Alternative Radiated Emission Measurements at Close Distance In Industry

Transcription

1 Alternative Radiated Emission Measurements at Close Distance In Industry Osman Şen, Bahadır Tektaş, Soydan Çakır, Mustafa Çetintaş Electromagnetic Laboratories, TUBITAK UME, Gebze, Kocaeli, Turkey Abstract Radiated emission measurements are always performed in laboratories at Open Area Sites (OATS) or in anechoic chambers at 10 or 3 m in accordance with the standard EN55022 other similar standards. However, it is not always possible to use laboratories because of some limitations. If the Equipment Under Test () has large dimensions or heavy, it is not possible to use laboratory facilities during tests. As a consequence, on-site radiated emission tests are required in industry. On the other hand, radiated emission tests have many challenges in industry such as background noise, uncontrolled reflections and narrow testing environment, for that reason EMC testing is very challenging at 3m/10m and the close distance testing at non-standard distances becomes unavoidable in industry. In this research, we have experimentally investigated the relation between measurement results at close distance and standard distance for certain types of antennas. Index Terms Alternative, Close Distance, Emission, On-Site, Radiated. I. INTRODUCTION All equipment placed on the European Market has to fulfill the essential requirements of the European EMC Directive. The normal approach is to show compliance with basic test requirements and testing electrical and electronic products is a must before entering the market. Actually, EMC measurement and validation are necessary during the whole period of development of products. However, development, implementation and maintenance of the EMC measurement facility in accordance with standards are heavy loads for industry. Using the facility in EMC laboratories is a solution but expensive and time consuming. In addition, most of the time, it is not always possible to use standard laboratory EMC methods for some s which are large, stationary or have high currents. Radiated emission test is one of the major tests for industry and widely performed in laboratories at OATS or anechoic chambers for the frequency range 30 MHz 6 GHz by the use of EMC antennas in accordance with CISPR22 [1] and CISPR11 [2] and other similar standards. However, it is not always possible to use laboratories because of the mentioned limitations and on-site measurements become unavoidable. On the other hand, on-site measurements have many challenges such as high background noises, narrow test environment and undesired severe reflections if standard measurement distances are chosen. Some of researches which include the investigation of correction factor between 3m and 10m radiated emission tests have been carried out in [3]. In [3], certain physical effects that complicate the use of radiated emission test data at one test distance to predict the results at a different distance were well studied. Another research on problems with radiated-emission testing at 3m distance according to CISPR11 and CISPR22 is presented in [4]. In [4], extrapolation factors for a personal computer that radiated the signal from a comb generator via its wiring harness and for a battery-powered reference radiator were experimentally determined. Results show that factors can vary between -18 db and -1 db. It is concluded in [4] that, based on the present specifications, compliance tests at 3 m distance should not be allowed. Finally, a good work on radiated emission measurements at a variety of distances from 1 to 30 m is accomplished in [5] in terms of theoretical comparisons of some standards. Although there are several researches especially on 3 and 10m radiated emission tests and correlations in literature, no sufficient information about more closer distances despite the fact that the closer distances have to be utilized in industry due to many limitations such as very large s and narrow places. In our research, we experimentally investigated the relation between measurements at non-standard close distances and standard distances. We utilized metal boxes and metal plates with radiator elements to simulate typical s in industrial environment. We investigated experimental relation between close distance and standard distance in three major setups and each setup also included different simulations. We selected 0.3m, 0.5m, 1m as alternative distances. On the other hand, as a reference distance we selected 3 m instead of 10m because there are already good experimental researches [5,6] that establish the link between 3m and 10m results. Despite the fact that we performed the measurement in an anechoic chamber, Normalized Site Attenuation (NSA) values are close to 4 db at 10m distance and around 2 db at 3m distance. Consequently, in this research, 3 m is the better choice as a reference in comparison by 10m in terms of measurement uncertainty. In our research, the main purpose was to investigate an experimental relation, which reasonably includes many complex parameters such as mutual coupling, near field effect and so on, between the close distance and the standard distance (3 m in our case) for three type popular basic antennas; biconical, log-periodical and horn antennas. In our research, simulations are composed of a cubic metalic box, additional metal plates, radiating elements and some slits to allow radiations. The metallic box that forms the center of the simulations has an internal transmitting small antenna which radiates through the slits to simulate emissions coming out of s. The other radiating element is a set of

