SE Measurement Comparison of a Conductive Plastic Modem Enclosure Employing GTEM cell or Mode-tuned Reverberation Chamber Methodology

Y. J. Wang¹, W. J. Koh¹, C. K. Lee² and Y. K. Tai¹

¹DSO National Laboratories, Singapore, 20 Science Park Drive, Singapore 118230, Email: wyajun@dso.org.sg

²Nanyang Technological University, Singapore, Nanyang Avenue, Singapore 639798, Email: ecklee@ntu.edu.sg

Abstract: The paper presents two methodologies for performing shielding effectiveness measurements of a USB modem enclosure shielded by conductive plastic material, by the aid of either a GTEM cell or a mode-tuned mini-reverberation chamber with two orthogonal and mechanical stirrers. The measurement set-ups and underlying mechanisms are described. Both shielding effectiveness measurement results using different facilities are evaluated and discussed. The methodology by the use of the reverberation chamber is proven to be preferable for assessing the shielding effectiveness of the shielding enclosure.
Key Words: SE, GTEM cell, reverberation chamber, conductive plastics, and EMC

1. Introduction

The rapid development and widespread proliferation of sophisticated electronic equipment, such as wireless communications, information technologies, and military industries, have accelerated interest in electromagnetic compatibility (EMC). One of major factors in achieving conformant EMC requirements is electromagnetic interference (EMI) shielding. Enclosing a circuitry in a shielding enclosure is a good way to control radiated emissions and improve immunity or reduce susceptibility to external EMI. Traditional shielding is based on the use of metal materials with well-understood electromagnetic properties. However, driven by economics, miniaturization and complexity, metals are increasingly replaced by conductive thermoplastics or composite materials for housing both commercial electronic equipment [1], such as computers and telecommunications, and military applications, such as spacecraft, aircraft, naval, transportation, and construction structures.

The conductive thermoplastic enclosures provide designers with lighter weight, lower cost, greater flexibility, more complex designs and more aesthetic appeal when compared with metallic counterpart [2], [3]. Most common thermoplastics available for shielding bases include polycarbonate (PC), acrylonitrile butadienestyrene (ABS), polystyrene, nylon, poly phenylene oxide, polypropylen and maleated polystyrene co-polymer. The key conductive coating processes applied to the thermoplastic substrates can be electroless coatings, electrolytic coating, conductive spraying, and vacuum metallizing, etc.

In general, both insertion loss and twin antenna techniques are employed in industry to assess shielding effectiveness (SE) of the conductive thermoplastic enclosure [4]. The insertion loss technique is widely used by manufacturers when evaluating the material shielding performance, while the twin antenna technique is utilized to assess the shielding effectiveness of an enclosure itself. The SE measurement methodology using a mode-tuned or mode-stirred reverberation chamber has been increasingly generating lots of interest recently [8]. Numerous test standards on the SE measurements of conductive materials or enclosures have been established, which result in the application of hybrid techniques to various shielding products. Some SE measurement techniques available to date are inclusive of ASTM 4935/89, ASTM ES7-83 dual chamber test fixture, MIL-STD-285, VG 95373, TEM-T cell, circular coaxial transmission line holders, dual TEM for near-field SE measurement, transfer impedance approach, time-domain approach, complex permittivity approach, and apertured TEM cell in a reverberation chamber [5]-[7].

The use of a mode-tuned or mode-stirred reverberation chamber for performing EMC measurements, such as shielding effectiveness evaluations and radiated susceptibility testing of equipment and subsystems, has become accepted practically. The concept used in the reverberation chamber is to excite available electromagnetic wave propagation modes to set up variable standing wave patterns in the chamber. The electromagnetic (EM) fields inside the chamber are regarded as statistically isotropic, randomly polarised, and uniformly homogenous within an acceptable uncertainty and confidence limit. The reverberation chamber method is cost-effective and time-efficient compared to some other conventional testing methods [11]-[13]. Extensive interest in introducing the reverberation chamber measurement technique into various standards has led to the attempt to develop performance-based criteria for reverberation chambers. These standards include IEC 61000-4-21, IEC 61000-4-3 (Annex), MIL-STD-461E, MIL-STD-1344A, EIA-364-66A, RTCA/DO-160D, SAE J1113/27, GM-9120P, CISPR 16-1, CISPR 16-2, FAA HIRF user guides, etc.

