Test & Measurement Achieving quality auio testing for mobile phones The auio capabilities of a cellular hanset provie the funamental interface between the user an the raio transceiver. Just as RF testing must be performe in a controlle environment, large-scale prouction testing of the mobile phones require special acoustical tests. By Joey Tun Mobile phone prouction testing has been increasing in complexity with the continue aition of enhance user features that require prouction verification. Together, with the push to reuce test time an cost, these trens are multiplying the challenges for toay s test engineers. Consequently, esigners are coming to rely more on the esign qualification process to simplify testing. However, most hanset manufacturers still operate uner 100% testing requirement. Aitionally, no esign qualification process, regarless of how rigorous it may be, can guarantee 100% efect-free components particularly if those components are electromechanical transucers, which can become efective uring the assembly or hanling process. Thus, the nee for auio quality measurements remains a funamental part of the ever-growing test suites for mobile phone manufacturers. Implementing a through-the-air auio quality test system generally involves some form of system level characterization, which can Resonance Mass Law also be useful for any application that requires auio characterization, such as entertainment systems an MP3 players. Coincience The nee for traitional auio quality testing There are two basic categories of auio quality tests: Auio only test: These tests can be one at preassembly, at the circuit boar level or after final assembly. This configuration usually sens the stimulus signal from the auio analyzer through auio components on the evice uner test (DUT) an back to the auio analyzer for measurement (e.g., an auio loopback moe test). The test measures auio quality parameters such as istortion an frequency response. Combine auio/rf test: This test is usually conucte after final assembly an involves verifying auio stimulus signal over the complete RF transmit-receive path. It requires a communications analyzer with auio analyzer capabilities. Combine auio/rf tests usually provie an accurate picture of the overall DUT, verifying correct moulation/emoulation, as well as the integrity of the auio signal. However, they on t necessarily reveal the conition of the electromechanical transucers, i.e., the speaker an the microphone. In fact, it s possible for a transucer with mechanical efects to prouce vali results in combine auio/rf tests. Traitional auio tests are still essential to establish the conition of the transucers. Through-the-air auio test measurement parameters, such as total harmonic istortion (THD) or total harmonic istortion plus noise (THD+N), can reveal subtle transucer efects that may be har to etect with other methos. They also provie the manufacturer with an objective measure of auio quality for total quality control purposes. Test fixture characteristics A goo test fixture is essential to attaining repeatable, meaningful results. The most important function of the fixture is shieling ambient acoustical noise, which means the fixture shoul be enclose in a material that acts as a noise shiel. To reuce noise as much as possible, the enclosure shoul be mae with material that woul prouce maximum transmission loss. Transmission loss is governe by mass law, approximate here as [3] : Magnitue of Soun Reuction (B) Frequency (Hz) Figure 1. A soun-reuction graph shows the effects of resonance an coincience. ƒ c Table 1. Material Table 2. Mass per unit area (kg/m2-per mm) Material thickness t (mm) fc (Hz) of Steel Material Constant A (Hz-mm) Aluminum 2.7 12900 Steel 7.7 12700 Glass 2.5 15200 Lea 11.0 55900 fc (Hz) of Aluminum 4 3175 3225 3 4233 4300 2 6350 6450 48 www.rfesign.com January 2006
