of the
   flutter blips created by the masking tape can then be
   compared with the unknown component to deter-
   mine if the two rates are identical.


6. Note any change in the flutter spectrum when each
   of the auxiliary rotating components such as guides
   and flutter idlers is stalled. Stalling the defective
   component will cause the offending flutter compo-
   nent to cease. A notable exception to this case is the
   scrape flutter idler. Stalling a scrape flutter idler
   should usually double or triple the scrape flutter
   amplitude. If little or no increase is noted, the idler
   is not functioning properly. Check for dirty or
   damaged bearings that would keep the idler from
   spinning freely.

The following procedure describes a flutter test

using a wide-bandwidth flutter meter, such as is shown

in Fig. 28-61. The general technique also applies to

other meters.

1. Connect the reference oscillator output (REF OSC)
   to the line input of the tape recorder.


2. Connect the demodulator input (INPUT) to the line
   output of the tape recorder.


3. Connect the demodulated output (MONITOR) to an
   oscilloscope and an audio monitor.


4. With the tape machine in the record mode, set the
   recorder’s input level control to achieve a playback
   level of –10 VU. The green Cal light should be illu-
   minated, indicating proper operating level.


5. With the FM/AM and Avg/Peak buttons both out
   and the 5 kHz and 1.0% FM buttons in, depress the
   Cal button. A reading of 0.68% indicates proper
   system operation. A 150 Hz square-wave tone will
   be seen on the oscilloscope and heard in the monitor.


6. To begin the actual test of the recorder, select the
   250 Hz filter and choose the meter sensitivity range
   that yields a reading near midscale. The meter

Flutter frequency

ST

Sd

= ------

Figure 28-61. Flutter meter. Courtesy MANCO.