- 1. Introduction Validation Emission measurement Conversion of results Calculation of conversion factors for the RVC method in accordance with CISPR 16-4-5 Dr.-Ing. Mathias Magdowski Chair for Electromagnetic Compatibility Institute for Medical Engineering Otto von Guericke University Magdeburg, Germany June 29, 2023 License: cb CC BY 4.0 (Attribution, ShareAlike) Mathias Magdowski Conversion factors for RVCs 2023-06-29 1 / 31
- 2. Introduction Validation Emission measurement Conversion of results Overview What is a reverberation chamber? How to validate the proper operation? How to measure emission? How to apply the CISPR-16-4-5? Mathias Magdowski Conversion factors for RVCs 2023-06-29 2 / 31
- 3. Introduction Validation Emission measurement Conversion of results Simple idea of a reverberation chamber Figure: Schematic setup of a reverberation chamber (top view) Mathias Magdowski Conversion factors for RVCs 2023-06-29 3 / 31
- 4. Introduction Validation Emission measurement Conversion of results Practical reverberation chamber Figure: Large reverberation chamber in Magdeburg Mathias Magdowski Conversion factors for RVCs 2023-06-29 4 / 31
- 5. Introduction Validation Emission measurement Conversion of results How to stir the field? Changes of the electromagnetic boundary conditions: ▶ mechanical stirrer(s) ▶ moving walls ▶ relocating the antenna(s) ▶ switching between several antennas Mathias Magdowski Conversion factors for RVCs 2023-06-29 5 / 31
- 6. Introduction Validation Emission measurement Conversion of results How to stir the field? Changes of the electromagnetic boundary conditions: ▶ mechanical stirrer(s) ▶ moving walls ▶ relocating the antenna(s) ▶ switching between several antennas Narrow band frequency changes: ▶ only for immunity testing Mathias Magdowski Conversion factors for RVCs 2023-06-29 5 / 31
- 7. Introduction Validation Emission measurement Conversion of results Vibrating intrinsic reverberation chamber (a) Demonstration with neon tubes (b) In-situ test on a ship Source: Prof. Leferink, University of Twente and THALES, Netherlands Mathias Magdowski Conversion factors for RVCs 2023-06-29 6 / 31
- 8. Introduction Validation Emission measurement Conversion of results Oscillating wall stirrer Figure: Reverberation chamber with an oscillating wall stirrer at the Laboratory of Electromagnetic Compatibility, School of Mechanical Engineering, Southeast University, Nanjing, China Source: https://dx.doi.org/10.1109/TEMC.2020.2983981 Mathias Magdowski Conversion factors for RVCs 2023-06-29 7 / 31
- 9. Introduction Validation Emission measurement Conversion of results Statistical properties of the field Homogeneity: ▶ uniformity over the space ▶ free placement of the EUT in the working volume Mathias Magdowski Conversion factors for RVCs 2023-06-29 8 / 31
- 10. Introduction Validation Emission measurement Conversion of results Statistical properties of the field Homogeneity: ▶ uniformity over the space ▶ free placement of the EUT in the working volume Isotropy: ▶ uniformity in all directions ▶ orientation of the EUT does not matter Mathias Magdowski Conversion factors for RVCs 2023-06-29 8 / 31
- 11. Introduction Validation Emission measurement Conversion of results Statistical properties of the field Homogeneity: ▶ uniformity over the space ▶ free placement of the EUT in the working volume Isotropy: ▶ uniformity in all directions ▶ orientation of the EUT does not matter Validity: ▶ only in the working volume ▶ minimum distance to the walls > λ 4 Mathias Magdowski Conversion factors for RVCs 2023-06-29 8 / 31
- 12. Introduction Validation Emission measurement Conversion of results Change my mind! Source: https://imgflip.com/i/6sa4e8 Mathias Magdowski Conversion factors for RVCs 2023-06-29 9 / 31
- 13. Introduction Validation Emission measurement Conversion of results Overview What is a reverberation chamber? How to validate the proper operation? How to measure emission? How to apply the CISPR-16-4-5? Mathias Magdowski Conversion factors for RVCs 2023-06-29 10 / 31
