SC-04-03 Richard Cabot posted an update in the group SC-04-03 3 weeks ago No folders found. Please create and select folder. Documents Folder Title Following special characters are not supported: \ / ? % * : | " < > Privacy Public All Members My Connections Only Me Cancel Create text-list.txt 321 B Text File - Click to view Options Copy Download Link DUA4:[STANDARDS.AESSC.SC_04.SC_04_03]AES19-EIA.TXT;1 16-JAN-1997 12:27:14.46 DUA4:[STANDARDS.AESSC.SC_04.SC_04_03]MEMBERS.LST;6 12-JAN-1997 14:41:30.67 DUA4:[STANDARDS.AESSC.SC_04.SC_04_03]TEXT-LIST.DAT;2 4-FEB-1997 13:10:06.27 Expand Suspension_IEC_draft_2008.pdf 368 KB PDF - Click to view Options Copy Download Link trns1.txt 3 KB Text File - Click to view Options Copy Download Link File: 146.lima.transdrft1 Date: 97:08:20 Issue: 1 J. M. Woodgate 3. Measurements: 3.1 Conditions Measurements, except where stated, are made with the transformer loaded with a (low inductance) resistive load of the rated value, and in the highest power transfer condition if there are 'power' tappings. NOTE - The reason for measuring the low limit of the EFR on-load is that it allows for the reduction of magnetic induction due to the voltage drop across the primary resistance. 3.2 Insertion loss A pply rated line voltage Vr at 1 kHz to the primary winding, with no load on the secondary winding. Measure the secondary voltage Va. Then connect the load resistor R to the secondary winding, readjust the input to rated line voltage if necessary and measure the secondary voltage Vb again. Express the insertion loss IL in decibels, 20 lg(Va/Vb). NOTES 1. With no load, the transformer losses are normally very low at 1 kHz: the on-load copper loss is much greater than the iron loss. 2. The power delivere d to the load resistor is Vb^2/R. From Note 1, the power input to the transformer is (very nearly) Va^2/R, and this latter value can be compared with the rated value. 3.3 Effective frequency range a) Low frequency end: Apply rated line voltage at 1 kHz and measure the input current (as voltage across a known low-value resistor in series if no suitable ammeter is available). Reduce the frequency until the current has risen by 11% (1 dB). Note the frequency as the low-frequency limit of the EFR. NOTE - Th is criterion for the low-frequency limit corresponds to typically 1% third-harmonic distortion due to core saturation. b) High-frequency end: Apply rated line voltage at 1 kHz and measure the secondary voltage. Increase the frequency until the secondary voltage falls by 3 dB. Note the frequency as the high-frequency limit of the EFR. 3.4 Input impedance at surveillance frequencies [QUESTION: What voltages are used for surveillance on 25 V line systems? Is 1.25 V enough?] 3.4.1 Input impedance at 30 Hz Apply one twentieth of rated line voltage at 30 Hz to the primary winding with the load resistance connected to the secondary. Measure the input current Ip and calculate the (modulus of the) input impedance Z30 = Vr/20Ip. 3.4.2 Input impedance at 20 kHz Apply one twentieth of rated line voltage at 20 kHz to the primary winding, with the secondary connected to the rated load resistance R in series with an inductor of value R/8 mH. Measure the input current Ip and calculate the (modulus of the) input imped ance Z20k = Vr/20Ip. NOTE - The inductance R/8 mH is an average value for the loudspeakers generally used. If the transformer is designed for use with a specified loudspeaker(s) , then the actual value of voice-coil inductance may be used, provided this is noted with the results. Expand x72-prince-030320.doc 25 KB Word Document - Click to view Options Copy Download Link x72hs-accentre-990324.pdf 114 KB PDF - Click to view Options Copy Download Link X86-fileformat_public_1.pdf 66 KB PDF - Click to view Options Copy Download Link X86-AES6631-20051015.pdf 746 KB PDF - Click to view Options Copy Download Link x103-klippel-000209.txt 7 KB Text File - Click to view Options Copy Download Link Initial Problem Discussion to "Large Signal Parameters of Low-Frequency Drivers" by Wolfgang Klippel, February 2000 The thermal power capacity P_E(max) and the linear displacement x_max (or the diaphragm peak displacement volume V_D=S_D*x_max) are very important large signal parameters which the system designer must know or specify to describe the maximal acoustic output. At moderately high frequencies, where the voice coil excursion is small, the power handling capability is limited by the ability of the driver to dissipate heat. At low frequencies much more diaphragm displacement is required and the maximal acoustic output is limited by the peak displacement in most cases. The measurement and specification of the displacement x_max is defined in point 4.3.2 (2) of AES2-1984 (r1997) as follows: "The Large Signal Parameters are ... (2) Voice-coil peak displacement at which the linearity of the motor deviates by 10%, x_max. Linearity may be measured by percent distortion of the input current or by percent deviation of displacement versus input current. Manufacturer shall state method used. The measurement shall be made in free air at f_s. ..." This recommendation goes in the right direction but there are still some weak points in it: What are the dominant nonlinearities of the driver ? The displacement limit x_max is closely related with the linearity of the driver. Clearly, the voice coil / gap geometry, the elastic limit of suspension and striking a mechanical stop limit the maximal displacement. The dominant nonlinearities can be modeled by variable parameters such as force factor b(x), inductance L(x) and