Disc Recording Equalization Demystified

By Gary A. Galo

The subject of disc-recording equaliza- and much more. You can obtain mem- a record is known as the recording char- tion has generated much confusion bership information from Peter Sham- acteristic, and a typical explanation for over the years. Many knowledgeable barger, Executive Director, ARSC, PO the 33-1/3-rpm LP record is illustrated in collectors and audio professionals Box 543, Annapolis, MD 21404-0543. Fig. 1. The recording curve shows the have been content with conventional In this article, italicized terms are bass rolled off (attenuated) and the treble explanations. Transfer engineers and defined in the Glossary on p. 52. boosted, with a flatter region in the mid- collectors are well aware that electri- dle of the curve. In order to obtain a flat cally recorded discs require a bass frequency response in playback, a com- boost, and sometimes a treble cut, in INTRODUCTION plementary equalization is necessary. playback. isc-recording equalization is The playback curve shows the bass They often assume that the playback often misunderstood by boosted and the treble attenuated. correction, or equalization, compen- audio professionals and hob- This equalization is normally accom- sates only for the method by which the Dbyists alike. Even the most ca- plished in the preamplifier, which also actual recording was made. If the bass sual collector of 33-1/3-rpm long-playing provides sufficient amplification of the is attenuated during the recording (LP) records has probably encountered relatively weak signal from the phono process, it must be boosted in playback; the term RIAA equalization. Most serious cartridge (which was known in the early similarly, if the treble is boosted when collectors involved with the playback of days of electrical playback as the the record is cut, it must be attenuated 78 rpm recordings are familiar with pickup, a term still used by the British). in playback. Close examination shows terms such as bass turnover and treble If the playback equalization curve is an that the recording and playback turnover, since any audio system suit- exact mirror of the recording curve, a process is more complex. able for the playback of “historic” potentially flat response will result. Selecting playback equalization recordings must have provisions for ad- Closer investigation shows that this must take account not only of the justing these parameters. explanation is an over-simplification, at recording characteristics, but also Currently available preamplifiers with best. The “record” curve of Fig. 1 is, in those of the playback cartridge. This tu- suitable adjustment capabilities include fact, not the recording curve at all, at torial will explain the methods of cut- the veteran Owl 1, distributed by Audio least not in terms of recorded amplitude. ting disc records, the characteristics of 78, the Resolution Series high-end pre- Rather, it is the frequency response of magnetic phono cartridges, and how amps from the Swiss firm FM Acoustics, the record when played back with a their combined response determines and the Esoteric Sound Re-Equalizer, magnetic phono cartridge. the required playback equalization. which is intended to correct a modern To add to the confusion, some sources This article is based on a paper I pre- preamp’s RIAA response to match older label the rising curve constant ampli- sented at the May 1996 conference of recording characteristics. tude, and the flatter region in the middle the Association for Recorded Sound Some collectors use vintage tube constant velocity, which, on the face of Collections (ARSC) in Kansas City, and units from the 1950s, either in stock or it, makes little sense.1 These terms de- was first published in the Fall 1996 modified form, since many preampli- scribe the two basic cutting methods, issue of the ARSC Journal. The Associa- fiers from that period offer flexible but the recording curve is a result of the tion is a nonprofit organization serv- equalization settings that reflect the lack disc cutting and the response of the mag- ing librarians, scholars, sound of standards at the time they were man- netic cartridge used in playback. The archivists, dealers, private collectors, ufactured. The best of these audio “clas- principles of disc recording and playback discographers, and reviewers. The sics” include the McIntosh C-8 and the equalization are the same whether the biannual ARSC Journal is devoted to re- Marantz Audio Consolette. record is lateral, vertical, or 45/45 stereo. search on sound-recording history, As most of you are probably aware, preservation and restoration of sound disc records are not cut with a flat fre- TWO CUTTING METHODS recordings, record and book reviews, quency response. The method of cutting Constant-amplitude (Fig. 2) is the cut- 44 The LP Is Back! Magnetic transducers are velocity-sen- sitive devices—they produce a flat fre- quency response only when the record- ed velocity remains constant as the fre- quency rises. Understanding the behav- ior of magnetic phono cartridges is the key to the mystery of disc recording and playback equalization. The horizontal line at the bottom of Fig. 4 shows the output of a magnetic cartridge playing a constant-velocity recording. The cartridge output is flat across the entire recorded spectrum. The slanting line in Fig. 2 shows the car- tridge’s output playing a constant-ampli- tude recording. Here, the cartridge’s out- put increases as the frequency rises, at a rate of 6dB/octave. In these illustrations, the straight lines not only illustrate the playback cartridge’s relative output, but FIGURE 1: Typical graph illustrating the RIAA recording characteristic for 33¹⁄₃-rpm also show relative recorded velocity. long-playing records. The record curve shows the bass rolled off and the treble boosted. A complementary equalization during playback restores a flat response. BASIC REQUIREMENTS The cutting of a phonograph record in- volves two seemingly contradictory re- ting method easiest to understand. If the speed in tracing each of the waveforms. quirements. First, the record must be cut signal being recorded is at the same level Figure 4 shows a constant-velocity at a level higher than its own residual for all frequencies, the recorded ampli- recording characteristic, with the ampli- surface noise, particularly in the high fre- tude will also be the same. Constant ve- tude progressively decreasing as the fre- quencies. This at first appears easy—sim- locity is somewhat more difficult to quency rises. ply cut the record at a very high level, grasp. The velocity of the playback stylus and any surface noise will be virtually in- is the speed with which it moves while MAGNETIC TRANSDUCERS audible. Unfortunately, doing so would tracing the record groove, and is directly Disc playback normally involves the use violate the second requirement—play- related to the physical distance the stylus of magnetic phonograph cartridges, back tracking and tracing ability. travels during a given time period. whether the playback system is modest or If a record is cut at too high a level, Referring to Fig. 2, if the stylus is trac- “state of the art.” A phonograph cartridge the playback cartridge and stylus will be ing at a very low frequency, say 20Hz, it is a transducer, since it converts mechani- unable to track the record. At both low must move back and forth 20 times each cal energy into electricity (a transducer is and high frequencies, the cartridge will second. As the frequency rises, the num- a device that converts one form of energy be unable to cope with excessively ber of times the stylus must move back into another). There are several variations wide groove excursions. The excursion and forth in one second also increases. on the magnetic-cartridge theme, includ- is the physical distance the stylus must At 10kHz, for example, the stylus must ing moving-magnet, moving-iron, and travel from the center of the groove move back and forth 10,000 times each moving-coil. In principle, all function the modulation (called the zero crossing) second. If the amplitude of the signal re- same way—a magnetic field is in motion to either peak. mains constant, it stands to reason that relative to a coil of wire. At high frequencies, there is an addi- the stylus velocity must increase