Before delving into record equalization, we should first look at all of the variables found when playing records of undetermined type:
Be aware that the materials in records made between 1939 and 1946 are substandard, due to shortages, caused by Japan holding the world's primary sources of shellac. You will encounter the fragile binder-filled shellac, and records with cardboard or glass cores and shellac surfaces. Do not use these records with record changers, and be careful not to bump the edges or let them get wet beyond a quick cleaning.
The recording curve for phonograph records was not standardized until 1958, and was not universally accepted until 1975. Before that, each company selected the curve that they thought would provide the best possible recording. Because of this disagreement between record companies, you must know how to equalize records made before 1958, and some re-issues and foreign records made after that. This article will show you how.
There are several phases in the history of recording phonograph records. Each phase has its own set of recording characteristics. The phases are:
THE ACOUSTIC ERA - 1877-1926: During this period, all recordings were made by collecting sound in a horn and funneling it to a mechanical groove cutter. Most of these cutters produced a constant amplitude cut, where the groove excursion is proportional to volume. The curves produced vary widely, and are certainly not matched to other recording curves. But they are similar enough to them to allow the same switchable reproducing pre-amp to handle them.
Note that record speeds were not standardized during this period. The direction the modulation moves the stylus was also not standardized, with some companies using vertical modulation (moving the stylus up and down) and some choosing lateral modulation (moving the stylus toward and away from the center spindle). One company used a 45 degree modulation so its records could play on any player. In a few cases, the stylus was a radically different size from the normal steel needle. Also, although all records turned clockwise, some were cut outward, starting at the center of the disc and ending at the rim.
THE EARLY ELECTRICAL ERA - 1926-1937: The development of the audio amplifier and the electric cutter produced a different basic cut, the constant velocity cut. In this type, the velocity of the stylus excursion is proportional to volume. This means that bass notes produce wider swings of the cutter stylus than treble notes of the same volume. The result was a very short playing time, because the groove spacing had to be increased.
To counteract this, the record companies rolled off the bass at 6 dB/octave below a selected frequency, called the turnover frequency. This frequency was selected to produce what the company thought was the best playback on an average acoustic machine. Each company selected its own turnover frequency, and it was considered a trade secret, so the average consumer didn't know what frequency to use, even if his set had a turnover control. This led to the introduction of the tone control on phonographs.
Records were also standardized to operate record changers during this period. By 1932, broadcast transcription discs were the only records still made by cutting outward. Many broadcasters alternated inward and outward cuts on long programs, so the radio listener could not tell that a new record had been started from sudden changes in quality. For voice broadcasts, the speed of 33.33 RPM became standard. In addition, all records for home use were modulated laterally after 1929, with vertical modulation used only for broadcast purposes.
THE MATURE 78 ERA - 1937-1947: Record companies found out that they could reduce surface noise by boosting the treble by 3 dB/octave or 6 dB/octave above a certain frequency. A corresponding reduction in treble (and surface noise) occurs on playback. This produced better sound than was ever possible before.
But again, they could not agree on how much to strengthen the treble. For some reason, instead of specifying a second turnover frequency for the boost, the treble boost was measured in dB at 10 KHz. Some of the records made with this method are as good as early LP records. This development also produced the first wide variation in recording curves, especially since most European companies did not strengthen the treble until after World War II. This new characteristic is called treble roll off. Again, the precise amount of treble rolloff to use for each brand of record was kept secret from the public. For the first time, some phonographs made during this period had treble and bass controls, or a tone control. Others had a switch with "American" and "European" on it - the first switchable record compensators.
THE BIRTH OF THE LP - 1947-1958: With the new slower rotation speed and a new material (vinyl) to make records out of, a new set of recording curves emerged. The first two, LP and NAB, were derived from the National Association of (Radio and Television) Broadcasters (NAB, or NARTB) curve, which was used to make acetate radio program transcriptions.
The NAB transcription curve always had a small rumble shelf, to reduce the effect of turntable rumble. Turntable rumble increased with slow rotation speeds and smaller grooves. The Columbia LP curve adds more to the rumble shelf. But the treble boost from the LP and NAB curves was so much that it caused mistracking in cheaper pickup cartridges. This caused some record companies to adopt more gentle recording curves.
