by David L. Josephson
RECENTLY, an ad for a major manufacturer of turntables appeared in this magazine, showing an LP record supported on a hand-held pencil; in the other hand was a phonograph cartridge. The ad was headlined, "This is all we want to do. But perfectly." That sums up the whole turntable game in one phrase. All of the arguments for the various types of motors, drive systems, and tone arm types are all toward one end ... spinning a record at a constant, specified speed with no vibration in any plane, and holding a phono cartridge exactly tangent to the record grooves, with just the right amount of downward pressure, and no other pressures or forces. If a turntable could do all those things, it would be the be-all and end-all of all turntables, and the arguments would end there. But as it is now, there are as many different kinds of turntables as there are of any other stereo component (excepting speakers -Editor), and no one can clearly be called best.
This article will attempt to clear up some of the confusion and mystery surrounding these record-playing devices, and should help you make your next turntable purchase with a little more insight into what is needed to do that spinning bit mentioned above.
Perhaps the first criterion for judging turntable quality is speed accuracy. The two standard speeds today are 33 1/3 and 45 rpm; 16 2/3 and 78.26 rpm however are sometimes encountered. A good turntable should be able to hold its set speed within 0.5% (at worst) over extended periods of time.
The accuracy of nearly all types of turntable motors depends directly or indirectly on the frequency accuracy of the incoming power line to keep the table at a precise speed. Power line frequency is almost always accurate to within about 0.04 %, so the speed of the turntable can be no more accurate than that.
An equally important turntable standard concerns more rapid changes in speed-wow and flutter. These are periodic variations in the speed of the turntable, producing a wavering sound. Flutter is defined as a fast variation in record speed; usually on the order of 10 cycles per second or more (not Hz, because it is a cyclic variation and not an a.c. signal). It is usually caused by mechanical irregularities in the motor itself (bumps on the drive shaft or magnetic problems in the motor) or in the drive mechanism (pulleys, idlers). Wow, also known as once-around flutter, is a variation which occurs only once for each revolution of the turntable. The most common cause of this is a warped record.
When the stylus is tracking the bumpy (and therefore stretched and/or compressed) record surface, the surface speed changes and thus the pitch of the reproduced sound changes. Wow can also be caused by variations in the turntable platter; whichever part of it is driven by the motor. A motor defect in a direct-drive turntable would be more likely to cause wow than flutter. In any case, the total wow and flutter, periodic variations in speed of any type, can be kept below 0.2% in a well designed turntable. Flutter and wow are especially bothersome in solo piano, flute, and guitar recordings ... flutter above 0.3% can be easily detected when listening to such music, and it rapidly becomes very annoying.
Rumble
Any periodic extraneous mechanical noise added by the turntable to the program material is called rumble. Rumble is usually caused by poor isolation of the motor from the rest of the turntable, the platter and arm in particular. Rumble is usually worst in a rim-drive turntable, since the platter is driven directly from the shaft of the high-speed motor through a relatively stiff rubber idler wheel. In belt-drive turntables, rumble is usually considerably reduced (for a given motor and speed) because more of the vibration is damped out in the stretchy rubber belt. Rumble is also directly related to motor speed ... the lower the motor speed, the lower will be the frequency of the rumble, and therefore the less objectionable (theoretically). The rumble frequency of the usual 1800-rpm, 4-pole induction motor is around 30 Hz, and most cartridges and amplifiers, and many speakers reproduce down to 30 Hz ... rumble of this frequency and higher is especially bothersome. The theoretical minimum rumble motor would be one operating at the record speed itself ... a direct drive machine. Nearly all of the rumble in this type of motor would come not from the motor itself, but from friction and vibration in the various motor and platter bearings.
