Disks have been manufactured in many different ways over the last hundred years, from early acoustic disks, (electrical) phonograph records, and various types of optical disks, the most popular of which is today's compact disc.
The single most important factor in restoring audio from early disks is proper playback, which includes disk cleaning, stylus selection, stylus tracking weight, correct speed and equalization.
Early disks were made on shellac, aluminum, glass and even paper, and were often comprised of a sandwich of materials. The earliest disks were recorded and played mechanically, while later disks were intended for electrical playback.
For many years, until about the 1960s, "acetate" or lacquer disks were made one at a time by physically cutting grooves in a lacquer which was coated on a metal or glass base. Except for the period during World War II when they were made on glass, most of these were on aluminum disks. These disks were made for demo recording purposes, delayed radio broadcast, or other limited purposes. "Acetate" is a misnomer, since they contain cellulose nitrate lacquer, and not acetate.
Until open reel recorders replaced them, these disks were the main method of recording audio for limited use and distribution. In the 1950's vinyl replaced the other materials for mass produced records, and in the US during the 1960's the 12" diameter stereo Lp became the most common format for recorded music.
back to top
Disk Cleaning
Vinyl disks may be washed in tepid water with a mild detergent, using the fingertips to lightly rub the grooves. A problem with washing is removing the soap and water completely in order to prevent drying marks.
In the case of lacquers, the surface is often coated with a white powdery substance which is most likely hexadecanoic acid. Since hexadecanoic acid is insoluble in water, as a very last resort you can attempt to dissolve the powdery white acid using N-propyl alcohol (do not use drugstore variety isopropyl alcohol). If you leave water on a lacquer disk, it can damage it since it is similar to lacquer used in wood finishes.
Always make a definitive transfer first before trying chemicals on lacquers, since the results may be unpredictable and even disastrous.
Many people prefer to play vinyl disks wet as a matter of course, and when wetting is needed, it is helpful to reduce the surface tension of the water by adding a few drops of glycerin to the water before wetting the disk. After playing, flush the surfaces thoroughly with distilled water to reduce the possibility of drying marks.
A piece of pure cotton velvet can be used to remove excess moisture, but always allow the disk to thoroughly air dry before putting it back into its protective sleeve.
Deteriorating (evidenced by cracks and delamination of the surface) lacquers, paper or sandwich type disks should not be washed at all.
Cotton velvet and a goat hair brush can remove surface lint and light soil from a dry disk.
Often a disk will pick up a clump of paper fiber from a deteriorating sleeve, and this can usually be popped off by very gently sliding a fingernail against it, always in the direction of the grooves.
Stanton made a small disk cleaning kit which contained fluid, a magnifier and a brush. The brush is especially useful; because its bristles are very densely packed, it can be used to clean a problem spot.
For the technically inclined, the Audio Engineering Society has published two anthologies entitled Disk Recording, Vols. 1 and 2.
back to top
Stylus Selection and Tracking Pressure
There are two major makes of vintage styli, those based on Stanton Magnetics cartridges (which may no longer be available in a wide selection) and those based on Shure Bros. cartridges (the latter styli may still be made by the Expert Stylus Company in the UK).
Selection of styli is no trivial matter, since a stylus that is too small will play at the bottom of the groove and will be quite noisy, while a larger styli will ride too high and cause distortion. In some cases of groove damage, selection of a marginally larger needle will allow the styli to ride higher in order to avoid damage, but not so high as to be distorted. Likewise, styli are available in a variety of shapes, viz. truncated, truncated elliptical, ball, and even bi-pointed.
As a general rule, lacquers use a stylus measuring 2.0 to 3.0 mils, while Lps and 45s use a 0.7 to 1.0 mils stylus (both figures are using Stanton measurement). These starting points are for truncated elliptical styli, but truncated and ball styli exist as well, and should be considered.
Bi-pointed styli are made for playing stampers and other metal parts which were used to manufacture the disk.
Stampers are negative images of the finished product and are often made of chrome plated copper. Instead of the stylus resting in the groove, the bi-pointed stylus straddles a ridge which is the negative image of the groove. This gives rise to all kinds of problems, not the least of which is that the sound is played backwards on normal turntables, making playback equalization and groove tracking a challenge. But when it works, the sound can be phenomenally good.