DMM SPL Meter must be place at the same location as test microphone for SPL measurements. Figure 2. When characterizing a subsystem, SPL meter placement must be carefully observe. TL = 20 log (ms f ) 48 Pre-Amplifier (20 B) SPL Meter where TL = ranom coincience transmission loss (B), ms = mass per unit area (kg/m²), f = frequency of the soun wave (Hz). A test fixture enclosure can be mae from a variety of materials. In general, the enser the material, the greater the transmission loss. Here, either steel or aluminum woul be the practical choice. Mass law is affecte by the coincience effect at higher frequencies an by the resonance effect at lower frequencies [3]. Both effects nee to be consiere when esigning the fixture. Coincience effect: High-frequency waves cause ripples or bening waves that travellongituinally along the wall of the fixture enclosure. The frequencies of these ripples iffer from those of incient waves, except at a certain frequency calle the coincience or critical frequency (fc). At this critical frequency, the soun energy is transferre efficiently through the walls of the fixture enclosure an the transmission loss escribe by mass law no longer hols. The value of fc is escribe by [3] : fc = A/t where A = constant of material (Hz-mm) (see Table 1 for values), t = material thickness in (mm). The goal is to have an fc that s beyon any frequency of interest in the test. Selecting a thin wall with a high A value material raises the critical frequency, as shown in Table 2. For the frequency range of 200 Hz to 4 khz, aluminum or steel walls that are 3 mm thick or less woul suffice. Resonance effect: Assuming the fixture is rectangular, consier the two facing walls of the fixture enclosure. Imagine that a soun wave originates from one wall as if there is a speaker in it. This soun wave is reflecte back by the facing wall [1]. If the istance between the two facing walls is half the wavelength, then the reflecte wave will be in phase with the incient wave, resulting in a classic resonant or staning wave. For each pair of facing walls insie the fixture, the resonant frequency f is calculate as [1] : f = (c/(2l)) n where c = spee of soun (~344m/s), l = istance between two facing walls (m) an n = 1, 2, 3, etc. (orer of harmonic). For a rectangular fixture, the resonance frequency is estimate as [4] : c f = 2 DMM 2 2 2 n n x y nz lx + l + y Enclose Fixture DUT Pre-Amplifier ( 20 B) Input Stimulus Output DUT (phone) in auio loop back moe DUT Acoustic Coupler Figure 3. Acoustic coupling isolates transmit an receive transucers uring over-the-air auio testing with the DUT in auio loop-back moe. where n x is the harmonic resonance between the x-wall an its facing wall, i.e., n x is 1 for the first harmonic, an l x, l y, an l z are the length, with an height of the fixture in meters. When choosing the fixture s esign, it s important to keep in min that the resonancefrequency epens on the compoun ratio of fixture length to height to with. Unfortunately, there s no one stanar ratio, although a perfect cube shoul be avoie because one frequency can resonate between any facing walls. One of the commonly use ratios (R. Walker, BBC, 1996) is the following [4] : ly l l x y 1. 1 < < 4. 5 4 Another major challenge in test fixture esign is to minimize cross-coupling, which is a result of the stimulus signal at the test speaker irectly coupling into the test microphone instea of going through the DUT. The best remey for this is to use an acoustic coupler to irect the soun from the test speaker to the DUT microphone. If the signal path through the acoustic coupler is seale properly, it can significantly minimize the stimulus signal leakage that causes cross-coupling. It s also important to minimize the ambient signal levels insie the fixture by aing material that will absorb or ampen soun. A variety of acoustic amping materials is commercially available with varying levels of soun absorption or reuction, epening on material type an thickness. Test system setup an characterization Transucer selection is the next important aspect of fixture setup. Frequency response is one of the most important specifications for the test speaker an the test microphone. Frequency response shoul be flat (within ±3 B) for the frequencies of interest, usually 200 Hz to 4 khz. The test microphone will generally require a pre-amplifier, which shoul also have a flat frequency response for the frequencies of interest an typically have a gain of aroun 20 B. Other specifications, such as power level, noise performance an soun pressure level shoul be consiere 50 www.rfesign.com January 2006