Rumble is measured in much the same way hum and noise is measured in purely electronic equipment. A standard 1-kHz tone recorded at a specified lateral groove velocity is played, and a notch filter takes out the 1-kHz tone; everything else is measured as rumble. It is expressed in decibels (dB) below the reference tone level. Most rumble measurements are made through a "weighting" filter which accentuates the higher frequency components (above 20 Hz) so that as the frequency of the rumble increases up to about 500 Hz, the indication of rumble level also increases. This system is fine except for the subsonic rumble which may be present in the turntable. We will discuss this later. The long-established National Association of Broadcasters standard for rumble in reproducing turntables is (for stereo) that the low-frequency rumble shall be 35 dB or more below a 100 Hz tone cut at 1 cm/sec peak velocity in either plane. High frequency rumble is to be at least 50 dB below 100 Hz, at a peak velocity of 5 cm/sec. These are not up to the performance standards of today's turntables and can be plainly audible. Any turntable aspiring to the high-fidelity market should have rumble down at least 45 dB, and the better units exhibit rumble down 60 dB or more. Many turntable manufacturers are switching to the DIN B method of measurement or the similar Japanese standard; both are somewhat similar to the NAB technique but are more stringent.
Poor design in under-deck components of a turntable can cause problems with hum. Inductive fields from the motor can induce hum in an unshielded cartridge or in the leads from the cartridge to the turntable output jacks, which are often unshielded for increased flexibility. The rumble figure for a given turntable almost always includes the hum and noise in the output as well as rumble from mechanical sources.
Two Basic Design Choices
Now that we have examined the various criteria and measurement standards by which turntables are judged, we can more easily understand the methods various manufacturers have used over the years to achieve these standards.
There are two: The method used for transferring the power from the motor to the turntable platter, and the type of motor used.
The simplest turntable motor, found in kiddie record players of the $29.95 variety as well as in a number of high-priced stereophile models, is the 1800-rpm, 4-pole induction or "squirrel-cage" motor. The 60-Hz line current is applied to the four field windings which form "poles" in such a manner that a rotating magnetic field is produced. In a 4-pole motor, this field spins at 1800 rpm. For a n-pole motor, the rotating field speed ("synchronous" speed) is found by using the equation:
(power line frequency)^2 / [n/2] = synch speed (rpm)
The rotating part in the middle of the motor, called the rotor or the armature, catches and is spun by the rotating field. Because of the losses in the air, and in the rotor itself, it never reaches full synchronous speed, but rather a speed about 3% lower. This loss in speed is called "slip." The induction motor is about equally sensitive to variations in power line frequency and to voltage. If the frequency changes, the speed of the rotating field will change, and if the voltage changes, the amount of slip will change.
Further, since it is not a constant-speed device, it will also change speed with variations in load (the amount of work it's required to do). Induction motors can be made very cheaply and are fairly stable in speed if the load, power line frequency, and voltage remain stable. They also produce a higher amount of torque for a given motor size (and cost), and thus are used to run all sorts of mechanical contraptions, not to mention record players.
The first constant speed motor for turntables was the hysteresis synchronous motor. In this type, the rotor's internal magnetic structure is changed so that the slip is reduced nearly to zero. This makes the synchronous motor dependent almost entirely on the accuracy of the incoming power line frequency for its speed stability. Since in the U.S. this is usually 0.04% or better, it is a reasonable standard upon which to hang something. The line voltage can vary as much as 10% before the slip in a synchronous motor will change enough to change its speed. Synchronous motors can be made constant in speed within about 0.1% for a given power line accuracy. The speed of a synchronous motor may be computed using the same formula as for the squirrel-cage motor. Thus we find that a direct-drive synchronous motor needs 216 poles to operate on 60 Hz line current.
Incidentally, a synchronous motor going that slowly, and supplying enough torque to drive a 12-in. turntable would have to be about a foot in diameter.
At least one company has combined the virtues and liabilities of the two motors described so far into one unit.
Garrard has used this type of motor for many years under the trade name Synchro-Lab. This motor has better speed regulation than a standard induction motor, and greater torque for a given size than a standard synchronous motor of similar size.