2 cable coming out from the slits and laid over the surface of the simulations in different configurations to simulate external cable emissions. The internal antenna and external cable configuration were connected to the same feeding point at the bottom of the metallic box. The metallic plates with different dimensions are also attached to the center metallic box in turn to simulate s with different dimensions and investigate the dependence of mutual coupling on dimensions. During the research, three different setups were mainly installed. Although all the three setups have the central metallic box, major difference between the setups was the different cable configuration, consequently different radiation patterns. In measurement, while the antennas were kept in fixed height at close distance measurements similar to actual tests in industry, the antenna was moved from 1m to 4m in height at the reference measurement at 3m distance as stipulated by the standard. In each distance, the maximum of horizontal and vertical measurements was recorded at close distance and these recorded maximum values were compared to the maximum results obtained in the reference standard distance in order for a relation. A. Setup1 II. EXPERIMENTAL SETUP The simulation used in Setup1 comprises a metallic radiator box at the centre and metallic plates around the metallic radiator box. The diagram of the metallic radiator box is shown in Fig. 1. As seen in Fig. 2, the metallic radiator box has an internal small transmitting rod antenna inside it and periodic slits on its front face to allow internal radiation. This radiator box has also external cables laid on it to allow external radiation. The external cables in setup1 were vertically placed in a zig-zagged arrangement on the front face over the slits. The external cables and the internal antenna are connected to the same feeding point inside the box. The photos of measurement setup are given in Fig. 3. The internal small transmitting antenna and external cables of the are supplied by a signal generator and an amplifier to produce a good amount of a signal level on the receiving antenna. In addition to the metallic central box, there are metal plates to be attached the metallic central box to increase the dimensions of the simulation in order to also check the effects of dimensions on the investigated relation. The dimensions of each metal plate are 0.3mx0.3m, 0.5mx0.5m, 0.7mx0.7m, 0.9mx0.9m, 1.1mx1.1m, 1.3mx1.3m, 1.5mx1.5m, 1.7mx1.7m, 2mx2m respectively. Each metal plate includes a hollow rectangle to accommodate the metallic central radiator box at the center. The metallic plates were so designed that they can tighten to the metallic radiator box securely to provide good earthling continuity. The measurements were performed in three frequency ranges; 30 MHz MHz, 200 MHz - 1 GHz and 1 GHz - 6 GHz in turn by using three basic antennas; biconical, log-periodical and horn antennas respectively. To make the measurement setup very stable, the Antenna distance was manually adjusted by moving the simulation on a special track system, so that negative effects caused by alignment and undesired movement were minimized. In the research, firstly the measurements at the close distance; 0.3m, 0.5m and 1m were completed at the fixed height 1.7m for each metallic plates attached to the central radiator metal box, in turn. Thereafter the same simulations were tested at the reference distance 3m by moving the antenna from 1m to 4m for both polarizations. The maximum of the measured values of both of the polarization was taken as the final result for the current configuration. Finally, the result taken for each configuration at the close distance was compared to the result of the same configuration measured at the reference distance at 3m. The difference in results between the current close distance measurement and the reference distance measurement was called Correction Factor (CF) for that close-distance measurement. In short, a correction factor for a close-distance measurement signal a link to the reference-distance measurement. The summary of the configurations under investigation and measurement distances/antenna heights is given in Table 1. Table.1 Configurations and Measurement Distances/Antenna Height Measurement Distance/Antenna Height configuration 0,3 m x 0,3 m 0,5 m x 0,5 m 0,7 m x 0,7 m Reference 0,9 m x 0,9 m d: 0.3 m d: 0.5 m d : 1 m d : 3 m 1,1 m x 1,1 m h: 1.7 m h: 1.7 m h: 1.7 m h: scan 1,3 m x 1,3 m from 1 to 4 m 1,5 m x 1,5 m 1,7 m x 1,7 m 2 m x 2 m Fig. 1. Diagram of metallic central radiator box and zig-zagged arrangement of external cables in Setup1 Fig. 2. Internal structure of metalic plates Fig. 3. Photo of measurement setup1

