G.1 Verification method for cal bration of mono ole anten as by the plane wave method from 5 MHz to 3 MHz.. 14 G.1.2 Un ertainty evaluation for the cal bration of monop le anten as by th
General
This standard outlines the calibration methods for antennas utilized in radiated disturbance measurements, focusing on the key calibration parameter known as antenna factor, \( F_a \), which can be converted to realized gain as detailed in Annex C.
It is recognized that when used in accordance with the test methods defined in CISPR 1 6-2-
The measurement antenna may be positioned above a metallic ground plane at heights ranging from 1 m to 4 m, rather than in free space Annex B outlines methods to measure the antenna factor as a function of height and polarization, denoted as \$F_a(h, p)\$ These methods help quantify the differences between \$F_a\$ and \$F_a(h, p)\$, contributing to the uncertainty in radiated disturbance measurements conducted above a metallic ground plane For further details on the necessity of a ground plane and the selection of calibration methods based on available test sites, state-of-the-art knowledge, and required uncertainty levels, refer to section A.1.
Each antenna calibration method is linked to a validation process for the calibration site, along with acceptance criteria that pertain to the required uncertainty of the antenna factor (AF) The pertinent site specifications and validation methods are outlined in CISPR 16-1-5.
The concept of antenna factor
The antenna factor F a, in dB(m −1 ), as defined in 3.1 2.1 , is determined by Equation (1 ):
E is the field strength in dB(μV/m) of the incident plane wave that illuminates the antenna;
V is the resultant voltage in dB(μV) across the output terminal of the antenna.
The antenna factor (AF) is influenced by the load impedance \( Z_0 \) connected to the antenna output and the impedance \( Z_0' \) observed from the gap between the radiating elements and the load Additionally, mutual coupling effects from nearby antennas, the ground, and buildings can impact the AF; however, these effects are minimized when defining AF in a free-space environment The voltage \( V \) is typically measured at the receiver end of the cable linking the antenna to the receiver, necessitating a correction for cable loss, as outlined in Equation (48) of section 7.4.3.1 Furthermore, a mismatch loss correction is essential for cables with poor matching, as discussed in section 6.2.2.
This standard stipulates that radiated disturbance measurements should utilize F a, as explained in the rationale of section A.4 In the context of radiated emission measurements, the incident field strength, E, can be determined from the voltage reading, V, obtained from a measuring receiver linked to the antenna.
The equation \$E = V + F_a\$ describes the relationship between electric field strength (E in dB(àV/m)), voltage (V in dB(àV)), and the influence factor (F_a in dB(m^{-1})) Any perturbations to F_a within the calibration environment are considered sources of uncertainty, as detailed in section 7.3 of CISPR 16-2-3:2010.
Measurements of radiated disturbance from an EUT are made at a fixed distance, e.g 3 m Where the distance from the EUT to the demarcated geometrical centre of an LPDA antenna
CISPR 16-1-6:2014 © IEC 2014 indicates that using the specified standards may lead to inaccuracies in electric field strength at frequencies that do not align with the center of the LPDA antenna Corrections for the electric field strength can be made by addressing the discrepancy between the actual phase center and the geometrical center, as outlined in A.6.2 and CISPR 16-2-3 [2] Alternatively, this correction can be integrated into the antenna factor.
The use of horizontally-polarized dipole, biconical, or hybrid antennas for measuring EUT radiated disturbances above a metallic ground plane introduces uncertainties due to mutual coupling with the antenna's image in the ground plane This interaction alters the impedance of the radiating elements at the balanced terminals of the balun, resulting in a height-dependent AF, denoted as F a(h, p) To assess the uncertainties arising from the variation of F a(h, p) compared to F a, it is permissible to derive this difference from generic measurements or simulations for each antenna model Antenna manufacturers may provide F a(h, p) data for heights ranging from 1 m to 4 m.
Calibration methods for 30 MHz and above
General
This subclause outlines essential considerations for antenna calibration methods specified in this standard The free-space antenna factor (AF), denoted as F_a, and the height-dependent AF, F_a(h, p), can be assessed using either the three antenna method (TAM) or the standard antenna method (SAM) Additionally, F_a can be evaluated through the standard site method (SSM) with an appropriate correction.
According to the reciprocity theorem, it is unnecessary to designate a specific antenna for transmission or reception To measure any Antenna Under Calibration (AUC), a second antenna, known as the "paired" antenna, is required, allowing the calibration laboratory to choose the most suitable connection to the signal source or receiver However, antennas equipped with preamplifiers must exclusively connect to the receiver All calibration techniques rely on the SIL measurement procedure outlined in section 7.2.
Antenna minimum separation distances
For accurate calibrations using the TAM, the separation of the antenna pair must be precisely known for calculating F a During dipole and LPDA antenna calibration, a fixed separation of 2λ at the lowest frequency of interest should be maintained between the phase centers In contrast, for biconical and hybrid antennas, a fixed separation of 10 m is sufficient, as the longest dipole elements act as short dipoles below 60 MHz For additional details, refer to section C.5.
At 30 MHz, a distance of 10 m corresponds to one wavelength (\$λ\$), while a biconical antenna measures only 0.14 \$λ\$ The mutual coupling between antennas is minimal Calibration at 10 m introduces an additional uncertainty of 0.1 dB in the antenna factor (AF), which decreases to 0.03 dB at 60 MHz.
General considerations for the TAM
The TAM is an antenna calibration method that requires no prior knowledge of the antenna factors (AF) for any of the three antennas used This technique utilizes three similar antennas, referred to as "paired antennas," which operate within a shared frequency range From these three antennas, three distinct pairs can be created, and the System Identification Level (SIL) for each pair is measured For more detailed information on the TAM, refer to sections 7.4.1.1, 7.4.1.2, 8.2, 9.2.4, 9.4.2, B.4.3, and B.5.3.
General considerations for the SSM
The SSM serves as the TAM when conducted over a ground plane, as outlined in sections 7.4.2 and 8.4 When the signal reflects off the ground plane, it experiences a 180° phase shift for horizontal polarization To achieve a signal maximum, the antenna heights and/or separation distances are carefully adjusted to ensure optimal combination of the direct and reflected signals.