So far we have been dealing with motors operating directly from the power line, and dependent on the power line voltage or frequency for their accuracy. The power line is often, at any given moment, at a quite different voltage from a moment previous, although in the U.S. the frequency accuracy is very good. It is possible to sample the actual speed of the turntable and compare it with a frequency standard generated within the turntable, and adjust the speed of the motor so that the two correspond. This general system, whether it uses an a.c. motor or a d.c. motor, is called servo-control. In the a.c. system a synchronous motor is fed the output of an amplified oscillator. The speed of the turntable is sensed and compared with the separate oscillator, and the frequency of the motor-driving oscillator is adjusted to stay in step. This eliminates speed fluctuations caused either by power line variations (within reasonable limits) or by mechanical changes in the motor drive parts.
It depends only on the accuracy of the standard oscillator and the response time of the servo-control. In a d.c. servo system, the speed variations are compared with the standard oscillator, and the difference is converted into a change in the d.c. power going to the motor.
Fig. 1--Exploded view of a servo-controlled d.c.-powered motor (Dual 701 turntable).
Right (A) house servo circuitry, middle (B) is a field coil assembly, left
(C) is rotor, atop which goes platter.
Standard d.c. motors use a commutator and brushes to mechanically switch the polarity of the magnetic fields in the armature. In motors required to have stable speed, this causes a problem in that the rotating field is not a continuously moving force, but rather a series of repeated impulses. Problems also arise when the commutator and the brushes wear to the point where they arc every time the brushes pass a particular part of the armature. Most d.c. operated motors used in high fidelity equipment are not really d.c. motors at all. Most are either synchronous motors driven by d.c.-powered oscillators, or else use electronic means to switch the power going to the various coils instead of brushes and a commutator. A series of coils around the circumference of the motor case is fed a signal genera ted by the servo-control circuitry. This produces the same rotating field used in synchronous and induction motors. In order to keep the output speed constant, the speed of the motor is sensed, either by photoelectric means or through another series of coils and magnets which generate an a.c. signal proportional to the motor speed as the motor turns.
This is then fed back to the oscillator in an inverse feedback loop to keep the motor speed constant. It is possible to achieve very high stability in a motor of this kind, as shown by wow and flutter figures for these turntables being typically less than 0.1%.
Several interesting developments have been made in motor design, one of which is the "inside out" synchronous motor. This uses the same principle as a standard synchronous motor except that the field coils are on the inside of the motor and the armature (rotor) spins around them.
This permits the rotor to be bigger for a given field coil size providing greater flywheel effect, hence less flutter.
Once the motor has gotten up to the right speed its power must be transferred to the turntable platter, to turn the record. There are three different ways to effect this transfer: idler-rim drive, belt drive and direct drive. Each method has advantages as well as disadvantages. You can best decide which system is appropriate for you when you know how each works.
Idler-Rim Drive
Idler-rim drive, the most common type, uses a relatively high-speed motor. This can be either an a.c. line-operated unit, or a servo-controlled motor. The motor has a step ground shaft, with various diameters corresponding to different speeds. When the turntable is turned on, a small rubber wheel (called an idler, puck, or tire) is engaged between the motor shaft (at the proper diameter for the desired speed) and some drivable surface of the platter. The turntable turns in the same direction as the motor shaft, at a speed dependent on the motor speed and the ratio between the motor shaft diameter and the diameter of the driven part. For higher speeds, the motor shaft diameter is larger, while the rotating speed of the motor remains unchanged. Flutter can be reduced to a very low level if the coefficient of friction between the motor shaft and the idler, and that between the idler and the platter drive surface is made high enough ... in other words, if the rubber is good and live. Very little torque is lost in this type of drive system, so record changing mechanisms operated from the platter instead of directly from the motor itself still get plenty of power to operate. Another advantage is that speeds can be changed easily, just by shifting the position of the idler vertically so that it Contacts different diameters of the motor shaft. Because of their high torque capability, rim-drive turntables also start quickly, making them useful in broadcast applications where instant cuing is required.