as the frequency rises. In order to keep the stylus velocity FIGURE 2: Constant- constant at all frequencies, it is necessary amplitude recording to reduce the amplitude of the recorded characteristic. The signal as the frequency increases. Figure recording amplitude is 3 illustrates this concept. The large held constant as the waveform is one cycle at an arbitrary fre- frequency increases. quency—the exact frequency does not The diagonal line matter for the purposes of illustration. shows the relative Each time the frequency doubles, the output of a magnetic amplitude must be cut in half to keep cartridge, and also il- lustrates the increase the velocity constant. in velocity as the fre- If each of the three waveforms in quency rises. Fig. 3 were made with a piece of string, the lengths of all three pieces would be identical. The stylus travels the same physical distance during the time period, and thus moves at the same rate of • THEORY AND CARE • The LP Is Back! 45 problems. The physical mass of the sty- FIGURE 3: Conceptual lus assembly limits the velocity at which illustration of con- the stylus can travel. stant-velocity record- ing. In order for the HYBRID APPROACH playback stylus to Maxfield and Harrison opted for a hybrid travel the same phys- ical distance in a recording characteristic that used both given time period, constant-amplitude and constant-velocity the amplitude must cutting to best advantage. Figure 5 be cut in half each shows the electrical-recording character- time the frequency istic used during the early years of the is doubled. 78-rpm era. The lower frequencies used a constant-amplitude characteristic, while that of the treble region was a con- stant-velocity type. The point where the transition from a constant-amplitude to a constant-velocity characteristic occurs is known as the bass turnover frequency, or simply the turnover frequency. The constant-amplitude characteristic tional problem, technically called the ra- does not refer to high or low pitch (i.e., limits the groove excursions in the bass dius of curvature, that is directly related frequency), but instead to the spacing of region, thus minimizing stylus-tracking to the physical size of the playback sty- the record grooves. Normally, groove problems and maximizing playing time. lus. As you can see in Fig. 2, the play- pitch is defined as the number of lines The constant-velocity characteristic mini- back stylus must change direction as it per inch across the radius of the record mizes high-frequency tracking and trac- passes through each peak of the groove surface. (Adjacent grooves are called ing problems by limiting groove excur- modulation. At low frequencies, the turn lines, since a record technically has only sion and preventing an excessively nar- is far less sharp than it is at high frequen- one groove—a continuous spiral from the row radius of curvature. A comparison of cies; i.e., the radius of the curve is beginning to the end of the side.) If the Figs. 2 and 4 shows that reducing the longer. excursions are extremely wide, the spac- amplitude at a given frequency increases As the frequency increases, however, ing must be increased to prevent bridg- the radius of curvature. the radius becomes shorter and shorter, ing of adjacent grooves, which would At high recorded amplitudes, the lines and a point is reached where the radius decrease the number of lines and thus re- connecting the peaks of the waveform of the playback stylus is actually greater duce the playing time. are steeper; consequently, the radius at than that of the curve. This results in The second problem with constant-ve- the peaks is smaller. Maxfield and Harri- tracing distortion. If the tracking force locity recording is that relative to the low son were not concerned about high-fre- is heavy enough, the playback stylus frequencies, the highs are recorded at an quency surface noise, since the high-fre- will cut its own path through the vinyl extremely low level, potentially lower quency response of the first electrical or shellac, destroying the original than the surface noise of the record. If a recordings was extremely limited. recorded vibrations. These problems wide frequency response is possible, Maxfield and Harrison initially set the were fully understood by Maxfield and constant-velocity cutting is not desirable bass turnover frequency at 200Hz. From Harrison,2 the inventors of electrical at extremely high frequencies, since the 200 to 4kHz, they employed a constant- disc recording. surface noise will mask the treble region. velocity characteristic. In their 1926 This was not a concern during the early paper, they describe using an approxi- UNIQUE ADVANTAGES years of electrical recording, since it was mate constant-acceleration characteris- Neither a constant-amplitude nor a con- not possible to achieve a wide frequency tic between 4k and 6kHz.2 For the pur- stant-velocity recording characteristic response at that time. pose of this discussion, all that need be can meet all of the requirements, yet The particular advantage of constant- said about constant acceleration is that, both have unique advantages. If a record amplitude cutting is that it works well at as the frequency rises, the amplitude de- is cut with a constant-velocity character- low frequencies. It holds groove excur- cays even more rapidly than with con- istic, playback equalization is unneces- sions to a reasonable level, since the stant velocity—6dB/octave faster, to be sary, since the magnetic cartridge has a recorded amplitude does not increase as specific.1 They used constant accelera- flat frequency response playing this type the frequency drops. It also minimizes tion to further minimize tracing prob- of recording. However, constant-velocity the high-frequency noise problem, be- lems in the treble region due to an exces- recording has two inherent problems. cause the recorded amplitude will al- sively short radius of curvature. First, since the recorded amplitude ways be higher than the surface noise of rises as the frequency decreases, groove the record. EARLY PHONOGRAPH DESIGN excursions at low frequencies become However, there is no free lunch here, Maxfield and Harrison initially designed too large. This not only makes the record either. Recording high frequencies at an acoustic phonograph for the playback difficult or impossible for the playback these levels will also cause stylus-tracing of electrical recordings. They sold the ex- stylus to track, but it also limits the play- distortion, in part due to the radius-of- clusive rights to this design to the Victor ing time on the record. curvature problem described above. Talking Machine Company, which then Playing time is directly related to the Wide excursions at high frequencies marketed it as the Orthophonic Victrola. groove pitch, which, in this context, also cause cartridge/stylus tracking Like previous acoustic phonographs, 46 The LP Is Back! turnover) frequency, the cutting charac- teristic became, once again, constant am- FIGURE 4: Constant- plitude. velocity recording This switch back to a constant-ampli- characteristic. The tude characteristic is often referred to as recorded amplitude treble preemphasis, but this term is mis- must decrease as the leading. The treble region is boosted frequency increases. The straight line only in terms of recorded velocity—the shows the relative amplitude is still lower than it is below output of a magnetic the treble transition frequency. cartridge, and also Since the amplitude in the treble re- shows the velocity gion is substantially lower than in the being held constant bass, high-frequency tracking and tracing as the frequency problems are minimal. By the late 1930s, rises. phono cartridges possessed improved high-frequency tracking abilities that made possible a constant-amplitude char- acteristic, provided the amplitude was held at a reasonable level. this machine used large steel needles and theoretical ideal. Due to variety of me- The constant-velocity characteristic a heavy, metal sound box. The only way chanical and electrical limitations, this between the bass and treble turnover to prevent destruction of the high-fre- ideal was never realized during the 78- points then functioned as a transitional quency information during playback was rpm or LP era. Even with modern, so- region between the higher amplitude in to attenuate the response above 4kHz— phisticated disc-recording and playback the bass and the lower amplitude of the hence the need to reduce the amplitude equipment, a perfectly flat response is treble. The solid line in Fig. 6 shows the even more rapidly above this frequency. rarely achieved. response of a magnetic cartridge, differ- Bell Laboratories subsequently modi- ing from Fig. 5 in that the cartridge out- fied the cutting apparatus to allow a con- NONSTANDARD TURNOVER put rises above the treble transition fre- stant-velocity characteristic up to FREQUENCIES quency. 