Horizontal polarization is favored due to its uniform H-plane pattern, which simplifies the calculation of the antenna factor (AF) and reduces reflections from masts and cables To ensure accurate measurement of AF in relation to F_a, one antenna is fixed at a height of 2 m in the far field of the paired antenna, which is adjusted in height to avoid signal nulls The recommended separation distance for the SSM is 10 m, but for tuned dipoles at 30 MHz, a 20 m separation is preferable to maintain far-field conditions If a 10 m distance is used, the associated uncertainty must be considered Additionally, the accuracy of F_a can be enhanced by applying a correction factor, as indicated in Equation (59).
General considerations for the SAM
For the Standard Antenna Measurement (SAM), a set of standard antennas (STAs) with precisely determined antenna factors is essential These STAs can be calibrated using either the Transfer Antenna Method (TAM) or by substituting with a calculable antenna Notably, the SAM requires only two measurements of Signal Input Level (SIL) for calibrating a single antenna, while the TAM necessitates three SIL measurements.
The SAM, also known as the reference antenna method, utilizes a reference antenna with a geometry, structure, and antenna factor (AF) that are precisely defined by standards like ANSI C63.5 and CISPR 16-1-5 This method is synonymous with the SAM, highlighting its importance in antenna measurement practices.
The SAM demonstrates greater tolerance to site non-uniformities compared to the TAM, allowing for a more relaxed site quality requirement This assumes that both the STA and AUC are of the same type, such as the classical biconical antenna, leading to similar mechanical dimensions and radiation patterns Ideally, the STA should match the AUC model, though calibration laboratories may not have multiple STAs available For guidance on antenna similarity in SAM calibrations, refer to section 8.3.3 A calculable dipole antenna serves as an ideal STA for calibrating dipole antennas, provided the field incident on the antennas is uniform, as outlined in section 4.9 of CISPR 16-1-5:2014.
Measurement uncertainties for antenna calibration measurement results
Measurement uncertainties must be estimated for each set of antenna calibration results, following the methods outlined in ISO/IEC Guide 98-3:2008 These uncertainties depend on the calibration measurement setups, instrumentation, and the overall characteristics of the Antenna Under Calibration (AUC) Calibration laboratories are responsible for quantifying their own measurement uncertainties, as detailed in section A.9.2 regarding typical measurement uncertainty ranges.
This standard provides example measurement uncertainty budgets for each calibration method, as detailed in Clause 8 and Annex B The column labeled "Value" represents the best estimated value of \(X_i\), while "Probability distribution" indicates the corresponding probability distribution function.
“Sensitivity” is the sensitivity coefficient c i , and u i is the uncertainty contribution u i (y) Numerical values are included for illustration purposes only The symbol u i signifies the
Laboratories should apply "standard uncertainty" to each term in their measurement uncertainty budgets, tailoring them to their specific needs, facilities, and instrumentation The provided examples encompass common influence quantities found at antenna calibration sites, though some laboratories may need to include additional factors not listed In such cases, these extra influence quantities should be incorporated into the model and the measurement uncertainty budget For further details on antenna characteristics relevant to uncertainty analysis, refer to Annex C.
The "law of propagation of uncertainty" is only an approximation when the dominant input quantity, such as site imperfection, exhibits a non-linear distribution Accurately predicting the potential error is challenging In cases of uncertainty, the ISO/IEC Guide 98-3:2008/SUP (GUM) recommends using a Monte Carlo method for the propagation of probability distributions.
Uncertainty budgets may frequently lead to an overestimation of expanded uncertainty due to the inclusion of numerous estimated component uncertainties To verify the accuracy of a budget, it is beneficial to compare the AF measured through at least two independent methods or conduct an international intercomparison A smaller difference in results indicates greater reliability.
The greater the number of methods used, the higher the confidence in the results If the difference in the AFs between method A and method B is less than or equal to the individual combined standard uncertainty for either method, it suggests that some larger components in the budgets may have been overestimated and require further investigation.
Summary of methods of measurement to obtain AF
Various antenna types are utilized for radiated disturbance measurements, calibrated using either the TAM or SAM for SIL measurements, or the SSM for SA measurements Calibration methods for frequencies above 30 MHz are outlined in Table 1, which includes subclause numbers for detailed procedures Specific measurement procedures for each calibration method are detailed in Clause 8 and Annex B, including the equations for determining the Antenna Factor (AF) relevant to each method.
Table 1 serves as a comprehensive lookup resource detailing antenna types and their corresponding frequency ranges, along with site specifications such as ground plane or absorber types, antenna heights, separation distances, and polarization methods For additional context on the various methods outlined in this standard, refer to section A.1.
The article discusses methods for measuring the height-dependent antenna factor, \( F_a(h, p) \), as outlined in section B.4 and referenced in section 4.2 These methods are relevant for horizontally-polarized tuned dipole antennas, biconical antennas, and the broadband dipole components of hybrid antennas.
Table 2 presents a streamlined approach to selecting calibration methods for different classes of antennas, functioning like a flowchart It highlights a primary method that is both straightforward and time-efficient; however, alternative methods are provided for laboratories lacking the necessary facilities For instance, if a laboratory does not have a Far-Field Antenna Range (FAR), horn antennas can be calibrated using the outdoor method outlined in section 9.4.2 Additionally, the column labeled "Method not requiring an STA" features methods with the lowest uncertainties that do not depend on having a Standard Test Antenna (STA) and utilize horizontal polarization, although these methods may require more time compared to others.