3 B. Setup2 The simulation Setup2 is very similar to the simulation in Setup1. It comprises the same radiator box and metallic plates with the same parameters given in Table 1. The only difference is the arrangement of external cables that provide external radiation. In Setup2, the external cables were horizontally placed in a zig-zagged arrangement to obtain a different radiation pattern as seen in Fig. 4. As the internal small transmitting antenna in the same position but the external cables in the different configuration, Setup2 is theoretically expected to have a different radiation pattern from the Setup1. Consequently, this gave us a change to check correction factors in a different radiation configuration from the Setup1. D. Setup4 Fig. 6. Metallic plates on which cables are placed A simple transmitting rod antenna used as a reference was measured in the same conditions as the other setups in order to simulate a very small. This final setup with the transmitting rod antenna as an was named Setup4 in the research. The rod antenna configuration is shown in Fig. 7. Fig. 4. Horizontal zig-zagged arrangement of external cables in Setup2 C. Setup3 Unlike Setup1 and Setup2, the metallic plates attached to the radiator central box have also external cables laid on them in addition the cables on the central box in order to increase the extent of radiation and significantly change the radiation pattern. The photos of measurement setup are given in Fig. 5 The metallic plates on which cables were laid are presented in Fig. 6. The cables placed on the metallic plates are connected to the cables which are on the metallic central radiator box via a connector close to one of edges of the box as seen in Fig. 6. As a consequence, in addition to increasing dimension of the simulation, the metallic plates have also function to contribute to the significant change in radiation pattern. Hence, this also gave us to check the correction factors for the close-distance measurement by using different s with different radiation pattern. Figure 7. Shape for Setup 4 III. EXPERIMENTAL RESULTS AND DISCUSSIONS CF is the subtraction between maximum measurement results ( at close distance ) with measurement results ( 3m ) for current configuration and current measurement distance. The maximum value of the horizontal and vertical polarizations has been taken for each frequency. The correction factors for the close distance 0.3m, 0.5m and 1m for the Setup1 for the frequency ranges 30 MHz MHz, 300 MHz - 1 GHz and 1 GHz - 6 GHz is calculated. Similarly, the correction factors for the Setup2, Setup3 and Setup4 for are calculated. Ultimately, we present the comparison of the CF results obtained for the same close distance and for the s with the similar dimension except the rod antenna (Setup4). The Setup 4 installed with the rod antenna is included in all the graphs as a general with a broad beamwidth. We also include the smoothed average graph to each graph to show the general tendency of the curves and present a rough single CF factor for a certain close distance. The comparison of all the correction factors obtained in all the four setups for 0.3m0.3m and 2mx2m is presented in Fig. 8 - Fig. 16. Fig. 5. Photo of measurement setup3 Fig. 8. 0,3-3 m CF for 30 MHz- 300 MHz 0.3mx0.3m, 2mx2m

4 Fig. 9. 0,3-3 m CF for 200 MHz MHz 0.3x0.3m, 2mx2m Fig m CF for 200 MHz MHz 0.3mx0.3m, 2mx2m Fig ,3-3 m CF for 1000 MHz MHz 0.3x0.3m, 2mx2m Fig CF for 1000 MHz MHz 0.3mx0.3m, 2mx2m IV. CONCLUSION In this paper, we took a first step to establish an experimental relation between close-distance and reference distances measurements. Fig ,5-3 m CF for 30 MHz- 300 MHz 0.3x0.3m, 2mx2m ACKNOWLEDGMENT This research is in the scope of the project IND60 Improved EMC test methods in industrial environment and financially supported by European Metrology Research Programme. The authors thank Savas Acak for his technical support. Fig ,5-3 m CF for 200 MHz MHz 0.3x0.3m, 2mx2m Fig ,5-3 m CF for 1000 MHz MHz 0.3mx0.3m, 2mx2m REFERENCES [1] CISPR22, Information technology equipment Radio disturbance characteristics Limits and methods of measurement [2] CISPR11, Industrial, scientific and medical equipment Radiofrequency disturbance characteristics Limits and methods of measurement [3] Ed Blankenship, David Arnett, Sidney Chan, Searching for the elusive correction factor between 3m and 10m radiated emission tests, IEEE International Symposium on Electromagnetic Compatibility, Austin, TX, USA, pp August [4] H.F Garn, E. Zink, R. Kremser, Problems with radiated-emission testing at 3 m distance according to CISPR 11 and CISPR 22, IEEE International Symposium on Electromagnetic Compatibility, Dallas, TX, USA, pp , August [5] [6] Victor H. Kee, The CISPR 20dB per decade Extrapolation Rule Explored Fig m CF for 30 MHz- 300 MHz 0.3mx0.3m, 2mx2m

5