Tracking pressure should be on the heavy side with older records. In line with this, when skips occur, it helps to examine the grooves with a loupe or cutting scope to determine the reason. If it's just a bit of embedded dirt it's easy to fix, but a scratch might require a tug on the anti-skate mechanism or gentle, steady force on the tone arm with a goat hair brush. Practice.
back to top
Speed Correction
Older disks were recorded nominally at either 33 or 78 rpm, although the actual speed for the 78 rpm records could vary from 60 to over 100 rpm. After 1940, however, speeds would be much closer to 78 rpm.
Although you will often hear the term "pitch correction" when applied to older recordings, it is more accurate to think in terms of speed correction.
Especially when preparing material for CD mastering, it is essential to mechanically play back the analog source material at the proper speed, because it is difficult to make speed corrections in the digital domain which sound as good.
In the digital world, time and pitch changes are different processes, and there are limits to the amount of changes that can be done before artifacts become noticeable.
Speed correction will affect tempo and pitch. Without knowledge of the exact playback speed, speed correction is often a musical decision, although in some historic recordings other assistance is available. This assistance is in the form of a trace of 60 Hz AC hum which can be use to calculate playback speed. Sometimes this is a pitfall if the source is a re-recording, or in rare instances where the power company was not supplying power at precisely 60 Hz.
Assuming the 60 Hz hum is accurate, there are at least two ways to proceed, one of which is to calibrate a parametric equalizer to notch out a known good 60 Hz signal source, and then to adjust the speed of the disk player to null the hum when playing the source material.
Another way is to perform a FFT analysis of the source material (containing the 60 Hz AC hum), and then to calculate the playback speed required to bring the hum to exactly 60 Hz. This requires care to avoid calibrating off of any hum that might be generated in the playback chain. Both methods require a relatively quiet section in order to examine the residual noise floor. If there is not enough 60 Hz AC hum to be useful, then a musician can do the job. Not all musicians are actually capable of this, especially where the material was not done to A440 pitch in the first place. This is true of much traditional jazz and blues, so in those cases it may be necessary to find a musician who knows what key a particular composition was performed in.
If a musician brings his own instrument, then the playback speed is simply adjusted until it falls into place. A chromatic tuner can also calculate how many half tones away the source material is. The adjustment in speed can then be easily made.
back to top
Equalization
There were numerous schemes used for equalization on early disks, and only after the early 1950's when the RIAA equalization curve became standard did playback equalization become predictable.
Any disk that did not use the RIAA equalization curve requires a special phono preamp with adjustable playback equalization. One model, the Owl 1 phono preamp, was designed by Tom Owen of Owl Investigations. The procedure for setting playback equalization is to put both the Turnover and Rolloff controls in the flat positions, and then to adjust Turnover for the proper bass response. Once this is done, the Rolloff control is set for proper treble response; in many cases the Rolloff setting will be left flat because there was no treble emphasis in some older recordings.
The Elberg MD12 Mk2 phono preamp has various fixed settings (shown at left) and is designed for stereo use.
If subsequent processing is needed for computer-based noise reduction, it is important to leave the high frequencies intact, and perhaps even slightly boost them. The reason is that computer-based noise reduction processors require the presence of broadband bursts of the higher frequency energy in order to trigger their repair algorithms. Pops and clicks are essentially bursts of white noise, and if they have been limited to say, 5 kHz, then automatically applied computer based noise reduction solutions are rendered essentially useless. Manual removal of impulse noise should only be used for cleanup purposes because it is extremely tedious and time consuming.
The use of any single-ended analog noise reduction processes typically destroys the high frequency response, and for this reason older disks should be played directly through CEDAR, NoNOISE, etc. before any analog process that rolls off high frequency material. This is often a problem with collector-supplied material, since a collector will want to make a dub instead of loaning the original disk.
back to top
Handling, Storage and Cleaning
Click here for more information from the Library of Congress on handling, storing and cleaning disk recordings.
back to top
Compact Disc