Radiated Emission Measurements in the Semi-Anechoic Chamber at 3, 5, and 10 m Distance: Results and Empirical Estimates

Radiated emission (RE) measurements in the frequency range 30 … 1000 MHz are usually performed at the open area test sites (OATS) or in the semi-anechoic chambers at the distances 3 or 10 m from the equipment under test (EUT).

If the RE limit in the applicable standard is given just for one measurement distance and the tests are performed at another distance it is a common practice to re-calculate the corresponding values, e.g. from 3 m to 10 m and vice versa. This is made in assumption that the field strength decreases inversely proportional to the measurement distance.

Nevertheless, this approach in many cases does not give correct results. Possible reasons are in particular:

measurement uncertainties, reflections from the reference ground plane, compliance of the measurement setup with the far-field conditions, uncertainties of the site validation, etc. There is also a concern regarding the RE measurements of the relatively large objects at 3 m distance as the results may be influenced by the near-field effects. The discussion is ongoing within the CISPR community.

Related problems, including the history of their development, are discussed by Daniel Hoolihan³. The interested reader is addressed to this excellent review.

In this article we make an attempt to improve our understanding of the following questions:

• Is it realistic to assume the inverse proportionality of the field strength to the measurement distance in a semi-anechoic chamber?
• Is it reliable to test relatively large objects at 3 m measurement distance?
• Can we predict the field strength when the near-field influences measurement results?

Measurements 

Radiated emission measurements in the frequency range 30 … 1000 MHz were performed in the accredited semi-anechoic chamber at the distances 3, 5, and 10 m from the EUT that was a stable reference source. The EUT had the largest linear dimension (height) equal to 2,3 m. Measurement procedure was according to ANSI C63.4-2009¹. For considered frequency range and distances the measurement uncertainties were estimated to
± (4,6 … 4,7) dB. 

Results and Discussion 

Measurement results are presented in Table 1.

Table 1: Field strength [dB(µV/m)] of radiated emissions from the reference EUT measured in a semi-anechoic chamber at 3, 5, 10 m distance with quasi-peak (QP) and peak (Pk) detectors 

It is usually assumed that the field strength in the far-field decreases inversely proportional to the measurement distance according to

E(d) = E(d_ref) + 20 • log(d_ref/d)(1)

where

E(d_ref) is the field strength, measured at the reference distance d_ref from the source;

E(d) is the field strength calculated at the distance d from the source;

E is expressed in dB(µV/m).

Using equation (1) and the data from Table 1 the field strength values were computed for our series of measurements with the reference distance d_ref set to 5 m. Calculation results are given in Table 2.

Table 2: Field strength E [dB(µV/m)] at 3 and 10 m distance estimated from equation (1) and compared with measured values
(Δ = Estimated – Measured). Reference distance d_ref = 5 m 

The difference Δ between the field strength values estimated using formula (1) and the measured values is positive when d_ref > d and negative for d_ref < d. The variations of Δ are rather big for both quasi-peak and peak measurement series and the absolute values |Δ| are up to 5.8 dB. For d_ref = 10 m (results not shown here) maximum Δ = 7.0 dB when the values are re-calculated to 3 m. Even much larger differences were observed in other studies, see, for example, David Weston’s article Egregious Errors in Electromagnetic Radiation Evaluation⁴.

In this way, application of equation (1) for semi-anechoic chambers may introduce uncertainties (not included in the uncertainty budget) leading to the erroneous judgement on compliance of the tested products with the radiated emission requirements.

In attempt to make the issue more clear we applied a semi-empirical model for prediction of the values that should have been measured at 3 m distance based on the reference results obtained at 5 m distance.

From the computational form of Maxwell’s Equations the real part of the field strength E at the distance d from the source at the time t = 0 may be expressed as

E(d)= R • [ ( 1/d²)cos (-βd) + (c/ ωd³)sin (-βd)] (2)

Derivation of equation (2) is similar to that considered by Glen Dash².

