This paper presents the design of Gigahertz Transverse Electromagnetic (GTEM) cell with a dimension of 2.1m x 1.2m x 0.81m over a frequency range from DC up to 1 GHz using finite-difference time-domain (FDTD) method in computer simulation technology (CST) microwave studio. Different design parameters such as tapering length of central conductor (septum), transition in the apex, septum height etc. were taken into consideration before finalizing the dimensions of the cell. The field uniformity inside the cell and volume of equipment under test (EUT) has been presented as per the standard IEC 61000-4-20. The GTEM cell can be used for EMC measurements i.e. Radiated Susceptibility and emission testing. The designed GTEM cell will be implemented for commercial use in EMI/EMC division laboratory of SAMEER, Mumbai.
All the electrical and electronic devices which are potential source of EMI i.e. electromagnetic interference must be electromagnetically compatible to ensure that simultaneous use of electronic devices does not degrade their performance . EMC measurements includes emission and immunity (susceptibility) tests. In emission test, measurements are made to discover if any unwanted signals being radiated from the system (radiated EMI) or appearing on the power lines or control/data lines of the system (conducted EMI) exceed the limits specified by standard agency. In case of immunity testing the system is exposed to selected levels of electromagnetic field at different frequencies to check the satisfactory performance of the system in its operational environment .
Suitable test environment is needed for EMC measurements and there are various measuring methods available with its own pros and cons. For radiated emission test the Open Area Test Site (OATS) is known as the reference test facility but it has limitations like large obstruction free space, unfavorable weather conditions, noise issues etc. A semi-anechoic chamber has electrically equivalent characteristics to OATS and have no reflections but the main limitation is its poor performance at low frequencies along with the high cost. Both of these methods are time consuming and they require antennas for measurements . The GTEM cell was developed in 1984 for the higher frequencies (GHz range) by Asea Brown Bovery Ltd. in Switzerland to avoid the limitation of TEM cells .
GTEM cell provides the electromagnetic environment that is free from ambient noise and disturbances from the external environment will not affect test results. There is no need for antennas as the field is created by the cell itself. The field generated will also not interface with external electrical devices .
Description of gtem cell
A. GTEM cell structure
The GTEM cell is a flared rectangular transmission line with a thin wide conductive central plate (septum) whic is terminated by a hybrid combination of resistors and RF absorbers.
The cell has opening angles of 20° and 30° in vertical and horizontal plane respectively. The inner height to width ratio is 2/3 and the angle between bottom plate and septum is 15°. The septum is typically situated at ¾ th of the cell height but can be set at some offset to increase the test volume as its effect on field uniformity is negligible and is also verified with the help of simulations. For feeding the cell, the tip of input section (apex) consist of standard 50Ω coaxial connector, the geometrical design of the cell should ensure matched characetristic impedance of 50Ω throughout the cell length. There are different approved numerical methods to calculate the impedance of the cell .
B. Test volume and maximum EUT size
The usable test volume (see fig. 2a and fig. 2b) within a TEM waveguide is the area in which uniform electromagnetic field is present as defined by the standard. The maximum size of EUT depends on the size of uniform area. The EUT should be not larger than 0.6 w times 0.6 L .
There is a standard procedure set by IEC 61000-4-20 to calculate the uniform area within TEM waveguides. According to it different grid sizes are chosen as test points example 1.5m x 1m 12-point grid; 0.5m x 0.5m 4-point grid with one central point; 1.5m x 1.5m 16-point grid etc. (see fig.3). Of these test points at least 75% of points shall be within 6 dB i.e. 12 of 16 points or 9 of 12 points have to be within 6 dB of field value. 
Design of gtem cell
The designed GTEM cell is for commercial purposes and accordingly the feasible dimensions are selected as per the requirements. Commercially this cell is available in different sizes. Using CST studio software, the model of GTEM cell has been designed with overall length of 2.2 m, width of 1.2 m and height 0.81 m. The outer body material of the cell is steel and septum is of aluminum. In the model the current is terminated by means of resistive network consisting of four parallel resistors each of 200Ω, thus overall resistance of 50Ω is achieved. The electromagnetic fields are terminated with the help of pyramidal microwave absorbers (see fig.4). The absorbers which shall be used for fabrication have highest absorption, up to -50 dB attenuation and height is 60.96 cm. To reflect the properties of these absorbers ECCOSORB BSRU-1 material has been used from CST library. The waveguide port is used to feed the cell.