Table 1 – Summary of calibration methods above 30 MHz for F a
Calibration site Antenna under calibration Calibration method
MHz Antenna set-upa Pol.b Subclause
SAC, utilizing the ground plane
TAM 30 to 1 000 d = 1 0 m,c h 1 , h 2 depend on f HP B.5.3 SAM 30 to 1 000 d = 1 0 m, h 1 , h 2 depend on f HP B.5.2 SAM with averaging 30 to 300 d = 1 0 m, h 1 , h 2 depend on f HP B.4.2 SSM 30 to 1 000 d = 1 0 m, h 2 = 2 m, h 1 = 1 m to 4 m
Biconical, (also biconical part of hybrid by 9 3)
SAM or TAM with averaging 30 to 300 d = 1 0 m, h 1 , h 2 depend on f HP B.4.2 SAM
FAR or minimizing ground reflections by height or absorber h 1 , h 2 , h 3 apply above ground
Tuned dipole SAM 60 to 1 000 d min dependent on frequency, use FAR HP
Biconical, (also biconical part of hybrid) SAM 30 to 300 d = 4 m minimum HP
(LPDA part of hybrid) Horn
LPDA (LPDA part of hybrid) using absorber
NOTE 1 The AUC is at height h 1
NOTE 2 See A.4.2 regarding the optimum crossover frequency between biconical and LPDA antennas
NOTE 3 If a set of standard antennas is available, the SAM may be preferable to SSM or TAM
NOTE 4 In a FAR, the result is expected to be the same whether antennas are oriented HP or VP; see also A.2.5
CISPR 1 6-1 -6:201 4 © IEC 201 4 – 25 – a d is the separation distance between the transmit and receive antennas h 1 is the height of an AUC h 2 and h 3 are the heights of other AUCs or paired antennas
For SSM normally the AUC shall be the height scanned antenna; by request of a calibration laboratory’s client the AUC can be the fixed height antenna [i.e see A.5 a)] b Pol is polarization
HP = All antennas positioned for horizontal polarization
All antennas are configured for vertical polarization (VP) To maintain a separation of at least 2 wavelengths (λ) and minimize additional uncertainty to approximately 0.25 dB, a distance of 20 m is recommended below 60 MHz An antenna height of 0 m indicates a monocone design with its feed at ground level Additionally, hybrid antennas can be calibrated using methods applicable to both biconical and log-periodic dipole array (LPDA) antennas The height of the antenna may be adjusted to meet site acceptance criteria, especially when lower uncertainties for the antenna factor (F_a) are necessary.
Table 2 – Calibration methods above 30 MHz by subclause number
Antenna type, frequency, MHz Suggested method Alternative method, frequency, MHz Method not requiring an STA Biconical, 30 to 300
9.1 SAM FAR B.4.2 HP SAM height- averaging
9.4.2 TAM high above ground 9.4.3 SAM high above ground 9.4.4 TAM or SAM in FAR or with absorber on ground 8.4 SSM CALTS,
9.5.1 3 TAM in FAR 9.5.2 SAM in FAR
Tuned dipole B.5.2 SAM at “free- space” height using calculable dipole
B.5.3 TAM at “free- space” height a The upper frequency depends on the manufacturer’s specification
5 Calibration methods for the frequency range 9 kHz to 30 MHz
Calibration of monopole antennas
Monopole antennas, sometimes called rod antennas, are used typically in the frequency range
Calibration frequencies ranging from 9 kHz to 30 MHz are outlined, with recommended increments specified in Table 3 Due to the long wavelengths at frequencies below 30 MHz, traditional calibration methods for higher frequency antennas are not suitable for monopole antennas A standard uncertainty of less than 1 dB can be attained using the plane wave method and the Equivalent Capacitance Substitution Method (ECSM), as detailed in the subsequent subclauses and Annex G.
Table 3 – Frequency increments for monopole antenna calibration
The plane wave method involves illuminating the entire antenna with a plane wave on a large ground plane, as detailed in section G.1 Typically, the matching unit is positioned above the ground plane, causing RF currents to flow within the housing Notably, an increase in housing height significantly impacts the antenna factor (AF).
The term "matching unit" refers to the metal housing that connects the monopole radiating element, such as a metal rod, to the input of a measuring receiver This housing may include a matching circuit and amplifier, and it is crucial for the metal base to maintain good electrical contact with the ground plane to ensure effective and consistent performance of the monopole antenna Monopole antenna models equipped with rubber feet typically feature a slightly taller metal foot (spacer) to facilitate this electrical connection In cases where a metal foot is absent, a short metal braid strip, at least 15 mm wide, should be used to establish electrical contact with the ground plane and the bottom of one vertical side of the matching unit, with screws employed as necessary to secure a reliable RF connection.
The ECSM method replaces the rod with a capacitor equivalent to the monopole's self-capacitance, as detailed in Annex G Each monopole antenna model necessitates the design of a dummy antenna, which must be verified during the design phase by comparing the antenna factors (AFs) obtained from the ECSM with those from the plane wave method This comparison allows for design improvements and can reduce potential uncertainty contributions by up to 4 dB Additionally, the antenna factor obtained from the ECSM can be validated against the factor derived from the plane wave method.
To verify the ECSM, one effective method involves comparing the field strength measured by an ECSM-calibrated monopole antenna on a large ground plane in the far field of a strong AM transmitter with that of a precisely calibrated loop antenna This approach, however, is limited to verifying AFs at specific spot frequencies At each spot frequency, it is essential to ensure that both the monopole and loop antennas receive the dominant direct wave from a single direction, while any other signals must be weaker by more than 30 dB.
NOTE 3 The ECSM can work well for antennas whose total length from the base of the matching unit to the top of the rod is shorter than λ /8 (i.e see G.2.1 )
A monopole antenna is often mounted on a counterpoise, such as a vestigial ground plane measuring 0.6 m by 0.6 m, on a tripod, which can result in an antenna factor (AF) that is several dB lower than when measured with the matching unit on the ground plane or by the ECSM For outdoor monopole antennas on tripods, accurate calibration can be achieved using the plane wave method Other configurations may involve connecting the vestigial ground plane to a conducting bench earthed to a wall in a shielded room, or placing the antenna on the bench, which necessitates more complex calibration methods While specific calibration methods for these configurations are not provided, following the manufacturer's instructions regarding the use of a counterpoise or ground plane, including bonding the antenna matching unit to the ground plane, can enhance the reproducibility of results.
The output from the matching unit is measured using the test configuration shown in Figure 1 or Figure 2 The antenna factor F ac in dB(m −1 ) (see 3.1 2.3) is given by Equation (2)
V D is the measured output of the signal generator, in dB(àV);
V L is the measured output of the matching unit, in dB(àV);
L h is the height correction factor (for the effective height), in dB(m)
The monopole antenna, typically utilized in EMC measurements, features a 1 m long rod with an effective height of 0.5 m It has a height correction factor of -6 dB(m) and a self-capacitance of 12 pF, based on a rod radius of 3.6 mm.