5.5kHz. Recordings made with the modi- One problem faced by those who play The dashed line shows the required fied characteristic were best suited for 78-rpm records is the lack of a standard playback equalization—the treble is now electrical playback.3 turnover frequency. During the course attenuated above the treble transition With the introduction of electrical of the 78-rpm era, turnover frequencies frequency. The dotted line shows the re- phonographs having magnetic car- varied widely, anywhere from 250Hz to sulting flat response. tridges, it was possible to equalize the as high as 1kHz. Even a given record signal below the turnover frequency to label may be inconsistent from year to TREBLE TRANSITION FREQUENCY restore a flat response in the bass region. year, or from one recording session to ADJUSTMENT The ability to do so was an advantage of the next. Playback of these recordings is problem- electrical playback that Maxfield and Har- In order to achieve a flat response in atic, since the treble transition frequency rison recognized. the bass region, an electrical 78 must be was not standardized, but was typically The solid line in Fig. 5 shows the out- equalized with the same turnover fre- somewhere between 2k and 3kHz. A put of a magnetic cartridge playing a 78- quency with which it was recorded. If preamplifier suitable for playback of rpm electrical recording. In the constant- the playback turnover frequency is set these old recordings must allow the in- amplitude region, beginning at the low- too high, excessive, “boomy” bass will sertion of a treble transition frequency, if est recorded frequency, the cartridge re- result; if it is too low, the bass region will needed, and must also provide for the sponse rises up to the turnover frequen- sound “thin.” In order to properly repro- variation of that frequency to match the cy. Above the turnover frequency, in the duce 78-rpm recordings, a preamplifier characteristics of each recording. If the constant-velocity region, the response is with adjustable bass turnover is absolute- playback transition frequency is set too flat up to the highest recorded frequen- ly necessary (see “Demos” sidebar). high, the playback will be excessively cy. (The graphs in Figs. 5, 6, and 7 are as- bright in the treble region; if it is set too ymptotic curves. The transition points TREBLE TRANSITION TO CONSTANT low, the high frequencies will be dull. from constant amplitude to constant-ve- AMPLITUDE Every preamplifier with adjustable locity are gradual, as you can see in Fig. As electrical recording advanced, it be- high-frequency equalization is calibrated 1. An asymptotic curve simply removes came possible to extend the high-fre- in decibels of attenuation at 10kHz, with the bends in the graph for the purposes quency response of the recording well the 0dB reference set at 1kHz. In fact, of illustration.) beyond Maxfield and Harrison’s 6kHz what you are adjusting is the treble Relative to the constant-velocity re- limit. In order to prevent surface noise turnover frequency. Each setting on the gion, the bass frequencies are attenuat- from burying the extended high frequen- preamplifier produces a 6dB/octave ed. The dashed line in Fig. 5 shows the cies, later electrical 78s were cut with rolloff, beginning at a specific turnover playback equalization necessary to the characteristic shown in Fig. 6. Here, frequency. achieve a flat response; the bass is boost- the constant-velocity characteristic was It is only the turnover frequency, not ed below the turnover frequency. The not used up to the highest recorded fre- the rate of rolloff, that you are adjusting. desired flat response is shown by the quency. At a predetermined transition The lower the turnover frequency, the dotted line. The term flat response is a point, the treble transition (or treble greater the 10kHz attenuation (Table 1). • THEORY AND CARE • The LP Is Back! 47 denser mikes. In order to achieve the FIGURE 5: Recording same recorded brightness levels, Victor characteristic for early engineers added equalization to the rib- 78-rpm electrical bon-mike preamplifiers.3 This electronic recordings. At the high-frequency boost was called voice bass turnover fre- effort equalization.4 quency, the character- istic changes from A NEW RECORDING-CHAIN constant amplitude to constant velocity. The DESIGN solid line shows the In 1938, RCA Victor redesigned the en- relative velocity and a tire recording chain. They removed the magnetic cartridge’s high-frequency equalization from the mi- output. The dashed crophone preamplifiers and added it to line shows the play- the disc-cutting equipment after the back equalization re- recording bus.3 At this point, the high- quired to produce the frequency boost, or preemphasis, be- flat response shown came part of the actual disc-recording by the dotted line. characteristic. Some authors and engineers consider the recording characteristic to be the combined response of the actual disc-cut- FIGURE 6: Later 78- ting equipment, the microphones, and rpm electrical record- any equalization applied in the recording ing characteristic. A bus, which includes the mike preamps. constant-amplitude characteristic is used But this does not appear to have been the above the treble official position at RCA Victor. transition frequency. On Oct. 30, 1935, in response to an in- The solid line shows quiry from J.M. Kaar of Menlo Park, CA, the relative velocity E.C. Forman, of Victor’s Recording and and a magnetic car- Record Sales Division, described the Vic- tridge’s output. The tor recording characteristic as follows: dashed line shows “From 300 cycles down to 30 cycles, the playback equal- the recording is made at constant ampli- ization required to tude and can be represented on the produce the flat re- sponse shown by the curve as a straight line down 20dB at 30 dotted line. cycles. From 300 cycles to 5,000 cycles, the curve is flat at zero level within ½dB. At 6,000 cycles, the curve is up 1½dB, (This turnover frequency is really the a peak about 5 or 6dB at 3.5kHz.”4 with zero level again reached at approxi- 3dB point for the equalization circuit. At In 1932, Victor began using ribbon mately 6,500 cycles. The curve then the turnover frequency, the response microphones. The conversion to these trails down to −30dB at approximately will be either 3dB down or 3dB up, de- was gradual, though, and some record- 7,500 cycles. pending on whether a cut or a boost is ings made through the mid ’30s still “The above should enable you to apply involved. This is why the transitions are used the older condenser mikes. Rib- the recording curve, and we believe that gradual, as mentioned previously.) bon microphones had a much flatter if you compensate your equipment in Determining the exact dates when var- high-frequency response, and conse- line with the above curve, you will have ious record companies began using tre- quently few people liked them as well the counterpart of our recorder. Howev- ble preemphasis is difficult, and beyond as the condensers—many missed the er, you should bear in mind that instru- the scope of this work. R. C. Moyer’s arti- presence and brilliance of the con- ment placement during recording plays cle3 sheds considerable light on the evo- almost as important a part as the lution of Victor’s recording curves. actual recorder characteristic, and Many Victor electrical 78s recorded TABLE 1 that even with proper compensa- from the mid-’20s through the late ’30s TURNOVER FREQUENCIES FOR tion, certain records may have too sound bright in the treble region, even COMMON ATTENUATION LEVELS many or too few lows or highs. though Victor was using a constant-ve- This particular situation can be cor- 10kHZ ATTENUATION