close

Вход

Забыли?

вход по аккаунту

?

piee.1969.0246

код для вставкиСкачать
Gramophone-record reproduction: development,
performance and potential of the stereophonic pickup
James H. Kogen, M.S.E.E.
Abstract
This paper describes the development of the stereophonic gramophone pickup since 1958. The basic
requirements for such a pickup are given. The characteristics of the early pickups and improvements over
the past ten years are discussed. Emphasis is placed on tracing and tracking distortion, mistracking and
frequency response. The paper concludes with a list of problems which must still be resolved.
Introduction
The improvement in quality of stereophonic gramophone reproduction, since its commercial introduction in
1957, has been truly remarkable. One of the most significant
contributors to this improvement has been the gramophone
pickup. Although the developments in pickups are not as
physically obvious as in record changers and tone arms, the
subjective effect of pickup improvement has probably been
more pronounced.
In this paper, the progress of the stereophonic gramophone
pickup is traced, from the production of the first units in
1958 to the present. We shall consider two eras; the initial
production stage, extending roughly to the early 1960s, and
the developmental stage since that time. During this second
stage, primary emphasis has been placed on attacking and
eliminating individual imperfections, some small and some
quite significant. At present, work continues along these lines,
and the paper will conclude with a discussion of several
important areas for future improvement of the gramophone
pickup.
1
Stereophonic reproduction: basic
considerations
T|ie primary purpose of a gramophone pickup is to
translate displacement, as presented by modulation cut into
the record groove, into electrical energy. The output of the
pickup must be of sufficient magnitude to provide acceptable
signal/noise level. The stereophonic pickup is required to
perform this function on two independent channels,
simultaneously.
Each groove wall in the stereophonic record contains an
independent channel of displacement information or modulation (Fig. 1). The modulation axes intersect at a 90° angle.
The pickup must respond to modulation in one groove wall
without producing electrical output in the other channel. The
only means available for separating the two channels is the
orientation of the modulation, and the corresponding physical
and electrical orientation of the pickup axes.
The performance requirements for a stereophonic gramophone pickup include:
time is considered reasonable, records have been
measured with modulation velocities in excess of
lOOcm/s (considered unreasonable),
(iii) The acceleration is typically as high as 1000#, and on
occasion as high as 3000#.
(b) Frequency response
The response must cover the audio range 20-20000Hz
and should optimally be flat in this range. In addition,
2
(a) Tracking the groove
This is the fundamental requirement of the pickup and
must be accomplished under the following conditions:
(i) The peak amplitudes range from a maximum of
approximately 0-005cm to a minimum of about
0-000005cm, or 500A, the minimum amplitude
requiring a detectable output signal with at least
20dB signal/noise ratio at that level. The minimum
amplitude is roughly one-tenth of the wavelength
of blue light.
(ii) The modulation velocity ranges from a maximum of
30cm/s to a minimum of about 0 03cm/s. Although
the maximum is stated here as 30cm/s, which at this
Paper 5880 E, first received 7th January and in revised form 9th April
1969. It is a subject review paper
Mr. Kogen is Vice-President Development & Design Engineering,
Shure Brothers Inc., Microphones & Electronic Components, 222
Hartrey Avenue, Evanston, III., USA
1338
Fig. 1
Stereophonic-record groo ves
there should be no excessive peaking at frequencies below
or above this range, so as to eliminate intermodulation
effects and possible record damage.
(c) Separation
The output from one channel should be no greater than
10% of the output produced from the other channel
when that channel is modulated. This separation should
exist over the complete audio spectrum.
(d) Output voltage
The level of the pickup must be sufficient to provide an
adequate signal/noise ratio for the total system. The
noise is usually produced by a source other than the
pickup, such as the noise in the record introduced by the
tape-recording process in mastering. At this time, a level
in the order of 0-6mV/cm per second (peak) of record
modulation is considered the minimum for a moving- or
induced-magnet cartridge feeding a 47000Q input.
(e) Dynamic range
The dynamic range of the pickup itself, i.e. the region
within which it operates linearly, should be at least 80dB
PROC. IEE, Vol. 116, No. 8, AUGUST 1969
(Fig. 2). The top end of the dynamic range, or what might
be called the overload limit, is determined by the record
modulation velocity above which the pickup can no
• maximum level: normally
set by trockinq ability
of pickup
usable
6 0 dB
dynomic
range
signal / noise
2 0 dB
level for minimum
usable signal
SOdBnronge of linear
operation of pickup
minimum level: normally
set by noise sources
external to pickup:
e.g. noise in record,
preamplifiers etc.
sistent fashion. The emphasis was placed on resolving
problems relating to frequency response and separation,
compliance (with emphasis on vertical compliance), tracking
force and reliability.