Formulas for calculating the effective height, height correction factor, and self-capacitance of monopole antennas with varying lengths and radii are provided in section 5.1.2.2 For instance, a commercial monopole antenna featuring a rod with a relatively large diameter exhibits a self-capacitance of 16 pF.
Two methods can be employed for measurements: using a network analyzer (5.1.2.3.2) or a measuring receiver with a signal generator (5.1.2.3.3), both utilizing the same dummy antenna For instructions on constructing a dummy antenna, refer to section 5.1.2.4 It is essential to conduct measurements at multiple frequencies, as outlined in Table 3, to create a smooth curve of antenna factor versus frequency across the antenna's operating range, which spans from 9 kHz.
The effective height, self-capacitance, and height correction factor of monopole antennas can be determined using specific equations These equations are applicable solely to cylindrical rods that are shorter than λ/8 For accurate self-capacitance values in ECSM calculations, the frequency-dependent ratio in Equation (4) can be disregarded, as it approaches unity under the λ/8 condition The relationship is expressed as \( h_e = \frac{2h}{\tan(\frac{\pi h}{\lambda})} \).
L h = (5) where h e is the effective height of the rod, in m; h is the actual height (i.e length) of the rod, in m; λ is the wavelength, in m;
C a is the self-capacitance of the rod, in pF; a is the radius of the rod, in m;
L h is the height correction factor, in dB(m)
Figure G.3 illustrates the graphical representation of Equation (4) for different rod diameters, while Figure G.4 presents Equation (5) with two antenna lengths The ECSM does not consider the matching unit's contribution to F ac; therefore, an empirical correction involves adding half the housing height to the monopole length for the height correction factor, as indicated in Equation (6) Typically, this results in a reduction of F ac by 0.8 dB For calibrating a monopole antenna in accordance with CISPR 25, where the matching unit is positioned beneath the ground plane, Equation (5) is applicable.
Equation (4) aligns with the specifications outlined in CISPR 16-1-4:2010-12 Corrigendum 1, which replaces Equation (B.2) from CISPR 16-1-4:2010-04 Future updates will eliminate Annex B of CISPR 16-1-4:2010-04 and will include a relevant cross-reference to section 5.1.2 of CISPR 16-1-6.
L h (6) where h b is the height of the matching unit, in m
Other details concerning Equation (3) are available in [28], [32], and [59], and for Equation (4) in [29], [39], [58], [59], and [68]
References [1, 3, and 36] present an alternative expression for C, where the term \(\ln(h/a) - 1\) in the denominator is substituted with \(\ln(2h/a) - 1\) However, both theoretical and experimental analyses [40] indicate that Equation (4) is the more suitable choice.
Two alternative calibration procedures are described: use of a network analyzer in 5.1 2.3.2 and use of a signal generator and receiver in 5.1 2.3.3
For optimal calibration, a 50 Ω termination is recommended to have a return loss exceeding 32 dB (VSWR < 1.05:1) to ensure minimal uncertainty contribution Additionally, the measuring receiver should be calibrated with a return loss greater than 20.9 dB (VSWR < 1.2:1) It is also essential that the output of the signal generator remains stable in both frequency and amplitude.
The dummy antenna specified in section 5.1.2.4 should be connected as close as possible to the antenna port on the matching unit, with the T-connector positioned near the dummy antenna It is essential to ensure that the outer conductor of the T-connector is electrically connected to the matching unit, potentially using a short flat wire braid Additionally, the matching unit must be grounded through the outer conductor of the coaxial cable to the measuring receiver If the return losses of both the receiver and signal generator are sufficiently high, additional pads may not be necessary at the input port of the receiver and the output port of the generator.
Calibration of loop antennas
General
Loop antennas used for EMC radiated disturbance measurements over the frequency range
Loop antennas with diameters ranging from 0.6 m are commonly used for frequencies between 9 kHz and 30 MHz These antennas are typically housed in a compact box that includes matching and tuning networks, and some models may also feature an amplifier To accurately measure the RF magnetic field strength with a loop antenna, it is essential to know its magnetic field antenna factor.
Various techniques for calibrating loop antennas and measuring their magnetic field antenna factor have been developed Reference [1 8] provides a comprehensive overview, while references [1 5] and [1 6] offer simplified versions of the standard field method The three-antenna method is detailed in reference [35] This article discusses two straightforward methods: the TEM cell method, which offers wide frequency coverage with a minimum uncertainty of approximately ± 0.5 dB, and the Helmholtz coil method, which achieves an accuracy of 0.7% (0.06 dB) up to 150 kHz and better than ± 0.5 dB up to 10 MHz The Helmholtz coil method serves as a valuable check for the TEM cell method When the TEM cell is used below its first resonant frequency and the loop is well shielded, validation of results up to 150 kHz using the Helmholtz coil method can confirm the calibration in the TEM cell up to 30 MHz.
A properly designed loop antenna features symmetry within its plane and is effectively shielded from electric fields Inadequate shielding can lead to inconsistencies in the antenna factor during the calibration of loop antennas.
TEM (Crawford) cell method
TEM cells are fully shielded, and intended not to emit energy that may be hazardous to personnel or cause interference with nearby electronic equipment The basic TEM cell is a
CISPR 1 6-1 -6:201 4 © IEC 201 4 – 33 – section of two-conductor transmission line that is operating in the transverse electromagnetic (TEM) mode, hence the name [34]
In a TEM cell, the field strength at the centre point between the centre conductor (septum) and outer conductor can be calculated from: b
E is the electric field strength, in V/m;
H is the magnetic field strength, in A/m;
V is the voltage at the input or output port of the TEM cell, in V;
Z 0 is the real part of the characteristic impedance of the TEM cell, in Ω;
P net is the net power at the input of the TEM cell, in W; b is the distance from the upper wall to the centre plate (septum), in m
The field strength equations are applicable solely at the center point of a well-matched TEM cell, with notable variations occurring near the septum Nevertheless, the average field strength across the loop area closely approximates the field strength at the center.
Increasing the frequency can lead to the generation of higher order modes Calibration should be conducted prior to the first cell resonance, identifiable by a sudden change in the field detected by an internal probe within the TEM cell When the cell is loaded, the resonant frequency may shift, so it is recommended to operate the cell significantly below this frequency, which may be lower than the maximum frequency indicated by the manufacturer.