- 14. Introduction Validation Emission measurement Conversion of results Chamber field uniformity and loading validation Goal: ▶ verification of a sufficiently small field inhomogeneity in the working volume ▶ determination of the lowest usable frequency (LUF) ▶ for the empty and maximum loaded chamber Mathias Magdowski Conversion factors for RVCs 2023-06-29 11 / 31
- 15. Introduction Validation Emission measurement Conversion of results Chamber field uniformity and loading validation Goal: ▶ verification of a sufficiently small field inhomogeneity in the working volume ▶ determination of the lowest usable frequency (LUF) ▶ for the empty and maximum loaded chamber Maximum chamber loading verification: ▶ simulate the chamber loading by the EUT ▶ using a sufficient amount of absorbers ▶ EUT loading ≤ maximum loading Mathias Magdowski Conversion factors for RVCs 2023-06-29 11 / 31
- 16. Introduction Validation Emission measurement Conversion of results Evaluation for each receiving antenna location: Net input power (for each stirrer position): PInput = PFwd − PRev (1) Mathias Magdowski Conversion factors for RVCs 2023-06-29 12 / 31
- 17. Introduction Validation Emission measurement Conversion of results Evaluation for each receiving antenna location: Net input power (for each stirrer position): PInput = PFwd − PRev (1) Average input power: PInput = PInput NTuner (2) Mathias Magdowski Conversion factors for RVCs 2023-06-29 12 / 31
- 18. Introduction Validation Emission measurement Conversion of results Evaluation for each receiving antenna location: Net input power (for each stirrer position): PInput = PFwd − PRev (1) Average input power: PInput = PInput NTuner (2) Average received power: PAveRec = ⟨PRec⟩NTuner (3) Mathias Magdowski Conversion factors for RVCs 2023-06-29 12 / 31
- 19. Introduction Validation Emission measurement Conversion of results Evaluation for each receiving antenna location: Net input power (for each stirrer position): PInput = PFwd − PRev (1) Average input power: PInput = PInput NTuner (2) Average received power: PAveRec = ⟨PRec⟩NTuner (3) Maximum received power: PMaxRec = ⌈PRec⌉NTuner (4) Mathias Magdowski Conversion factors for RVCs 2023-06-29 12 / 31
- 20. Introduction Validation Emission measurement Conversion of results Evaluation For the empty and maximum loaded chamber: Antenna validation factor: AVF = PAveRec PInput 8 positions (5) Insertion loss: IL = PMaxRec PInput 8 positions (6) Mathias Magdowski Conversion factors for RVCs 2023-06-29 13 / 31
- 21. Introduction Validation Emission measurement Conversion of results Evaluation For the empty and maximum loaded chamber: Antenna validation factor: AVF = PAveRec PInput 8 positions (5) Insertion loss: IL = PMaxRec PInput 8 positions (6) For the chamber loaded with the EUT: Chamber validation factor: CVF = PAveRec PInput p positions (7) Chamber loading factor: CLF = CVF AVF (8) Mathias Magdowski Conversion factors for RVCs 2023-06-29 13 / 31
- 22. Introduction Validation Emission measurement Conversion of results Overview What is a reverberation chamber? How to validate the proper operation? How to measure emission? How to apply the CISPR-16-4-5? Mathias Magdowski Conversion factors for RVCs 2023-06-29 14 / 31
- 23. Introduction Validation Emission measurement Conversion of results Radiated emission measurement Based on the measurement of the average received power: Prad = ηTX · PAveRec CVF (9) Advantages: ▶ lower statistical uncertainty when using averages ▶ influence of the EUT validation only, empty chamber validation does not matter Mathias Magdowski Conversion factors for RVCs 2023-06-29 15 / 31
- 24. Introduction Validation Emission measurement Conversion of results Radiated emission measurement Based on the measurement of the average received power: Prad = ηTX · PAveRec CVF (9) Advantages: ▶ lower statistical uncertainty when using averages ▶ influence of the EUT validation only, empty chamber validation does not matter Based on the measurement of the maximum received power: Prad = ηTX · PMaxRec CLF · IL (10) Advantages: ▶ better signal-to-noise ratio for weak emitters close to noise floor ▶ easy to use the maximum hold (or peak hold) function of the measuring instrument Mathias Magdowski Conversion factors for RVCs 2023-06-29 15 / 31