During this period, the frequency response of a good highfidelity pickup was reasonably smooth, up to approximately
15 kHz. A peak in the curve was commonly found in the
12-15kHz region. Fig. 3 shows the response of a cartridge
•10
20
100
1000
frequency.Hz
20000
Fig. 3
Fig. 2
Frequency response and separation of 1960 stereophonic cartridge
Dynamic range of gramophone pickup desired at any given frequency
longer maintain proper contact with the groove. The
lower end of the dynamic range is determined by the
noise level. This noise level is not usually produced by the
pickup, but by other sources, such as the preamplifier
or the noise which has been cut into the record. A
minimum signal/noise ratio of 20 dB is required when the
signal from the cartridge is at its lowest level. Hence,
as shown in Fig. 2, the usable dynamic range of the
cartridge should be in the order of 60dB.
These are the primary functional requirements of the
gramophone pickup. It is assumed that these requirements
must be fulfilled with minimum distortion. To optimise the
device, however, the following additional objectives are most
significant:
(a) Minimum wear
The pickup must meet the functional requirements, while
minimising wear of the record, as well as wear of its own
tip. The most effective practical method yet found to
minimise wear is that of reducing the tracking force.
Thus, the functional requirements must be met at a
minimum tracking force.
(b) Weight
The cartridge should be light in weight, to minimise
dynamic problems encountered with the cartridge-tone
arm system. Included in these problems are the effects
of record warp and structure-borne noise.
(c) Ruggedness and stability
The device is normally used in a nonlaboratory situation
and must therefore stand reasonable abuse. In particular,
the temperature and humidity conditions will not be
controlled, and the pickup is expected to operate in
extremes of temperature ranging from at least 50° F to
IOO°F. The humidity conditions might range from as low
as 2% relative humidity, such as might be found in
Pheonix, Arizona, to nearly 100% humidity, such as might
occur in New Orleans, Louisiana.
(d) Noise
The pickup must be reasonably immune to stray hum
fields and other sources of electrical radiation. It must
not generate significant amounts of noise within its own
structure.
(e) Price
Price must be such that the device can be purchased by a
reasonable number of consumers. The reasons for this
should be obvious.
3
Pickup development: from 1958 to the
early 1960s
This period covers the initial phase of the development
and production of the stereophonic gramophone pickup. The
basic problem at this time was primarily that of producing
something that would work and do so in a reasonably conPROC. IEE, Vol. 116, No. 8, AUGUST 1969
introduced in. 1960. The variation from a flat frequency
response was generally not considered serious, because of the
lower average-energy content of program material at higher
frequencies, as well as the influence of masking.
The problem of maintaining adequate separation across the
audio spectrum was of major concern during the 1958-1960
period, since this, after all, is the criterion which separates
monaural from stereophonic reproduction. This was also the
era in which great emphasis was placed, by record manufacturers and the users, on demonstrating the stereophonic
effect. Records were made with isolated sources in the two
channels, and they required a comparable separation capability in the pickup.
The retention of adequate separation across the frequency
spectrum has proved to be both a manufacturing and a design
problem. From the manufacturing standpoint, one must
retain the perfect orthogonal relationship in the transducer
element or elements, as well as in the mechanical driving
members. It is also necessary to maintain minimum electrical
crosstalk between the two channels. Since the economics of
the situation precludes the use of adjustments, it has been
necessary to maintain extremely close tolerances in all parts
affecting separation.
From the design standpoint, the problem of separation is
primarily one of eliminating resonances within the audiofrequency spectrum. It is practically impossible to retain
good separation when the mechanical system resonates.
Rotational modes are set up in the stylus and the tone arm,
which crosscouple information from one channel to the other.
Although this fact was known during the 1958-1960 era,
most cartridges had a resonance in the 12-14kHz region, and,
as a consequence, the separation became minimal at these
frequencies. Fig. 3 shows the separation of a cartridge which
was typical of the performance of top-quality cartridges
during this period.
Low-frequency resonance of the cartridge-compliance/tonearm inertia system can contribute to a lack of separation.
The separation at low frequencies tends to be less significant
than at midfrequencies, because of the lack of directionality
of the low-frequency sound. For other reasons, however, it
was essential that the effect of resonance in the low-frequency
region be eliminated; and, as a consequence, the design of
tone arms, in conjunction with the available cartridge, was
such as to place the tone-arm stylus resonance below 20 Hz
for most systems.
The provision of an adequate measurement of separation
proved to be a problem. Ideally, the two modulation axes
on the record should be at a 45° angle to the surface of the
record (Fig. 4). This provides an included angle between the
two modulation axes of 90°. To provide maximum separation,
the pickup axes should exactly coincide with the modulation
axes.
It has been found, in comparing test records, that differences
exist in the included angle between the modulation axes.
Differences have also been found in the angle between the
1339
modulation axes and the surface of various test records. A
pickup optimised for one test record has been found to
mismatch another test record, causing a change in the
separation reading of as much as 13dB. Separation is also
affected by a mismatch in the vertical tracking angle of the
pickup and that of the record,17 although this is a relatively
minor effect for vertical tracking angle errors of less than 5°.
modulation axis
view
modulation axis
have put significant efforts into the minimisation of distortion
caused by poor tracking.