The system's characteristic impedance is assumed to be 50 Ω; however, the impedance at the loop's position, when loaded, may vary by approximately 2% To minimize the uncertainty associated with \( Z_0 \) at the center of the loaded cell, it is advisable to measure the impedance at this location and incorporate that value into Equation (11) A time domain reflectometer can be utilized for this impedance measurement.
Polystyrene foam blocks are utilized to centrally support the loop between the plates, effectively insulating it from the TEM cell conductor surface To achieve a sufficient output signal from a passive loop, an amplifier is necessary to enhance the field strength within the TEM cell Calibration involves switching the direct line to the measuring receiver via attenuators before connecting the loop output In contrast, an active loop, which includes an amplifier module in a tripod support, does not require an amplifier This loop is lowered through a hole in the upper plate of the cell, allowing the tripod to remain outside, while ensuring that the metallic part does not contact the outer conductor Accessing the connections from the top is more convenient, although some loop antenna models can be placed entirely inside the cell The routing of the cable through the top or side of the TEM cell has minimal impact, provided a high-quality screened cable is used.
Only loops shielded from the electrical component of a TEM wave can be calibrated using a TEM cell If the loop is unshielded and the two halves are unbalanced, it will respond to the electric component of the electromagnetic field To test the effectiveness of the shielding, the loop can be rotated 180° about its vertical diameter, resulting in a 180° phase change between the electric and magnetic field pickups; any output signal change must be included in the measurement uncertainty budget, representing the "electric field rejection" component in Table 5.
In a TEM cell with a plate separation of 0.915 m, loops with a diameter of up to approximately 0.63 m can be effectively calibrated, demonstrating efficient use of the spacing between the plates.
Figure 4 – Block diagram of TEM cell set-up for passive loop antennas
Figure 5 – Block diagram of TEM cell set-up for active loop antennas
An example measurement uncertainty budget for the magnetic field antenna factor at frequencies exceeding 9 kHz is presented in Table 5 The uncertainty in measuring a loop is primarily influenced by the ratio of the septum height to the loop size, with measurement uncertainties being less than a specified threshold.
1 dB are achievable for small loops in large cells, but can become worse for loops with size greater than two-thirds of the septum height
Table 5 –Example measurement uncertainty budget for F aH of a loop antenna measured in a TEM cell
Source of uncertainty or quantity X i Value dB Probability distribution Divisor Sensitivity u i dB Note
Expanded uncertainty , U ( k = 2) 0,84 a For receiver sensitivity, refer to 6.2 4 and A.8 1 If the S/N ratio is above 1 7 dB (average detection) the influence is below 0,1 dB
6 Frequencies, equipment and functional checks for calibrations at or above
Calibration frequencies
6.1 1 Calibration frequency ranges and increments
Broadband antennas operating at or above 30 MHz must have their F a measurements conducted using swept frequency methods or equivalent maximum frequency increments as specified in Table 6 Additionally, discrete frequencies for the calibration of tuned dipoles are outlined in section B.3.
Table 6 – Frequency increments for broadband antenna calibration
When conducting measurements at an outdoor location, certain frequencies can lead to considerable interference from ambient signals To mitigate this, identify a frequency, denoted as \$f_2\$, where the ambient signal is at least 30 dB lower than the received measurement signal at a specific frequency \$f_1\$ Measurements at \$f_2\$ should be taken within the frequency range of \$f_1 \pm \Delta f\$, where \$\Delta f\$ represents the designated bandwidth.
In the frequency range of 30 MHz to 150 MHz, a tolerance of 1 MHz is acceptable, while a tolerance of 3 MHz applies from 151 MHz to 300 MHz, and 5 MHz from 301 MHz to 1,000 MHz Any deviation from the specified frequency, denoted as f2, must be documented in the calibration report For frequencies above 1 GHz, ambient signals can be mitigated by conducting measurements in a shielded anechoic chamber.
Some antennas display resonances as spikes in the antenna factor versus frequency plot, as illustrated in Figure 6 A calibration increment of 2 MHz can indicate a sharp resonance's presence, but it may not accurately capture its peak To ensure precision, calibration should be conducted with a smaller frequency increment, or the calibration certificate must include a statement indicating that uncertainty may be higher in the frequency range of \$0.985 f_{res} < f < 1.015 f_{res}\$, where \$f_{res}\$ represents the resonant frequency.
Certain models of LPDA or hybrid antennas may experience resonances caused by the deterioration of RF contact in some dipole elements, particularly if these elements are not welded or protected from oxidation.
F a by between 2 dB and 5 dB This occurs particularly for LPDA antennas that have also been used for transmitting relatively high power levels for immunity testing
Figure 6 – Example of resonant spike due to poor biconical element connections, using 2 MHz increment
6.1 2 Transition frequency for hybrid antennas
The hybrid antenna design integrates a conventional biconical antenna with a log-periodic dipole array (LPDA) to achieve a frequency range of 30 MHz to 1 GHz or higher For calibration, a hybrid antenna can be paired with two hybrid antennas, but if lower uncertainty in the factor \( F_a \) is needed, a two-stage calibration process using the tapered antenna method (TAM) is recommended This involves initial calibration against two biconical antennas, followed by calibration against two LPDA antennas, with data from both stages combined at a suitable transition frequency Alternatively, the first stage can utilize a single biconical standard antenna (STA) up to 240 MHz, as the larger wavelength allows for non-identical configurations, while the TAM should be used for non-identical LPDAs.
Frequency (MHz) An te nn a fa ct or (d Bm -1 )
When utilizing the SSM, it is important to note that for frequencies exceeding 1 GHz, alternative calibration procedures outlined in Clause 8 or Clause 9 must be followed Additionally, refer to section 5.3.2 of CISPR 16-1-5:2014 for guidance on validating a FAR for this calibration method.
The optimum transition frequency is determined by analyzing the plots of F a against frequency and identifying the frequency band where the two plots closely align Research on various hybrid antenna models indicates that this alignment typically occurs between 140 MHz and 240 MHz, with a common frequency around 180 MHz At lower frequencies, mutual coupling with the ground plane can cause discrepancies, while at higher frequencies, antenna directivity may lead to variations in the results.