- 25. Introduction Validation Emission measurement Conversion of results Further discussion on radiated emission measurements “Monte Carlo Simulation of the Statistical Uncertainty of Emission Measurements in an Ideal Reverberation Chamber” by Mathias Magdowski and Ralf Vick published at the EMC Europe 2017 in Angers/France https://ieeexplore.ieee.org/ document/8094737 Monte Carlo Simulation of the Statistical Uncertainty of Emission Measurements in an Ideal Reverberation Chamber Mathias Magdowski and Ralf Vick Chair for Electromagnetic Compatibility Otto von Guericke University Magdeburg, Germany Email: mathias.magdowski@ovgu.de Abstract—This paper describes the results of a Monte Carlo simulation to quantify the statistical uncertainty of emission measurements in the scope of electromagnetic compatibility in a reverberation chamber. Such a measurement can be performed based on the average or maximum received power at the reference antenna. Depending on whether value is measured, different parameters as the chamber validation factor CVF, the chamber loading factor CLF or the insertion loss IL are needed to calculate the total radiated power of a device under test. Each parameter as well as the measurement itself features its own statistical uncertainty, which are combined into a joint statistical uncertainty of the total radiated power, which is analyzed in this paper. I. INTRODUCTION A reverberation chamber (or mode-stirred chamber) is an alternative test environment for radiated immunity tests and emissions measurements in the scope of electromagnetic compatibility. It consists of a shielded room that forms an electromagnetic cavity resonator and a rotating metallic object (known as the mode stirrer) to change the boundary conditions and to obtain a statistically homogenous and isotropic field. Reverberation chambers are rather popular for immunity testing [1, Annex D], as their high quality factor allows to reach high field strengths with only low input powers. The utilization for emission measurements [1, Annex E] is unfortunately not that prevalent, as the measured quantity is the total radiated power of some device under test (DUT) that cannot be easily compared with standard limits, which are usually given in terms of a maximum emitted field strength in a certain distance. A conversion between both quantities would be possible, if the maximum directivity of the device under test is known, which is usually not the case. One solution of this problem is to model the maximum directivity of a generic device under test as a function of the frequency and its size [1, Annex E.8]. A more sophisticated model including the statistical distribution and uncertainty of the directivity and the conversion of emission limits is given in [2]. Another solution is to establish conversion factors between different test environments on an empirical basis as described in [3]. For a correct application of radiated emission limits, also the uncertainty of each test environment has to be known. While such uncertainties are quite well known for the established test environments like open area test sites or semi-anechoic chambers, no plausible values are available for reverberation chambers. The goal of this paper is to determine this uncertainty for emission measurements. Section II repeats and summarizes the standard procedure of emission measurements in reverberation chambers. The assumptions and simplifications of the utilized simulation model are given in Sec. III. The following Sections IV, V and VI contain the probability density functions, cumulative distribution functions and percentiles of the quantities of interest, respectively. The main conclusions are summarized in Sec. VII. II. MEASUREMENT OF THE TOTAL RADIATED POWER The total radiated power of any device under test can be calculated from [1, Eq. (E.1)] Prad = PAveRec · ηTx CVF (1) in terms of the average received power PAveRec or from [1, Eq. (E.2)] Prad = PMaxRec · ηTx CLF · IL (2) in terms of the maximum received power PMaxRec. The efficiency of the transmitting antenna used during the chamber validation is denoted with ηTx. The chamber validation factor CVF = PAveRec PInput Np (3) and the chamber loading factor CLF = CVF AVF (4) are both determined from the validation of the chamber performance with the DUT in place [1, Eqs. (B.11) and (B.12)], where PAveRec is the average received power at the reference antenna and PInput is the average input power into the chamber, 978-1-5386-0689-6/17/$31.00 © 2017 IEEE Mathias Magdowski Conversion factors for RVCs 2023-06-29 16 / 31
