Measurement of Semi-Anechoic Chamber Using Modified VSWR method above 1GHz

Transcription

1 Measurement of Semi-Anechoic Chamber Using Moifie VSWR metho above 1GHz M. Bittera, K. Kováč, J. Hallon Department of Measurement, Faculty of Electrical Engineering an Information Technology, Slovak University of Technology, Ilkovičova 3, Bratislava, , Slovakia Abstract. Most of alternative test places as semi-anechoic chambers are esigne to use them for EMI measurement in frequency range MHz. Nowaays it is necessary to measure raiate emission also at higher frequencies in some cases. However, no vali stanar exists to etermine the process of their verification. In this paper moifie site VSWR metho was use to obtain behaviour an possible utilization of semi-anechoic chamber of EMC Laboratory for measurement at higher frequencies up to 10 GHz in measuring istance 1 m. Keywors: semi-anechoic chambers, site VSWR, verification of test place 1. Introuction The basic stanar referring verification of test places for electromagnetic interference (EMI) measurement is [1]. This stanar eals with verification only in frequency range 30 MHz 1 GHz. Accoring to [1] measure values of normalise site attenuation of the test place are compare with ieal ones. The only official ocument assuming a testing above 1 GHz is a technical specification [2] of ETSI, which assumes also NSA measurement but only anechoic chambers. However, at frequencies above 1 GHz irectional broaban antennas shall be use, so effect of absorbers manifests itself in results at minimally 10 m measuring istance [3]. An alternative metho replacing NSA measurement is escribe in [4]. The metho is base on measurement of voltage staning-wave ratio (VSWR) that inicates ifferences in fiel istribution insie test place. This metho is convenient also for anechoic chambers, but there are no measuring istance conitions. Therefore this metho was use to survey properties of semi-anechoic chamber of EMC Laboratory of our university in frequency range 1 10 GHz using some moifications. 2. Subject an Methos Site VSWR metho is base on moving transmitting source along measurement axis to sample staning wave of electromagnetic fiel. The VSWR value is given as ratio of maximum an minimum signal for measurement series: E V VSWR = max max = (1) Emin Vmin where E max an E min are maximal an minimal value of measure E-fiel an similarly for values of measure voltages on receiving antenna terminals V max an V min. The value of VSWR etermines fiel uniformity instea of space attenuation, while it oes not epen on antenna factor an cable loss accuracies. 42

2 To verify the test place it is necessary to measure values of E-fiel in volume, which is bigger than λ λ λ (λ is wavelength at given frequency). At least measurements along all axes of rectangular coorinate system shall be performe to obtain the quality of the test place. VSWR shoul be below 3.5 B that is the value erive from ±4 B NSA criterion. Accoring to [4] this metho is even better than NSA measurement because of better sensitivity of VSWR measurement, while it is simpler an less time-consuming. At measuring istance 1 m an more, the receiving antenna is situate in far zone for frequencies above 1 GHz. Therefore, the level of measure E-fiel escens linearly with increasing istance from transmitting antenna. Not to influence the value of VSWR by this effect, a correction has to be applie on measure values of E-fiel or voltages on receiving antenna terminals. The E-fiel of arbitrary antenna with gain G supplie by power P in istance can be compute: PG E where β is the phase constant. If E 1 is the value of E-fiel in istance 1, in istance 2 its value E 2 is change accoring Eq. 2: jβ = 30 e (2) 1 E 2 = E1 (3) 2 To apply the site VSWR metho in the semi-anechoic chamber it is necessary to line the reference groun plane by aitional absorbers to minimize its effect. This is suppresse also by utilization of monopole with its groun plane as transmitting antenna an therefore only effect of absorbers are surveye. 3. Results Such a moifie site VSWR metho was use to verify semi-anechoic chamber of our EMC Laboratory, which is line by ferrite tile absorbers. Values of E-fiel are measure in 1 m measuring istance from centre of turntable an in 1 m height, where teste equipment is situate. Horn antenna was use as receiving antenna an tune monopole, which has isotropic irectional pattern, as transmitting one. During the measurement only the position of transmitting monopole was change along the irection of interest. Excitation signal is obtaine by means of hf generator an hf power meter is use as measuring equipment. In orer to verify our chamber in frequency range 1 10 GHz some measurement were performe. Because uring measurements at one iscrete frequency the same antenna was use, it is sufficient to obtain the voltage measure at antenna terminal to know the E-fiel istribution. At first variation of E-fiel along measuring axis was measure. Accoring to [4], values of E-fiel were sample along a 40 cm long line. The measure values escent with increasing measuring istance from 1 to 1.4 m, hence correction was applie accoring to Eq. 3. Also these values have oscillating tenency as it is shown in Fig. 1. The measure values of voltage V are influence also by attenuation an mismatching of transmission path. Maximal VSWR oes not excee 3 B in case of all measure iscrete frequencies. To know character of staning wave in the chamber an to ientify potential sources of oscillations, another measurement was performe in the same irection but only along 2λ long line. Such a measurement was realize at two frequencies 3 GHz (Fig. 2) an 10 GHz. As one can see there are two maximums an minimums within a wavelength. Hence fiel istribution is influence significantly by one reflection from surrouning walls. Consiering 43