Measurement instrumentation requirements for antenna calibrations
Equipment types
All instrumentation must maintain a nominal impedance of 50 Ω The ideal tool for antenna calibrations is a network analyzer, which includes a swept-frequency signal source and a tracking receiver Other options include a spectrum analyzer with a tracking generator or a computer-controlled setup that combines a signal generator with a receiver.
“measuring receiver” (i.e 3.1 4.1 ) is used for the receiving part of these suitable instrument types
To establish the sweep time, follow a procedure that ensures adequate dwell time for accurate signal measurement at the receiver, particularly for long cable runs Set up antennas on a CALTS at fixed heights of 2 m with a separation of 10 m for calibration Configure the measuring receiver for antenna calibration, adjusting parameters such as sweep time, RF bandwidth, number of averages, and frequency points, which can be optimized for a quicker sweep time in a FAR Gradually increase the sweep time to T min until the response stabilizes, allowing T min or longer to be used for calibrations.
At frequencies where a signal null occurs, response comparisons may be unreliable It is advisable to only observe response differences when the signal-to-receiver noise ratio exceeds 17 dB for average detectors, or 34 dB for vector network analyzers (VNAs) For additional considerations, refer to section A.8.1.
NOTE 1 Decreasing the RF bandwidth increases the dynamic range and reduces the influence of ambient signals, but also may necessitate a longer dwell time
NOTE 2 When measuring SA (i.e 8.4) it is assumed that the speed of the motorized mast allows the sweep of the frequency range to complete before the mast has moved more than 2 cm for frequencies above 300 MHz, and 5 cm for frequencies below 300 MHz If the mast speed is too fast, errors can occur in the calculated AF caused by the height at which the signal maximum is actually recorded differing from the height at which it is calculated in e 0( i , j )|max in Equation (C.27) (see C.3.3) In a height scan where there are multiple maxima lobes, particularly at the higher frequencies, the calculated and measured maxima can be from different lobes and the error in AF can be of the order of 0,5 dB, showing as an uncharacteristic step in a plot of AF against frequency
Calibrating an antenna using the TAM or SSM requires two paired antennas that cover the frequency range, which can be either other AUCs or antennas from the calibration laboratory When dealing with three unknown antennas, it is advisable to conduct a suitability check to ensure accurate measurements and stable performance This can be achieved by temporarily substituting one antenna with a Standard Test Antenna (STA) at the beginning of the calibration session To ensure traceable results, it is essential to include at least one of the three antennas with a known Antenna Factor (AF) based on a history of multiple calibrations.
Paired antennas play distinct roles in measurements conducted by the TAM and SAM systems For the TAM, the antennas must be identical, as specified in section 3.1.1.1.2 In contrast, the SAM requires paired antennas to create a uniform field across the AUC and STA apertures When utilizing a ground plane, the radiation patterns of both the AUC and STA must closely resemble each other Additionally, section 3.1.4.3 introduces the concept of SIL, which categorizes antennas as either receive or transmit, offering an alternative to the AUC and paired antenna classification This "receive and transmit" designation simplifies operations by eliminating one degree of freedom, as the measurement results remain consistent regardless of which antenna is connected to the receiver or signal generator.
At appropriate intervals, for traceability the linearity of the instrumentation is verified using calibrated attenuators Measurements are performed with calibrated attenuators, e.g 1 0 dB,
30 dB, or 50 dB, and mismatch standards The results are compared to the values from the calibration certificate by the E n criterion [8]
To accurately measure antenna separation and height, a traceable distance-measuring instrument, such as a tape measure or laser meter, is essential The RF measurement is more sensitive to height above a ground plane than to separation distances For the TAM, both absolute heights and separation distances are critical, while for the SAM, the positioning of the AUC relative to the STA is paramount, with position tolerance specified in section 8.3.2 The height should be maintained within a tolerance of ±10 mm, as greater deviations can increase measurement uncertainties These tolerances and uncertainties can be assessed through a sensitivity test of SIL magnitude by varying the height.
The precision required for measuring separation is directly related to the absolute distance; smaller separations necessitate tighter tolerances For instance, a 10 m separation with a 10 mm positional error for both antennas results in a field strength uncertainty of 0.017 dB Conversely, a 0.1 dB uncertainty corresponds to a 114 mm separation error This separation error becomes even more critical when considering ground plane reflections, as seen in the SSM, due to the involvement of signal phases Height scanning can help mitigate distance errors by seeking in-phase conditions, with a preferred separation error of less than ± λ/30 At the upper frequency of 1 GHz for SSM, this translates to a λ/30 of 10 mm.
To minimize signal generator and receiver drift, as well as the impact of temperature changes, it is essential to keep the time interval between AUC and STA SIL measurements as short as possible Additionally, the transmission loss of calibration cables can fluctuate with temperature, so it is advisable to use cables with white skins or a white sheath when they are exposed to direct sunlight This helps mitigate the effects of temperature variations caused by changing cloud cover.
When transferring cables from a warm environment to a cold outdoor location, or when allowing cables to warm up outdoors during the transition from night to day, it is essential to allow time for the cable temperatures to stabilize The maximum stabilization time should be determined based on local conditions by monitoring changes in the received signal, ensuring that neither the source nor the measurement setup, including cables and antennas, is disturbed during this process.
Mismatch
This standard's measurement uncertainty budgets encompass mismatch uncertainties, addressing the discrepancies between the receiver, signal source, cables, attenuators, and antennas The relevant equations for calculating these mismatch uncertainties can be found in Annex F.
NOTE 1 Despite that the use of slotted lines is obsolete, VSWR is often used to denote mismatch of equipment such as attenuators and measuring receivers, but it is more common to cite measurements of mismatch as return loss The quantity “return loss” is favoured over VSWR for the measured magnitude of mismatch of antennas
To ensure optimal performance, the receive antenna must present a return loss greater than 20.9 dB to the receiver This can typically be accomplished by adding a 6 dB attenuator to the cable linking the antenna and the receiver It's important to consider that the cable's attenuation contributes to the overall padding attenuation, necessitating a balance between maintaining an acceptable signal-to-noise ratio (SNR) and minimizing mismatch uncertainty.