TURNOVER FREQUENCY locity characteristic up to the highest rected only by the use of adjustable −5dB 6.8kHz recorded frequency. The brightness −8.5dB 4.0kHz low and high-frequency tone con- more likely has its source in the micro- −10dB 3.3kHz trols to give the desired balance.” phones or microphone preamps. Early −12dB (AES) 2.5kHz − Victor electrics used Western Electric 13.75dB (RIAA) 2.122kHz VICTOR’S OFFICIAL POSITION −14dB 2.05kHz condenser microphones, which had an −15dB 1.8kHz The above letter is extremely sig- elevated response in the higher frequen- −16dB (NAB and 1.6kHz nificant in that it clearly states cies. Frayne, in his interview with Columbia LP) RCA Victor’s official position on Sutheim, notes that “The condenser had −20dB 1kHz what constituted the recording 48 The LP Is Back! equipment, a variety of factors can FIGURE 7: Modern RIAA make a record sound too bright. There recording characteristic for is nothing fundamentally incorrect 33¹⁄₃-rpm long-playing about using a preamp’s treble rolloff records. Bass turnover and switches to correct these problems. treble transition frequen- However, it is important to understand cies are standardized at what is being corrected. The actual disc- 500Hz and 2.122kHz. A recording characteristic should not be constant-velocity charac- confused with other factors that can teristic is used below 50Hz. cause brightness in playback. The solid line shows the During the late ’40s and early ’50s relative velocity and a magnetic cartridge’s there were several attempts to arrive at output. The dashed line a standardized recording characteristic. shows the playback equal- In 1950, the Audio Engineering Society ization required to produce (AES) endorsed a recording characteris- the flat response shown by tic specifying a 400Hz bass turnover and the dotted line. a 2.5kHz treble transition frequency. The AES standard was originally pro- posed as a compromise playback curve that would yield satisfactory results with a variety of recording characteristics. characteristic—it is the characteristic of high-frequency boost into the recording Some very late 78s as well as many early the disc-cutter head and its associated process in order to compensate for the LPs were recorded using the AES bass electronics. The 1½dB peak at 6kHz is a high-frequency rolloff that can occur at and treble turnover frequencies.6 cutter-head resonance, but the frequen- the inner grooves of a disc record. This cy-response characteristics of micro- was called diameter equalization. THE MODERN LONG-PLAYING phones and mike preamplifiers are not Moyer also notes that additional pre- RECORD described as part of the actual recording emphasis was applied when lacquer re- Columbia’s introduction in 1948 of the characteristic. Interestingly, Mr. Forman placed wax as the disc-mastering medi- modern long-playing record (33-1/3 does not discuss microphones and mike um. This was necessary to compensate rpm) brought another turnover frequen- preamps at all, but he clearly recognizes for high-frequency losses that occurred cy into the equation. Now there were that factors other than the recording when lacquer was cut with a cold stylus three such frequencies, and no industry characteristic can affect the tonal bal- (a heated stylus was not used until standards to govern them. In 1953, RCA ance of a recording, which may make 1950).3 Moyer does not give a specific Victor introduced the recording charac- additional compensation desirable. date for the introduction of lacquer, but teristic shown in Fig. 7 as the New Or- It is also possible that some record Powell implies that it was around 1937.5 thophonic curve. This characteristic is manufacturers introduced a moderate On modern wideband playback similar to that shown in Fig. 6, with the

DEMOS

When I presented my paper on this subject to the ARSC con- was thin and lacking in warmth. Finally, I played the Stokowski ference, I played a recording to demonstrate the effects of dif- recording with RIAA equalization. The resulting playback was ferent bass turnover frequencies in playback (a cassette tape bass-heavy with a dull, muffled treble. of the recorded examples is available; see ordering informa- I also demonstrated varied approaches to acoustical play- tion at the end of the article). The recording was an excerpt back at the ARSC conference, using an excerpt from George from the 4th movement of Beethoven’s Seventh Symphony, M. Cohan’s “Over There,” recorded by Enrico Caruso on July performed by the Philadelphia Orchestra conducted by 11, 1918 (Victor 87294). I used an original, single-faced, ten- Leopold Stokowski and recorded on April 24, 1927 (Victor inch “wing label” pressing for the transfer. First, I played the 6674-A in Album M-17). recording completely flat, with no phono equalization. Caru- The source was an original “scroll label” Orthophonic press- so’s voice was rather thin, lacking in warmth on this transfer. ing, transferred using a Technics SP-15 turntable, SME 3012R The second transfer used a flat phono equalization, with the tonearm, and a Stanton 500AL cartridge with a 2.5 mil truncat- addition of the parametric equalizer set for a 4dB boost, 0.7 oc- ed elliptical stylus. My modified McIntosh C-8 preamplifier pro- tave wide, centered around 180Hz. The vocal warmth was con- vided the equalization, and I used a modified Phoenix Systems siderably improved, yet there was no increase in low-frequency P-94SR parametric equalizer for additional rumble filtering. I rumble. The third transfer used a 300Hz bass turnover, with made the transfers directly to DAT, using a Panasonic SV-3700 the parametric equalizer removed. The vocal warmth re- DAT recorder, with no high-frequency filtering in order to mained, but low-frequency rumble was clearly audible. avoid adding another variable to the equation. Finally, I played the recording with a 1kHz bass turnover, I first played the recording with a 300Hz bass turnover, which made Caruso’s voice assume an unnatural heaviness, which resulted in a realistic balance between bass and treble, and the level of low-frequency rumble was extremely distract- and I then repeated it with a 500Hz bass turnover, which pro- ing. At higher playback levels, the excessive rumble could duced a slightly heavy, “tubby” bass reproduction. I then cause loudspeaker damage. played the excerpt with no bass equalization, i.e., a completely flat playback. The bass disappeared and the resulting playback

• THEORY AND CARE • The LP Is Back! 49 addition of the third turnover frequency The NAB curve used the same bass stant-velocity characteristic. This charac- in the low bass region. and treble turnover frequencies as the teristic yields a flat frequency response Below 50Hz (actually 50.05), the Columbia LP curve, but the NAB charac- from a magnetic cartridge, but that re- characteristic is constant velocity. The teristic did not have the low bass transi- sponse is flat over a very narrow range. bass and treble turnover frequencies tion to constant velocity described Only those who adhere to a strictly sci- have been set at 500Hz (actually 500.5) above. The 100Hz low bass turnover entific approach to restoration of his- and 2.122kHz. By 1956, the Recording was added by Columbia.5,7 toric recordings use a completely flat Industry Association of America (RIAA) Both Victor and AES found the high- playback. had adopted the RCA Victor standard, frequency amplitude to be too high with (Due to horn resonances and other and the entire industry followed suit. a 1.6kHz treble turnover, since it might mechanical limitations, acoustical The reason for the switch to constant cause mistracking and tracing distortion recordings have many irregularities. velocity below 50Hz is related to low- on wideband high-fidelity recordings. By These prevent them from having a per- frequency noise in the playback 1950 improved microphones, electron- fect constant-velocity characteristic—and process. As the turntable speed progres- ics, and disc-cutting equipment made it hence a true flat response—when played sively decreases, from 78 to 33-1/3 rpm, possible to record frequencies up to an with a magnetic cartridge.) turntable rumble and record-warp noise unprecedented 15kHz.6 Most listeners, including myself, find also become lower in frequency. The Columbia/NAB curve had origi- that acoustical