Portal de Periódicos da CAPES

Latest developments on the shielding effectiveness measurements of materials and gaskets in reverberation chambers

2019; Institution of Engineering and Technology; Volume: 14; Issue: 4 Linguagem: Inglês

10.1049/iet-smt.2019.0242

1751-8830

Angelo Gifuni, Gabriele Gradoni, Christopher Smartt, Steve Greedy, Armando Méndez-Villalon, Luca Bastianelli, Franco Moglie, Valter Mariani Primiani, Stefano Perna, David Thomas,

Geophysical Methods and Applications

IET Science, Measurement & TechnologyVolume 14, Issue 4 p. 435-445 Research ArticleFree Access Latest developments on the shielding effectiveness measurements of materials and gaskets in reverberation chambers Angelo Gifuni, Corresponding Author Angelo Gifuni angelo.gifuni@gmail.com Dipartimento di Ingegneria, Università di Napoli Parthenope, Centro Direzionale di Napoli, Napoli, ItalySearch for more papers by this authorGabriele Gradoni, Gabriele Gradoni George Green Institute for Electromagnetics Research, University of Nottingham, NG7 2RD UK School of Mathematical Sciences, University of Nottingham, NG7 2RD Nottingham, UKSearch for more papers by this authorChristopher Smartt, Christopher Smartt George Green Institute for Electromagnetics Research, University of Nottingham, NG7 2RD UKSearch for more papers by this authorSteve Greedy, Steve Greedy George Green Institute for Electromagnetics Research, University of Nottingham, NG7 2RD UKSearch for more papers by this authorArmando M. Villalón, Armando M. Villalón George Green Institute for Electromagnetics Research, University of Nottingham, NG7 2RD UKSearch for more papers by this authorLuca Bastianelli, Luca Bastianelli orcid.org/0000-0003-2042-5252 Dipartimento di Ingegneria dell'Informazione, Università Politecnica delle Marche, Via Brecce Bianche 12, Ancona, ItalySearch for more papers by this authorFranco Moglie, Franco Moglie orcid.org/0000-0003-4421-4396 Dipartimento di Ingegneria dell'Informazione, Università Politecnica delle Marche, Via Brecce Bianche 12, Ancona, ItalySearch for more papers by this authorValter Mariani Primiani, Valter Mariani Primiani Dipartimento di Ingegneria dell'Informazione, Università Politecnica delle Marche, Via Brecce Bianche 12, Ancona, ItalySearch for more papers by this authorStefano Perna, Stefano Perna Dipartimento di Ingegneria, Università di Napoli Parthenope, Centro Direzionale di Napoli, Napoli, ItalySearch for more papers by this authorDavid Thomas, David Thomas George Green Institute for Electromagnetics Research, University of Nottingham, NG7 2RD UKSearch for more papers by this author Angelo Gifuni, Corresponding Author Angelo Gifuni angelo.gifuni@gmail.com Dipartimento di Ingegneria, Università di Napoli Parthenope, Centro Direzionale di Napoli, Napoli, ItalySearch for more papers by this authorGabriele Gradoni, Gabriele Gradoni George Green Institute for Electromagnetics Research, University of Nottingham, NG7 2RD UK School of Mathematical Sciences, University of Nottingham, NG7 2RD Nottingham, UKSearch for more papers by this authorChristopher Smartt, Christopher Smartt George Green Institute for Electromagnetics Research, University of Nottingham, NG7 2RD UKSearch for more papers by this authorSteve Greedy, Steve Greedy George Green Institute for Electromagnetics Research, University of Nottingham, NG7 2RD UKSearch for more papers by this authorArmando M. Villalón, Armando M. Villalón George Green Institute for Electromagnetics Research, University of Nottingham, NG7 2RD UKSearch for more papers by this authorLuca Bastianelli, Luca Bastianelli orcid.org/0000-0003-2042-5252 Dipartimento di Ingegneria dell'Informazione, Università Politecnica delle Marche, Via Brecce Bianche 12, Ancona, ItalySearch for more papers by this authorFranco Moglie, Franco Moglie orcid.org/0000-0003-4421-4396 Dipartimento di Ingegneria dell'Informazione, Università Politecnica delle Marche, Via Brecce Bianche 12, Ancona, ItalySearch for more papers by this authorValter Mariani Primiani, Valter Mariani Primiani Dipartimento di Ingegneria dell'Informazione, Università Politecnica delle Marche, Via Brecce Bianche 12, Ancona, ItalySearch for more papers by this authorStefano Perna, Stefano Perna Dipartimento di Ingegneria, Università di Napoli Parthenope, Centro Direzionale di Napoli, Napoli, ItalySearch for more papers by this authorDavid Thomas, David Thomas George Green Institute for Electromagnetics Research, University of Nottingham, NG7 2RD UKSearch for more papers by this author First published: 01 June 2020 https://doi.org/10.1049/iet-smt.2019.0242Citations: 1AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat Abstract In this study, the authors present the latest developments on the measurements for the shielding effectiveness (SE) of gaskets and materials in reverberation chambers (RCs). A variant method, where the insertion loss of the fixture is achieved from the SE of the fixture with no sample in the aperture is found; it is appropriate for gaskets and for any flat material having a sufficiently high reflectivity at least on one side. A simple and usable condition under which the simplest method for the SE measurements of gaskets and materials in RCs can be used is also given, as well as particular cases where it can be directly applied are shown. This method, whose applicability is enhanced in this study, can be used for gaskets and any flat sample. Such developments simplify measurement setup and associated procedures. Comparisons of results support the methods for the SE measurements of gaskets and material in RCs shown in this study. 