Author: S. K. Pramanik [Engineer, R & D (Audio), Bang & Olufsen;
Struer, Denmark]
The function of a phono cartridge was described in an earlier article (Audio,
March, 1979) in which the tonearm was assumed to be virtually ideal.
Its only influence on cartridge performance was taken to be the effect
of a single parameter, the equivalent mass. We can perform a similar exercise
with tonearms, this time assuming the cartridge to be virtually perfect.
To understand the function of a tonearm and its contribution to the overall
quality of a hi-fi system, it is necessary to remember one basic fact.
Electrical signals are generated in the cartridge when, and only when,
a relative movement occurs between the record surface in contact with the
stylus tip and the surface on which the cartridge is mounted at the end
of the tonearm. Such signals will be generated regardless of the cause of
the relative movement.
To be strictly accurate, signals are generated when there is relative movement
between the vibrating section (the armature) and the static sections of
the transducer elements within the cartridge. However, if the cartridge
is perfect, the stylus tip will move in a manner identical to the record
surface so that an identical movement is transferred to the armature. Also,
the static section of the generating elements will be rigidly connected
to the end of the tonearm.
Thus, a relative movement between the stylus tip and the end of the tone
arm will generate signals, whether audible or inaudible (except induced
signals such as hum or r.f. breakthrough.) To analyze the performance of
the tonearm and its contribution to audio quality, it is necessary to examine
the effect on the tonearm of the various sources causing movement in a record
playing system, and thereby to deduce the movement of the tonearm relative
to the record surface. This analysis concerns the dynamic properties and
includes what may be called the performance of the tonearm. But unless the
cartridge is correctly aligned relative to the record, relative movement
of the transducer elements will not be the same as the recorded signal.
The tonearm also has static properties which make a contribution, and these
causes are cartridge mounting and tonearm geometry.
Cartridge Mounting
The cartridge is mounted on a tone arm in a fixture which in turn is mounted
either permanently or so that it is removable at the end of the arm tube.
Incorrect mounting can lead to two types of errors.
First, unless the cartridge stylus coincides with its designed position,
the effective length of the arm is altered, leading to tracking error due
to tonearm geometry.
The other error arises when the static transducer planes in the cartridge
are not parallel to the planes of modulation on the record. This happens,
for example, if the cartridge is mounted so that it is rotated on its horizontal
axis.
In this case, a small component of the signal from an undesired channel
is, picked up and mixed with the desired signal, as the relative movement
of the generator elements is not identical to the signal engraved in the
record grooves, as shown in Fig. 1. Since crosstalk is introduced, the stereo
image suffers. The same result occurs if the transducer elements are rotated
because the tonearm is incorrectly mounted; any other mounting error leads
to the same result. In all cases the necessary correction may be applied
at the cartridge.
For the best stereo image, the cartridge should be mounted to give equal
separation in both channels. Rotating the cartridge on its horizontal axis
will increase separation in one channel at the cost of separation in the
other channel, but the rule about equal separation remains unaltered.
Records themselves are often less than perfect, and the two channels may
not be recorded at right angles or may be tilted with respect to the vertical.
In such cases it is generally not possible to get the optimum degree of
separation from the cartridge by conventional means.
Tonearm Geometry
The mathematics of the geometry for conventional or radial tracking tonearms
has been known for many years and is well documented, most recently in a
comprehensive article by Kessler and Pisha (Audio, January, 1980). It is
therefore sufficient to repeat a few of the most important points here.
Distortion due to tracking error, the angular error between the cantilever
axis and a true tangent to the record groove, is not dependent on the error
angle alone, but also on the speed of the record groove moving past the
stylus.
Fig. 1-Effect of crosstalk on stylus movement.
Figure 2 shows an arm of effective length (I), mounted at a distance (d)
from the turntable spindle. The arm length is the sum of the mounting distance
and the effective overhang. At a radius (r) from the center, the angle between
the line connecting the stylus and the tonearm bearing and the tangent to
the groove is (x), and the angle with the radius is (90 x). By the cosine
law of triangles:
If the offset angle is (y), then the angular tracking error is (y x). The
distortion due to tracking error at radius (r) is proportional to (y x)/r.
Distortion due to tracking error is quoted in specifications as the error
factor, which is the maximum angular error over the recorded surface divided
by the radius at which this occurs (degrees/cm). Unless qualified, this
specification is misleading as it fails to specify the limits of groove
radii for which the arm is designed. While the maximum radius is effectively
fixed by the outside diameter of the record, and in any case is less critical,
a small change in the minimum radius considered can alter the calculated
error factor of a tonearm appreciably.
Fig. 2-Geometry of radial tonearm movement across a record.
Fig. 4-Geometry of side thrust on the stylus tip.
Fig. 3-Error factors for different tonearm layouts.
The minimum modulated groove diameter is not the same on all records, as
it depends on the playing time, groove width, etc., and in practice can
vary by 2 cm or more on commercial records. Further, although national standards
for the minimum permitted groove radius do exist, the various world standards
do not specify the same radius. In other words, arms can differ slightly
in their geometries simply because the designers have assumed different
values for the maximum and minimum limits of groove radii.