Series 2015 an 2016 total harmonic istortion (THD) an auio analyzing 6 ½ igit multimeters Keithley s moels 2015-P an 2016-P auio analyzing multimeters an 2015 an 2016 THD multimeters combine auio ban quality measurements an analysis capability with a full-function 6 1 / 2 igit igital multimeter (DMM). In aition to harmonic istortion measurements an measurements of iniviual harmonics, test engineers can make a broa range of measurements incluing voltage, resistance, current an frequency. The units generate pure tone signals for istortion measurements, an the -P versions compute peaks in the spectrum of the measure signal. The series 2015 an 2016 multimeters are esigne for the emans of auio evice test engineers in high-spee prouction test applications. They can perform a 30-point frequency response test an simultaneously measure THD, THD+noise, or SINAD at each point in 1.1 secons, incluing the time to process a GPIB comman an to transfer the ata to a PC. In aition, the series 2015 an 2016 can measure narrowban noise with the use of internal, programmable igital filters. They also can etermine characteristics of the signal spectrum such as harmonics an spectral peaks. With a complete DMM in the instrument, the series 2015 an 2016 total harmonic istortion an auio analyzing DMMs can function also as the basic measurement instrument in any test system, eliminating the nee for a separate DMM. base on the specifications of the DUT. Because mobile phone transucers are by necessity physically small, their soun prouction quality is somewhat limite. Therefore, specialize expensive transucers usually aren t necessary. A variety of aequate, lowcost transucers is reaily available. The size of the test microphone an the test speaker shoul generally be small to allow for convenient mounting insie the test fixture. The soun pressure level ecreases 6 B for every oubling of the istance, so it s critical to pay attention to the istance () between the test speaker an the DUT microphone, an the test microphone an the DUT speaker when mounting these items insie the fixture. These istances shoul be equal an, in most cases, shoul be from 2 mm to 15 mm. Although each component in the system has a set of manufacturer s specifications, the overall system shoul be treate as a black box an characterize as a system. Characterizing a selecte group of parts (e.g., speaker an microphone of the test fixture, without the DUT) can also reveal useful insights into the system s overall behavior by verifying the performance of the subsystem. Three basic proceures are generally applicable for any type of characterization: 1. Determine system noise. 2. Determine a test signal level for a chosen mi-ban frequency. 3. Use this signal level as a reference level to quantify frequency response. As an example, consier part of the 52 www.rfesign.com January 2006
Figure 4a. This 10-point frequency sweep measuring THD+N(%) an frequency response (Vrms) can verify transucer quality an monitor total system noise. system, the speaker an the microphone (with pre-amplifier), as shown in Figure 2. The following test equipment will be neee for characterization: A DMM/auio analyzer. Soun pressure level (SPL) meter (preferably with 30 B to 120 B range). A simple way to start is to tape the speaker on a wall an align the microphone irectly in front of it. The microphone can be mounte on a supporting stan, but avoi using a microphone support that coul eflect soun, e.g., a facing wall. To ensure the same soun pressure levels, the SPL meter s microphone shoul also be place at istance (in the same physical location as the microphone occupie) when measuring SPL. System noise: With source of the auio analyzer off, measure VRMS using the DMM/ auio analyzer an the soun pressure using the SPL meter. This provies a measure of the noise level in the system an the environment. In a reasonably quiet room, the soun pressure shoul be aroun 40 B, an VRMS can be a few millivolts as the microphone picks up an amplifies ambient noise. Test signal levels: Generally, it s safer to start with low signal levels to avoi amaging the evice [2]. Also, note that the higher the test signal, the more potential it has to create cross-coupling effect an prouce resonance conitions insie the test fixture. Set the auio analyzer output to 1 khz, 50 mv, an measure VRMS an SPL. Then, increase the stimulus by 6 B (ouble to 100 mv). Notice that the VRMS measure shoul be ouble an the SPL measure shoul be increase by 6 B inicating the linear operating region of this subsystem. Repeat this process at least three or four times to verify the linear operating region. If we arbitrarily assume the DMM/auio analyzer output has no limitation an keep repeating this process, we will reach a point where a 6 B increase in the stimulus is no longer met by a 6 B increase in measure signal, inicating a non-linear region. In practice, we may not reach this region because the instrument s output has finite levels. However, this is generally not a concern as we