1 Introduction Reverberation chambers (RCs) are used for shielding effectiveness (SE) measurements [1-9]. In particular, SE of gaskets and materials are made by using nested RCs (NRCs) [1, 2, 4, 8, 9]; contiguous reverberation chambers (CRCs) can also be used [8, 9]. It is meant that materials and gaskets include any flat electronic system such as printed circuit board (PCB), PCB assembly (PCBA) etc. According to a recent study, the standard procedure for the SE measurements of gaskets and materials in RCs, which is shown in IEC 64000-4-21 [1], can be improved [10]. In this paper, some developments of the improved measurement method shown in [10], are shown in order to simplify the measurement setup and associated procedures. A variant of the improved method [10] using a single antenna in the fixture is found. For such a variant procedure, insertion loss (IL) of the fixture is achieved from the SE of the fixture with no sample in the aperture; therefore, the mechanical stirring in the fixture is not strictly required. The single antenna in the fixture works only as a receiving antenna. This variant procedure is appropriate for gaskets and for any flat material having a sufficiently high reflectivity at least on the side where the field impinges from the fixture. Below, the expression 'sufficiently high reflectivity' will be quantitatively made clear. This case generally includes complex PCBs and complex PCBAs. The SE for this variant procedure is denoted by SE5 in this paper. We also show and discuss a simple and usable condition under which the simplest model for SE measurements of gaskets and materials can be used. This procedure was formally introduced in [4, (14)], where it is denoted by SE4; it is denoted in the same way in this paper. When the condition for its validity is met, it can be applied for any type of flat sample. It uses only two antennas and two transmission measurements; it is rapid and accurate. The only drawback is a possible moderate reduction of the measurement dynamic range (MDR). However, the applicability of SE4 is strongly enhanced by this paper. It is specified that a single antenna is also used in the fixture in [10]; but, the method of measurement is different as it uses reflection measurements and an enhanced backscatter constant [11-13], and it assumes that = 2 in the fixture itself. However, reflection measurements are critical with respect to transmission measurements, especially when the calibration plane has to be taken at end of a long cable and when values to be measured are very low, as it is frequently the case for SE measurements by NRCs. In fact, movements of cables and stress to cable and connector considerably affect reflection measurements. The validity of the method SE5 and of the simple and usable condition for the applicability of SE4, as well as the validity of an intermediate method, which is seen below, is shown by measurements. Note that the symbols SE4, SE5, as well as other similar symbols are used to denote the concerning measurement methods. The paper is organised as follows: in Section 2, background is shown; in Section 3, the theory and an in-depth analysis are shown; in Section 4, results are shown; finally, in Section 5, the conclusions are drawn. 2 Background In [4], a systematic method for SE measurements of gasket and materials was introduced, which is valid under a specified condition of isolation between the chambers [9]. It is denoted by SE3. It clarifies the differences in the results from the two previous and incomplete methods [2, 3], which are denoted by SE1 and SE2. Note that the methods SE1, SE2, and SE3 are denoted in the same way both in [4] and in [10], as well as in this paper [In this connection, we specify that in the formula expressing SE1 after (7b) in [10], an equals sign have to be put after the symbol ; its lack was clearly due to a clerical error.]