In this equation E – is in V/m; d – in m; ω = 2πf; β = ω / c; f – frequency in Hz; c – speed of light. R is an unknown function describing the radiation pattern of the source expressed in V • m. Physically R may be interpreted as the product

current • resistance • length

where all three parameters are integrated through the whole structure of the radiating source.

From equation (2) we obtain

R =E(d)• [ (1 /d²)cos (-βd) + ( c/ ωd³) sin (-βd) ] ^-1(3)

We assume that the radiation pattern R calculated from (3) for d = 5 m using the values from Table 1 is also valid for measurement distance d = 3 m, i.e. R(3 m) = R(5 m). That allows predicting the field strength E at 3 m from the model (2).

The calculations were performed for low frequencies where the far-field condition d >> λ / (2π) is not fulfilled. The results are shown in Table 3.

Table 3: Comparison of the field strength E [dB(µV/m)] predicted by the semi-empirical model (2) with the values measured at
3 m distance (δ = Predicted – Measured)

In this way, for frequencies 30,6 … 34,0 MHz the average difference between predicted and measured numbers is 6,85 dB and 7,2 dB for quasi-peak and peak values respectively. Thus, the near-field effects should not be ignored if the RE measurements are performed for large test objects at short distances. In our case the scaling factor, i.e. the relation between the measurement distance and the largest linear dimension of the EUT, is 3 m / 2,3 m = 1,3.

Conclusions 

Commonly used assumption that in a semi-anechoic chamber the field strength decreases inversely proportional to the distance from the test object is not correct and should be considered with caution to avoid erroneous judgement on compliance of the tested products with the radiated emission requirements. Corresponding errors may be up to 5,8 dB and even more.

Near-field effects contribute to the values of RE from the relatively large objects measured in a semi-anechoic chamber at 3 m distance when the far-field condition d >> λ / (2π) becomes weak. Using the semi-empirical model it is estimated that for frequencies 30,6 … 34,0 MHz the value of 7,0 dB should be added to the results obtained at 3 m. The number may be recommended as the measurement correction factor.

References 

1. ANSI C63.4-2009, American National Standard for Methods of Measurement of Radio-Noise Emissions from Low-Voltage Electrical and Electronic Equipment in the Range of 9 kHz to 40 GHz, Institute of Electrical and Electronic Engineers, September, 2009.
2. Dash, Glen. A Maxwell’s Equations Primer, Chapter V – Radiation from a Small Wire Element, Ampyx LLC, Copyright 2000, 2005 Ampyx LLC.
3. Hoolihan, Daniel. Radiated Emission Measurements at 1/3/5/10/30 Meters, Interference Technology, 2010.
4. Weston, David A. Egregious Errors in Electromagnetic Radiation Evaluation, 2010 EMC Directory & Design Guide, Interference Technology 2010.

Vladimir Bazhanov received the M.Sc. degree from Kharkov State University of Radio Electronics, Ukraine, 1974, and PhD from the Institute of Applied Geophysics, Moscow, Russian Federation, 1988. He has a wide scientific experience from the university research including numerous expeditions. From 2001 his work was focused on RF and EMC compliance verification, test sites validation and development of test methods. Vladimir is a Radio Standardization Manager at Ericsson, Sweden, a member of the New York Academy of Sciences, IEEE Sweden Section Board and ANSI C63/SC1 Wireless Working Group. Contact e-mail: vladimir.bazhanov@ericsson.com

Bruno Liska received his M.Sc. degree in electro engineering and radio communication from Czech Technical High School in Prague, in 1979. He has 25 years’ experience working with EMC and lightning protection. Currently Bruno is a Standardization Manager and EMC Senior Specialist at Ericsson, Sweden, working with radio base station products in cellular systems. He is a member of the Swedish EMC standardization committee and the ETSI ERM EMC standardization working group. Contact e-mail: bruno.liska@ericsson.com