In this paper, a SE measurement methodology using a GTEM (gigahertz transverse electromagnetic mode) cell is employed initially to quantitatively evaluate the SE of a stylish modem enclosure made of conductive plastic material, within the frequency range of interest. Another novel SE measurement methodology by the aid of a mode-tuned mini-reverberation chamber with two orthogonal and mechanical stirrers is also proposed. Both results are compared and discussed.

The studied device under test (DUT) is a commercially available USB (universal serial bus) modem with a stingray-shaped appearance. Its maximum external dimensions are around 17cm´13cm´4cm. There is no complete perforation except some shallow holes simulating the stingray skin in the plastic top-cover of the modem, while several holes are perforated in the plastic bottom-cover of the modem, which are mainly used for the top-bottom closure purpose and are adequately sealed with a conductive gasket ring. The internal conductive parts of both the top-cover and the bottom-cover of the modem are specially designed such that they can be properly linked to the ground planes of both sides of the printed circuit board (PCB). No inner gasket is installed between the top-cover and the bottom-cover of the modem.

2. Shielding Effectiveness Measurement Set-ups

2.1. GTEM Cell

A GTEM cell is a single-taper development of an asymmetrical TEM (transverse electromagnetic mode) cell with an offset septum plate for increased working volume. It has a current load connected to the septum and distributed wave termination in the form of RAM (radio absorbent material) wall at the end of the enclosing taper. It may be viewed as a careful combination of aspects of a TEM cell and an anechoic chamber. The SE measurement made with a GTEM cell [9] is based on the use of coaxial transmission lines supporting TEM mode propagation: the DUT is placed between the tapered coaxial lines and is immersed in an almost uniform plane-wave field.

The experimental set-up is shown in Fig.1. The source of RF (radio frequency) signal generator feeds the input of the GTEM cell via RF amplifier and RF coupler. The function of the RF coupler is to make sure that the injected power into the GTEM cell is stable throughout the measurements. A properly designed monopole antenna is placed within the modem shielding enclosure after removing its internal circuitry. The monopole is supported by some tiny nonconductive substrate to avoid touching the inner conductive layer of the enclosure material. All the connecting cables with high SE are chosen to avoid externally ambient EMI; the coaxial cable linking the monopole and the outlet of the GTEM cell is properly shielded with film copper to minimize possible EM coupling from the plane wave. The monopole requires adequate grounding connection to the internal conductive layer of the modem enclosure to ensure accurate results. The appropriate positioning of the monopole with respect to the field polarization of the plane wave is essential to the acquisition of SE measurement data.

The ability of a shield to screen out EM fields is defined quantitatively in MIL-STD-285 [10] as the attenuation or the ratio of the received power on both sides of the shield when it is illuminated by EM radiation. Similarly, in the case of SE measurement of a shielding enclosure, the shielding effectiveness can be expressed as

Or

(2)

Where P1, E1 and P2, E2 are the received powers and electric field strengths of the receiving monopole antenna without and with the modem enclosure, respectively. It is well known that SE measurement result of a practical enclosure is widely influenced by the enclosure material, material thickness, enclosure shape, field incidence, and functional apertures, unintentional leakage, seams and joints, etc.

The measurement procedure using this methodology demands the initial received power measurements without the enclosure and the received power measurements with the enclosure, within the frequency range of interest. Three different enclosure orientations are studied to cover possible polarizations and directions of the EM coupling into the enclosure: enclosure top-cover normally facing the GTEM RF inlet, enclosure bottom-cover normally facing the GTEM RF inlet, and enclosure side horizontally facing the GTEM RF inlet.