compliance C_MS(x) which are not assumed as constant but depend on the instantaneous displacement x. These parameter variations are not negligible but might exceed factor 5 and more without causing a permanent destruction of the driver. The nonlinear parameters which are functions of x can be represented as a curve or may be approximated by a power series expansion. To specify x_max, which corresponds with the allowed working range of the driver, it might be sufficient to ensure that the maximal parameter variations do not exceed a critical value and the nonlinear effects are still acceptable for the particular application. What are the problems in assessing driver nonlinearity by distortion measurements ? The displacement varying parameters generate additional signal components in the reproduced sound as well as in the state signals of the driver (current, displacement, velocity...) which can be measured as harmonic and intermodulation distortion for a multi-tone input. The nonlinearities are the physical cause and the distortion are the measured effects for a special excitation signal. The relationship between distortion and nonlinearities is predictable but complicated. From loudspeaker modeling we know that each nonlinearity generates characteristic distortion components having particular spectral properties. For example, the distortion components generated by stiffness are restricted to the lower frequency band and are perceived as a changed bass sound. Nonlinearities of the force factor and inductance cause broad-band components which can be perceived as disturbances affecting the overall sound quality. By performing a harmonic distortion measurement it is hardly possible to see a difference between stiffness and force factor distortion. At the resonance frequency the harmonics caused by the force factor have sometimes a local minimum due to the cancellation of nonlinear damping and nonlinear excitation. The measurement of the intermodulation distortion generated between a low-frequency and high-frequency tone shows a clear difference between suspension and motor distortion which are more valid to predict the subjective sensation. Considering two loudspeakers with comparable harmonic distortion, the speaker which has a linear motor coupled with a bad suspension will usually sound better than the less expensive speaker with nonlinear motor but having a linear suspension. Thus, the assessment of the driver linearity by measuring harmonic distortion only is quit questionable. What is required to improve the current practice ? Considering the results of nonlinear loudspeaker modeling the maximal displacement x_max can be specified much more precisely to the benefit of driver and system designers. Assess dominant driver nonlinearities The nonlinear parameters force factor and suspension can be calculated by using numerical methods (FEM) and measured on the final product by performing static, quasi-static or dynamic measurements. This information are quit valuable to the driver designer to optimize the system in respect with quality, price, weight and size. Having this information the driver designer can easily fix asymmetric parameters variations (without increasing the cost) to reduce second- and higher-order distortion. The system designer takes also benefit from this information to design an active or passive loudspeaker system with optimal properties. For him the maximal variations of the nonlinear parameters are most important three single value parameters. Define allowed limits of parameter variations Using the nonlinear parameters as criteria for specifying x_max we need the thresholds of allowed parameter variations. David Clark suggested for force factor variations an allowed decay down to the limit b_lim=70.7% of the rest value and a compliance limit C_lim = 25%. This proposal is quit practical but this thresholds should be a subject of broad discussions of course. The results of subjective investigations about the audibility of the different kinds of loudspeaker distortions would be helpful in this process. Listening tests with synthesized loudspeaker distortions have been performed and are an interesting field for further research. Derive maximal displacement x_max from parameter variations The large parameter x_max can be derived from the nonlinear parameter characteristics b(x) and C(x) compared with the allowed limit values the b_lim and C_lim. The nonlinear parameter which equals the allowed limit first will determine x_max and the range of operation x_max < x < x_max. Thus we will find drivers which are limit by motor or by suspension nonlinearities. One or the other type might be optimal for a particular application such as car, hifi, multimedia, active system (with nonlinear compensation ?). How can we proceed in this matter ? The problems addressed in the previous two points are only examples of the issues related with the Large Signal Loudspeaker Parameters. There is much more to say and of course there are different points of view. I think it is time to start with a broad discussion to collect the input from many engineers having a theoretical or practical background. They suggestions presented above might be a starting point. Our target should be to improve the standard or practical recommendations eventually. Rationale: Current activity: .s Liaisons: .!</blockquote><hr> Expand x129-cabot-backgroun.pdf 1 MB PDF - Click to view Options Copy Download Link 0 Comments Public All Members My Connections Only Me PublicAll MembersMy ConnectionsOnly Me Public All Members My Connections Only Me
