Обзор 6" мидвуфера Seas "Excel" W18NX003 (E0096-08)

Раз уж в индустрии сложилась такая вольность определения параметров Тиля-Смолла, то интересно посмотреть как на это смотрят другие ведущие производители. Информация взята с официальных сайтов компаний:

Wavecor measures using a semi-constant current source, nominal level 2 mA. After burn-in specifications are measured 12 hours after exiting the transducer by a 20 Hz sine wave for 2 hours at level 10/14.1 VRMS (4/8 ohm version).

SB Acoustics - "A lot can be said about how to measure loudspeaker Thiele/Small parameters. There are different methods of measurement and they are not all equally good (this will not be discussed here). Due to drive unit non-linearities, there is not one final set of parameters for any drive unit. Several of the parameters are a function of cone/voice coil excursion, hence they depend on the input voltage/current level. At SB Acoustics we use the delta mass method (as opposed to delta compliance). We use a constant voltage source and a 150 Ω resistor in series with the drive unit. Hence, this is neither a constant voltage nor a constant current measurement (the value of the resistor is not nearly large enough to approximate constant current, which is not the intention either). Prior to measuring T/S parameters, the drive unit should be broken in – for two reasons. If the suspension of the drive unit is not broken in, its compliance will slightly increase during the measurement, thus affecting/distorting the results. Furthermore, eventually the drive unit will end up being broken in (it actually does not take that long), which is why it is recommended to use T/S parameters that apply for a broken-in drive unit when calculating box volumes and tuning frequencies. The free air resonance frequency typically drops about 10-15% during break-in, as the suspension compliance increases. This directly affects the Q-factors and the equivalent volume. To break in a drive unit, you are going to need a sine wave or a noise generator with adjustable output voltage (the former is recommendable) and a power amplifier. Using a sine wave generator, adjust the frequency somewhere below the expected free air resonance frequency of the drive unit (typically about 80% of this value). Slowly turn up the voltage until the suspension reaches maximum displacement. Keep it below clipping level. Usually you can hear when the suspension goes into clipping mode - there will be some kind of mechanical noise. Adjust the voltage slightly below this level. During the process of breaking in the drive unit, it might be necessary to turn down the voltage a little to keep it from clipping, as the suspension softens. It needs to run for about ten minutes. Before measuring, let the drive unit cool to ambient temperature. The voltage that should be applied to the terminals of the drive unit depends on its size/type. For a typical mid-woofer, the voltage should be about 1 V (rms) at the resonance frequency. 

Scan-Speak - "Measurement of Thiele/Small parameters can lead to quite a bit of discussion within the loudspeaker community. The following is a description of the method used at Scan-Speak. The initial description of measurement and calculation of these Small Signal Parameters comes from Richard Small’s papers in the Journal of Audio Engineering in 1972, “Direct Radiator Loudspeaker Analysis”. Both Neville Thiele’s and Richard Small’s papers on this subject dealt with alignment of bass response for loudspeaker systems. It is really important to understand that we are working with “small signal” parameters, which means that the excitation of the moving parts is so small that the mechanical properties can be considered to operate in their linear range. It is also important to realize that any small change of measurement method will result in different results. In fact, there are different ways of measuring these parameters in the market, and they are being used by different systems manufacturers to align or tune loudspeaker systems. So there is no one “correct” set of parameters, but several sets that could potentially be used and the key is to stick with one method and evaluate whether the resulting system response matches your simulation results.

We use the classical simplified model and the measurement by Scan-Speak’s R&D follows Richard Small’s derivation of the parameters. We use very low excitation and obtain constant current by using a compressor to maintain a constant voltage over a series resistor. We use pure sine tones from an analog generator and search the resonance frequencies, the side frequencies and the voltages manually. We use the “added mass” method, and choose an added mass big enough to get a large enough difference in resonance frequency. 

Before starting measurement of the parameters, the drive unit must be burned-in by exposing it to a (preferably) sine tone at a frequency slightly below or at the initial resonance frequency and then applying a voltage large enough for the driver to move almost to its maximum excursion. If the driver starts clipping, which is clearly audible, turn back the voltage. The T/S parameters should always be measured on a burned-in driver, because the compliance of a completely new driver will change quite a bit in the first few minutes of excitation. So if the data is to be used for system design, the burned-in dataset is the useful one.

B&C Speakers - Applied RMS Voltage is set to 2.83 V for 8 ohms Nominal Impedance. Thiele-Small parameters are measured after a high level 20 Hz sine wave preconditioning test.

Eighteen Sound - Applied RMS Voltage is set to 2.83 V for 8 ohms Nominal Impedance. Thiele-Small parameters are measured after a high level 20 Hz sine wave preconditioning test.

From the "Summary" part of the W. Klippel, U. Seidel, KLIPPEL GmbH Germany "Fast and Accurate Measurement of Linear Transducer Parameters" article - Despite linear parameters are straight forward and appealing they are meaningless in the large signal domain. As some parameters (see figure 13) are not constant and vary even at small amplitudes the question arises: “What is the small signal domain ?”. Linear modeling of the mechanical suspension seems to be an inadequate simplification of the reality. The dynamic behaviour of the suspension depends not only on the instantaneous displacement but also on the peak displacement occurred in the last period of time. Although it is convenient and common practice to express compliance CMS and resonance frequency fs by single numbers this information shows only a small part of the whole picture. The “single number” linear parameters can be interpreted as weighted mean values of the corresponding nonlinear parameter curves. However there is no contradiction between linear and large signal parameters. There is in fact a smooth connection as large signal parameters preserve and generalize the linear parameters. The large signal parameters give additional information valuable for assessing the permissible working range (maximal displacement, maximal power) and the permissible mechanical and thermal load. Furthermore the large signal parameters reveal the dominant sources of distortion.

Как видите, не всё так просто, как может показаться на первый взгляд.