- 26. Introduction Validation Emission measurement Conversion of results Overview What is a reverberation chamber? How to validate the proper operation? How to measure emission? How to apply the CISPR-16-4-5? Mathias Magdowski Conversion factors for RVCs 2023-06-29 17 / 31
- 27. Introduction Validation Emission measurement Conversion of results Electrical size of an equipment under test a Definition as k · a: k: wave number, k = 2πf c = 2π λ a: radius of the smallest sphere surrounding the EUT Questions: ▶ What belongs to the EUT (case, cables, . . . )? ▶ Which cable length has to be considered? Mathias Magdowski Conversion factors for RVCs 2023-06-29 18 / 31
- 28. Introduction Validation Emission measurement Conversion of results Electrical small EUTs Condition: k · a ≤ 1 Figure: Radiation pattern of a small dipole (Source: Wikipedia) Mathias Magdowski Conversion factors for RVCs 2023-06-29 19 / 31
- 29. Introduction Validation Emission measurement Conversion of results Electrical large EUTs Condition: k · a 1 Figure: Radiation pattern (planar cut) of a practical EUT (Source: Magnus Höijer, FOI) Mathias Magdowski Conversion factors for RVCs 2023-06-29 20 / 31
- 30. Introduction Validation Emission measurement Conversion of results Electrical large EUTs Condition: k · a 1 Figure: Radiation pattern (planar cut) of a practical EUT (Source: Magnus Höijer, FOI) Mathias Magdowski Conversion factors for RVCs 2023-06-29 20 / 31
- 31. Introduction Validation Emission measurement Conversion of results Electrical large EUTs Condition: k · a 1 Figure: Radiation pattern (planar cut) of a practical EUT (Source: Magnus Höijer, FOI) Mathias Magdowski Conversion factors for RVCs 2023-06-29 20 / 31
- 32. Introduction Validation Emission measurement Conversion of results The higher the frequency, the larger the problem! Source: https://imgflip.com/i/6sagu4 Mathias Magdowski Conversion factors for RVCs 2023-06-29 21 / 31
- 33. Introduction Validation Emission measurement Conversion of results Directivity Definition: D(ϑ, φ) = Φ(ϑ, φ) Prad/4π (11) ▶ Φ is the power density radiated per solid angle ▶ Prad is the total radiated power ▶ Prad/4π is the average radiated power Maximum directivity Dmax: ▶ Directivity in the main beam direction ▶ electrically short dipole: Dmax = 3/2 = 1.76 dBi ▶ electrically large EUT: Dmax ≈ 10 = 10 dBi Mathias Magdowski Conversion factors for RVCs 2023-06-29 22 / 31
- 34. Introduction Validation Emission measurement Conversion of results Free space or fully anechoic room (FAR) Relationship between power and field strength: E2 max = Dmax η0 4πr2 Prad (12) ▶ η0 is the free space impedance ▶ r is the distance ▶ derivation with the help of a short electric dipole ▶ valid in the far field Mathias Magdowski Conversion factors for RVCs 2023-06-29 23 / 31
- 35. Introduction Validation Emission measurement Conversion of results Half space or semi-anechoic chamber (SAC) Relationship between power and field strength: E2 max = Dmax η0 4πr2 Prad g2 max (13) ▶ gmax is an additional geometry factor ▶ value range between 0 and 2 ▶ consideration of the reflection at the ground plane ▶ interference of the direct and the reflected wave Mathias Magdowski Conversion factors for RVCs 2023-06-29 24 / 31
- 36. Introduction Validation Emission measurement Conversion of results How to estimate or model the directivity? “Statistical Analysis of the Correlation of Emission Limits for Established and Alternative Test Sites” by Hans Georg Krauthäuser published in the IEEE Transactions on EMC, vol. 53, no. 4, Nov. 2011 https://ieeexplore.ieee.org/ document/5715864/ IEEE TRANSACTIONS ON ELECTROMAGNETIC COMPATIBILITY, VOL. 53, NO. 4, NOVEMBER 2011 863 Statistical Analysis of the Correlation of Emission Limits for Established and Alternative Test Sites Hans Georg Krauthäuser, Member, IEEE Abstract—This paper discusses the correlation of radiated emis- sion measurements on open area test sites, in TEM waveguides, fully anechoic rooms and