Using single cables is preferred over multiple interconnected cables due to the mismatch uncertainty introduced by each connection For precise measurements, particularly above 1 GHz, a Vector Network Analyzer (VNA) can measure complex reflection coefficients, aiding in the correction of mismatch uncertainties when impedances deviate from 50 Ω at the transmit and receive antennas It is crucial to apply error correction techniques with caution in setups where cable movement may undermine the corrections.
When conducting SIL measurements between two antennas, two distinct cable sections are utilized: one (T) connects the signal source's output port to the transmit antenna's port, while the other (R) links the receive antenna's port to the measuring receiver's input port The equations provided in Annex F, specifically equations (1 3) and (1 4), pertain to a cable section with an attached attenuator, measured using a fully calibrated two-port VNA Additionally, this cable section may incorporate a padding attenuator, typically positioned at the end connected to the antenna.
The limit for mismatch uncertainty in power transfer from the signal source to the transmit antenna is defined by Equation (1 3), while the limit for mismatch uncertainty in power transfer between the receiver and the receive antenna is specified in Equation (1 4).
The reflection coefficients and S-parameters are assessed, including the reflection coefficient of the receive antenna port (\( \Gamma_{aR} \)), the reflection coefficient of the transmit antenna port (\( \Gamma_{aT} \)), the reflection coefficient of the output port of the signal source (\( \Gamma_{T} \)), and the reflection coefficient of the input port of the measuring receiver (\( \Gamma_{R} \)).
S 1 1 is the reflection coefficient of the cable section R or T that connects with the ports of the receive and transmit antennas respectively, applied to Equation (1 3) or Equation (1 4), respectively;
S 21 is the transmission coefficient (i.e loss) of the cable section R or T that connects with the ports of the receive and transmit antennas respectively, applied to Equation (1 3) or Equation (1 4), respectively;
The S 22 parameter represents the reflection coefficient of the cable section R or T, which connects to the input port of the measuring receiver or the output port of the signal source, as described in Equation (13) and Equation (14).
For example, if an antenna and the measuring receiver have a return loss of 20,9 dB (i.e
|Γ aR | = |Γ R | = 0,091 ), and the connecting cable has |S 1 1| = |S 22| = 0,024 and |S 21| = 0,5, then Equation (1 4) yields an error bound of M R ± =0,056dB
NOTE 2 A correction could be applied to reduce an error, but where no correction is applied, the value of the error is taken to be the value of the uncertainty, which is M R / 2
In cases where S 1 1 and S 22 of the cables are negligibly small, e.g at frequencies below
200 MHz, Equations (1 3) and (1 4) can be simplified by setting all terms multiplied by S 1 1 and
Dynamic range and reproducibility of SIL measurement
Accurate measurement of Signal Interference Level (SIL) between two antennas requires amplitude linearity, also known as dynamic accuracy, with a target of better than ±0.1 dB per decade It is essential to consider the measurement uncertainty budget based on the actual linearity achieved Antenna measurements typically necessitate a dynamic range of 60 dB or more; alternatively, a null detection method, or substitution method, can be employed alongside a precision step attenuator For further details on the null detection method, refer to section 4.4.4.3.2 of CISPR 16-1-5.
Coaxial cables must not be bent beyond their specified minimum bend radius to avoid introducing mismatches Additionally, excessive bending in non-permanently fixed cables can negatively impact performance consistency If cables are excessively bent, it is essential to repeat the cable-through measurement to ensure that the Signal Integrity Loss (SIL) has not changed by more than 0.2 dB.
The typical attenuation between two antennas 10 m apart is approximately 40 dB, necessitating the use of two fixed attenuators (e.g., 6 dB) to minimize mismatch uncertainties A signal-to-receiver noise ratio of at least 34 dB is essential for achieving low measurement uncertainties, leading to a total attenuation of around 90 dB when including cable loss The attenuator value can be adjusted based on the cable's attenuation Dynamic range is defined as the ratio between the maximum reading and the noise floor To further enhance mismatch uncertainty, employing a calibrated VNA and removing the attenuators can be beneficial if greater dynamic range is required Calibrated VNAs typically offer an effective source and load match with a return loss better than 30 dB, which may involve full two-port calibration (12-term correction) and is best suited for methods utilizing shorter cables.
Signal-to-noise ratio
The signal generator must deliver sufficient output power to ensure that the signal reaches the measuring receiver input well above its noise level When utilizing a Vector Network Analyzer (VNA), the signal-to-receiver noise ratio should be at least 34 dB; however, for receivers with an average detector, this ratio can be lowered to 17 dB or more Additionally, to mitigate the effects of sinusoidal ambient signals, the signal-to-interference ratio should be maintained at 30 dB or higher Receiver noise can be further reduced by adjusting the resolution bandwidth as necessary For additional considerations, refer to section A.8.1.
Power amplifiers are essential for boosting signals at the output of signal generators, ensuring they surpass ambient and receiver noise levels It is crucial to adhere to radio regulations when utilizing these amplifiers.
Preamplifiers can enhance the signal at the measuring receiver input, but it is crucial to prevent overloading both the preamplifier and the receiver The linearity of preamplifiers must be checked, especially in the presence of high ambient signals To mitigate issues with out-of-band signals, filters should be employed, and any errors related to out-of-band signals and saturation must be assessed and factored into the uncertainty analysis of antenna calibration Utilizing a shielded FAR or SAC can effectively eliminate problems caused by ambient interference during antenna calibration.
During the calibration process, the presence of ambient signals within the measuring receiver's bandwidth can introduce errors that depend on the characteristics of these signals Specifically, if the ambient signals include sinusoidal components, such as those from analogue broadcasts, a higher signal-plus-ambient-to-ambient ratio may be necessary For instance, a sinusoidal ambient signal that is 20 dB lower than the test signal can lead to an uncertainty contribution of approximately 0.9 dB To mitigate this uncertainty, it is essential to increase the level of the test signal accordingly.
CISPR 1 6-1 -6:201 4 © IEC 201 4 – 41 – larger uncertainty contributions than noise-like or broadband signals; see also the second paragraph of 6.1 1 about avoiding measurements on ambient frequencies.
Antenna masts and cables
Unwanted reflections from antenna support structures and cables can introduce systematic uncertainties in antenna calibration To limit this uncertainty to within ± 0.5 dB, it is recommended to use lightweight non-metallic masts, with motorized masts for the SSM being a more robust option, as outlined in section A.2.3.