. A sketch of NRCs and CRCs, where all four antennas are present, is shown in Figs. 1 and 2, respectively, in order to improve the readability of the necessary detailed background. Fig 1Open in figure viewerPowerPoint NRC when all four antennas are present: two outer antennas and two inner antennas Fig 2Open in figure viewerPowerPoint Contiguous RC when all four antennas are present: two left antennas and two right antennas In the standard IEC 64000-4-21, the procedure and concerning formula [1, eq. (G.5)] is not supported by the theory [4, 10]. The improvement of the procedure for SE includes two cases [10, Sections 2 and 4]: case A, where the condition on the isolation between the chambers, that is formally expressed in [10, (6)], has to be met with the sample in the aperture and where the aperture size could also be of the order of a full side of the fixture; case B, where the condition on the isolation has to be met with no sample in the aperture. In the latter case, the aperture size is limited by the condition of isolation itself even though the aperture has to however be electrically large. Considering the above, it is recalled that the method in [10, (1)] can be applied for both cases A and B according to the isolation conditions to be met [Note that in [10, (1)], TFVF1 was wrongly written as 'TFVN1'. It was clearly a clerical error.]. Note that the condition on the isolation for the case A in [10] is the same as that in [4] (with the sample in the aperture), and that for the case B in [10], it is the same as that in [4] (with no sample in the aperture). In any case, the average transmission cross section (TCS) of the aperture is well-approximated by the geometrical optics as it is electrically large. It is equal to , where is the geometrical area of the aperture. The isolation between the chambers is required to be greater than or equal to 10 dB both with the sample in the aperture and with no sample in the aperture, [1, 9]. Such a condition, which is shown in detail in [9], is certainly met when the sufficient condition is met. The sufficient condition is formally expressed in [10, (5)] for the sample in the aperture. The corresponding condition for no sample in the aperture is overall similar. The latter depends on the aperture dimensions and on the volume of the fixture, as well as on its load. When the isolation with no sample in the aperture is not met and the method to be used requires it, the error in the SE measurements, which depends both on the isolation with no sample and on the isolation with sample, can be read in [9, Fig. 2]. For better readability of this paper, the conditions on the isolation between the chambers are shown below, as well as the corresponding sufficient conditions. They are written also in terms of ILs. We can write [9] [Note that some authors use a definition of IL such that its value in dB is negative [10], as is the case here.]: (1) (2) (3) (4) The subscripts s and ns mean 'with sample' and 'with no sample' in the aperture, respectively; therefore, the interpretation of the corresponding parameters is consequential. It is specified that is the average power received by an antenna in the outer chamber when the nested system is fed by the outer chamber; is the average power received by an antenna in the inner chamber when the nested system is fed by the outer chamber; is the average power received by an antenna in the inner chamber when the nested system is fed by the inner chamber; is the average power received by an antenna in the outer chamber when the nested system is fed by the inner chamber; is the IL of the outer chamber with no sample in the aperture; is the IL measured between the outer and inner chambers with no sample in the aperture; is the IL of the fixture; is the SE of the fixture with no sample in the aperture. The notation for all parameters with sample in the aperture is obvious; for example, we have , where is the IL of the outer chamber with sample in the aperture and is the IL measured between the outer and inner chambers with the sample in the aperture. It is important to note that the ratios of received powers can be read as ratios of ILs, as the corresponding transmitting powers can on average be considered constant. For sake of simplicity, hereinafter, the ratios of power are written in terms of ILs only. Note also that and from reciprocity. 2.1 Background for SE5 using a single aperture Two essential conditions have to be met to apply the method SE5: (a) a fixture having an electrically large aperture (ELA) that meets the condition (1); (b) a sample for which it turns out that , where is the IL of the fixture when the aperture is totally covered by a metallic plate. The condition clarifies the meant of 'sufficiently high reflectivity' for a sample. Such a condition is considered met from the choice of the samples; this rationale is supported by the fact that is not very sensitive for small absorptions of the sample in the aperture and/or for the leakage from its edge. For symmetrical samples from the point of view of the reflectivity, the implication could be used even though it becomes practically weak for apertures physically very small and very large outer chamber. Finally, a preliminary absorption measurement [1, 14-20] or a preliminary reflectivity measurement could be made [21]. The condition (1) is necessary only to achieve by the SE of the fixture (enclosure) with no sample in the aperture, which in turns allows the SE of the sample to be obtained by using a single antenna in the fixture when the condition (b) is also met, as further specified in the next section. Measurements to verify the necessary condition (1) are included in those necessary to obtain the SE of the sample. It is specified that the ELA is arranged with a sample holder which depends on the sample type; it minimises the leakage from the edge of the sample and consequently the MDR increases. 