«PREVIOUSLY ISSUED NUMBERS OF BRUEL & KJ/ER TECHNICAL REVIEW 1-1956 Noise Measurements and Analyses. 2-1956 Use of Resistance Strain Gauges to ...»
PREVIOUSLY ISSUED NUMBERS OF
BRUEL & KJ/ER TECHNICAL REVIEW
1-1956 Noise Measurements and Analyses.
2-1956 Use of Resistance Strain Gauges to determine Friction
3-1956 Determination of Acoustical Quality of Rooms from
4-1956 Electrical Measurements of Mechanical Vibrations.
1-1957 Strain Gauge Measurements.
2-1957 Sound Analysis in Industrial Processes and Production.
3-1957 Measurement on Tape Recorders.
4-1957 Measurements of Modules of Elasticity and Loss Factor for Solid Materials.
Surface Roughness Measurements.
1-1958 Measurement of the Complex Modulus of Elasticity.
2-1958 Vibration Testing of Components.
Automatic Level Regulation of Vibration Exciters.
3-1958 Design Features in Microphone A m p l i f i e r Type 2603 and A. F. Spectrometer Type 2110.
A true RMS Instrument.
4-1958 Microphonics in Vacuum Tubes.
1-1959 A New Condenser Microphone.
Free Field Response of Condenser Microphones.
2-1959 Free Field Response of Condenser Microphones (Part II) 3-1959 Frequency-Amplitude Analyses of Dynamic Strain and its Use in Modern Measuring Technique.
4-1959 Automatic Recording of A m p l i t u d e Density Curves.
1-1960 Pressure Equalization of Condenser Microphones and Performance at Varying Altitudes.
2-1960 Aerodynamically Induced Noise of Microphones and Windscreens.
3-1960 Vibration Exciter Characteristics.
4-1960 R.M.S. Recording of Narrow Band Noise w i t h the Level Recorder Type 2305.
1-1961 Effective Averaging Time of the Level Recorder Type 2305.
2-1961 The Application and Generation of Audio Frequency Random Noise.
3-1961 On the Standardisation of Surface Roughness 4-1961 Artificial Ears.
1-1962 Artificial Ears part 2.
2-1962 Loudness Evaluation.
TECHNICAL REVIEWNo. 3 - 1962 Testing of Stereophonic Pick-ups by means of Gliding Frequency Records by Denis Dion, Ing. E.S.E.*) and Ole Hejlsberg, El. Eng.
ABSTRACTAfter a survey of the principles of stereophonic disk reproduction, there f o l l o w s a short analysis of the problem of production testing of stereophonic pick-ups, f o r m i n g the technical background of the B & K G l i d i n g Frequency Records. These new test records are critically described and the different associated measuring methods are discussed mainly w i t h regard t o the automaticity and rapidity of operation. Finally, some experimental results are s h o w n.
SOMMAIREApres un rappel des principes de la reproduction stereophonique sur disques, le probleme des essais en serie des tetes lectrices («pick-ups») est etudie de maniere a mettre en evidence les specifications essentielles relatives aux enregistrements de reference utilises. Les caracteristiques techniques des nouveaux enregistrements etalons produits par Bruel et Kjasr sont ensuite discutees et les methodes de mesure associees sont decrites principalement sous I'angle de leur rapidite et automatisme d ' e m p l o i. Des exemples de resultats experimentaux obtenus avec I'equipement standard B & K sont presentes en f i n d'article.
ZUSAMMENFASSUNGNach einer ubersicht uber die Wiedergabeverfahren fur stereophonische Schallplatten werden Prufprobleme von Stereophonie-Aufnehmern anhand der B & K-Frequenzgang-MeBschallplatten behandelt. Die neuen MeBschallplatten und die damit verbundenen MeBmethoden werden unter Berucksichtigung der rationellen: Prufarbeit eingehend beschrieben. Zum SchluB werden einige experimentelle Ergebnisse angegeben.