4.1
Tracking distortion
The problems of vertical-tracking-angle error (Fig. 5)
were primarily those of erroneous information on cutting
angles and no standardisation. Stereophonic records were
being mastered with a variety of cutters, including those of
Westrex, Neumann, Teldec and Fairchild. The geometric,
vertical cutting angles of these cutters ranged from 0 to 30°.
It was well known by the pickup manufacturer that the
cartridge tracking angle should match that of the record.
The Westrex cutter was predominant in the United States,
and most US pickup manufacturers tried to match the 23°
perpendicular to record
in plane perpendicular to groove
Fig. 4
cutting
Modulation axes as related to record surface
To this day, there does not exist a standard record for
measuring separation. Consequently, the pickup manufacturer
must idealise his cartridge for a particular record of his choice.
The compliance of the pickup was a factor which emerged
as an extremely important requirement in the stereo cartridge.
It had been known previously that, for monaural reproduction, the vertical compliance was important to minimise
record damage caused by the vertical motions resulting from
the pinch effect. Some vertical compliance was designed into
the monaural gramophone cartridge; however, the requirement was not nearly as severe as that which emerged when
the cartridge actually had to reproduce the vertical modulation of the 45° stereo groove. The emphasis during this
period was placed on providing proper reproduction at low
frequency, and increasing vertical compliance was essential to
achieve this goal. Dynamic compliance ratings of the order
of 6 x 10~6cm/dyne were common by 1960. This provided
for reasonably good reproduction in the 20-500Hz frequency
range, at tracking forces of the order of 3 gf and above.
Little was said during this period about the fact that the
compliance of most cartridges varies with frequency. The
static compliance is usually much larger than the dynamic
compliance. Dynamic compliance measurements depend
largely on the method used and the frequency at which it is
made. The lack of a standard measurement and the usual
commercial pressure led to a compliance 'race', which was
not significantly counteracted until the recent emphasis on
the trackability criterion.
Tracking force reached a minimum of approximately 1 -5gf
during this era. By today's standards, it is probable that the
cartridge rated at 1 -5gf in 1960 would have to be played at
2 • 5-3 gf to be acceptable. Nevertheless, by the end of this era,
serious consideration was being given to the need for lower
tracking force. Some evidence was available as to the improvement in record wear and tip life as the tracking force was
reduced, and this obviously offered a great incentive for
minimising tracking force.
As the 1960s began, the major problems of manufacuring a
reliable, reasonably good cartridge had been resolved. The
optimisation needed to reduce tracking force, while maintaining, or even possibly improving, the functional performance,
is undoubtedly the major effort that has pervaded development
work on gramophone pickups from 1960 until the present.
4
Pickup development: 1960 until the
present
By the early 1960s, the question was no longer whether
the gramophone cartridge could meet the functional requirements, but how well. Major attention was soon turned to
reducing distortion. The first type of distortion to be emphasised was that produced by vertical-tracking-angle error, but,
shortly thereafter, the emphasis was placed on tracing
distortion. Somewhat later, pickup manufacturers rediscovered the fact that the gramophone pickup still had to
track the record properly, and, in the last few years, they
1340
stylus
record surface
cutting-stylus motion in air
playback
stylus
record surface
* playback- stylus motion
Fig. 5
Schematic of production of vertical-tracking-angle error
6 = effective cutting angle on record
a = effective vertical tracking angle of stylus
6 — a = vertical tracking-angle error
tracking angle which the Westrex cutter purportedly produced.
Bauer16 and Woodward17 discovered, in 1962, that stereo
cutters did not, in fact, produce the vertical tracking angles
which had previously been specified. As a consequence, considerable emphasis was placed on standardising the vertical
tracking angle, and 15° has been generally accepted throughout the industry. New pickups were subsequently introduced
to fulfil this requirement.
Unfortunately, a standard method of measuring vertical
tracking angle has not as yet evolved, and it is not possible
to guarantee an exact match of pickup and record. It appears
that if the vertical tracking angle of the pickup is reasonably
close to that at which the record is cut, say within ±5°, the
resulting distortion is not serious, at least in terms of the
distortion which is acceptable at this time. Records are
probably being cut within a ±3° range of 15° at 400Hz, and
the majority of more recent pickups are within a ±5° range
of this median angle. Statistically speaking, most cartridges
play the majority of records within ±5° vertical-trackingangle error.
Halter and Woodward18 describe measurements of the
distortion caused by vertical-tracking-angle error. For an
error of 5°, second-harmonic distortion, in one example,
changes from 1 % to 2%. The intermodulation distortion, in
terms of the sum of the two sidebands divided by the 4kHz
carrier, changes from about 18% to 28%. No attempt is
made to correlate these measurements with subjective effects,
but it is the author's opinion that this distortion is not
significant in terms'of the present overall system distortion.
This distortion, however, might very well become significant,
as other sources of distortion are diminished.