2.2 Background for SE5 using an 'Auxiliary ELA' In the previous subsection, since the condition (1) is necessary only to achieve as mentioned above, in the procedure for SE5, the size of the aperture with no sample can also be less than that of the aperture with sample. In fact, when the aperture with a sample holder, i.e. the aperture where the sample is mounted, does not meet the condition (1), then it can appropriately be reduced by covering it by an aluminium tape or an aluminium sheet. Such a reduced aperture is called the 'auxiliary ELA (AELA)' in this paper. Clearly, the AELA, which is always an aperture with no sample, has to meet the condition (1). In this case, the aperture with sample, which has different size than the aperture with no sample (AELA), has to meet the condition (3). The condition has to however be met. Measurements to verify the necessary conditions (1) and (3) are included in those necessary to obtain the SE of the sample also in this case. In particular, it is specified that when the sufficient condition (4) is not met, then is replaced with in (3), which can be obtained by , as shown below. In the standard, it is expected that the smallest dimension of the fixture aperture should be at least at the lowest usable frequency (LUF) in order to minimise the cutoff effect; is the wavelength of the electromagnetic radiation. Note that the blocking effect depends also on the thickness of the sample holder, which in turn depends on the specific sample: the greater the thickness (in terms of ), the greater is the blocking effect. Normally, the thickness of a sample holder is much less of the maximum value. However, we note that no sample holder is necessary for an AELA. In Section 4, we will show by measurements that a square aperture, whose side is a little less than long, could be used as a large aperture. In [22] a square aperture of the side length of about is used. Analytical model for of transmission coefficients of aperture for a specific polarisation and incident direction are available in the literature [23, 24]. Finally, note that the frequency range (FR) is determined by the AELA. However, samples greater than the AELA can be used according to the size of the aperture with sample. 2.3 Brief considerations for both methods SE5 The fixture is always randomly fed from the outer chamber for both versions of the method SE5; therefore, the mechanical stirring in the fixture is not strictly necessary, i.e. only frequency stirring (FS) can be used. It is stressed that for both versions of the method SE5, a single antenna is necessary in the fixture. If a load is added in the fixture, in order to achieve the requested isolation conditions, it has to remain throughout the whole measurement procedure. Moreover, it has to be carefully chosen in order to avoid unacceptable non-uniformity of the field inside the fixture itself. Clearly, the SE of any ELA with no sample (clearly, including any AELA) considered in this paper is assumed to be equal to 0 dB. 2.4 Background for the methods SE4 and SE6 Measurements of SE of gaskets and materials can be made by using only two antennas when the quality factors and of the two chambers (outer and inner chambers, respectively) remain practically constant with no sample and with sample in the aperture. In this case, the simplest method SE4 can be used. From theory in [4, 9], we find a simple and usable condition for the applicability of SE4. For such a derivation, an intermediate method is also found for SE measurements; this method is denoted by SE6 in this paper. It is appropriate in cases where the ILs of the outer chamber with sample and with no sample in the aperture are practically equal and the IL of the fixture with sample in the aperture is practically equal to that of the fixture with a metallic slab in the aperture. We will see that SE6 is less useful than SE5 and SE4. 3 Theory In this section, the theory for the method SE5 and for the simple and usable condition for the applicability of SE4 is shown. The theory for SE6 is also included. Since the theory for SE5 is the same both when the AELA is used and when it is not used, as shown above, no difference is highlighted in the nomenclature between the two cases. However, the cases where the AELA is necessary and those where it is not necessary are well specified according to the considerations in the previous section. 