The fundamental superiority of stereophonic sound reproduction over the ordinary monophonic process is universally recognized and stereophony has become a common feature of mass-produced sound reproducing equipment.
Basically, the word "stereophonic" means that the sound is considered, in one particular location, not only in intensity but also in direction, i.e. a vector quantity is involved. This vector can be resolved in two components of fixed directions. Stereophonic sound reproduction consists in the recording, reproducing and transmitting of these components with their original amplitude and relative phase.
Conservation of the correct phase relation between the two components is of particular importance when considering the recording-reproducing process.
*) This article follows the Contributed Paper presented by the author at the 4th International Congress on Acoustics.
In the case of tape recorders, this is obtained without difficulty by recording the components on different tracks of the same tape. In the case of disk recording, however, the use of two grooves would rise a problem of correct tracking, and would at least halve the playing time, which is an unacceptable drawback for commercial use. For this reason, a rather delicate process is employed for stereophonic disk recording: this is called the 45°/45° process.
Here the two components of information are recorded in the same groove in two directions at 90° angle with each other and at 45° to the disk surface (Fig. 1). Physically, the two components of the original signal are picked up by two or more microphones, amplified separately, and then recombined in the recorder to give the cutting stylus, controlled by an electrodynamic device, a resulting velocity representing the amplitude and direction of the sound. The sound is reproduced stereophonically by tracing the groove on a u copy of the recording with a suitable reproducing head, or pick-up". The stereophonic pick-up resolves the movement of its stylus tip into two components at ± 45° to the disk surface and then transduces these components into electrical signals to be amplified through the two channels of the stereophonic reproducing chain.
Other methods of stereophonic disk recording have been developed, but the 45°/45° process described above is one of the simplest, and offers the great advantage of symmetrical conditions for both channels. It is now accepted and utilized the world over.
The 45°/45° process was proposed by Blumlein of Electrical and Musical Industries Ltd. already in 1932, together with a vertical-horizontal system which was the only system compatible with the vertical record players used at that time. Stimulated by the constantly increasing interest of the public for disk records, a considerable amount of research has since been devoted to the design of stereophonic disk recording and reproducing equipment.
Recent results achieve a high degree of fidelity in recording, but good performance is more difficult to attain in the reproduction. One reason for this is that since at the present time commercial recordings are available on plastic materials with limited mechanical compliance, such as "Vinylite", it is essential for true reproduction that the loading of the record by the reproducing stylus by extremely small. This means a correspondingly low mechanical input level for the transducing head and the necessity for high precision manufacture.
Reproducing pick-ups capable of following the best quality recordings available today without appreciable distortion have been designed and realised in laboratories. However, when considering pick-ups which are series produced at competitive price, important deviations from the prototype specification may often be expected.
The complete check of a stereophonic pick-up requires, as explained below, controlled variation of several parameters. The operation tends to be long and tedious if it is not made automatic, and consequently it is often neglected in serial production for economical reason. This renders any comparison between the nominal characteristics of different pick-ups rather doubtful and if high performances are claimed, individual precision test of stereophonic pick-ups is a necessity.
A demand has thus arisen for an automatic system to test disk reproducing equipment and especially pick-ups which usually are the weak link of "HiFi" chains, using the Briiel & Kjser audio frequency automatic response recorder.
This equipment, whose accuracy and reliability in operation has been continuously improved, has been in production for about 15 years and has long been recognised as an excellent basis for any such test-system. This adaptation is now available, together with suitable calibrated gliding frequency recordings.
The technical back-ground of these new recordings and some examples of applications are presented in the following pages.
Principles of Stereophonic Pick-up Test.
The stereophonic pick-up, considered as a mechanical-electrical transducer, presents a two-dimensional mechanical input and two electrical outputs. The ideal transfer function consists in the resolution of the two-dimensional input into two components along two axis making ± 4 5 ° to the disk surface, and the transducing of these mechanical quantities into proportional electric voltages. In practice this function will be dependent upon various parameters, the most important of which are here the amplitude and the direction of the mechanical input, and the frequency.