2.2. Mini-reverberation Chamber

The mode-tuned mini-reverberation chamber applied in the course of the SE measurement has the dimensions of 113.6cm (length)´77.0cm(width)´54.5cm(height). The LUF (lowest usable frequency) of the chamber was determined to be around 0.74GHz; the field uniformity within the frequency of interest satisfies the specific requirements proposed by IEC standards (IEC-61000-4-3 or IEC-61000-4-21). The reverberation chamber itself is a shielded rectangular enclosure made of aluminium material. A rectangular door cut in the chamber top is of double-layer structure: one layer of transparent wire-meshed conductive plastic sheet above the lower layer of removable aluminium sheet. High quality finger gaskets are lined evenly along the door perimeters to avoid radiation leakage when the door is closed. The door is sealed shut during measurements by the use of a series of forth latches surrounding the four sides of the door. There are totally two orthogonal Z-shaped stirrers (rotating paddle wheels) inside the chamber, which are mutually at right angles. The vertical stirrer reaches from the floor to the ceiling of the chamber, while the horizontal stirrer is parallel to the width edge of the chamber at the opposite top. The extended lengths of the vertical stirrer and the horizontal one are 48.0cm and 60.0cm, respectively. Both have the same width of 12.0cm.

The schematic of the SE measurement is plotted in Fig. 2. A small horn antenna, attached to a RF amplifier and RF generator, is placed nearby the corner of the chamber to transmit RF signal into the chamber. A monopole antenna, identical to the one used in the GTEM cell methodology, is utilised to detect the received power. A step motor assembly, which is operated by a laptop with a GPIB port, is connected to both stirrers. The rotations of both stirrers are controlled by a specific operating program developed in the laptop.

A spectrum analyzer is used to collect data at discrete sampling frequencies, which are spaced at 0.25GHz interval. The receiving monopole will monitor the received powers to the spectrum analyzer, which are then recorded into the same laptop by the aid of some instrumental measurement software. With these collected raw data, the software will go through a series of post-processing calculations. A three-axis isotropic E-probe is used as a reference antenna to ensure the same field strength inside the chamber in two cases with and without the modem enclosure immersed. Due to the loading effect of the modem enclosure, it is essential to increase the RF power injected to the chamber appropriately, based on the calibration results obtained by the E-probe.

3. Results and Analyses

Fig.3 compares the SE measurement results of the USB modem shielding enclosure by using two different methodologies. The frequency band was limited to 0.1GHz - 1.0GHz in the GTEM cell methodology due to the facility constrains. The frequency band for the mini-chamber was chosen to be 0.5GHz - 1.0GHz. Since the LUF of the chamber was originally designed to be around 0.74GHz, the field uniformity at much lower frequencies than the LUF is supposed to be unacceptable for practical applications [12].

Three different orientations of the modem enclosure were investigated due to the vertical polarization of the plane wave in the GTEM cell: the top-cover, bottom-cover and wide side of the enclosure normally facing the GTEM power inlet, respectively. Similarly, three orthogonal orientations of the modem enclosure were also tested in the case of the mini-reverberation chamber methodology, the average received power of all the three orientations at one specific frequency was used for the SE post-processing.

It can be seen that big fluctuations of SE measurement results are obvious in the GTEM methodology, within the frequency range of interest, as shown in Fig.3. The SE variation with the enclosure top-cover facing the GTEM RF power inlet is similar to that with the enclosure bottom-cover facing the GTEM RF power inlet except at few frequencies. This may be due to the quasi-symmetrical structure and the similar material characteristics of both the top and bottom of the modem enclosure. The SE results with the enclosure side facing the GTEM RF power inlet show better performance compared to the above two cases, which may be associated with the perfect contact between the peripheral conductive parts of both the top and bottom enclosure covers. Most of the SE values in the GTEM methodology are within the range of 20 - 40dB, while the poorest SE at 0.35GHz and 0.825GHz might be related to the enclosure resonant modes.