are more intereste in lower signal levels. After verifying the linear region or at least part of it, choose a stimulus signal that will prouce measure signals at least 20 B above noise; i.e., if the noise SPL is 40 B, the stimulus signal level chosen shoul prouce at least 60 B of SPL. In most manufacturing test applications, the test signal ranges from tens of millivolts to a few hunre millivolts, proucing SPLs from 65 B to 85 B. Frequency response: Once the stimulus or Figure 4b. Ten-point frequency sweep measuring THD+N(b) an frequency response (Bm). test signal level is chosen, we can characterize the frequency response of the system. Performing a frequency sweep with the auio analyzer over the chosen frequency range will prouce a picture of how the system behaves at those frequencies. In some cases, the system may exhibit frequency-epenent gain. In that case, it s important to verify that it isn t cause by saturation or resonance of any component. The sweep may be repeate at ifferent signal levels. Generally, for the entire linear region, the gain shoul remain constant for a given frequency. Having this frequency response ata is crucial to choosing the test frequency range an interpreting the test results. Overall test system Every test system is unique, so a golen phone is essential as a benchmark against which to compare subsequent test results. The golen phone is a DUT with known goo performance an functionality, or which is otherwise etermine to be a suitable reference DUT. The overall test system can be characterize following the same general proceure escribe in the speaker an microphone example. During the system characterization proceure, the cross-coupling shoul be verifie by performing a frequency sweep 54 www.rfesign.com January 2006
with the DUT power off or auio loopback moe off. Test parameters Once the characterization process is complete, esign an test engineers can typically work together to efine the specific test criteria. Having a complete set of characterization ata with a golen phone will simplify this task significantly. If a system s transfer function is perfectly linear, its response to a sinusoial stimulus signal will be ientical in shape with the stimulus; i.e., in the frequency omain, both the stimulus an the response signals will have the same frequency (f). However, if the system isn t perfectly linear, any nonlinearity will show up as energy at harmonics of the funamental stimulus frequency [2]. Distortion measurements are one of the most wiely use methos of measuring nonlinearity, which, in the case of a system with electromechanical transucers, coul mean a efective transucer. THD+N is the most commonly use istortion parameter, because it measures the linearity of the DUT while taking into account the effects of both harmonic istortion an noise [2]. A low THD+N value not only inicates that the harmonic istortion of the DUT is low, but that noise in the system is also low. The frequency-epenent nature of the system an the DUT makes it avisable to use a frequency sweep to measure istortion over a selecte frequency range. In the example escribe in this article, a 10-point frequency sweep took only 542 ms to complete, incluing the GPIB bus-transfer time. Figure 4a illustrates simultaneous THD+N an VRMS frequency response measurements. Conclusion Some basic steps nee to be followe to create an auio test system capable of fast, multifrequency auio quality measurements using inexpensive components. Although conceptually simple, this type of traitional auio test can reveal subtle component efects that other test methos can overlook. Although not aresse here, it s esirable to incorporate an esign the same test fixture with aitional RF esign parameters to take RF measurements as well as auio. This will reuce the evice hanling time an can significantly reuce overall test time. RFD References 1. D. Halliay, R. Resnick, an J. Walker, Funamentals of Physics, 4th e., Hoboken, NJ; John Wiley & Sons, 1993. 2. B. Metzler, Auio Measurement Hanbook, 1st e., Beaverton, OR; Auio Precision Inc., 1993. 3. A.C.C. Warnock. (1985, July). Factors Affecting Soun Transmission Loss; Canaian Builing Digest [Online]. Available: http://irc. nrc-cnrc.gc.ca/cb/cb239e.html. 4. S. Linkwitz. (2005, June). Room Acoustics; Linkwitz Lab, Corte Maera, CA [Online]. Available: http://www.linkwitzlab. com/rooms.htm. ABOUT THE AUTHOR Joey Tun is applications engineer at Keithley Instruments Inc., Clevelan, Ohio. 56 www.rfesign.com January 2006