reverberation chambers. Distributions of all essential factors have been derived and mean and quantile values are given. Influence of polarization mismatch and directivity is discussed. Based on the analysis of uncertainties, an alternative data treatment on open area test sites and in fully anechoic cham- bers is proposed. A statistical link has been established between the spherical wave expansion model and the array of point sources model of directivity. Index Terms—Directivity, fully anechoic room (FAR), gigahertz transverse electromagnetic (GTEM), open area test sites (OATS), polarization mismatch, radiated emission limits, reverberation chamber (RC), unintentional emitters. I. INTRODUCTION HISTORICALLY, emission limits have been developed for measurements on open area test sites (OATS). In order to be independent of weather conditions and to reduce ambient noise, most OATS-like emission measurements are nowadays done in semianechoic chambers (SAC). No distinction between OATS and SAC will be made within this paper. Alternative methods claim to have advantages regarding one or more of the aspects: speed, costs, reliability, frequency range, and ap- proximation of reality. These alternative methods include the reverberation chamber (RC) method and the TEM waveguide method that are already covered by international standards, e.g., [1], [2], and also the fully anechoic room (FAR) method for which an international standard is in development [3]. This paper will not address the validity of these claims. In- stead, it focuses on the correlation of the physical quantities measured by the methods in the case of unintentional radiators. The analysis is strongly based on the discussion of the statistical properties of the quantities involved. Also, this paper will not address the intrinsic uncertainties of the measurement methods itself. This study is focused on uncer- tainties of the prediction of a measurement result for a certain measurement site based on the (supposedly correct) measure- ment results from another site. Understanding the directivity patterns for electrically large sources turns out to be a key topic for the application of alter- Manuscript received September 15, 2010; revised December 6, 2010; accepted December 21, 2010. Date of publication February 17, 2011; date of current version November 18, 2011. Review of this manuscript was arranged by Department Editor H. Garbe H. G. Krauthäuser is with the TU Dresden 01069 Dresden, Germany (e-mail: hgk@ieee.org). Digital Object Identifier 10.1109/TEMC.2010.2102764 native radiated emission and immunity measurement methods and is already addressed in various papers, e.g., [4]–[15]. A second important quantity for the correlation of uninten- tional emissions is the polarization mismatch factor μ. A. Directivity Directivity D is defined as the ratio of the radiated power per unit solid angle Θ(θ, φ) in direction (θ, φ) and the average radi- ated power 1/(4π) · 2π 0 π 0 Θ(φ, θ) sin θdθdφ = 1/(4π)·PT , D(φ, θ) = 4π · Θ(φ, θ) PT (1) where PT is the total radiated power. Most often, one is not interested in the whole radiation pattern, but in the maximum of the directivity Dmax which is the directivity of the largest radiation: Dmax = 4π · max [Θ(φ, θ)] PT . (2) It is clear that a measured value of Dmax depends on the angular sampling. The true value is only achieved for very fine angular resolution on the whole surface of a sphere surrounding the source region. In this paper, a sampling of the complete surface of a sphere is denoted as a 3D scan or a 3D case. In contrast to this 3D case, a 1D case will also be investigated. The 1D case describes a situation where the samples are located along a great circle of the sphere surrounding the sources. An intermediate case occurs in the context of OATS measurements. Directivity models for unintentional radiators are discussed in Section III. B. Polarization Mismatch The polarization mismatch factor μ ∈ [0, 1] is the fraction of the measured power to the emitted power in a certain direction μ(θ, φ) = Pmeasured(θ, φ) Pradiated(θ, φ) . (3) Linearly polarized antennas are used on OATS or in FAR to pick up equipment under test (EUT) radiation. Measurements are done in horizontal and vertical polarizations. For uninten- tional emitters, having emissions with uniformly distributed po- larization angles, none of the orientations of the receiving an- tenna will completely pick up all radiated power. Thus, also the maximum of both measurements is an uncertain quantity (a random variable, RV). The influence of this additional source of uncertainty has been neglected until now. It will be addressed in Section II-A. 0018-9375/$26.00 © 2011 IEEE Mathias Magdowski Conversion factors for RVCs 2023-06-29 25 / 31