3.1 Theory for the method SE5 In this section, we show that when the condition (3) and (b) are met, can be achieved by . If the condition (3) is met, then we can write [10]: (5) Note that the first term in (5) is equal to . The average TCS, i.e. , is equal to a perfectly absorption cross-section (ACS) [14-20]; is the average effective area of an antenna in an RC [14, 15, 19]. Note also that in this paper, the parameters are shown both in dB and in absolute value; the context makes clear if a parameter is in dB or in absolute value. Under the abovementioned hypotheses, the SE of the aperture with no sample, which is denoted by , can be assumed equal to one. By (5), we can write [25, 26]: (6) By using (6), the total ACS of the fixture can be calculated. We can write (7) where is the total ACS of the fixture with no sample in the aperture [16, 23]. By using (6) and (7), we can achieve . We can write: (8) It is specified that when (5)–(8) are referred to an AELA, the symbol is meant as ; the subscript 'aux' denotes the AELA. Note that (8) can be substituted with in (3), as specified above. The reflecting samples, which can be tested by the method SE5, include gaskets, complex PCBs, fabric shield [8], graphene shields [27], band-stop frequency selective surface [28, 29] etc. In such cases, (8) can be replaced in (5) when it is written for a sample in the aperture; that is, we can write [10]: (9) It is highlighted that in (9) is always referred to as the aperture with the sample. By using (6)–(8), (9) can be written as follows: (10) where , which [under condition (1)] is always greater than one, is referred to the AELA when it is used. Clearly, is always referred to the aperture with sample. When an AELA is not necessary, then and (10) simplifies. The ratio is practically always greater than 1. Equation (10) is simple to implement as it does not require directly the application of (6)–(8), even though (9) has the same form of (1) in [10], which is similar to the form of the SE of gaskets and material in [1]. The form (10) highlights the fact that the procedure for SE5 requires only three antennas. Considering the necessary conditions for the applicability of the method SE5, we note that cases where the AELA is not necessary correspond to the case B in [10, Sections 2 and 4], whereas cases where the AELA is necessary correspond to the case A in [10, Sections 2 and 4] once is achieved. 3.2 Theory for a simple and usable condition for the applicability of the method SE4 When the condition (1) is met, the model SE3 for the SE of gaskets and materials is [4, 9]: (11) Note that the ratio of the net power supply can be considered equal to one. We achieve the required usable condition for the applicability of SE4 by an intermediate step, which results in the method SE6. If the conditions (12a) (12b) are met, then we can write as follows: (13) The condition (12a) always holds when the total ACS of the outer chamber including the fixture with no sample in the aperture, which is denoted by , is much greater than the ACS of the sample obtained when the field impinges from the outer chamber, whose maximum value is ; namely, (12a) always holds when . It also holds when the ACS of the sample obtained when the field impinges by the outer chamber is about the same as that of the aperture with no sample. The ILs , , and , which are required to apply (13), could be measured by using only two antennas; such ILs determine the SE of the fixture [13]. However, such a procedure requires reflection coefficient measurements, which could become critical, as mentioned above. Similar considerations can be made for the method SE5. We are interested in a simple usable condition under which . Note that does not necessarily imply as mentioned above, whereas (14) In fact, if , then and a fortiori ; it follows that the condition on the left side of (14) is met. In this case, (11), as well as (13), results in (15) It is reaffirmed that SE4 can be applied for any flat sample. By considering the condition on the left side of (14), one notes that (15), i.e. the method SE4, can certainly be applied to samples having low reflectivity. For such a class of samples, the ACS of both sides is about the same as that of the aperture with no sample; such samples can also have a considerable SE value. Note that the condition of isolation expressed in (1), or that expressed in (2), is a necessary condition to apply SE4 as it comes from SE3. The implication (14) is true also for two equal CRCs. By (6) and (7), we can write (16) The ranges from to . We are interested in a simple and usable condition that approximates the ratio to 1, i.e. . Note that the condition , which is requested in the procedure for SE4, is more stringent than the , which is requested in the procedure for SE5. In fact, the