PROC. IEE, Vol. 116, No. 8, AUGUST 1969
4.2
\ objective lens of microscope /
without allowing the tip to touch the bottom of the groove
(Fig. la). With a correctly made elliptical tip, the area of
contact between the tip and the record looks something like
that shown in Fig. 7c, this area of contact being similar to
that achieved with a 0 0005in-radius spherical tip. The
greatest benefit of the elliptical tip is obtained with the
smallest playing radius, which, in turn, requires the use of low
tracking force, to minimise tip and record wear. The use of
the 0 0002in-radius elliptical tip requires that the tracking
force does not exceed 1 • 5gf. A playing radius of 0 0004in
can be employed with a tracking force of up to 4gf, but it does
not offer as large a reduction in tracing distortion as the
smaller 0 • 0002-radius tip
A great deal has been written on tracing distortion (see
References 1-15), and some truly large numbers have been
quoted for distortion from this source. It is impossible to
describe, in a simple fashion, what might be termed 'typical
state-of-the-art' distortion figures, since tracing distortion is
affected by a large number of variables. Total harmonic distortion, measured at 1kHz and 25cm/s modulation velocity,
using elliptical tips, runs to the order of 2-3 %. This probably
does not constitute a subjective problem, but tracing distortion
at higher frequency and modulation levels run very much
higher and can definitely constitute a subjective problem.
plone of
light rays
4.3
Tracing distortion
Tracing distortion has been somewhat unique, because,
unlike many of the problems of gramophone pickups, it is,
at least in its simplest form, amenable to mathematical analysis.
Piere, Hunt, Corrington, Cooper and others3*4>5>" have
made contributions in terms of the mathematical analysis of
tracing distortion, assuming no indentation of the tip into the
record. Along with Woodward and Fox, and others,7-8'9
Cooper has also analysed the effects of predistortion cut
into the records to balance out the inherent effects of tracing
distortion. At least two record companies now use this technique to help minimise tracing distortion.
The reduction of tracing distortion has been accomplished
by providing pick-ups with elliptical tips. Although Edison's
phonograph employed an elliptical tip, or, more properly, a
'biradial tip', this configuration was not in common use for
many years, primarily because of the great difficulty in producing a diamond tip within acceptable dimensional tolerances. In addition to manufacturing such a tip, it was
necessary to develop means for accurately measuring the
critical dimensions. The reflected-light method shown in
Fig. 6 offers one means of determining the radius of the
plane of
light rays
*•-—o
light
source
Fig. 6
Method of using high-power microscope
to measure elliptical tip
a Front view
b Top view
portion of the tip that contacts the groove. With the development of manufacturing and measuring techniques, good tips
are now being made in large quantities.
The main advantage of the elliptical tip is that it provides
a small playing radius (Fig. 76), of the order of 0-0002 in,
tip
groove
Distortion from mistracking
When the pickup does not track the record modulation
correctly, severe distortion results. The ability of the pickup
to track properly is called 'trackability'; improvements in
pickup trackability have centred around the reduction of the
tracking force and the need for tracking higher modulation
levels.
In addition to achieving good trackability, it is considered
essential that the tracking force be minimised, to improve
tip life. The life testing of large numbers of diamond tips
on vinyl records clearly indicates a reduction of tip wear at
lower tracking forces. A brief description of some of these
life tests is given in Appendix 8. Although similar test data
are not available on record wear, experience indicates that
this will also decrease as the tracking force is lowered. Proper
operation at minimum tracking force is clearly an objective
for top performance.
In order to determine the minimum usable tracking force,
it is necessary to determine reaction forces to be overcome.
As the stylus is displaced, a reaction force is produced which
is proportional to the magnitude and frequency of the displacement and the mechanical impedance of the stylus. It is
important, therefore, to develop knowledge of both the input,
which is the record modulation in this case, and the load,
which is the stylus impedance.
The input, or displacement, of the stylus, is proportional to
the frequency and amplitude of the modulation on the record.
No standard exists for the modulation amplitude, and none
appears to be forthcoming in the foreseeable future. The broken
curve shown as a dotted line in Fig. 8 has been published a
100
playing radius
groove
>, 10 -
groove wall
maximum velocity: based 700Hz 8OOOH2
on measurements on
/^
.N
v
commercially accepted/''
\
\
N
33^- rev/min v
/.'
V
33-312 in records
•' ^theoretical maximum.based on geometrical
considerations
io
0-1'0
,
10
100
1000
frequency. Hz
20000
Fig. 8
Indentation area
of elliptical tip
Fig. 7
Views of'elliptical'
biradial tip
a Front view of elliptical tip
b View at 45° to record surface
c View perpendicular to groove (at 45° to record surface)
PROC. IEE, Vol. 116, No. 8, AUGUST 1969
71 E25
Maximum modulation velocity against frequency, for 33^revjmin
12 in-diameter gramophone record
number of times 10 ' 24 and depicts the theoretical limits of the
modulation velocity. The boundaries are based on geometrical
factors, including the distance between grooves in the
low-frequency region, the cutting-stylus geometry in the
1341
mid-frequency region and the playback-tip radius in the highfrequency region. If these boundaries actually existed, life
would be much easier for the gramophone-pickup manufacturer. The examination of records shows that these
boundaries are exceeded quite frequently, even on very good
quality 33-^-rev/min discs. Pickup manufacturers must make
cartridges that will play these highly modulated records.