recordings sound ex- The late-78 recording characteristic, nally been optimized for 16″ discs. The tremely thin in the bass when played with its playback bass equalization ex- radius-of-curvature problem is not near- back without any bass equalization. One tending down to the lowest recorded ly as severe on a 16″ record, since most solution is to set the phono preamplifi- frequency, also boosts turntable rumble of the playing surface has a high linear er for an unequalized, flat response, and and record-warp noise. If the low-fre- recording speed—and consequently then apply a discreet amount of boost in quency noise is severe enough, the ex- longer physical wavelengths—than a 12″ the 100–200Hz region with an external cessive boost can cause amplifiers and record. At any given frequency, the equalizer, preferably parametric rather loudspeakers to operate in a nonlinear physical wavelengths will be shortest than graphic. This can “warm up” the manner, producing an unacceptable near the center of the record, where the upper bass/lower midrange region, re- amount of distortion. In the worst case, linear recording speed is slowest. sulting in a more realistic sonic presenta- damage to these components, especially The RIAA, AES, NAB and Columbia tion, assuming there is any information loudspeakers, is possible. LP playback characteristics are unac- on the record at these frequencies. If a constant-velocity characteristic is ceptable for nearly all 78-rpm recordings. Another approach is to set the pream- used below 50Hz, the output of a mag- Playback of most electrically recorded plifier to a low turnover frequency, netic cartridge is flat up to this frequen- 78s with any of these curves invariably 250–300Hz; the bass boost will add cy, as shown by the solid line in Fig. 7. results in dull, muffled sound in the some warmth to the recording.8 This ap- Consequently, no bass boost below this upper midrange and treble region. Early proach can be problematic, however, frequency is needed in playback, as electric discs also sound bass-heavy, since the preamplifier’s bass equaliza- shown by the dashed line. This mini- since the turnover frequency is too high tion will also boost low-frequency noise mizes low-frequency noise problems, in- for many of these recordings. and rumble well below the usable range cluding turntable rumble, record-warp At least one source, Tremaine, has er- of the recording. It makes little sense to noise and AC power-line hum. roneously listed the RIAA bass turnover boost the low frequencies down to The flat response resulting from the frequency as 1kHz. The front panel of 20Hz if the record contains virtually no playback equalization is shown by the one of FM Acoustics’ Resolution Series musical information below 150Hz. One dotted line. Columbia had set the low preamps is also labeled this way in an ad restoration equipment dealer actually bass turnover at 100Hz for the first long- in Stereophile (May 1996, p. 80). This is recommends using a turnover frequency playing records, which was even more incorrect—1kHz is simply a 0dB refer- of 1kHz for acoustic records. effective at minimizing low-frequency ence point to which the record and noise problems in playback. RCA Victor playback equalization levels at all SUMMARY decided that this frequency was too other frequencies are compared; it is Electrical recordings are not cut with a high, since it could result in excessively not the turnover frequency. flat frequency response, but convention- wide groove excursions at the lowest al descriptions of the recording charac- recorded frequencies. Hence, a frequen- ACOUSTICAL RECORDINGS teristics used for 78-rpm and long-play- cy of 50Hz was accepted as a sensible The first commercial electrical record- ing records are often misleading. Disc compromise between reasonable ings appeared in 1925. Before then, all records are not cut with the bass attenu- groove excursions and noise levels. disc records and cylinders were made ated and the treble boosted, at least not with the acoustical recording process. in terms of recorded amplitude. Electri- MODIFIED NAB CURVE Although there was no electrical means cal recordings are made using a combi- The original Columbia LP record also by which record companies could ad- nation of constant-amplitude and con- used a bass turnover frequency of just the recording characteristic, the stant-velocity cutting. 500Hz, but its treble transition frequen- acoustical process did possess an inher- Early electrical 78s were cut with a cy was 1.6kHz. The Columbia recording ent mechanical equalization. Even the constant-amplitude characteristic in the characteristic was a modification of the best acoustic records were limited to a bass region, and a constant-velocity char- National Association of Broadcasters frequency range of 150–4kHz, and most acteristic in the treble. Later electrical (NAB) standard, a curve in use for 33- were even narrower. 78s and all modern 33-1/3-rpm LPs were 1/3 rpm broadcast transcriptions since Within their limited frequency range, cut with a constant-amplitude character- 1942. acoustical recordings exhibited a con- istic in the treble region, as well. 50 The LP Is Back! Since a magnetic cartridge is a veloci- In order to achieve a flat playback re- myriad of nonstandard recording charac- ty-sensitive device, its response rises with sponse, a phono preamplifier must apply teristics, adjustable preamplifier equaliza- frequency when playing a constant-am- equalization that is the complement of tion is a necessity. plitude characteristic, and is flat when the cartridge’s output. For optimum play- playing a constant-velocity characteristic. back of historic recordings, with their

MORE ABOUT CARTRIDGES

AMPLITUDE-SENSITIVE CARTRIDGES ciated electronics? I have always taken the latter view, and I At the conclusion of the conference presentation, a number of am grateful to Doug Pomery, an independent audio engineer interesting questions and points were raised. Joe Salerno, presi- and the proprietor of Pomeroy Audio, for sharing the letter dent of the Texas chapter of ARSC, asked if there were phono from E. C. Forman with me. Forman’s letter shows RCA Vic- cartridges which were amplitude-sensitive. Crystal phono car- tor’s official position on this issue to be in line with my own. tridges (sometimes called “ceramic” cartridges, if they are made with ceramic crystals) are amplitude-sensitive, and were CRYSTAL CARTRIDGES usually found on inexpensive portable equipment during the Dennis Rooney also pointed out that I give the impression that LP era. They were often supplied with a “flip-over” stylus as- 78-rpm recording characteristics were developed with magnet- sembly, with one side labeled “33/45” and the other “78.” Due ic cartridges in mind, noting that crystal cartridges were widely to heavy tracking forces and poor high-frequency tracking abili- used from the mid-1930s through the 1950s. It is true that Max- ty, they are equally adept at destroying both types of records. field and Harrison, and their successors, did not develop their Crystal cartridges have two advantages, however. First, they recording characteristics specifically for magnetic cartridges. have very high output (0.5-1V) and do not require a phono pre- However, in every source I have encountered, graphs illus- amplifier. Second, they produce a flat frequency response trating recording characteristics show recorded velocity, like when playing a constant-amplitude recording characteristic, the record curve in Fig. 1 and the solid lines in my other illus- and exhibit a high-frequency rolloff when playing a constant- trations. Since the record curves illustrate velocity, they also velocity characteristic. show the response of a magnetic cartridge. I have never seen a When they are playing records with high-frequency preem- graph of a recording curve that showed recorded amplitude. phasis, equalization circuits are not absolutely necessary. The re- When an electrical recording is played, the bass frequencies sults are less than optimum, though, since there is no compensa- are boosted and the treble region is sometimes attenuated be- tion for the rolloff in the constant-velocity region. On the other cause magnetic cartridges are used in playback. If an ampli- hand, the reduced high-frequency output tends to mask the