In the mini-reverberation chamber methodology, the SE results are relatively more stabile compared to the GTEM methodology. Two additional important aspects are observed. Firstly, the SE values within the frequency range of 0.5GHz - 0.725GHz are apparently lower compared to those within the frequency range from 0.75GHz to 1.0GHz; there are big differences (up to 20dB) in the SE measurements by using the GTEM cell or the mini-reverberation chamber set-ups at the frequency range of 0.5GHz Ð 0.725GHz. This is principally due to the LUF, i.e. 0.74GHz, of the mini-reverberation chamber, where lower mode density and non-uniform field strengths will exist under evanescent modes. The discrepancy suggests that the mini-reverberation chamber will not be suitable for frequencies lower than the LUF of 0.74GHz. On the contrary, this phenomenon could be regarded as the possible verification of the LUF of the chamber. In the second aspect, the dynamic range of the SE results is limited to about 18 - 27dB from 0.75GHz to 1.0GHz, which is close to that measured applying the GTEM methodology except at the possible enclosure resonant frequencies. The mini-reverberation chamber methodology may imply the worst case for the SE measurements, as the EM coupling into the enclosure are highly due to the random polarization within the chamber.

It should be noted that the correlation between the SE results using the two different methods is not investigated. As the radiation conditions in the GTEM cell are not satisfied everywhere, the presence of the DUT enclosure would modify the field distribution and polarization and introduce the loading effect. Therefore, great attention should be paid to the use of Formula (1) or (2) for the SE calculation in the case of the GTEM methodology. As for the loading effect of the DUT enclosure, one could increase the input RF power accordingly by introducing and monitoring a small sensor inside the GTEM cell. The similar problem occurs in the mini-reverberation chamber methodology, where a three-axis isotropic E-probe, as shown in Fig. 2, is applied as a reference antenna to ensure the same field strength within the chamber with or without the DUT enclosure immersed.

4. Conclusions

Two methodologies, employing either a GTEM cell or a mode-tuned mini-reverberation chamber, for experimentally measuring the shielding effectiveness of a USB modem, made of conductive plastic material, have been presented and compared. In the case of the enclosure SE measurement within the frequency range of interest, the reverberation chamber method offers better testing repeatability, more stable outcome, more aspect angles and significant time-effectiveness over the GTEM cell counterpart.

Acknowledgements

The first author would like to recognize L. S. Lam of DSO National Laboratories, Singapore for her valuable assistance in the course of the shielding effectiveness measurements.

References

1. B. C. Jackson and G. Shawhan, “Current Review of the Performance Characteristics of Conductive Coatings for EMI Control”, Proceedings of 1998 IEEE International Symposium on EMC, vol.1, pp. 567-572.
2. H. Rahman, P. K. Saha, J. Dowling, and T. Curran, “Shielding Effectiveness Measurement Techniques for Various Materials Used for EMI Shielding”, IEE Colloquium on Screening of Connectors, pp. 9/1-9/6.
3. R. D. Semidi, J. Pintelon, G. Spildooren, and S. Criel, “Practical Experience with the Shielding Effectiveness of Small Coated Plastic Boxes”, Proceedings of 2001 IEEE International Symposium on EMC, vol. 3, pp. 245-260.
4. A. Ogunsola, “A Test Programme for the Shielding Effectiveness Evaluation of Conductive Plastic Enclosures”, Proceedings of 2000 IEEE International Symposium on EMC, pp. 851-854.
5. B. A. Kinningham and D. M. Yenni, “Test Methods for Electromagnetic Shielding Materials”, Proceedings of 1988 IEEE International Symposium on EMC, pp. 223-230.
6. P. F. Wilson, M. T. Ma, and J. W. Adams, “Techniques for Measuring the Electromagnetic Shielding Effectiveness of Materials: Part I-Far-field Source Simulation”, IEEE Transactions on EMC, vol. 30, No. 3, Aug. 1988, pp. 239-250.
7. P. F. Wilson and M. T. Ma, “Techniques for Measuring the Electromagnetic Shielding Effectiveness of Materials: Part II-Near-field Source Simulation”, IEEE Transactions on EMC, vol. 30, No. 3, Aug. 1988, pp. 251-259.
8. M. O. Hatfield, “Shielding Effectiveness Measurements Using Mode-stirred Chambers: a Comparison of Two Approaches”, IEEE Transactions on EMC, vol. 30, No. 3, Aug. 1988, pp. 229-238.
9. A. Manara, “Measurement of Material Shielding Effectiveness Using a Dual TEM Cell and Vector Network Analyzer”, IEEE Transactions on EMC, vol. 38, No. 3, Aug. 1996, pp. 327-333.
10. MIL-STD-285, “Method of Attenuation Measurement for Enclosures Electromagnetic Shielding for Electronic Test Purposes”, 1956.
11. G. J. Freyer and M. O. Hatfield, “An Introduction to Reverberation Chambers for Radiated Emission/Immunity Testing”, ITEM 1998, pp. 86-95.
12. M. L. Crawford and G. H. Koepke, “Design, Evaluation, and Use of Reverberation Chamber for Performing Electromagnetic Susceptibility/Vulnerability Measurements”, NBS Technical Note 1092, Apr. 1986.
13. K. Goldsmith, “Reverberation Chambers-What Are They?” IEEE EMC Society Newsletter, Fall 1999.