A further point which must be emphasised in this regard is
that the peak tracking requirements are always of a transient
nature. Sounds such as the hard 's' in 'since', muted trumpets,
castanets, high-frequency bells, high-level low-frequency drum
beats and harpsichords are typical of those which produce
serious tracking problems. In some recordings, these sounds
may occur only infrequently; nevertheless, the distortion can
be most disturbing, since the sounds often occur at an
important part in a musical passage, such as a cresendo.
Even though the average modulation level on the record may
be very moderate, these high-level transients do occur and,
unless handled correctly by the gramophone pickup, can be
most distressing.
Since no standard exists on modulation amplitude, and the
geometric criteria just described are easily circumvented, what
can be used as a proper criterion for the modulation amplitude? A practical objective is the reproduction of almost
anything that is worth reproducing. The 'solid-line' trackability curve shown in Fig. 8, would at this time, in our
judgment, cover this objective. A recently discovered record
with a modulation amplitude of 104cm/s at 7kHz would
fall outside this criterion. Some 45rev/min records, cut at
very high average-modulation levels, but certainly not of
high quality, would also be considered outside this objective.
As stated above, the reaction force is a function of both
the record modulation and the mechanical impedance of the
stylus. Since the pickup manufacturer has no control over
record modulation, his only means of reducing the reaction
force is that of decreasing the mechanical impedance of the
stylus. This requires an optimisation which has generally
been based on a reduction of stylus mass. Simply reducing
mass is not sufficient, however, since one must consider the
effects of shank compliance, ruggedness and other factors.
In total, this means optimising the dynamic characteristics
of the pickup, to achieve suitable mechanical impedance
across the audio-frequency spectrum.24
In the past few years, there have been two major developments with regard to trackability. First, reliable gramophone
pickups with vastly improved trackability have been made
in production quantities. Secondly, the trackability specification is now becoming accepted as a significant criterion for
determining gramophone-pickup performance.
The trackability specification defines the maximum modulation which the stylus can track, through the complete audiofrequency spectrum, at a specified tracking force. This
measurement has been known for some time and has been
used, to some degree, in evaluating cartridges.2'21 Various
forms of the trackability measurement have appeared in the
literature under the titles of 'impedance measurements',
'admittance measurements' and 'measurements of force
required to ensure tracking for constant modulation velocity
as a function of frequency'. The most significant advantage
upper-limit response^
CD 0
•a,
2-io
/
lower-limit response
. separation
typical response
i-20
20
100
1000
frequency.Hz
20000
10000
Fig. 9
Limits of frequency response and separation of 1968 stereophonic
cartridge
of this measurement is that it points out major differences
among pickups that may appear to be similar when measured
in terms of frequency response. A frequency response shows
a characteristic of the pickup in its linear region. The
1342
trackability test measures the overload limits or, synonymously, the level at which the pickup enters a nonlinear
region, all of these being as a function of frequency.
4.4
Frequency response
During the past five or six years, there has been a
significant improvement in frequency response in the 1020kHz region. Gramophone pickups with essentially flat
frequency response, within the limits shown in Fig. 9, are
available today. This has been achieved principally through
decreases in mass and optimisation of the stylus dynamics,
which have moved the lowest high-frequency resonance to
beyond 20kHz. This has also resulted in vastly improved
separation in the 10-20kHz region, as shown in Fig. 9.
5
Problems yet to be solved
Although not intended to be all-inclusive, the following
problems are cited as being of major importance at this time.
5.1
Specifying distortion in a manner which is
related to subjective evaluation
This is a fundamental problem which obviously
involves much more than the gramophone pickup. Specifically,
how can a measurement be made which will tell us, in numbers,
something about the subjective evaluation of the sound? An
example of the problem is the work on the measurement of
vertical tracking angle. On a given test record, the frequencyintermodulation distortion becomes a minimum when the
vertical tracking angle of the pickup is 23°. On the same
record, the second-harmonic distortion becomes a minimum
when the vertical tracking angle is 12°. Should the pickup
employ a vertical tracking angle that minimises the secondharmonic distortion, or one that minimises the frequency
intermodulation distortion? The desired answer is that the
vertical tracking angle minimises the subjective distortion.
Unfortunately, it is not certain whether harmonic distortion,
intermodulation distortion, or some other criterion best
characterises subjective distortion. It is of great importance
that some way be found to provide an objective measurement
for subjective distortion.
5.2
Measurement of vertical tracking angle
At present, there is no standard method of measuring
vertical tracking angle. One of the reasons for this has just
been cited. Secondly, additional work must be done in
determining the variation of vertical tracking angle with
frequency. Tests indicate that such a variation exists, and
that a standardisation of the vertical tracking angle must be
made at some defined frequency. The variation of the vertical
tracking angle with frequency must also then be specified.