Another source of small differences in tonearm geometry can occur because
of different design philosophies.
An arm can be designed so that the error factor is equal at the minimum
and maximum record diameters, as well as at the point of maximum error somewhere
in between, as shown in Fig. 3, Curve A. One could with justification say
that since tracking distortion also rises with modulation level, and most
recordings have high modulation levels at the end of the record, tracking
error at minimum radius should be zero, as shown in Curve B. Small errors
in mounting distance from the center of the platter or in overhang adjustment
can make comparatively large differences in angular errors, as well as the
groove radii at which maximum and minimum errors occur. It is therefore
important that the arm support is mounted correctly relative to the turntable
spindle and that the effective arm length with the cartridge mounted is
correct. This also means that where overhang is adjustable, an arm can be
optimized to track any particular record. If the minimum and maximum radii
of the modulated grooves on the record being played are measured, optimum
overhang can be calculated for the offset angle of the arm, and the necessary
adjustment in effective arm length made.
Note, however, that distortion generated by tracking error consists mainly
of second and higher order even harmonics, which are the least objectionable
of the various types of distortion. Also, tracking error distortion increases
and decreases smoothly across the record surface, and no sudden changes
occur to make the distortion more obvious. Thus, for any correctly mounted
rational design of tonearm, small differences in geometry do not give rise
to large audible differences.
Side Thrust or Skating Effect
The expression "skating effect" goes back to the days when a
smooth disc was used to demonstrate this effect or to check the compensation
applied.
Since this method gives incorrect compensation, the term may also be said
to be misleading.
Side thrust occurs due to the offset angle necessary for optimum angular
tracking in radial arms. Friction between the stylus and the rotating record
pulls the stylus in a direction tangential to the groove, as shown in Fig.
4. Since the force is not in line with the arm bearing, a rotating force
is generated forcing the arm towards the center of the record. This force,
called side thrust, depends on the instantaneous magnitude of the frictional
force between the stylus and the groove walls.
Fig. 5--Analysis of side thrust.
In the absence of side thrust, pressure is equal on both walls of the groove
at the points of contact due to the vertical tracking force (VTF), as shown
in Fig. 5A. If side thrust acts in conjunction with the VTF, the two forces
combine to give a result which is not vertical but is inclined towards the
center of the record, as shown in Fig. 5B. The effect of this inclination
is that there is more pressure on the inner groove wall (left channel) than
on the outer, as shown in Fig. 5C. It is obvious that where this happens
the cartridge cannot have ideal working conditions. The effect of side thrust
can be compensated by applying a force at the tonearm bearing. If an otherwise
correctly mounted and adjusted tone arm distorts due to mistracking on the
right channel only on high-level signals, the cause can be inadequate side-thrust
compensation; mistracking on the left channel only indicates overcompensation.
Side-thrust compensation can be applied in many forms and ideally should
give exactly the same force outwards' that side thrust causes inwards. Various
methods are used in commercial applications, such as a thread and weight,
springs, opposing magnets, etc.
It should be remembered that side thrust is affected not only by relatively
constant factors such as the shape and polish on the stylus and by vertical
tracking force, but also by the material from which the record is made and
by such varying conditions as groove radius and modulation. Since side thrust
varies more or less at random over the surface of the record, it can never
be compensated exactly. A good compromise is to adjust compensation to cope
with the highest level of modulation likely to be met on records. Although
overcompensated for all lower levels of modulation, no mistracking will
occur at any modulation level because the pressure will always be higher
than the minimum required for the stylus to maintain contact with the groove
walls.
Tangential Tracking
The conventional tonearm is pivoted on fixed axes to allow movement in
the vertical and horizontal planes. Another approach to the design of tone
arms is the tangential tracking tone arm.
In the tangential arm the horizontal bearing (which allows tonearm movement
in the vertical plane) is a conventional bearing, but the vertical bearing
is replaced by a carriage which moves as required to keep the cartridge
tangential to the groove, as shown in Fig. 6. Such tonearms are also called
parallel or straight-line tracking arms.
The most obvious advantage is that tracking error can theoretically be
zero, and the cartridge is allowed to track the record with geometry identical
to that when it was cut. In practice this advantage is less important than
it might seem, as the errors in correctly designed and mounted radial tonearms
are so small that a tangential tonearm provides only minor audible improvement.
Also, small tolerance errors in mounting a tangential tonearm can lead to
a constant tracking angle error over the whole of the record surface, virtually
negating its advantages (a 1-mm error in length gives approximately a 1-degree
error at minimum radius).
Assuming that the moving carriage functions ideally, the tangential arm
offers two major advantages. Since the minimum length required by radial
arms for adequate tracking becomes unnecessary, the tangential arm can be
short and straight. The short arm weighs very little and leads to a fairly
large reduction in the effective mass of the tonearm. Further, since the
bearings are always directly behind the stylus in line with the cantilever,
the frictional force at the stylus is taken up by the bearing. Thus, there
is no side thrust under any conditions, and no compensation is necessary.
The contact force on both groove walls provided by the VTF is the same over
the whole surface of the record, and it does not vary.