- 37. Introduction Validation Emission measurement Conversion of results Conversion SAC to RC w. r. t. f for a = 0.75 m and r = 10 m 107 108 109 1010 0 2 4 6 8 10 12 14 Frequency, f in Hz Conversion factor E 2 max P rad in V 2 W m 2 95. percentile average 50. percentile 5. percentile https://octave-online.net/bucket~Lk6YAigoBeuKfmDvu2sWDr Mathias Magdowski Conversion factors for RVCs 2023-06-29 26 / 31
- 38. Introduction Validation Emission measurement Conversion of results Conversion FAR to RC w. r. t. f for a = 0.75 m and r = 3 m 107 108 109 1010 0 5 10 15 20 25 Frequency, f in Hz Conversion factor E 2 max P rad in V 2 W m 2 95. percentile average 50. percentile 5. percentile https://octave-online.net/bucket~MovSav14VA5nnoHKoo7s3A Mathias Magdowski Conversion factors for RVCs 2023-06-29 27 / 31
- 39. Introduction Validation Emission measurement Conversion of results Conversion between FAR and RC with respect to ka for r = 10 m 10−1 100 101 102 103 0 0.5 1 1.5 2 2.5 3 Product of wavenumber and EUT size, ka Conversion factor E 2 max P rad in V 2 W m 2 95. percentile average 50. percentile 5. percentile https://octave-online.net/bucket~NWBmAdyjm8aqLKaBRMNbzS Mathias Magdowski Conversion factors for RVCs 2023-06-29 28 / 31
- 40. Introduction Validation Emission measurement Conversion of results Which type of measurement would you like to have? Mathias Magdowski Conversion factors for RVCs 2023-06-29 29 / 31
- 41. Introduction Validation Emission measurement Conversion of results Which type of measurement would you like to have? Environment: deterministic Mathias Magdowski Conversion factors for RVCs 2023-06-29 29 / 31
- 42. Introduction Validation Emission measurement Conversion of results Which type of measurement would you like to have? Environment: deterministic EUT: random Mathias Magdowski Conversion factors for RVCs 2023-06-29 29 / 31
- 43. Introduction Validation Emission measurement Conversion of results Which type of measurement would you like to have? Environment: deterministic EUT: random Mathias Magdowski Conversion factors for RVCs 2023-06-29 29 / 31
- 44. Introduction Validation Emission measurement Conversion of results Which type of measurement would you like to have? Environment: deterministic EUT: random Environment: random Mathias Magdowski Conversion factors for RVCs 2023-06-29 29 / 31
- 45. Introduction Validation Emission measurement Conversion of results Which type of measurement would you like to have? Environment: deterministic EUT: random Environment: random EUT: deterministic Mathias Magdowski Conversion factors for RVCs 2023-06-29 29 / 31
- 46. Introduction Validation Emission measurement Conversion of results Back to a more fundamental question What is a good measurand for emission? ▶ field strength in V m (in a certain distance) ▶ power flux density in W m2 (in a certain distance) ▶ total radiated power W (independent of the distance) Mathias Magdowski Conversion factors for RVCs 2023-06-29 30 / 31
- 47. Introduction Validation Emission measurement Conversion of results Back to a more fundamental question What is a good measurand for emission? ▶ field strength in V m (in a certain distance) ▶ power flux density in W m2 (in a certain distance) ▶ total radiated power W (independent of the distance) In which environment is the measurement performed? ▶ reflection-free environment ▶ environment with reflections ▶ highly reflective environment Mathias Magdowski Conversion factors for RVCs 2023-06-29 30 / 31
- 48. Introduction Validation Emission measurement Conversion of results Thank you very much for your attention! Are there any questions? Mathias Magdowski Conversion factors for RVCs 2023-06-29 31 / 31