tude-sensitive crystal cartridge is used to play an electrically high-frequency distortion caused by poor tracking. In any case, recorded 78, the bass must be played without equalization, i.e., crystal cartridges are not suitable for high-fidelity applications. completely flat, below the bass turnover frequency. One high-end audio manufacturer, Win Research, makes an Above the bass turnover, the cartridge’s output should actu- amplitude-sensitive cartridge with outstanding performance ally be boosted in order to compensate for the decreasing am- characteristics. Its model FET-10 retails for $3,500, and does not plitude as the frequency rises. If a recording contains high-fre- use piezoelectric crystals for operation. Instead, the stylus as- quency preemphasis, the high-frequency boost must stop at sembly is connected to the gate of a field-effect transistor (FET), the treble transition frequency. Above that frequency, the play- controlling the current flow between the source and the drain. back response must again be flat, without any equalization. The current through the transistor is an electrical representation Crystal cartridges were a cost-effective choice for 78s with of the recorded amplitude rather than the recorded velocity. treble brightness, whether due to microphone/preamp charac- The cartridge is supplied with a power supply and preampli- teristics or preemphasis in the disc cutter, since the cartridge’s fier, the latter containing equalization circuits that provide cor- output was much flatter, and considerably higher, than that of rect response in the constant-velocity regions of the RIAA a magnetic transducer. For this reason, phonograph manufac- curve. The replacement stylus, which costs $450, is not user- turers could dispense with both playback equalization and the replaceable, and is available only in one size, for stereo LP preamplifier required for magnetic cartridges. records. Therefore, it is not suitable for archival work. The Win To achieve a truly flat response, it would be necessary to Labs cartridge is actually an FET version of the strain-gauge car- equalize the constant-velocity region as described above. I have tridge, which is also amplitude-sensitive. not researched electric phonographs from the 1930s and ’40s in any detail, so I am not sure whether any manufacturers actu- RELEVANT QUESTIONS ally went to the trouble of providing proper equalization for Dennis Rooney, producer of Sony Classical’s Columbia Mas- crystal cartridges. terworks Heritage reissue series, asked when Victor and other The point, however, is that all phono preamplifiers for high- record labels began using treble preemphasis, suggesting that fidelity use, whether straight RIAA or with adjustable equaliza- it was earlier than I had originally stated. Related to this issue, tion, are designed for use with magnetic cartridges. The play- Michael Biel pointed out the switch to ribbon microphones back-equalization curves provided by these preamps—as shown by Victor in the early 1930s, and mentioned the use of micro- in Fig. 1 and the other illustrations—are incorrect for amplitude- phone preamp equalization to achieve the same tonal balance sensitive transducers. The velocity-sensitive nature of the mag- as condenser mikes. For readers interested in further informa- netic cartridge is the reason we apply bass boost and treble tion on the specific recording characteristics used by various rolloff to electrically recorded discs. record companies, I highly recommend the books by James I am grateful to Dennis Rooney, Michael Biel, and Doug R. Powell (see References and Bibliography). Pomeroy for the points they raised and the information that These questions also raised the issue of what constitutes they subsequently provided. I believe this is a stronger article the actual recording characteristic—is it the combined re- thanks to their input. I would also like to thank C. Victor Cam- sponse of the microphones and the disc-cutting equipment, pos for his help in proofreading the manuscript and for his or is it strictly the characteristic of the disc cutter and its asso- input on several technical matters.

• THEORY AND CARE • The LP Is Back! 51 GLOSSARY turnover frequencies of some electrical compensate for inner groove high-fre- Terms italicized in the definitions 78s fall outside the 300–500Hz range. quency losses caused by tracing errors below are also listed as independent Columbia LP Recording Characteris- (see tracing distortion). Glossary entries. tic: Recording characteristic used by Co- Electrical Playback: Playback involving lumbia for the modern long-playing a phono cartridge, amplifier, and loud- Acoustical Playback: A purely mechan- record introduced in 1948. Bass and tre- speaker. The phono cartridge is a trans- ical playback process whereby a play- ble turnover frequencies are set at ducer that converts the mechanical vi- back stylus connected to a diaphragm 500Hz and 1.6kHz, respectively. A low brations in the record groove into elec- traces the mechanical undulations in the bass turnover frequency of 100Hz is tricity. The electrical signal is then ampli- record groove. The diaphragm is at- also specified. The record is cut with a fied and fed to a loudspeaker. The loud- tached to a horn that provides mechani- constant-velocity characteristic below speaker is also a transducer, converting cal amplification of the sound vibrations, this frequency. The Columbia LP charac- the amplified electrical signal into sound like a megaphone. Acoustical phono- teristic is a modification of the NAB vibrations. graphs were manufactured well into the recording characteristic. The 100Hz low Electrical Recording: Recording involv- era of electrical recording. bass turnover was added by Columbia to ing the use of a microphone, an amplifi- Acoustical Recording: A purely me- reduce low-frequency noise in playback. er, and an electrical disc-cutting head. chanical recording process whereby Constant Acceleration: A recording The microphone is a transducer that sound vibrations are collected by a characteristic in which the amplitude of converts sound vibrations into electrici- recording horn, and funneled to a di- the recorded signal decreases at a rate of ty. The electrical signal is then amplified aphragm that is, in turn, connected to a 12dB/octave as the frequency rises, or and fed to the disc-cutting head. The cut- cutting stylus. The cutting stylus is 6dB/octave faster than a constant-veloci- ting head, also a transducer, converts the moved by sound power alone. The ty characteristic. This characteristic was amplified electrical signal into mechani- acoustical method of recording was used used above 4kHz on early electrical 78s cal vibrations that cut the record groove. until 1925, when it was replaced with to prevent destruction of the high-fre- Electrical disc recording was invented by electrical recording. Acoustical record- quency information by heavy acoustical Maxfield and Harrison of Western Elec- ings have very limited frequency re- reproducers. tric (Bell Laboratories) and introduced sponse, 150Hz to 4kHz at best, and are Constant Amplitude: A recording commercially in 1925. plagued with resonance problems. With- characteristic in which all frequencies Equalization: The deliberate alteration in their limited frequency range, acousti- are recorded at the same level. Recorded of the frequency response of a recording cal recordings approximate a constant- velocity increases as the frequency rises, or playback system. Disc records are not velocity recording characteristic. at the rate of 6dB/octave. recorded with a flat frequency response. AES Recording Characteristic: Charac- Constant Velocity: A recording charac- In order to achieve a flat response, equal- teristic proposed by the Audio Engineer- teristic in which the stylus speed is held ization that is the complement of the ing Society in 1950. Bass and treble constant at all recorded frequencies. recording characteristic and phono-car- turnover frequencies were set at 400Hz Recorded amplitude decreases as the tridge response must be applied in play- and 2.5kHz, respectively. It was original- frequency rises, at the rate of 6dB/oc- back. ly proposed as a compromise playback tave. Excursion: The physical distance be- curve that would yield satisfactory re- Crystal Phono Cartridge: A