5.3
Tracing distortion and record deformation
Much of the mathematical analysis of tracking distortion has been based on the assumption that the tip does not
deform the record surface. Since the tip does, in fact, indent the
record surface to a significant degree, future work in the area
of tracing distortion must take into account this indentation.
Previous investigations have been in two areas. Hunt, Barlow,
Walton26-27>31-32 and others have examined the physical
characteristics of the record material and how it indents.
Studies have involved both elastic and plastic deformation
of the record, and other investigations have covered the
dynamic effects of indentation. Practical distortion measurements have been made by Walton,25 and a theoretical analysis
of a somewhat simplified case has been made by Shiga.28
A thorough understanding of the effects of indentation will
be necessary to further optimise the gramophone pickup in
terms of decreasing distortion. At the risk of being repetitious,
however, the question of defining distortion must be restated.
It will be necessary to learn what kind of .distortion to
minimise, before the objective of optimising the pickup design
can be achieved.
5.4
Improvement of trackability tests
The major problem with trackability testing is the lack
of a simple test procedure. At the present time, trackability
PROC. IEE, Vol. 116, No. 8, AUGUST 1969
measurements are somewhat laborious. Optically, a single test
record, which can be used for measuring trackability in a
manner similar to the frequency-response test procedure, is
desired. The major problem at this time seems to be that of
cutting a record which will have very high modulation
velocities, so that the 'trackability capability' of the better
pickups may be reached.
5.5
Elimination of noise sources
It has been said that the groove of a gramophone
record is one of the smoothest surfaces known to man. The
inherent noise floor of the record is usually considered to be
that imparted to the record from the master tape. The noise
levels on records can be kept very low, provided that the
records are kept clean. Thus, one of the major objectives in
the future is that of providing a practical means for minimising
the dirt problem on records. At the same time, it must be
assumed that minimisation of the electrostatic problem will
also be necessary.
5.6
Reduction of record and tip wear
Wear of diamond tips and of records remains one of
the major problems in gramophone reproduction, and the
reduction of tracking force offers one significant means of
reducing such wear. It is expected that work on materials
for both tips and records will continue and will eventually
result in improvements in life of both the tip and the record.
5.7
Development of measurement standards
Ultimately, it will be necessary to develop standard
measuring techniques for the important variables associated
with gramophone-pickup performance. The most difficult
aspect of this problem is that of producing and reproducing
standard records. At this time, standard records can be
produced for the measurement of frequency response and
separation. Much more work must be done before an acceptable standard can be produced for the measurement of distortion and trackability through the complete audio spectrum.
More must be learned about the nature of the phenomenon
before settling on a standard test procedure.
Consider the multitude of variables which must be measured
and/or controlled in order to provide a repeatable frequencyresponse measurement. A number of these variables are cited
below, and the control of most of these factors is essential,
to guarantee the absolute accuracy of the measurement:
Gramophone pickup
(a) tracking force
(b) tilt, as viewed from the front
(c) vertical tracking angle at tracking force employed
(d) trackability
Record
(a) distortion in modulation
(b) vertical tracking angle
(c) modulation velocity as a function of frequency
(d) Groove radius to centre of record
(e) included angle in groove and relationship of groove to
record surface
(/) noise in record
(g) warp in record
(h) cleanliness
Turntable
(a) speed of rotation
(b) Wow,flutterand rumble
(c) horizontal positioning
Tone arm
(a) lateral tracking error
(b) tone-arm mass, as related to stylus-tone-arm resonance
(c) friction; viscous and Coulomb
(d) antiskating adjustment
Instrumentation
(a) distortion and noise in preamplifiers
(b) Accuracy and repeatability of measuring instruments,
such as wave analysers, total-harmonic-distortion
analysers, voltmeters etc.
PROC. IEE, Vol. 116, No. 8, AUGUST 1969
Environmental
(a) electromagnetic and electrostatic pickup
(b) temperature
(c) humidity
(d) structure-borne noise.
In addition, one must consider repeatability. As an experiment to determine the repeatability of existing frequencyresponse measurements, a series of tests on three different
high-quality reproducing systems has been made. The simplest
of these tests consisted of running a frequency-response
measurement repeatedly on the same record, with the same
equipment and using the same cartridge. A statistical analysis
showed that 99-7% of the measurements ( ± 3 standard
deviations) would fall within a range of 1 dB at frequencies
of 1-lOkHz. This would apply to any one of the setups tested.
In comparing measurements among the three systems,
it was found that the ±3 standard deviation range was 2dB.
At higher frequencies, particularly in the neighbourhood of
cartridge resonance, the repeatability becomes much worse
and might even approach 6dB near resonance. This lack of
repeatability near resonance is probably caused by poor
tracking, which obviously must be controlled if one is to
achieve a reasonable frequency-response measurement.
