Broadband Calculable Dipole Antennas
Simply use the links to learn more about each product area.
Dipoles antennas can be modelled very precisely, both analytically and numerically, both in free-space and when the antenna is coupled to a large flat ground plane. NPL developed a calculable dipole antenna and achieved unprecedented accuracy over a broadband by incorporating the complex S-parameters of the antenna balun in the calculation of antenna factor. The uncertainty in antenna factor was better than 0.15 dB for a resonant dipole, and 0.3 dB for a dipole over a broad band such that 4 dipoles covered the frequency range 30 MHz to 1 GHz used in EMC testing. The software to accomplish this, CAP2010, has been made freely available on the NPL website.
Introduction
Dipole-like antennas are used in Electromagnetic Compatibility (EMC) radiated emission testing [1]. A biconical antenna is a broadband version of a dipole antenna, typically used over the frequency range 30 MHz to 300 MHz. It has two main advantages, one being that its length is
1.4 m in contrast to the 4.8 m length of a resonant dipole at 30 MHz. The other advantage is that it is broadband and does not need tuning at each frequency, enabling a huge time saving in testing.
Because of the uniform H-plane pattern of dipole and biconical antennas, very costly measures would be needed to avoid reflections of the radiated signal from the ground, especially at frequencies as low as 30 MHz at which the wavelength is 10 m. The solution adopted is to do the measurements above a flat conducting ground plane, typically larger than 30 m x 20 m [1], which ensures reproducible measurements.
However a ground plane significantly affects the gain of a horizontally polarised antenna because of coupling to its image in the ground plane. Field strength is the required parameter for EMC emissions and in place of antenna gain it is more convenient to use antenna factor which gives the E-field strength when multiplied by the voltage at the antenna output.
Dipole antennas
The method of moments makes it possible to calculate the antenna factor of a dipole antenna over a broad frequency band with very low uncertainties.
The NPL Broadband dipole antenna
NPL has gained much experience since 1989 in the calculation and use of calculable dipole antennas. NPL’s first paper published in 1991 [2] described combining the Numerical Electromagnetic Code (NEC) [3] and the complex impedances of the antenna balun. The principles of this method were included in Reference [1]. This enabled traceability of antenna factor to unprecedented low uncertainties. The impedances were obtained by measuring the S-parameters of the 3-port balun between the input and the two outputs very close to the feed points of the two halves of the dipole elements.
A pair of NPL calculable dipole antennas is shown in Figure 1. NPL obtained uncertainties in antenna factor of better than 0.3 dB for fixed length dipoles used over the frequency bands listed in Table 1.
Figure 1 Pair of calculable antennas showing baluns and 4 pairs of dipole elements with resonant lengths at 60, 180, 400 and 700 MHz.
Table 1 – Resonant frequency of fixed-length dipole antenna and the frequency range over which it operates
| Resonant frequency MHz |
Frequency band MHz |
| 60 | 30 to 100 |
| 180 | 100 to 300 |
| 400 | 300 to 600 |
| 700 | 600 to 1000 |
Calculation provides independent verification of antenna factor as measured by the three-antenna method [5]. Measurements to uncertainties of tenths of a decibel require very precise antenna setup on a very high quality ground plane. Also special measures are required to minimise the reflections from antenna supports and feed cables and repeat measurements are needed to obtain an average. Antenna coupling, and hence antenna factor, can be calculated more precisely than can be achieved by a single measurement.
Calculation provides instant results in free space or above a ground plane. Furthermore validation of a antenna can be performed by comparing measured and computed results in the near-field; for example, two antennas can be measured at 30 MHz using a separation of 10 m and the uncertainty of the physical antenna established in this near-field can be applied to the computed antenna factor in the far-field. Likewise the uncertainty for antenna factor measured at 2 m height, for example, above a ground plane, can be applied to the calculated antenna factor at other heights and in free-space.
A pair of calculable antennas is used for the validation of calibration sites as described in CISPR 16-1-5 [1]. This is an independent method of site validation based on the high precision with which the coupling between antennas can be calculated, which is much more accurate than another published method of site validation based on antenna factors measured on a partially validated site.
Software
The executable software CAP2010 calculates the coupling between two wire antennas, and their antenna factors, over an ideal ground plane or in free space. The software is based on a subset of NEC, but is distinguished from other versions of NEC-based software by the ability to incorporate the measured antenna balun S-parameters.
The software is available for free download from:
http://www.npl.co.uk/software/calculable-antenna-processor
NEC models for dipole, biconical and LPDA antennas are included in the download.
Acknowledgements
The development of the NPL calculable dipole and original software was funded by the UK National Measurement System. Acknowledgements are due to Dr. Stanley Baker who developed the user-friendly Windows version of the software, CAP2010, and to the originators of the NEC code, a subset of which is incorporated into CAP2010.
Specification:
Frequency range: 30 MHz to 1000 MHz
Set of 4 dipoles resonant at 60, 180, 400 and 700 MHz
Return loss: 30 – 50 MHz > 15 dB, 51 - 1000 MHz > 20 dB (typically >25 dB above 80 MHz)
Insertion loss of balun 1.5 dB
Power handling 10 W
Uncertainty in site insertion loss (SIL) at resonant frequencies: 0.3 dB
Uncertainty of antenna factor (AF) at resonant frequencies: 0.15 dB 30 – 600 MHz, 0.2 dB 601 – 1000 MHz
Uncertainty of antenna factor over the frequency bands shown in Figure 2: 0.3 dB
The above uncertainties apply to the deviation of the computed SIL and AF from the AF measured over a continuously welded metal plate of minimum area 30 m x 20 m and flatness better than 0.6 mm. Free-space AF does not depend on a ground plane.
Balun length: 1.3 m
Longest dipole length (60 MHz): 2.38 m
(dipoles resonant at 30, 35, 40, 45, 50, 70, 80, 90, 100, 120, 140, 160, 200, 250, 300, 500, 600, 800 and 1000 MHz available on request. Option down to 20 MHz)
Figure 2 Typical free-space antenna factors for the 4 dipole elements with resonant lengths at 60, 180, 400 and 700 MHz.
References:
[1] CISPR 16-1-5, Specification for radio disturbance and immunity measuring apparatus and methods – Antenna calibration test sites for 30 MHz to 1000 MHz. www.iec.ch/searchpub.
[2] SALTER, M.J. and ALEXANDER M.J., EMC antenna calibration and the design of an open-field site, Jnl. of Phys. E, Meas. Sci. Technol., 1991, 2, 510-519.
[3] LOGAN, J.C. and BURKE, A.J., Numerical Electromagnetic Code, 1981, Naval Oceans Systems Centre, CA, USA.
[4] Alexander M J, Salter M J, Loader B G and Knight D A, Broadband calculable dipole reference antennas, IEEE Trans. EMC, Vol. 44, No.1, pp 45-58, February 2002.
[5] CISPR 16-1-6, Specification for radio disturbance and immunity measuring apparatus and methods – EMC-antenna calibration (expected publication in 2013)