Fig. 2--Dual 1219 uses typical idler-rim drive. Note stepped motor shaft at
right of the idler which contacts both the shaft and the inside of the platter
when installed.
The main problem with the idler-rim drive system is rumble. First of all, the motor must run at a relatively fast speed in order to drive the turntable at a proper speed through the reduction of the idler. This makes the rumble frequency higher to begin with than it would be with a slow speed motor, and therefore more objectionable. Second, the motor is connected directly to the platter by means of the solid idler. Thus any vibration in the motor is transmitted to the platter and comes out as rumble, even if the motor is totally isolated from the turntable in its mounting. If the idler is made too spongy, to eliminate the rumble transmission, it gets slippery. Make it stiffer, to transmit more torque, and you're back to lots of rumble. With proper design of the motor, support fixtures and drive mechanisms, the rumble of rim drive turntables can be reduced to an acceptable level. One interesting approach to the rumble problem inherent in idler-rim drive turntables is used by only one manufacturer, the Swiss Lenco Company. Instead of a vertical motor shaft and horizontal idler, the motor is suspended horizontally by springs, and the idler wheel engages between the motor shaft and the underside of the platter. This seems to make the rumble mostly vertical, rather than lateral or radial, so that the standard stereo reproducing cartridge does not pick up as much noise as with conventional idler-drive systems.
Belt Drive
The first commercially built turntables which merited the phrase "high fidelity" were belt-driven ones from Components Corp. during the mid-Fifties. Interestingly, Emile Berliner, who developed the flat phonograph disc, also used belt drive. With this drive system, the motor has a pulley on the end of its shaft that drives a flexible rubber belt. The motor can be outside the platter and the belt around the platter's outer circumference, or it can be inside, as with a rim-drive turntable, driving an inner pulley machined into the platter. Motors and speeds used are often similar to those in rim-drive turntables, as the ratio of motor diameters to drive-surface diameters is similar. Nearly all of the rumble in the motor shaft is damped out by the elasticity of the belt. Once the motor and platter get up to speed, flutter and wow are reduced to very low levels. The main problems with belt drive are functions of the belt itself. Quite a' bit of torque is lost in stretching the belt while bringing the" relatively massive platter up to speed; thus belt-drive turntables usually take at least a full revolution to get up to the playing speed at 33 1/3 rpm. Further, this torque loss limits the use of complex record-changing devices which use the main turntable motor for power. Another problem with belt drive is the difficulty in changing speeds. You either have to electrically switch the speeds of the motor, which requires a d.c. or oscillator-driven a.c. motor (because the speed ratios commonly available in ordinary a.c. motors do not correspond to the ratios of record-playing speeds), or else mechanically change the belt from one section of the driving pulley to another.
Direct Drive
Perhaps the most promising new type of turntable design for home use is direct drive. With this system, the power from the motor does not pass through any speed-reducing ...
[...missing content ...]
...weight (1/2 to 1 1/2 grams), while others will not stay in the groove accurately ("track") at less than 3 grams. In most cases the cartridge weighs considerably more than the recommended tracking force, so the arm, by means of a spring or counterweight, balances this out except for the recommended tracking force.
Tangent error is that error in the angle of the cartridge with respect to exact tangency with the circumference of the record. When this error is too great, the edges of the stylus can grind off the delicate ridges of the record grooves, making the sound permanently muddy. Smaller amounts of tangent error will simply cause distorted reproduction. For a conventional tone arm, tangency only occurs at one point along the arm's arc across the record, however, there have been two methods developed to reduce this error to near zero, and these are described later on.
The bearings in the pivot of the tone arm are quite important, since their design (and condition) determine how much force the stylus must exert on the outside groove edge in order to stay in the groove. Kiddie record players and inexpensive changers use very simple sleeve bearings, which require quite a bit of force to move compared to the ball, point, knife, and magnetic bearings used in the better manual turntables and separate arms.