Philosophically speaking, standards which would prohibit
or misdirect the development of new products should not be
instituted. At this time, a strong movement to set up standards is not advocated, and a concerted effort to develop
better measuring techniques and, in particular, good test
records, is needed first. As a user of large quantities of test
records, we feel that considerable effort is required to
guarantee the reproducibility of such records from pressing
to pressing.
6
Conclusions
In reviewing the efforts of ten years, one must of
necessity touch only on highlights; there have undoubtedly
been many other areas of development which might be given
recognition. These include such things as reduction in hum
pickup, improvements in ruggedness and ease of mounting,
weight reduction and so on. Many of these items are related to
practical considerations that make the product easier for the
customer to use.
It is essential to continue to evaluate the gramophone
pickup in a critical manner. Today's product is a good one,
but there are still many things to learn and many defects to
eliminate. The ultimate objective is the provision of a highfidelity reproducing system that will accurately reproduce the
sound of a live performance in a fine symphony hall. We
must continue our efforts to achieve this goal, but we still
have a long way to go.
7
References
Papers relating to tracing distortion
1 KANTROWITZ, P.: 'High frequency stylus-groove relationships in
phonocraph cartridge transducers',/. Audio Engng. Soc, 1963, I I ,
pp. 250-262
2 BAYLIFF, R. w.: 'Some aspects of gramophone pick-up design',
Proc. IEE, 1962, 109B, pp. 233-243
3
CORRINGTON, M. s., and MURAKAMI, T.: 'Tracing distortion in
stereophonic disc recording', RCA Rev., 1958, 19, pp. 216-231
4
PIERCE, J. A., and HUNT, F. v.: 'Distortion in sound reproduction
from phonograph records', J. Acoust. Soc. Amer., 1938,10, pp. 14—28
5
LEWIS, w. D., and HUNT, F. v.: 'A theory of tracing distortion in
sound reproduction from phonograph records', ibid., 1941, 12,
pp. 348-365
6 COOPER, D. H.: 'Tracking distortion as phase modulation', IEEE
Trans., 1963, AU-11, pp. 41-46
7 FOX, E. c , and WOODWARD, J. C : 'Tracing distortion: its cause
and correction in stereo disk recording systems', J. Audio Engng.
Soc, 1963, 11, pp. 294-301
8
REDLICH, H., and KLEMP, H.: 'A new method of disc recording for
reproduction with reduced distortion: the tracing simulator', ibid.,
'
9
1965, 1 3 , p p . 111-118
COOPER, D. H. : 'Compensation for tracing an'd tracking error', ibid.,
1963, 11, pp. 318-322
10 COOPER, D. H. : 'Integrated treatment of tracing and tracking error',
ibid., 1964, 12, pp. 2-7
11 COOPER, D. H . : 'Tracking and tracing error measurements', ibid.,
1964, 12, pp. 312-326
12
KOGEN, J. H., and SAMSON, R. s.: 'The elliptical stylus", Audio, 1964,
48, (5), pp. 60-61
13 BASTIAANS, c. R.: 'Factors affecting the stylus-groove relationship
in phonograph playback systems', J. Audio £/i?/r?. Soc, 1967, 15,
pp. 389-399
1343
14 COOPER, D. H.: 'Interaction of tracing and tracking error', ibid.,
1965, 13, pp. 138-141
15 WOODWARD, J. c : 'Reducing distortion in stereo phonograph
systems', Audio, 1964, 48, (2), pp. 34-40, 77
Papers relating to vertical tracking angle
16 BAUER, B. B.: 'Vertical tracking improvements in stereo recording',
ibid., 1963,47,(2), pp. 1-4
17 UASTIAANS, c. R.: 'Further thoughts on geometric conditions in the
cutting and playing of stereo discs', J. Audio Engng. Soc, 1963,
11, pp. 6-15
18
HALTER, J. B., and WOODWARD, j . c : 'Measurement of distortions
due to vertical tracking angle errors in stereo disc systems', ibid.,
1964, 12, pp. 8-14
19 BAUER, B. B.: 'The vertical tracking angle problem in stereophonic
record reproduction', IEEE Trans., 1963, AU-11, pp. 47-55
20 WOODWARD, J. c : Techniques for measuring the vertical tracking
angle of stereo phono pickups', / . Audio Engng. Soc, 1965, 13,
pp. 241-247
Papers relating to trackability
21 KANTROWITZ, P . : 'Mechanical impedance measurements at the
stylus of stereo phonograph cartridge transducers', ibid., 1962,
10, pp. 323-328
22 KOGEN, J. H.: 'Tracking ability specifications for phonograph
cartidges', ibid., 1967, Preprint 537
23 KOGEN, i. H.: 'Trackability', Audio, 1966, 50, (11) and (12),
pp. 19-22, 26-27
24
ANDERSON, c. R., KOGEN, J. H., and SAMSON, R. s.: 'Optimizing the
dynamic characteristics of a phonograph pickup', J. Audio Engng.