3 character of measure values one can account only the reflection from back wall, which is situate the closest to transmitting antenna an perpenicularly to measuring axis GHz B B 2.13B 4GHz 6GHz 8GHz vertical 1.88B 1.33B B 2.75B 10GHz horizontal Fig. 1. Depenence of measure voltage V at antenna on raising measuring istance for ifferent frequencies in 1 m measuring istance Fig. 2. Depenence of measure voltage V at antenna on raising measuring istance for both polarisations of antennas at 3 GHz. The same measurements were performe also for horizontally polarize antennas. As one can see in Fig. 2, in comparison to vertically polarize wave the staning wave of horizontally polarize E-fiel is move in volume of the chamber ue to ifferent conitions. On the other han values of VSWR are very similar, so in next only vertically polarize wave are surveye. Also properties of fiel staning wave in other irections were surveye. The transmitting antenna was eflecte perpenicularly to measuring axis within the range <+λ; -λ> in horizontal an <0; 4λ/3> in vertical irection. Chosen ranges of antenna movement represent the maximal imensions of potentially teste evice. Results of measurement, which was performe for three frequencies, are present in Tab. 1. The values VSWR obtaine in this manner are lower than in previous analysis. This confirms the iea about influence of reflection from the closest wall. The measure values of E-fiel at horizontal eflection are not symmetrical ue to non-symmetry of the analyse test place equippe by measuring equipments. Table 1. VSWR values of analyse fiels at eflection of monopole from measuring axis Deflection in horizontal irection in vertical irection f (GHz) VSWR (B) As was mentione in previous the only influencing estimate effect is unesire reflection from the back wall. It is because use absorbing material ferrite are not esigne to work in such high frequencies. Therefore this wall was line with aitional pyramial foam absorbers to suppress this reflection. As it is shown in Fig. 3 an Fig. 4, pyramial absorbers help us to minimize the values of site VSWR. Their influence is more significant at higher frequencies, at frequency 10 GHz aitional absorbers suppress site VSWR from 2.81 B to 44

4 only 1.05 B (Fig. 4). It is because resonant frequency of use ferrite tile absorbers is at value of 160 MHz [5], which is far from analyse range of frequencies B with aitional absorbers -54 with aitional absorbers B B without aitional absorbers B -57 without aitional absorbers Fig. 3. Influence of aitional absorbers on site VSWR at 3 GHz. Fig. 4. Influence of aitional absorbers on site VSWR at 10 GHz. 4. Discussion The semi-anechoic chamber of EMC Laboratory was buil for EMI measurements in frequency range MHz. Moifie site VSWR metho was use to obtain overview about possible extension of operation frequency range up to 10 GHz. It is known that ferrite absorbers use in analyse chamber are not intene to use them at higher frequency up to 1 GHz. In spite of this fact, it is possible to use this test place for measuring also in frequency range 1 10 GHz an in measuring istance 1 m, because values of site VSWR oes not excee 3 B in all realize measurements. It is evient from the measure values that E-fiel istribution is influence especially by unesire reflection from the nearest wall that is perpenicular to the measuring axis. Resulting from performe analysis, using aitional pyramial foam absorbers on mentione wall leas to reuction of site VSWR to only 1.5 B. Acknowlegements This research was financially supporte by the project VEGA VG 1/3101/06 an by the Ministry of Eucation of the Slovak Republic uner grant 2003SP References [1] CISPR Specification for raio isturbance an immunity measuring apparatus an methos - Part 1-4: Raio isturbance an immunity measuring apparatus - Ancillary equipment - Raiate isturbance. Eition 1.1, May [2] ETSI TS Electromagnetic compatibility an raio spectrum matters; Normalize site attenuation an valiation of a fully line anechoic chamber up to 40 GHz. Version 1.1.1, May [3] Barron, M.: Measure an theoretical NSA above 1 GHz. In proceeings of IEEE Symposium on Electromagnetic Compatibility, August 2002, Montreal, [4] Nothofer, A. et al: A practical analysis of test site valiation methos for raiate RF measurements above 1 GHz. In proceeings of Conference EMC 2005, February 2005, Zurrich,

5 [5] Bittera, M., Smieško, V., Kováč, K.: Moelling of ferrite line semi-anechoic chambers. In proceeings of 16 th International Conference on Electromagnetic Disturbances EMD 2006, Kaunas, September 2006,