crystal tween the zero crossing and the peak of sults with a variety of recording charac- phono cartridge has a stylus attached to the groove modulation on a disc record. teristics. Some very late 78s and many a thin slab of piezoelectric crystal. As the Flat Frequency Response: A recording early LPs were recorded with this charac- stylus traces the record groove, the crys- or playback system is said to have a flat teristic. tal slab is twisted back and forth, gener- frequency response if all frequencies Amplitude: The level, or volume of a ating electricity. Crystal cartridges have across the entire audible spectrum can be sound vibration or recorded signal. On a very high output, and do not normally recorded and reproduced at the same disc recording, the higher the amplitude, require a preamplifier. They are also amplitude. Disc records are not cut with the wider the groove modulation or ex- amplitude-sensitive, and were often a flat frequency response, but by applying cursion. used without any playback equaliza- correct equalization in playback, a flat Amplitude-Sensitive: If a phono car- tion, though the results were usually less response can be achieved. tridge is amplitude-sensitive, its output is than optimum. They require heavy track- Frequency: The number of times per proportional to the recorded amplitude ing forces and are not suitable for high-fi- second that a sound vibration repeats it- at all frequencies. Crystal cartridges are delity applications. self. It is measured in hertz, abbreviated amplitude-sensitive. A crystal cartridge dB/Octave (decibels per octave): The Hz. One Hz equals one vibration, or will produce a flat frequency response standard method of specifying the rate of cycle, per second. The human ear, in when playing a constant-amplitude rolloff or boost relative to a specific fre- good condition, can perceive sound vibra- recording characteristic. When playing a quency, usually the turnover frequency. tions between 20Hz and 20kHz. Higher constant-velocity recording, its output For example, if the recorded amplitude frequencies are often specified in kilo- decreases as the frequency rises. decreases at the rate of 6dB/octave hertz, abbreviated kHz. 1kHz equals Bass Turnover Frequency: The fre- above 500Hz, the level of a 1kHz signal 1000Hz, 20kHz equals 20,000Hz, and so quency where the transition from a con- will be 6dB lower than a 500Hz signal. A on. Frequency is directly related to pitch. stant-amplitude to a constant-velocity 2kHz signal will be 6dB lower than the The lower the frequency, the lower the characteristic occurs, usually between 1kHz signal, and so on. pitch; conversely, the higher the frequen- 300–500Hz. The recording characteristic Diameter Equalization: A high-fre- cy, the higher the pitch. The note A is constant amplitude below this frequen- quency boost applied during the disc above middle C on the piano, used as a cy, and constant velocity above. The bass recording process. The purpose is to tuning reference by musicians, is normal- 52 The LP Is Back! ly 440Hz, or 440 vibrations per second. velocities and/or groove excursions that for amplitude-sensitive transducers, such The piano has a range of 27.5Hz (the low- exceed the physical and/or mechanical as crystal phono cartridges. Preamplifiers est A) to 4186Hz (the top C). Harmonics capabilities of the stylus assembly. are also used in the electrical recording produced by the piano and other musical Modern Long-Playing (LP) Record: chain to amplify the relatively weak signal instruments extend well beyond the The 33-1/3-rpm record invented by Gold- from the microphone. A mike preamplifi- range of human hearing. mark, Bachman, and Snepvangers, and in- er will not have disc playback equaliza- Groove Modulation: The side-to-side troduced by Columbia in 1948. The Co- tion, but it can be designed to alter the movement of the record groove, as lumbia LP used microgroove cutting, re- frequency response of the microphone. formed by the cutting stylus, on either quiring a playback stylus with a much Preemphasis: The high-frequency veloc- side of the zero crossing. An unmodulat- smaller tip radius than those used for 78- ity boost above the treble turnover fre- ed groove remains at the zero crossing at rpm records. Victor introduced a 33-1/3 quency in later 78-rpm records, and all all times. Modulating the groove with rpm record in 1931, but the groove width modern long-playing records. Records sound vibrations causes the recording sty- of the Victor LP was similar to that of the cut with preemphasis are constant ampli- lus to move back and forth on either side 78-rpm record. The early Victor LP was tude above the treble turnover frequency. of the zero crossing, cutting a physical not a success, technically or commercial- The term “preemphasis” is misleading picture of the original musical waveform. ly, and should not be confused with the since it implies a boost in recorded am- A vertically cut recording causes the sty- modern LP record. (See Columbia LP plitude. In reality, only the recorded ve- lus to move up and down, rather than recording characteristic.) locity is boosted. The amplitude above side-to-side. A 45/45 stereo recording pro- NAB Recording Characteristic: Estab- the treble transition frequency appears duces diagonal stylus motion. lished by the National Association of boosted only when the record is played Groove Pitch or Pitch of Grooves: The Broadcasters for 33¹⁄₃ rpm transcription with a magnetic phono cartridge. physical spacing of the record grooves discs in 1942. Bass and treble turnover Radius of Curvature: The radius of the across the surface of the disc. Groove frequencies were set at 500Hz and curve at the peak of the groove modula- pitch is usually specified in “lines per 1.6kHz, respectively. tion, where the stylus reverses direction. inch” rather than “grooves per inch,” New Orthophonic Recording Charac- The higher the frequency, the shorter the since a record has only one groove which teristic: Proposed by RCA Victor in 1953 radius. If the radius of the playback stylus spirals continuously from the beginning and subsequently adopted by RIAA. (See is larger than the radius of this curve, to the end. 78-rpm records have a coarse RIAA recording characteristic.) high-frequency tracing distortion and groove pitch, typically about 75 lines per Orthophonic Victrola: An acoustical record groove damage may result. inch. Modern long-playing records aver- phonograph designed by Western Elec- Recording Characteristic: The method age about 225 lines per inch, but can be tric for playback of the first electrically used for cutting a phonograph record, de- as high as 300. recorded discs. The exclusive rights to fined by the relationship of recorded am- Hertz (Hz): See frequency the design were sold to the Victor Talking plitude, recorded velocity, and frequency. Low Bass Turnover Frequency: The Machine Company, which marketed it as The two most common recording charac- low-frequency turnover used on modern the Orthophonic Victrola. teristics are constant amplitude and con- long-playing records, set at 50.05Hz by Phono Cartridge or Pickup: The trans- stant velocity. Practical disc records are RIAA. The recording characteristic is ducer that converts the mechanical vibra- made using a hybrid of these two charac- constant velocity below this frequency. tions in the record groove into an electri- teristics. This minimizes turntable rumble and hum cal signal. The term “pickup” has fallen RIAA Recording Characteristic: En- in playback, since there is no bass boost out of favor in the United States, but it is dorsed by the Recording Industry Associa- in playback below this frequency. On the still used by the British. tion of America in 1956. It was originally first Columbia LP records, the low bass Playback Equalization: The electronic proposed by RCA Victor as the New Or- turnover frequency was set at 100Hz. compensation provided by the preampli- thophonic recording characteristic in Magnetic Phono Cartridge: In a mag- fier that is the reverse of the recording 1953. Bass and treble turnover frequen- netic phono cartridge, the playback sty- characteristic and the phono cartridge’s cies are set at 500.5Hz and 2.122kHz, re- lus is attached to a magnet placed in close output. Correct playback equalization spectively. A low bass turnover