Design and experiment of a directional coupler for X-band long pulse high power

microwaves
Zhen Bai, Guolin Li, Jun Zhang, and Zhenxing Jin

Citation: Review of Scientific Instruments 84, 034701 (2013); doi: 10.1063/1.4789782

View online: http://dx.doi.org/10.1063/1.4789782
View Table of Contents: http://scitation.aip.org/content/aip/journal/rsi/84/3?ver=pdfcov
Published by the AIP Publishing

Articles you may be interested in

Multichannel microwave interferometer for the levitated dipole experiment
Rev. Sci. Instrum. 80, 043502 (2009); 10.1063/1.3095684

Power-dependent switching of microwave signals in a ferrite-film nonlinear directional coupler

Appl. Phys. Lett. 89, 172511 (2006); 10.1063/1.2362576

In-line sensor for accurate rf power measurements

Rev. Sci. Instrum. 76, 106107 (2005); 10.1063/1.2090287

Two-channel active high-power X-band pulse compressor

AIP Conf. Proc. 569, 741 (2001); 10.1063/1.1384400

High power continuous wave microwave system at 3.7 GHz

Rev. Sci. Instrum. 72, 1869 (2001); 10.1063/1.1347373

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
134.176.129.147 On: Tue, 09 Dec 2014 09:44:05
 REVIEW OF SCIENTIFIC INSTRUMENTS 84, 034701 (2013)

Design and experiment of a directional coupler for X-band long pulse high
power microwaves
Zhen Bai,a) Guolin Li, Jun Zhang, and Zhenxing Jin
College of Optoelectronic Science and Engineering, National University of Defense Technology,
Changsha 410073, People’s Republic of China
(Received 8 November 2012; accepted 14 January 2013; published online 4 March 2013)
Higher power and longer pulse are the trend of the development of high power microwave (HPM), and
then some problems emerge in measuring the power of HPM because rf breakdown is easier to occur
under the circumstance of high power (the level of gigawatt) and long pulse (about 100 ns). In order to
measure the power of the dominant TM01 mode of an X-band long pulse overmoded HPM source, a
directional coupler with stable coupling coefficient, high directivity, and high power handling capacity
in wide band is investigated numerically and experimentally. At the central frequency 9.4 GHz, the
simulation results show that the coupling coefficient is −59.6 dB with the directivity of 35 dB and
the power handling capacity of 2 GW. The coupling coefficient is calibrated to be accordant with the
simulation results. The high power tests are performed on an X-band long pulse HPM source, whose
output mode is mainly TM01 mode, and the results show that the measured power and waveform of
the directional coupler have a good consistency with the far-field measuring results. © 2013 American
Institute of Physics. [http://dx.doi.org/10.1063/1.4789782]

I. INTRODUCTION for decades, there are still some problems of directional cou-
plers working with long pulse high power microwave source.
The high power microwave (HPM) sources have been
First, rf breakdown is easier to occur in the measuring device
developed rapidly in the past 4 decades, and many meth-
due to the feature of long pulse duration (about 100 ns) and
ods have been applied to improve the microwave power.1–7
high power level (the level of gigawatt), so the power han-
Overmoded devices are such method to increase the power
dling capacity of the coupler should be high enough. Second,
handling capacity of HPM devices.8–11 Usually, the power of
considering the disparity between the output power of the mi-
HPM is measured by far-field method, which is regarded to
crowave source and the power handling capacity of the trans-
be accurate when the far-field criterion r>2D2 /λ is satisfied,
mission system such as the waveguide-coaxial transformer
where D is the diameter of the larger of the transmitter or re-
and microwave cable, the coupling coefficient of the coupler
ceiver, r is the distance between the transmitter and receiver,
should be designed carefully. Third, since the operation fre-
and λ is the wavelength of microwave in free space. How-
quency of the X-band HPM source under test ranges in about
ever, sometimes the D is large or the phasing of the trans-
80 MHz band, the coupler should work in a wider band. Fi-
mitting antenna is poor,12 the far-field criterion can hardly
nally, in order to avoid the effect of the wave reflected from
be satisfied. In this case, the far-field method is not conve-
the backward wave, which is not absorbed entirely by the
nient. And some other problems influencing the inaccuracy of
matching load, the directivity of the coupler is supposed to
this method exist, such as the reflective effect of the environ-
be higher than 10 dB.
ment, the errors in measuring the effective area of the receiv-
ing antenna, and so on. Online power measurement method
(Fig. 1 shows the schematic of online measuring system.) can
measure power online and dynamically without the problems
mentioned above. Directional coupler is such an online mea-
suring device based on coupling principles.13–16 In practice,
some directional couplers have been presented.17–20 Accord-
ing to L. M. Earley’s research, to diagnose the 3.15 GHz mi-
crowaves at power of approximately 100 MW with the pulse
length of about 40 ns, a TM01 circular/rectangular waveguide
directional coupler with 3 coupling slots is constructed, which
has an operational bandwidth of 2.9–3.4 GHz with a coupling
coefficient of −63 ± 0.5 dB and a directivity that varied be-
tween 13 and 20 dB.19
Although the coupling theory is very mature and direc-
tional couplers have been employed to measure power online

a) Electronic mail: baizhen206@gmail.com. Tel.: +86 731 84576333. FAX:

+86 731 84515535. FIG. 1. Schematic of online measuring system.

0034-6748/2013/84(3)/034701/5/$30.00 84, 034701-1 © 2013 American Institute of Physics

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
134.176.129.147 On: Tue, 09 Dec 2014 09:44:05
 034701-2 Bai et al.
FIG. 2. Structure of the X-band overmoded O-type Cerenkov oscillator.
According to the demands mentioned above, a directional
coupler with high power handling capacity for measuring the
power of TM01 mode of an X-band long pulse (100 ns) over-
moded HPM source is investigated numerically and experi-
mentally in this paper. When the central frequency is 9.4 GHz,
the simulation results show that the coupling coefficient and
the directivity are −59.6 dB and 35 dB, respectively, and the
power handling capacity of the coupler is about 2 GW.
This paper is arranged as follows. In Sec. II, the design
and simulation results are presented. The results of the cali-
bration and high-power test are presented in Sec. III. Conclu-
sions are given in Sec. IV.
II. DESIGN AND SIMULATION RESULTS
The HPM source under test is an X-band long pulse
overmoded HPM source, which has been developed in our
laboratory.8, 11 The overmode ratio of this device is about 3
and its dominant mode is TM01 mode.8 The structure of the
X-band overmoded O-type Cerenkov oscillator designed in
Ref. 11 is shown in Fig. 2, and its measured microwave ra-
diation pattern is well in good agreement with the theoretical
pattern of the symmetric TM01 mode, indicates the dominate
mode of the generator is TM01 . Actually, at angles more than
30◦ , the power of radiation is higher than that of a single TM01
mode shows that other higher symmetric modes are radiated,
but it was just a very small fraction of the radiation and could
be attributed to the overmoded property of the generator. The
mode content of the generator has been analyzed in Ref. 21.
Table I shows the particle-in-cell (PIC) simulation results and
mode composition.21
TABLE I. PIC simulation results and mode composition.
Diode Beam Dominant
voltage (kV) current (kA) frequency (GHz)
570 7.2 9.30
630 8.4 9.34
690 9.6 9.37
750 10.8 9.40
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
134.176.129.147 On: Tue, 09 Dec 2014 09:44:05
Rev. Sci. Instrum. 84, 034701 (2013)
FIG. 3. Schematic diagram of the HPM directional coupler structure.
Table I shows that the percentage of total output mi-
crowave power carried by the TM01 mode is higher than 95%
in the X-band overmoded Cerenkov oscillator at the diode
voltage ranging from 570 kV to 750 kV, which means the
higher symmetric modes only account for less than 5%. Cou-
plers designed for a specific mode would also couple a little
of other modes, especially if the guide wavelength of the un-
wanted mode is close to that of the coupling mode. Consider-
ing the higher symmetric modes are indeed a very small frac-
tion in this case, we can take no account of the little effect of
the higher modes when designing the TM01 mode directional
coupler.
The schematic of the directional coupler is shown in
Fig. 3. It is composed of a circular overmoded waveguide as
the main waveguide and a rectangular waveguide as the sub
waveguide. In the main waveguide exist transverse magnetic
modes and in the sub waveguide exists the fundamental mode
of TE10 . In order to couple the specified mode of TM01 in the
main waveguide into the coupling port of the sub arm when
equal-interval coupling, the distances between adjacent cou-
pling holes should be satisfied that16
λ10 λw −
S= i (iw− = 0, 1, 2, ...), (1)
λ10 + λw w
where S indicates the distance between adjacent coupling
holes, λ10 is the guide wavelength of the TE10 mode in the
rectangular sub-waveguide, λw is that of the coupled mode in
the main waveguide.
In order to achieve stable −60 dB coupling of TM01
mode in the overmoded circular waveguide and work in wide
band, an equal-interval Chebyshev distribution16 is employed.
For the sake of improving the power handling capacity of the
couple, both main waveguide and sub waveguide are cho-
sen with relative larger dimensions for X-band. The main
waveguide is a circular overmoded waveguide with diame-
ter of 112 mm matching with the output waveguide of the
HPM source under test, and the sub waveguide is a standard
Percentage of the total
Output power carried by TM0n modes
power (GW) TM01 TM02 TM03
1.0 98.5% 0.3% 1.2%
1.3 98.2% 0.7% 1.1%
1.7 97.0% 1.8% 1.2%
2.0 95.0% 3.6% 1.4%
 034701-3 Bai et al. Rev. Sci. Instrum. 84, 034701 (2013)

TABLE III. The results of calibration.

Frequency Coupling Isolating Directivity

(GHz) coefficient (dB) coefficient (dB) (dB)

9.36 −59.4 −71.3 11.9

9.38 −59.2 −74.4 15.2
9.40 −59.1 −82.8 23.7
9.42 −59.4 −76.5 17.1
9.44 −59.1 −77.9 18.8

FIG. 4. E-field distribution at the edges before and after chamfering.

rectangle waveguide of WR137. The coupling holes are on the

common wall formed by one of the wide walls of rectangular
waveguide and a part of the wall of the circular waveguide.
The edges of the coupling holes are chamfered to weaken
the intensity of electric field there lest the rf breakdown oc-
curs. Figure 4 shows the E-field distribution at the edges be-
fore and after chamfering. When the input power is 1 W, the FIG. 5. The result of simulation of the directional coupler.
intensity of electric field at the holes’ edges declines from
267 V/m to 251 V/m, which means the intensity of elec-
tric field at the edges of coupling holes would decline from
119.4 kV/cm to 112.3 kV/cm when the input power is 2 GW.
According to the abovementioned structure, simulation
investigation is performed. The optimized dimensions of the
TM01 directional coupler are shown in Table II. L the length
of the coupler; n is the number of the coupling holes; S is the
distance between adjacent coupling holes; r1 , r2 , r3 are the ra-
dius of the coupling holes, respectively; and t is the thickness
of the common wall between the main waveguide and the sub
waveguide.
The exciting mode in input port is TM01 mode, and cen- FIG. 6. E-field distribution at 9.4 GHz.
tral frequency is 9.4 GHz. Figure 5 shows the simulation re-
sult. The coupling coefficient is −59.6 dB with the directivity
of 35 dB.
The electric field distribution in the coupler is shown
in Fig. 6. The frequency at which the field is calculated is
9.4 GHz. The simulation results show that the maximum
electric field 277 V/m when the input power is 1 W, which
means if the input power is 2 GW, the maximum electric field
would be 123.9 kV/cm, far below the threshold of vacuum
breakdown.

TABLE II. The optimized dimensions of the TM01 directional coupler.

L (mm) n S (mm) r1 (mm) r2 (mm) r3 (mm) t (mm)

420 2 42.3 4.15 3.8 3.1 5.9

FIG. 7. The scheme of calibration system.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
134.176.129.147 On: Tue, 09 Dec 2014 09:44:05
 034701-4 Bai et al.
FIG. 8. Typical measured waveforms of the coupler.
III. EXPERIMENTAL RESULTS
A TM01 directional coupler was fabricated according to
the optimized configuration, and its coupling coefficient and
directivity are calibrated with network analyzer in an anechoic
chamber. Figure 7 is the schemes of calibration of the cou-
pling coefficient. A TM01 mode exciter is employed to gener-
ate the input microwave into the main waveguide.
The calibrated results are shown in Table III. The mea-
sured and simulated coupling coefficient can match accord-
ingly. The difference between the measured and simulated
isolating coefficient is likely the results of the errors of ma-
chining the coupler and the reflective effect of the transmit-
ting antenna and environment. So, the measured directivity is
less than the simulated one.
The high power tests are performed on an X-band long-
pulse overmoded HPM source, whose output mode is mainly
TM01 mode. The TM01 mode content is about 95% and mi-
crowave frequency is 9.4 GHz. The device generates 1.8 GW
microwave power with an pulse duration of about 110 ns at
20 Hz repetition mode. At far-field zone four WR112 open-
end waveguides as receiving antennas are utilized to measure
FIG. 9. Typical measured waveforms of the receiving antennas.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
134.176.129.147 On: Tue, 09 Dec 2014 09:44:05
Rev. Sci. Instrum. 84, 034701 (2013)
the radiation power in contrast with the directional coupler in
the experiment.
Figure 8 shows the typical measured waveform of the di-
rectional coupler. Waveform 1 is the microwave power mea-
sured online by the directional coupler. Figure 9 shows the
typical measured waveform of the receiving antennas. In the
experiments, four receiving antennas are placed in the main
lobe region. Waveform 2–5 are the radiated microwave power
measured by four receiving antennas. The measured pulse du-
rations of Waveform 1 is 111 ns, and ones of Waveform 2–
5 are 114 ns, 116 ns, 112 ns, and 110 ns, respectively. The
waveform of the directional coupler is similar to ones of the
receiving antennas and the power amplitude measured online
is 1.7 GW, which is a little lower than the power measured by
far-field method. These results of high power tests confirm the
existence of higher symmetric modes, and indicate that no rf
breakdown occurs in the directional coupler. So, the coupler
can be utilized to measure power of the TM01 mode of the X-
band HPM source with high power and long pulse duration.
IV. CONCLUSION
A TM01 directional coupler for an X-band long-pulse
overmoded high power microwave source is investigated nu-
merically and experimentally. In order to work with a high
directivity and stable coupling coefficient in wide band, an
equal-interval Chebyshev distribution is employed. The cold
test reveals that the calibrated coupling coefficient agrees with
the simulation result, which varies in the range of −59.06 dB
to −59.38 dB with the frequency from 9.36 to 9.44 GHz. The
high power test is performed on an X-band long pulse over-
moded HPM source. The results show that the waveform and
the measured power of the directional coupler are similar to
that of the receiving antennas and the power handling capacity
of the coupler is higher than 1.8 GW.
ACKNOWLEDGMENTS
This research was supported by the National Natural Sci-
ence Foundation of China.
1 S. K. Nam, C. Lim, and J. P. Verboncoeur, Phys. Plasmas 16, 023501
(2009).
2 C. Chang, G. Z. Liu, X. X. Zhu, H. B. Chen, and J. Y. Fang, Phys. Plasmas
16, 033505 (2009).
3 G. L. Li, C. W. Yuan, J. Y. Zhang, T. Shu, and J. Zhang, Laser Part. Beams
26, 371 (2008).
4 G. L. Li, T. Shu, C. W. Yuan, J. Zhang, Z. X. Jin, and J. H. Yang, Phys.
Plasmas 17, 033301 (2010).
5 Y. W. Fan, H. H. Zhong, T. Shu, Z. Q. Li, H. Zhou, W. H. Zhou, and L.
Guo, Rev. Sci. Instrum. 80, 024701 (2009).
6 Q. Zhang, C. W. Yuan, and L. Liu, Laser Part. Beams 28, 377 (2010).
7 Q. Zhang, C. W. Yuan, and L. Liu, Phys. Plasmas 18, 083110 (2011).
8 J. Zhang, Z. X. Jin, J. H. Yang, H. H. Zhong, T. Shu, J. D. Zhang, B. L.
Qian, C. W. Yuan, Z. Q. Li, Y. W. Fan, S. Y. Zhou, and L. R. Xu, IEEE
Trans. Plasma Sci. 39, 1438 (2011).
9 J. Zhu, T. Shu, and J. Zhang, Phys. Plasmas 18, 053101 (2011).
10 Y. B. Cao, J. D. Zhang, and J. T. He, Phys. Plasmas 16, 083102 (2009).
11 J. Zhang, H. H. Zhong, and L. Luo, IEEE Trans. Plasma Sci. 32, 2236
(2004).
12 E. H. Braun, Proc. IRE 41,109 (1953).
13 H. Bethe, Phys. Rev. 66, 163 (1944).
14 S. E. Miller, Bell Syst. Tech. J. 33, 661 (1954).
 034701-5 Bai et al. Rev. Sci. Instrum. 84, 034701 (2013)

15 R. E. Collin, Foundations for Microwave Engineering (McGraw-Hill, New 19 L. M. Earley, W. P. Ballard, L. D. Roose, and C. B. Wharton, Rev. Sci.
York, 1966). Instrum. 57, 2283 (1986).
16 W. X. Wang, Microwave Engineering Technology (National Defense Indus- 20 W. X. Wang, L. N. Yue, P. Y. Jiang, and Y. B. Gong, Int. J. Electron. 95,

try Press, Beijing, China, 2009). 99 (2008).

17 G. Janzen and H. Stickel, Report IPF-83-8, University of Stuttgart, 1983. 21 D. Zhang, J. Zhang, H.-H. Zhong, and Z. X. Jin, Phys. Plasmas 19, 103102
18 W. P. Ballard and L. M. Earley, Rev. Sci. Instrum. 56, 1470 (1985). (2012).

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
134.176.129.147 On: Tue, 09 Dec 2014 09:44:05