B. Transition in the apex
The design of apex is very crucial part and in order to avoid reflection the transition from centre pin of N connector to septum shall be smooth enough. Normally the apex length it is taken as 10% of overall cell length . The two cases for transition from N connector to septum were considered. In case one the septum is kept as square transition as seen in the fig.5 with length of 12 mm while in next case (see fig.6) round corner transition with radius 2.7 mm for corners was taken into consideration.
The effect of these two transitions can be seen from fig.7. The difference is quite clear; for square transition the return loss obtained for lower frequency range is less than 20 dB which is much better as compared to that of round corner transition in the same range of frequency.But with increase in frequencies we observe that round transition graph shifts downwards giving good return loss values whereas the square transiton graph is shifyed upwards and thus reflections are increased. Finally we conclude that in order to get low reflections at higher frequencies the round corner transition works better;as we desire to operate GTEM cells for higher frequencies.
C. Septum tooth length
The septum inside the cell is used for absorbing the electromagnetic radiations emitted by EUT.The tapering length of the septum i.e. the tooth length was changed and observations were noted for length of 20 cm, 30 cm, 40 cm.
Graph of return loss vs frequency was plotted. When there is no reflections the GTEM cell is said to be perfectly matched since complete absorption of radiated waves is achieved.From the graph, for lower frequency range upto 0.45 GHz, when the tooth length is 20 cm most of the signal is reflected back (Red curve).Return loss level decreases as we increase the septum length (black and blue curve).Thus we observe that return loss is inversely proportional to the tooth length.But the tooth length can be increased only upto some certain limits since we need uniform field area below the septum.So, in the final model the tooth length chosen was 30 cm.
GTEM Simulation results
Considering all the design factors mentioned in the above section the final model of the GTEM cell was simulated for frequency range of DC up to 1 GHz and the return loss obtained is less than 16 dB while the VSWR is less than 1.4 as seen in the fig.10 and fig.11 respectively. The impedance value obtained from simulations is 50.20Ω. More accurate values for impedance of 50Ω can be achieved by adjusting the ratio of width of septum to that of outer conductor.
The electric field is measured with the help of electric-field probes placed in the form of two layers as shown in the fig.12. The maximum difference between the field values should be less than 6 dB as per the standard IEC-61000-4-20. It can be seen from the simulation results that all the field values measured by the probes are within 6 dB thus satisfying the standard criteria.
Conclusion and future scope
The design of GTEM cell in CST microwave studio has been presented. We have observed that tooth length of septum affects the return loss. The transition in the apex should be smooth enough to avoid the reflections and so round corner transition was found out to be best. The test volume wherein field should be within 6 dB was found out with the help of field probes. The final dimension of EUT was determined with the help of uniform area. From the simulation results return loss value obtained is less than 16 dB and VSWR is less than 1.5.The design part is successfully completed and the cell is under construction. In future work the simulated and experimental results will be verified.
- K. K. Mukherji, ‘EMI and EMC-relevance on electronic power supplies,’ Proceedings of 1995 International Conference on Power Electronics and Drive Systems. PEDS 95, Singapore, 1995, pp. 423-426 vol.1.
- M.T. Ma, M. Kanda,“ Electromagnetic Compatibility and Interference Metrology,” U.S. DEPARTMENT OF COMMERCE, Malcolm Baldrige, Secretary, Natl. Bur. Stand. (U.S.), Tech Note 1099, 178 pages July 1986.
- Jong-Hwa Kwon, Hyun Ho Park, Ae-Kyoung Lee and Hyung-Do Choi, ‘Comparison of correlation algorithms between GTEM cell and semi anechoic chamber,’ 2002 IEEE International Symposium on Electromagnetic Compatibility, Minneapolis, MN, USA, 2002, pp. 481-485 vol.1.
- S. Pasakawee and V. Sittakul, ‘Implementation and characterization of GTEM cell using ferrite tile absorber,’ 2017 IEEE Conference on Antenna Measurements & Applications (CAMA), Tsukuba, 2017, pp. 65-68.
- D. C. Pande and P. K. Bhatt, ‘Characterization of a gigahertz transverse electromagnetic cell (GTEM cell),’ Proceedings of the International Conference on Electromagnetic Interference and Compatibility ’99 (IEEE Cat. No. 99TH 8487), Hyderabad, India, 1997, pp. 31-38.
- D. Bozec, L. M. McCormack, A. C. Marvin, A. Nothofer and M. J. Alexander, ‘A good practice guide for the use of GTEM cells in EMC measurements according to IEC61000-4-20,’ 2004 International Symposium on Electromagnetic Compatibility (IEEE Cat. No.04CH37559), Silicon Valley, CA, USA, 2004, pp. 660-665 vol.2.