former includes the latter. By using (16), (6), and (8), we can write (17) If the ELA of the fixture, the volume, and the total losses are such that the SE of the enclosure is greater than or equal to 4 (6 dB), then the conditions expressed by (14) are met and the ratio is less than or equal to 0.75 (–1.2 dB). Therefore, the model (15) can be applied under the condition as follows: (18) We can write (19) We note that under the condition (18), or equivalently (19), SE4 can be applied with a maximum error of about 1 dB. It is noted that this error is a systematic error which is only a component of the measurement uncertainty (MU) [30], as it will be discussed in Section 3.4. Note that when the sufficient condition on the isolation is met with no sample in the aperture, (18) is roughly met. It is specified that the method SE4 was used in [22] but there it was not adequately justified; however, we have found a simple and usable condition to check its applicability. It is specified that the abovementioned considerations on a possible load added inside the fixture are valid also in this case. All measurements made with no sample can be considered calibration measurements for the fixture; therefore, they are made only once in a series of measurements where the configuration of the chambers changes only for the different sample types in the aperture. 3.3 Further in-depth analysis on the proposed methods and their measurement dynamic range When (18) is met, we have an isolation with no sample of at least 12 dB; this occurs when the chambers are equal in volume, see (1) and (2). It is the minimum value of isolation (worst case) under the condition (18). Therefore, for the NRC, the isolation is normally >12 dB. Such an isolation causes just an error of some tenth of dB in the SE measurements [9, Fig. 2]. When (18) is met, (15) is equivalent to (11) and to [10, (1)]. Hence, when (18) is met and no AELA is used for SE5, the methods SE5 and SE6 are also applicable apart from the sample absorption and reflectivity, and they give the same SE value as that given by SE4 excepting for the approximation in (17) due to the value in (19). As a consequence, the comparison of results from SE5, SE6, and SE4 is an intrinsic validation for the three models. In Section 4.2, this criterion is used to validate the three methods. However, it is highlighted that in such cases it is convenient apply SE4 for its greater simplicity. The method SE6 is applicable to samples for which (12a) and (12b) are met. Both methods for SE5 and SE6 require three antennas, one of which is transmitting and two are receiving. One receiving antenna is placed inside the fixture; the transmitting antenna and the other receiving antenna are placed inside the outer chamber. The method SE4 requires only two antennas and two IL measurements. The method SE6 requires one fewer measurement with respect to SE5, which is the measurement of . However, the method SE5 has no reduction in MDR with respect to the corresponding methods improved in [10] and SE6 has no reduction in MDR with respect to SE3. When the method SE6 can be applied for a given sample and measurement setup, then the method SE5 can certainly be applied to the same sample and measurement setup. It is specified that measurements for SE6 and SE4 are self-calibrating, whereas measurements for SE5 (9) or equivalently (10) are not strictly self-calibrating, except for cases where the two receiving antennas are equal, as well as the total length of the cables connecting them to the instrumentation, as further specified in Section 4.2. In any case, SE6 and SE4 do not require corrections for antenna efficiency, as well as SE5 when the two receiving antennas (in the outer and inner chambers) are equal. It is important to note that for a given NRC system and under the condition , the simplest model for SE4 has a reduction of 6 dB in MDR with respect to the cases where the is about 0 dB. These can be the procedures where is about 10 dB, see (1), or those where the condition of isolation between the chambers has to be met with sample in the aperture. If the > 6 dB, the MDR is reduced accordingly. We stress that the MDR can be increased by a reduction of the volume of the greater chamber. One notes that SE measurements of an effective gasket represent reference measurements for SE of flat shields (materials). They represent the maximum SE measurable including the leakage of the fixture, which is meanly due to the sample holder, gasket, and closing system of the aperture (intensity and uniformity of the pressure for the electric contact) being the sample a metallic slab. SE measurements where the antenna inside the fixture is replaced with a well-shielded termination represent the maximum SE values that are potentially measurable, where no leakage from the fixture is present. Such maximum SE values represent the MDR when the minimum SE measurable is about 0