Soc, 1966, 14, pp. 145-152
Papers relating to indentation {deformation)
25 WALTON, J. : 'Stylus mass and distortion', Wireless World, 1963,
69, pp. 171-178
26 BARLOW, D. A.: 'Groove deformation in gramophone records',
ibid., 1964, 70, pp. 160-166
27 WALTON, J.: 'Gramophone record deformation', ibid., 1961, 67,
pp. 353-357
28 SHIGA, T.: 'Deformation distortion in disc records', J. Audio
Engng. Soc, 1966, 14, pp. 208-217
29 IIARL, DOWA.: 'Limiting tracking weight of gramophone pickups
for negligible groove damage', ibid., 1958, 6, pp. 216-219
30
31
32
33
34
actual playing would be preferred, the physical measurement
was chosen because of better repeatability.
Test 1:
In this test, the pickup was run continuously on the same
record. The record was cleaned at least once a day. Ten
pickups were run at each of three tracking forces, 1, 2 and 3g.
Test 2:
This was identical to test 1, except that the records were
changed after 20 plays and were then discarded. Tests were
run for tracking forces of 0-75, 1-5, and 3-0g.
Test 3:
This test was similar to test 1, except that sapphire tips
were used, and tests were run at 0-75, 1-5, and 3 0g. In this
test, five styluses were tested at each tracking force.
8.2
Results and conclusion
The results of the test are shown in Fig. 10, and the
significant conclusions are as follows:
(a) Tip life increases rapidly with lower tracking force at
tracking forces below about 1 -5g for diamond.
(b) Diamond wears at least twice as long as sapphire for
tracking forces of 3gf, and almost four times as long for
tracking forces of 3-4 gf.
(c) Continuous use of the same record in life testing
apparently accelerates wear of the tip.
Fig. 10 points out the great value of employing a low
3r
FLOM, D. c , and HUGGINS, C. M.: 'The deformation of plastics with
hard spherical indentors', ibid., 1959, 7, pp. 122-124
HUNT, F. v.: 'Stylus wear and surface noise in phonograph playback
systems', ibid., 1955, 3, pp. 2-18
BARLOW, D. A. : 'Comments on the paper "Stylus wear and surface
noise in phonograph playback systems"', ibid., 1956, 4, pp. 116-119
KORNEI, o.: 'On the playback loss in the reproduction of phonograph records', J. Soc. Motion Picture Engrs., 1941, 37, p. 569
MILLER, F. G.: 'Stylus-groove relations in phonograph records'.
Ph.D. thesis, Harvard University, 1950
~ 2
Q.
en
o
Miscellaneous papers
35
WOODWARD, J. c , and WERNER, R. E. : 'High frequency intermodula-
tion testing of stereo phonograph pickups', / . Audio Engng. Soc,
1967, 15, pp. 130-142
36 GILBERT, j . c. G.: 'The assessment of the performance of gramophone pickups', Brit. Kinematogr. Sound Televis., 1965, 47, pp. 4-10
37 WOODWARD, J. G., and FOX, E. c.: 'A study of program level overloading in phonograph recording', J. Audio Engng. Soc, 1963, 11,
pp. 16-23
0
1
2
3
tracking force,gf
8
Appendix
Fig. 10
Life of tip against tracking force
8.1
Life tests of diamond and sapphire tips
x — x — x diamond: test 2 (20 plays per record)
O—O—O diamond: test 1 (continuous play)
A—A—A sapphire: test 3 (continuous play)
The objective of the measurements described in this
section was the determination of the relationship between the
tip wear and the tracking force. Life tests were run on a
sufficient quantity of pickups to provide a reasonable statistical
sample. The following were standard throughout the test:
(a) gramophone pickup
(b) turntable and tone arm
(c) record
(d) temperature: 75° ± 5°F.
The following variables were not controlled:
(a) humidity: ranging from 15% to 90% relative
(b) crystal orientation of diamond (random).
The end of the life of a tip was arbitrarily chosen to occur
when a flat of length 0-0008 in appeared on either side of the
tip. Tips were checked every 25h, under a microscope.
Although it was felt that a measurement of distortion in
1344
4
tracking force. It must be borne in mind, however, that
reduction in tracking force must be accompanied by good
trackability. Experience indicates that poor trackability leads
to a reduction in record life, as well as poor sound reproduction. This means that there is some optimum tracking force
which provides a maximum tip life as a result of the lowering
of the tracking force, with maximum record life as a result
of maintaining good trackability.
One final point which may be of interest is the result of
the life testing of diamond tips on blank-groove records. The
tests run at 3gf on a groove without modulation show an
average life approximately nine times that of similar tests
run with modulation. This indicates that the wear is caused
by the effects of modulation, such as impact forces, more
than by the friction between the diamond tip and the vinyl
surface.
PROC. IEE, Vol. 116, No. 8, AUGUST 1969
Документ
Категория
Без категории
Просмотров
3
Размер файла
1 152 Кб
Теги
1969, 0246, piee
1/--страниц
Пожаловаться на содержимое документа