One final consideration is arm resonance. This is the frequency or frequencies at which the arm will vibrate to some extent when the stylus comes across a note of this frequency on the record. This can cause strange peaks or dips in the frequency response of the system near the resonant frequency. Really bad effects can occur if the rumble frequency of the turntable motor is at or near the arm resonant frequency. The rumble would be magnified many times and could cause severe intermodulation distortion of the audio.
Several means have been used to lower the resonant frequency of arms below the audio range. The arm can be a metal tube, filled with a wood dowel or fiberglass wool, or the entire arm can be made of wood, or duraluminum, which is substantially less likely to resonate than other metals.
Fig. 6--Simplified drawing of servo-controlled straight line arm, Rabco SL-8.
Fig. 7--Pantograph action of articulated arm developed by Garrard.
Minimizing Tangent Error
Four types of tone arm designs have been used over the years, none of them completely ideal. The main difference between them has been in their approach to the problem of tangent error.
The straight arm with the cartridge mounted at an angle on the end has the largest tangent error. Straight arms are inexpensive to make, and with proper selection of the angle at which the arm pivot and cartridge shell are mounted, they can be perfectly adequate for many listening purposes.
One obvious and simple solution to the tangency error of a straight arm is simply to make the arm longer and bend it into an "S" shape. One can in this way significantly decrease the tangent error without resorting to an arm which is longer overall.
The first approach to the absolute solution of the tangent error problem was the straight-line pickup arm, and it wasn't really a new approach for the disc process at all. All professional recording lathes use this design to drive the cutter head across the surface of the disc by a spiraling lead screw. The straight-line reproducing arm can, be of two types. The first is free to move across the surface of the record, being suspended on one or more metal rails. This presents the obvious problem of friction. The other straightline reproducing arm design uses a servo motor to drive the cartridge across the record. The straight-line arm was first introduced in the fifties under the name Ortho-Sonic. Since then a number of companies, including Rabco and Marantz, have made arms using this technique.
Finally, the pantograph arm presents an entirely different solution to the tangent error problem. In this type of arm, also introduced in the 50s, some part of the arm is constructed so that the angle of the cartridge head is continuously varied with respect to the tangent of the record circumference. In the original Ortho-Vox arm, the entire arm would change angle as it progressed across the surface of the record. In the more recent Garrard design, the cartridge head is pivoted, one corner being connected by a rod parallel to the main arm to a series of levers and cams at the main arm pivot. This continuously changes the tangent angle of the cartridge shell with respect to the record.
Both straight-line and pantograph designs reduce the tangent error to a point where it ceases to be a factor. It can't ever be exactly zero, but either design effectively eliminates the problem.
It has been argued that the infinitesimal betterments of sound quality achieved with these various rumble-reducing and tangent-error-reducing systems are for naught; that the average commercial record quality does not even approach that of the finer home audio systems. This is true; the average commercial record quality is pretty poor compared to the best. But the aim of audiophiles is to extract every bit of realism possible. Thus every little bit of garbage one can remove from the signal helps, even with the worst commercial records. And, of course, you can only derive the fullest enjoyment from the best recordings when you have a system that will allow every bit of the recorded sound to get out of the record and into your ears.
(Adapted from: Audio magazine, June 1975)
Also see:
Automatic or Manual turntables--which one to buy? (Jun. 1973)
Trackability--1973 (Aug. 1973)
Linear Tracking Turntables (June 1981)
Which Tracks Best--A pivoted or a radial Tonearm? (Jun. 1982)
A New Standard in Turntable Speed Constancy (June 1978)
How Vinyl Records Are Made (June 1980),
Understanding Tonearms (June 1980)
How Phono Cartridges Work (Mar. 1982)
Construct a Magnetic Cartridge Preamp (Jun. 1974)
= = = =