One such curve was designed by the Audio Engineering Society (AES), but the surface noise on records cut with that curve was much more audible. Then RCA stuck its nose into the mix with the 45 RPM record, with an entirely new curve as well as a new speed and center hole size. As a result, several different curves were adopted by different record companies.
By this time, hi-fi enthusiasts clamored for the record companies to publish this information, so the user could adjust his set to handle each curve correctly. Most 33 RPM records made after 1951 had information printed on the label or jacket telling which curve to use. But most 45 and 78 RPM records didn't display this information.
THE RIAA YEARS - 1952-present: Hi-fi enthusiasts called for a single curve that all records would be made in. In 1952, RCA developed the New Orthophonic curve with all of the solutions to problems listed above compounded into one set of record and reproduce curves. Within a year, the Recording Industry Association of America (RIAA) adopted this curve. By 1955, most American labels had adopted the RIAA standard.
The only changes to the curve were definitions adopted in 1974 to set the ultrasonic and infrasonic characteristics of records and playback pre-amps. This change was prompted by the introduction of the CD-4 quadraphonic disc system with its ultrasonic carrier. Note that some European companies did not make the change to RIAA until the 1960s, and some Russian and Asian companies didn't standardize until 1975.
STEREO - 1958-present: When stereo was introduced, many different ways of putting two audio tracks on the same record were discussed. Cook developed a disc with two sets of grooves, played with a forked tonearm. Minter tried to put a radio-frequency carrier into the record groove, a process that failed just as badly back then as it did with CD-4 in the 1970s. A worn record sounded like sandpaper, and the record could not be slip-cued. The Zenith system finally chosen records the left channel at a 45 degree angle from horizontal in one direction, and the right channel at a 45 degree angle in the other direction, in the same groove. A sound centered between the speakers is recorded laterally. And sounds to the back of the room in quadraphonic and Dolby Surround systems are recorded vertically.
Now let's look at the actual components of the RIAA curve and other curves used for reproducing records:
At some point (undefined until 1974, where it was set at 15 Hz) is the low-frequency cutoff of the system. Usually it was defined by the limits of the cutter, amplifiers, and loudspeaker, until extended low frequency response became available with the use of transistors. |
The flat midrange section comes next, between the turnover frequencies of the bass boost and the treble rolloff (500 Hz and 2150 Hz in RIAA). This is the original flat portion of the recording curve around which the rest of the curve is built. |
Above the low frequency cutoff point is the rumble shelf. This band of flat response (between 15 Hz and 50 Hz in RIAA) was inserted to keep from boosting rumble frequencies too much with the rest of the bass that is boosted in the next section. Not all records have rumble shelves. |
The next part of the curve is the noise reducing treble roll-off (from 2150 Hz to 22 KHz in RIAA). This compensates for the treble boost in the recording curve. This is expressed in dB of reduction at 10 KHz (-13.7 dB for RIAA), and is usually a 6 dB/octave slope with more reduction at higher frequencies. |
Next comes a section between the rumble shelf and the bass turnover frequency (50 Hz to 500 Hz in RIAA). In this section the bass is boosted at a 6 dB/octave slope, with the lower frequencies being stronger. This boost compensates for a roll-off in the bass during recording. This is done to allow more recording time on the record. |
Finally comes the upper limit of the system's frequency response. Until recently, this was usually a function of the cutter head. The advent of the aborted CD-4 quadraphonic system forced a need for a standard cutoff frequency, which was set at 22 KHz in 1974. |
The low and high frequency response limits are fixed at 15 Hz and 22 KHz. The other values are dependent on which recording curve is used. The curve is usually specified by three values, listed in this order:
The RIAA curve is identified as 500R-13.7, meaning a 500 Hz turnover, a 20 dB high rumble shelf, and a 13.7 dB rolloff at 10 KHz.
Other common curves: |
||
|
|
|
Use these table pages for the complete list of recording curves used, and which record companies used which curves:
Tips to finding the correct playback equalization of a record of unknown vintage:
LINKS: