22-Elec-B9 Electromagnetic Field, Transmission Lines, Antennas, and Radiation · May 2016
Question 4 of 8: Maximum Power in an Air-Filled Rectangular Waveguide
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Notes on this paper
Paper format. National Exams, May 2016 — 07-Elec-B9, Electromagnetic Field, Transmission Lines, Antennas, and Radiation. Three hours; closed book; any non-communicating calculator permitted. Eight questions, each of equal value (20 marks); the paper states that five questions constitute a complete exam paper and that the first five appearing in the answer book are the ones marked. The only constants printed on the cover page are the free-space permittivity and permeability, and a Smith chart is bound in as page 4 for use with Question 3. All eight questions are worked here, because this set is a study resource rather than a three-hour sitting.
Reference texts. D. M. Pozar, Microwave Engineering (transmission lines, the Smith chart, waveguides, resonators); M. N. O. Sadiku, Elements of Electromagnetics (Maxwell's equations, plane waves, guided waves); W. H. Hayt and J. A. Buck, Engineering Electromagnetics (field theory and wave propagation); C. A. Balanis, Antenna Theory: Analysis and Design (short elements, arrays, radiation resistance); F. T. Ulaby, Fundamentals of Applied Electromagnetics (radar, link budgets, applied propagation).
Units and constants. The paper works in SI throughout. Where a question supplies a rounded propagation velocity (for example 3×108 m/s) that value is used as given; where it does not, the exact free-space constants are used, giving $c = 2.998\times10^{8}\ \text{m/s}$ and $\eta_0 = 376.7\ \Omega$.
Question 4: Maximum Power in an Air-Filled Rectangular Waveguide (20 marks)
Given. An air-filled rectangular guide of internal broad dimension $a = 2.25\ \text{cm}$ and narrow dimension $b = 1.00\ \text{cm}$, operating at $f = 10\ \text{GHz}$, matched at both ends so that only a forward travelling wave exists, with an electric-field ceiling of $E_0 = 10^{6}\ \text{V/m}$ peak anywhere inside.
Find. The largest time-average power the guide can carry without exceeding that field.
Waveguide cross-section with the TE10 field. The half-sine profile means the 10^6 V/m ceiling is reached only on the centre line, and the mean of E-squared over the cross-section is half the peak value.
Approach. Identify which mode or modes can propagate at 10 GHz, write the transmitted power of the dominant mode in terms of its peak field, and evaluate.
Find the cut-off frequencies and confirm single-mode operation. For a rectangular guide, $f_{c,mn} = \dfrac{c}{2}\sqrt{(m/a)^{2} + (n/b)^{2}}$, so with $a = 22.5\ \text{mm}$ and $b = 10\ \text{mm}$:
$$f_{c,10} = \frac{c}{2a} = 6.662\ \text{GHz}, \qquad f_{c,20} = \frac{c}{a} = 13.32\ \text{GHz}, \qquad f_{c,01} = \frac{c}{2b} = 14.99\ \text{GHz}.$$
At 10 GHz only TE10 is above cut-off; every other mode is evanescent and carries no average power. Note that $a/b = 2.25 > 2$, so TE20 and not TE01 is what closes the single-mode band — a ranking worth checking rather than remembering, since it reverses for $a/b < 2$.
Write down the TE10 field. The single transverse electric component varies as a half sine across the broad wall and is uniform across the narrow one:
$$E_y(x) = E_0 \sin\!\left(\frac{\pi x}{a}\right), \qquad H_x = -\frac{E_y}{Z_{TE}}.$$
The stated ceiling therefore applies at the centre line $x = a/2$, where the sine peaks.
Evaluate the wave impedance at the operating frequency. The dispersion factor appears in every guided quantity, so compute it once:
$$\sqrt{1 - \left(\frac{f_c}{f}\right)^{2}} = \sqrt{1 - \left(\frac{6.662}{10}\right)^{2}} = 0.7458,$$
$$Z_{TE} = \frac{\eta_0}{\sqrt{1 - (f_c/f)^{2}}} = \frac{376.73}{0.7458} = 505.2\ \Omega.$$
The same factor gives the guide wavelength $\lambda_g = \lambda_0/0.7458 = 4.020\ \text{cm}$, which is useful for checking that the guide is being operated sensibly.
Integrate the Poynting vector over the cross-section. The time-average power is
$$P = \frac{1}{2}\int_0^{b}\!\!\int_0^{a} \frac{|E_y(x)|^{2}}{Z_{TE}}\,dx\,dy = \frac{E_0^{2}\,b}{2Z_{TE}}\int_0^{a}\sin^{2}\!\left(\frac{\pi x}{a}\right)dx = \frac{E_0^{2}\,a\,b}{4 Z_{TE}},$$
the factor of four collecting one half from the time average and one half from the mean of $\sin^{2}$ across the guide.
Sanity-check the result physically. The stated ceiling of $10^{6}\ \text{V/m}$ is about a third of the dry-air breakdown strength of roughly $3\times10^{6}\ \text{V/m}$ at sea level, so it is a plausible engineering limit with a safety margin rather than an arbitrary number. Real guides derate further for humidity, altitude, surface roughness and any discontinuity that concentrates the field, and a standing wave of ratio $S$ raises the peak field by the factor $S$ — which is exactly why the question specifies "with no reflections".
Question 4 — results
Quantity
Symbol
Value
Dominant-mode cut-off
$f_{c,10}$
$6.662\ \text{GHz}$
Next two modes
$f_{c,20}$, $f_{c,01}$
$13.32\ \text{GHz}$, $14.99\ \text{GHz}$ — both evanescent at 10 GHz