frequen- proximity to a coil of wire. The move- will result in a flat frequency response. cy of 50.05Hz is also specified. The ment of the magnet generates electricity Preamplifier: The control center of an record is cut with a constant-velocity in the coil. Moving-magnet cartridges are audio playback system. A preamplifier characteristic below this frequency, the most popular, since they can be man- with a phono input contains circuitry to which reduces low-frequency noise in ufactured relatively inexpensively and boost the very low output of a magnetic playback. Since 1956, all LP records have have user-replaceable stylii (essential for phono cartridge, as well as to apply cor- been made with the RIAA Characteristic. playback of historic recordings). Varia- rect playback equalization. Modern pre- Tracing Distortion: Distortion pro- tions on the same principle include mov- amplifiers are normally designed to pro- duced when the playback stylus cannot ing-coil and moving-iron cartridges. Mag- vide only RIAA playback equalization. follow the exact path cut by the record- netic cartridges are velocity-sensitive Preamplifiers suitable for the playback of ing stylus. In extreme cases, record transducers. They normally have very 78s and early LPs must have variable groove damage can result from this mis- low output, which must be boosted by equalization to accommodate the variety match. (See radius of curvature.) the preamplifier. of recording characteristics encountered Transducer: A device that converts one Mike: Short for microphone (see trans- on old records. The equalization circuits form of energy into another. In disc ducer). in phono preamplifiers are designed for recording and playback, a transducer Mistracking: A situation that occurs use with magnetic phono cartridges, such as a microphone or phono cartridge when the stylus loses contact with the which are velocity-sensitive. These play- may convert mechanical energy into elec- record groove. Causes include recorded back equalization curves are not suitable tricity, or vice versa in the case of a disc- • THEORY AND CARE • The LP Is Back! 53 cutter head or a loudspeaker. These are REFERENCES BIBLIOGRAPHY called electromechanical transducers. 1. Eargle, John, “Disc Recording and Reproduction,” 1. Bauer, B. B., “The High-Fidelity Phonograph Trans- There are also electromagnetic transduc- Sound Recording, 2nd edition, New York: Van Nos- ducer,” Journal of the Audio Engineering Society, Vol. ers, such as the magnetic tape head, that trand Reinhold Company, 1980, pp. 276–317. 25, No. 10, 11, Oct./Nov. 1977, pp. 729–748. convert magnetic information into elec- 2. Maxfield, Joseph P. and Harrison, Henry C., “Meth- 2. Bogantz, Gregory, and Ruda, Joseph C., “Analog tricity, or vice versa. ods of High Quality Recording and Reproduction of Disk Recording and Reproduction,” Audio Engineering Treble Turnover or Treble Transition Music and Speech Based on Telephone Research,” Handbook, ed. K. Blair Benson, New York: McGraw- The Bell System Technical Journal, Vol. V, June 1926, Hill Book Company, 1988, pp. 8.1–8.39. Frequency: The frequency at which the pp. 493–523. Abridged version published in AIEE 3. Davies, Sean, “Disc Cutting,” Sound Recording transition from a constant-velocity to a Transactions, February 1926. The AIEE version is also Practice, 3rd edition, ed. John Borwick, New York: Ox- constant-amplitude characteristic takes the one reprinted as an appendix in From Tin Foil to ford University Press, 1987, pp. 392–411. place, usually between 2k and 3kHz. The Stereo, 2nd edition, by Oliver Read and Walter L. Welch, Indianapolis: Howard W. Sams and Co., 1976. 4. Denman, Roderick, and George, Peter, “Gramo- recording characteristic is constant-veloc- phone, Encyclopædia Brittanica, 14th edition, Vol. 10, ity between the bass-turnover and treble- 3. Moyer, R. C., “Evolution of a Recording Curve,” London: The Encyclopædia Brittanica Co., Ltd., 1929, Audio Engineering, July 1953, pp. 19–22, 53–54. turnover frequencies, and constant-ampli- pp. 616–622. 4. Sutheim, Peter, “An Afternoon With John G. tude above the treble turnover. The 5. Esoteric Sound, Instructions for Operation of Eso- Frayne,” Journal of the Audio Engineering Society, teric Sound Re-Equalizer, c. 1985. switch to a constant-amplitude character- Vol. 38, No. 7, 8, July-August 1990, pp. 597–599. istic in the high frequencies is sometimes 6. Goldmark, Peter C., Snepvangers, René, and Bach- 5. Powell, James R., “The Audiophile’s Technical man, William S., “The Columbia Long-Playing Record- called treble preemphasis. Guide to 78-rpm, Transcription, and Microgroove ing System,” Proceedings of the I.R.E., August 1949, Velocity: The speed at which the stylus Recordings,” Portage, MI: Gramophone Adventures, pp. 923–927. 1992. moves while tracing the record groove. 7. Lipshitz, Stanley P., “On RIAA Equalization Net- It is directly related to the physical dis- 6. McProud, C. G., “AES Standard Playback Curve,” works,” Journal of the Audio Engineering Society, Vol. tance the stylus must travel in a given Audio Engineering, January 1951. Reprinted in Audio 27, No. 6, June 1979, pp. 458–481. Anthology, Vol. 2, Peterborough, NH: Audio Amateur time period. Publications, 1989, pp. 65–66. 8. Maxfield, Joseph P., “Phonograph,” Encyclopædia Velocity-Sensitive: If a phono cartridge Brittanica, 14th edition, Vol. 17, London: The Ency- 7. Powell, James R., “Playback Equalizer Settings for clopædia Brittanica Co., Ltd., 1929, pp. 776–777. is velocity-sensitive, its output is propor- 78-rpm Recordings,” Portage, MI: Gramophone Ad- tional to the recorded velocity at all fre- ventures, 1993. 9. McIntosh Model C-8 and C-8P Instruction Manual and Record Compensation Chart, McIntosh Laborato- quencies. Magnetic phono cartridges are 8. Lane, Michael R., “Sonic Restoration of Historical ry, c. 1956. velocity-sensitive. A magnetic cartridge Recordings,” Parts 1 and 2, Audio, Vol. 75, No. 6, 10. McProud, C. G., “Recording Characteristics,” will produce a flat frequency response June 1991, pp. 34–44, and Vol. 75, No. 7, July 1991, pp. 26–37. Audio Engineering, Jan. 1950. Reprinted in Audio An- playing a constant-velocity recording thology, Vol. 2, Peterborough, NH: Audio Amateur characteristic. When playing a constant- Publications, 1989, pp. 67–68. amplitude recording, its output increases 11. Owen, Tom, and Fesler, John C., “Electrical Re- as the frequency rises. A CD-R of the eight recorded exam- production of Acoustically Recorded Discs and Cylin- Voice Effort Equalization: A high-fre- ples is available for $10 in the US, $12 ders,” presented at the 70th Convention of the Audio Engineering Society, AES preprint No. 1854, 1981. quency boost added in the microphone in Canada, $15 for all other countries. 12. Installation and Operating Instructions for Owl 1, preamplifiers used with ribbon mikes. Prices include postage (air mail outside the USA). Total time on the CD-R is Owl Audio Products, c. 1983. This was done to give the ribbon mikes 6:45. Send payment in US funds to: 13. Powell, James R., “The Audiophile’s Guide to the same level of brightness as the con- Gary Galo, 72 Waverly St., Potsdam, Phonorecord Playback Equalizer Settings,” ARSC denser mikes they had replaced. NY 13676 USA. Journal, Vol. 20, No. 1, Spring 1989, pp. 14–23. Zero Crossing: The position of the 14. “RIAA Standard Recording and Reproducing Char- recording or playback stylus at rest. acteristic,” Bulletin No. E-1, New York: Recording In- When a record is cut, the recording stylus dustry Association of America, Nov. 6, 1976. moves back and forth on either side of 15. Tremaine, Howard M., “Cutting Heads,” The Audio the zero crossing, cutting a physical pic- Cyclopedia, 2nd edition, Indianapolis: Howard W. Sams & Co, 1969, pp. 693–710. ture of the original musical waveform. The playback stylus traces the path cut by 16. Villchur, Edgar, “Disc Recording,” The Reproduc- tion of Sound in High-Fidelity and Stereo the recording stylus. The zero crossing is Phonographs, 2nd edition, New York: Dover Publica- at the center of the groove modulation. tions, 1965, pp. 26–33. 17. Williamson, Reg, “About Equalization,” Audio Ama- teur, Vol. IV, No. 3, July 1974, pp. 18–19. 18. Williamson, Reg, “Understanding the RIAA Curve,” Audio Amateur, Vol. 21, No. 2, April 1990, pp. 18–24.

54 The LP Is Back!