Fig. 6--Geometry of tangential tracking tonearm.
Many forms of tangential tracking can be devised, from the purely mechanical
to contactless electronic servo systems. While the method used is a matter
of engineering application, the requirement for the mechanism remains the
same for all kinds of tonearm, i.e. that the mechanism should have no (or
as little as possible) influence on the function of the cartridge.
Bearings
The primary requirement for the bearings is that they be free from friction,
a force that always opposes motion. As the stylus tracks and the arm is
moved inwards in a slow spiral by the groove acting on the stylus tip, friction
in the vertical bearing (allowing horizontal tonearm movement) exerts a
force attempting to stop the arm from moving. The outer wall of the groove
must therefore exert a larger force to move the arm, resulting in a higher
pressure on the outer groove wall (right channel) with a corresponding decrease
on the inner wall. If a correctly mounted and adjusted tonearm mistracks
on the left channel only on high-level signals, the cause can be friction
in the vertical bearing.
If a record is eccentric, there will also be an outward motion of the tonearm
for every half-rotation (if the eccentricity is very small, it may just
compensate for the inward motion of the groove at that point with no net
outward movement). On the other half-rotation the inward motion will be
larger by the amount of eccentricity. Bearing friction will have the same
effect, but the increase and decrease in contact pressure will alternate
between groove walls.
Friction in the horizontal bearing (vertical movements) has a similar effect
when tracking a warped record. In this case, however, the arm is stopped
from moving with the warp, resulting in an increase or decrease in the effective
VTF on both groove walls simultaneously, depending on whether the stylus
is being forced up or down by the warp. The effect will also occur when
the stylus is lowered onto the surface to play a record, with a consequent
decrease in the effective VTF. Undesirable effects also occur if there is
play in the bearings. Such effects do not lend themselves to easy mathematical
analysis, because they depend on the complex interplay between the design
of the remainder of the tonearm and the amount and type of play. It can
be seen that under the influence of external forces (such as friction) between
the stylus and the groove, the whole arm will move. If the movement is in
the line of the arm tube, it will result in wow, similar to warp wow explained
below. But movement can also be sideways or up and down, with results which
will depend on a combination of many factors. All that can be said in general
is that the result will not be an accurate reproduction of the recorded
signal, and the effect is likely to be unpleasant.
Fig. 7--Effect of different bearing heights on stylus movement with record
warp.
Two bearing types require a little elaboration. The first of these is called
a knife edge, sometimes used for the horizontal bearing. The entire weight
of the tonearm rests on the bearing "edge," which is supported
in a suitably shaped slot. In bearings of this type, there is theoretically
no friction; vertical movement of the tonearm results in one of the surfaces
rolling on the other, rather than sliding. The rolling action will alter
the effective length by a small amount as the end of the tonearm rises,
but this can be used to advantage to compensate for warp wow. The other
type of bearing is the unipivot, in which a single point, on which the arm
rests, is used for both vertical and horizontal movements.
While the complexities of the design are a subject in itself and cannot
be discussed here, a correctly designed unipivot can provide extremely low
friction with minimal side effects.
Finally, an important aspect of bearing design is a consideration of the
lead wires from the cartridge. These have to pass through or around the
vertical bearing, and they are attached to a point below the surface of
the turntable base. As the tonearm moves horizontally, the lead wires twist
and may generate torque to move the tonearm either inwards or outwards.
Both the leads and their layout can therefore play an important part in
overall performance and may even be a determining factor in the choice of
the type of bearing used.
Warp Wow
The height of the horizontal bearing above the record surface
is important for reproduction quality. As the tone arm moves up and
down under the influence of record warps, the stylus moves in an arc with
the axis of the horizontal bearing as its center. If the bearing is
well above the record surface, as in the exaggerated sketch of Fig. 7A,
the stylus will move forward simultaneously with its upward movement, and
the speed of the record groove relative to the stylus will decrease. The
effect is identical to a decrease in turntable speed, and it lowers the
pitch of the reproduced signal. The reverse occurs as the stylus moves down
the warp and the pitch returns to its nominal value. This variation in pitch
due to the combination of bearing height and warped records is known
as warp wow.
The ideal bearing position is at a height above the record surface equal
to the height of warps, which results in minimum changes in relative speed,
as shown in Fig. 7B.
Fig. 8-Effect of undesired axis in the horizontal bearing on tonearm movement
with warps.
The axis of the horizontal bearing can lead to another undesired effect
unless it is perpendicular to the axis of the cartridge cantilever. As the
tone arm moves up under the action of warps, the arm "twists" simultaneously
with respect to the record surface, as shown in Fig. 8A. The result is a
changing crosstalk pattern as the arm moves up and down, depending on the
angle between the transducer elements and the record surface, and this leads
to an unsteady or unstable stereo image. In an ideal case the horizontal
bearing will be perpendicular to the axis of the cantilever so that the
cartridge does not twist, as shown in Fig. 8B.
cont. in part 2 >>
( Audio magazine, June 1980)
Also see:
Tone Arm Damping--James
Brinton--This overlooked feature can offer better sound for a small investment
(High Fidelity, Jul. 1975) |