The amplitude of the mechanical input is characterized by the modulus of the two components of the instantaneous velocity (Fig. 2). For a sinusoidal excitation of definite direction, as used for the tests, the r.m.s. values-of the velocity components can be employed. By calling these A and B respectively,
the definition can be written symbolically:
V = V2 (A + jB) sin cot
The inner and outer side wall of the groove are called "Left" and "Right" sides respectively and their modulations correspond to the left and right hand loudspeakers or recording microphones, as viewed from the audience.
A very important characteristic of a stereophonic pick-up is its frequency response, described for each channel by different curves for different directions of the mechanical input. Practically, it is sufficient to consider the curves relative to the four principal directions, i.e.: 45° left, 45° right, lateral (parallel to the disk surface), and vertical (perpendicular to the disk surface). The frequency range in which the response has to be determined should of course include the whole audible range and attain as high frequencies as possible for the determination of the stylus resonance. A range covering three decades (10 octaves), typically 20 c/s to 20 kc/s, is consequently desirable. Measurements under 20 c/s are also sometimes necessary for tone arm resonance investigation.
In addition to the frequency response of each channel, another important factor is the separation between channels. When exciting the stereophonic pick-up with a movement parallel to one of the axis of projection defined above, it is an essential condition that the component on the other axis should be equal to zero, or in practice that the cross-talk in the unmodulated channel be negligible with respect to the modulation in the other channel.
This property which is called the separation between channels, is in fact the basis of stereophony and it is essential to control it accurately throughout the audible range and if possible at different modulation levels. It happens quite often that it is satisfactory only in a small part of the audible range for mass-produced pick-ups, and sometimes it has only been controlled at
one fixed frequency, e.g. looo c/s. Separation is usually expressed in decibels:
by calling b, or a, the r.m.s. cross-talk signal in the unmodulated channel when the other channel is carrying a signal of r.m.s. value A, or B, respectively. Separation is generally considered as satisfactory when over 20—30 dB (i.e. the cross-talk signal is less than a few percents of the main signal).
The behaviour of the stereophonic pick-up as a function of amplitude, i.e.
the dynamic properties, is investigated by means of calibrated recordings of fixed frequency and varying amplitude. However, these properties are generally not critical, regarding only the pick-up, since the harmonic distortion in the pick-up is normally much lower than the inherent tracing distortions..
Finally, a parameter inherent in disk reproduction should be mentioned: the groove speed, which is continuously variable from the periphery to the center of the disk, in the proportion of 2.5 to 1 for 30 cm (12") records.
However, the influence of the groove speed is mainly a matter of relative magnitudes of the mechanical curvature of the recorded signal and of the stylus tip radius, the latter being fixed within rather narrow limits by the standards and usually pre-checked accurately by optical methods. There exists nevertheless a relationship between the influence of the groove speed on the response and the mechanical impedance of the input. A check at various groove speeds is therefore useful for controlling this impedance, if correct interpretation is available. See Appendix A.
All of the previously mentioned characteristics should be known for assessing the quality of a stereophonic pick-up. F u r t h e r m o r e, both channels should behave identically. Appreciable deviations in frequency response for example would cause some phase distortion when the stereophonic sound is reconstituted at audition.
Reference Test Records.
The different characteristics outlined above are conveniently checked by using calibrated disk recordings of sinusoidal tones covering the frequency range of interest and with different directions of modulation. These recordings are obtained on specially selected recording equipment, with the maximum care, from the output of a precision audio generator. See Appendix B.
If the pick-ups to be tested are intended to be used for reproducing commercial records, the test records should be manufactured out of a similar material. The elastic properties of the "Vinylite", which is so far very generally employed, play a certain role in disk reproduction. This is easily understood when considering that the acceleration which is communicated by the record to the stylus tip is of the order of a few hundred g for the higher tones, and quite often presents peaks of more than looo g.