Biographical Notes

Ya Jun Wang (M’02) was born in Ningbo, Zhejiang, P.R. China. He received B. Sc. (Best Hons) from Physics Department of Xiamen University, P. R. China in 1992, and M. Eng. (by research) in electrical & electronic engineering from Nanyang Technological University, Singapore in 2000. From 1992 to 1998 he was an electrical engineer/a senior electrical engineer and assistant to General Manager at China Import and Export Commodities Inspection Bureau (CCIB), Fujian, P.R. China. Since November 2000, he has been working with DSO (Defence Science Organisation) National Laboratories, Singapore as a member of technical staff and an EMC (electromagnetic compatibility) engineer. He is an NARTE-certified EMC engineer and a committee member of IEEE Singapore EMC chapter for year 2002. Some areas of his interest are devoted to microstrip antennas, mobile antennas, numerical methods, computational electromagnetics, test chambers, EMC protection and measurement techniques.

Wee Jin Koh (M’90) was born on May 13, 1957 in Singapore. He received his B.Sc. from UMIST, UK in 1979, M. Sc. from Naval Postgraduate School, California USA in 1987 and PhD from Ohio State University, Ohio USA in 1995, all in electrical engineering. He has been working at DSO National Laboratories since 1981. He worked as an EMC engineer from 1982 to 1985, and headed an EMC Group from 1987 to 1991. He was appointed Head of Research in EM in 95 and became Head of EM Centre in 1999. His area of interest is in RCS and EMC.

Ching Kwang Lee was born in Malaysia. He received his B.Sc. and PhD degrees from the University of Kent at Canterbury, United Kingdom, in 1982 and 1987 respectively. He was a research fellow in the areas of microwave antenna majoring in frequency selective surface, at the above university between 1988 and 1990. He joined the Electro-Optic Group, Division of Radiophysics (now renamed as Telecommunications and Industrial Physics), and Commonwealth Scientific Industrial Research Organisation (CSIRO) in Australia as a research scientist from Oct. 1990 to Jul. 1991 working on near field range. He is currently an associate professor in the school of Electrical and Electronic Engineering, Nanyang Technological University, Singapore. His research interests include frequency selective surface, microstrip antenna and electromagnetic inverse scattering.

Yeow Kwang Tai was born in Singapore. He received his B. Eng. (Hons) in Electrical Engineering (telecommunication) from the University of Edinburgh, UK in 1998. He has been working at DSO National laboratories, Singapore as a Member of Technical Staff since his graduation. His specialisation includes reverberation chamber characterisation, radio frequency interference and field to cable coupling.