Antenna cables must be positioned orthogonally to the dipole elements and should run horizontally for a minimum of 1 meter behind the antenna before transitioning vertically to the ground In the case of a vertically-polarized antenna, the cable should also extend horizontally behind the antenna for at least the same distance.
For vertical routing to the ground, measurements should be taken at least 5 meters away If the distance is less than 5 meters, it is essential to assess and apply the measurement uncertainty accordingly Refer to section A.2.3 for methods to quantify this effect These guidelines primarily pertain to dipole and biconical antennas, as the impact is reduced for more directive antennas like LPDA antennas.
Functional checks of an AUC
General
Before calibration, it is essential to verify the integrity of the AUC A visual inspection of the antenna is necessary, especially if it is not new, to check for any mechanical or structural damage and oxidation on the electrical contact surfaces Although measuring the return loss is optional, it is a quick and highly recommended test prior to measuring the antenna factor, as significant deviations from the manufacturer's data can indicate whether further measurements are warranted.
If the return loss is not initially measured and the measured antenna factor (AF) significantly deviates from previous calibrations or manufacturer specifications, it is advisable to measure the return loss This deviation can indicate a faulty antenna and assist in diagnosing the issue A measurement method for return loss is outlined in section A.8.7 Additionally, while optional, it is recommended to check the connector pin-depth (refer to A.8.2) and to look for sharp resonances in the frequency response (see A.8.6).
Balance of an antenna
In antenna calibration and radiated disturbance measurements, common-mode currents on the cable attached to the receive antenna can arise from the receive antenna balun, leading to electromagnetic fields that may introduce systematic uncertainties in measurement results The method for assessing this imbalance is detailed in section 4.5.4 of CISPR 16-1-4:2010 If the balun exhibits poor balance, resulting in common-mode currents on the cable's outer conductor, the use of ferrite clamps can help mitigate these currents.
Achieving optimal balance on high power baluns can be challenging, making their use in receive antennas generally discouraged When the balance of a receive antenna deteriorates, especially at lower frequencies, it is advisable to carry out repairs.
Cross-polar performance of an antenna
When an antenna is aligned with the electric field vector of a linearly-polarized plane-wave field, it is considered co-polarized, and upon rotating the antenna 90°, the cross-polar rejection should be at least 20 dB This alignment occurs when the antenna's mechanical reference line is parallel to the field vector, which varies depending on the type of antenna; for dipole, biconical, LPDA, and hybrid antennas, the reference line corresponds to the physical dipole axes, while for horn antennas, it is represented by a physical plane For instance, in a vertically-polarized field, the side-wall of a horn antenna or the central vane of a DRH antenna must be vertically aligned.
While it is not always necessary to measure cross-polar performance during antenna calibration, it is essential to obtain cross-polar rejection data from the manufacturer Typically, dipole, biconical, and horn antennas meet the 20 dB requirement; however, many LPDA antennas, particularly those with echelon dipole elements, may not achieve this standard, especially at the higher end of their frequency ranges It is important to calculate the additional uncertainty associated with cross-polar rejection below 20 dB for inclusion in the measurement uncertainty budget for EMC disturbance testing.
To achieve a cross-polar response measurement better than 20 dB, it is essential to use a paired antenna with a cross-polar rejection exceeding 40 dB, such as a standard waveguide horn antenna Typically, the cross-polar performance of Log-Periodic Dipole Array (LPDA) antennas deteriorates with increasing frequency; for instance, an LPDA rated for a maximum frequency of 2 GHz may only meet the 20 dB cross-polar rejection below 1 GHz Therefore, it is crucial to measure the cross-polar response above 1 GHz, as outlined in section 4.5.5 of CISPR 16-1-4:2010 While a linear dipole antenna can be utilized, horn antennas are preferred due to their superior directivity, which significantly reduces reflected signals and generally offers a broader frequency range Additionally, one of the antennas must be rotatable through slightly more than 90°, as detailed in reference [14] regarding test procedures.
Radiation patterns of an antenna
The radiation pattern is crucial for antenna performance, particularly in calibrations using ground-plane reflection methods Antennas designed for EMC testing typically feature a wide main lobe directed along the antenna's boresight However, certain antennas, such as horns, may exhibit a narrower main lobe or a dip in the boresight direction as they approach the upper limits of their operating frequency range These variations in radiation pattern can significantly impact the precision needed for accurate antenna alignment during calibration.
In free-space setups without a ground plane, the antenna pattern has minimal impact on the uncertainty of the antenna factor during calibration, except for specific cases like DRH antennas However, when calculating the antenna factor with ground-plane reflection, as outlined in Equation (23) and Equation (C.22), significant uncertainty can arise for antennas with narrow-beam patterns Equation (23) simplifies Equation (C.22) by assuming uniform radiation in all directions, which is applicable in calibrations where the direct transmission between antennas has the same amplitude as the transmission via the ground plane This assumption is crucial for simplified antenna factor calculations, such as in Equation (39) An example of this is the uniform H-plane pattern of a horizontally-polarized biconical antenna.
In horizontally-polarized LPDA antennas with a 10 m separation, the uncertainty from the radiation pattern is minimal However, at a 3 m separation or with more directive antennas, this uncertainty can become significant For instance, if the reflected signal is 2 dB below the peak of the beam, the uncertainty in the antenna factor (AF) is 0.46 dB.
The uncertainty magnitude can be assessed by understanding the radiation pattern and the antenna setup geometry By analyzing the separation and heights of the antenna pair, one can determine the angles from boresight for both direct and reflected rays Incorporating radiation pattern data into the calculation of the antenna factor (AF) can significantly reduce this uncertainty, as demonstrated in Equation (C.29).
Omnidirectional antennas, like biconicals, necessitate stricter site validation criteria compared to directional antennas, such as horns, due to the latter's ability to suppress unwanted reflections in low directivity directions.
Radiation patterns are assessed by rotating the antenna around its phase center in a free-space setting, typically in a horizontal plane, known as azimuth rotation During this process, the amplitude response is recorded in relation to the angle of rotation from the boresight.
7 Basic parameters and equations common to antenna calibration methods for frequencies above 30 MHz