skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: THE MEASURED PERFORMANCE OF A MILLIMETER WAVE BEAM SPLITTER

Abstract

An essential component of any high power transmission system is a directional coupler that provides a sample of the forward and reflected power when this power is being delivered to the intended load. In the case of millimeter power delivered through a highly oversized corrugated waveguide, there is the much more complex issue of mode purity. It is possible to design an effective mode selective branch guide directional coupler in smooth wall overmoded waveguide. In the typical highly overmoded corrugated waveguide propagating the HE{sub 11} mode, however, obtaining an adequate coupling factor can be difficult, and branch guide attenuation and phase velocity matching over several meters become concerns. A more practical approach for large diameter corrugated waveguide is to obtain a sample of the propagating beam at a miter bend mirror. At low power, the mirror could be a thin metal screen. At the megawatt level, however, heat removal must be considered. For example, at 110 GHz at 1 MW, taking the surface resistance of copper to be 0.10 {Omega}, the dissipation on a 45{sup o} copper mirror would be 750 W or 1500 W for H or E plane reflection, respectively. With a peak to average power ratio ofmore » 3.7 for the circular HE{sub 11} mode, in 31.75 cm diameter corrugated waveguide the peak dissipation can be as high as 500 W/cm{sup 2} at the center of the mirror. An edge cooled thin metal screen is not therefore practical, but a thick plate containing a single narrow channel, at the bottom of which is a row of holes in the remaining thin wall, can be adequately water-cooled on its face. To maintain vacuum and focus the radiation from the holes, the narrow channel is filled by a fused quartz plate, the shape of which is a 45{sup o} sector of a circle having a truncated apex at the coupling holes. These are being used as power monitors on the DIII-D ECH system and on other systems. Since this single row of holes samples only part of the wave field, however, interference among higher order modes, even when they are of low amplitude, can produce noticeable variations in received power, depending on the exact gyrotron frequency and temperature of the waveguides. The only certain way to avoid this problem is to use a true beam splitter that provides a sample of the fields in the entire cross section of the waveguide. Our proposed means of overcoming the heat removal problem is to make the mirror an edge cooled CVD diamond disk having a very thin copper coating on the high power side of the mirror. A two dimensional array of holes is etched through the copper. An image of the incident wave free of aliasing is launched on the opposite side of the copper film. This process is easily scaled to frequencies greater than 170 GHz. We have analyzed the thermal response of such a plate bonded to water-cooled heat sinks on the periphery of both faces. In our case, the heat sinks have 32 mm by 45 mm apertures. The gray diamond we are using is 60 mm diameter with two flats 48 mm apart. For a diamond 1.1 mm thick having an assumed thermal conductivity of 1300 W/m-K and heated with a total of 1500 W (times a correction for the resistivity increase with temperature) on one face with the circular HE11 distribution, the steady state temperature at the center is 210 C, starting at 35 C. The temperature reaches steady state by 1 s, while the 1/e time is about 160 ms. This temperature is slightly higher than the steady state gyrotron window temperature. If necessary we can reduce the peak temperature, at greater cost, by using a thicker gray diamond, say 1.5 mm, or by using a white diamond 1.1 mm thick having 1800 W/m-K conductivity. Either would bring the final central temperature down to <160 C. Since we are using edge cooling, this result is independent of waveguide diameter, so this device can be made for any diameter.« less

Authors:
; ;
Publication Date:
Research Org.:
GENERAL ATOMICS (US)
Sponsoring Org.:
(US)
OSTI Identifier:
804737
DOE Contract Number:  
AC03-99ER54463
Resource Type:
Conference
Resource Relation:
Conference: 27th International Conference on Infrared and Millimeter Waves, San Diego, CA (US), 09/22/2002--09/26/2002; Other Information: THIS IS A PREPRINT OF AN INVITED PAPER TO BE PRESENTED AT THE 27TH INTERNATIONAL CONFERENCE ON INFRARED AND MILLIMETER WAVES, SEPTEMBER 22-26,2002, SAN DIEGO, CALIFORNIA, AND TO BE PUBLISHED IN THE ''PROCEEDINGS''; PBD: 1 Sep 2002
Country of Publication:
United States
Language:
English
Subject:
24 POWER TRANSMISSION AND DISTRIBUTION; BEAM SPLITTING; COPPER; CROSS SECTIONS; HEAT SINKS; MICROWAVE AMPLIFIERS; MIRRORS; PERFORMANCE; PHASE VELOCITY; POWER TRANSMISSION; QUARTZ; RADIATIONS; THERMAL CONDUCTIVITY; WAVEGUIDES

Citation Formats

MOELLER, C P, LOHR, J, and DOANE, J L. THE MEASURED PERFORMANCE OF A MILLIMETER WAVE BEAM SPLITTER. United States: N. p., 2002. Web.
MOELLER, C P, LOHR, J, & DOANE, J L. THE MEASURED PERFORMANCE OF A MILLIMETER WAVE BEAM SPLITTER. United States.
MOELLER, C P, LOHR, J, and DOANE, J L. Sun . "THE MEASURED PERFORMANCE OF A MILLIMETER WAVE BEAM SPLITTER". United States. https://www.osti.gov/servlets/purl/804737.
@article{osti_804737,
title = {THE MEASURED PERFORMANCE OF A MILLIMETER WAVE BEAM SPLITTER},
author = {MOELLER, C P and LOHR, J and DOANE, J L},
abstractNote = {An essential component of any high power transmission system is a directional coupler that provides a sample of the forward and reflected power when this power is being delivered to the intended load. In the case of millimeter power delivered through a highly oversized corrugated waveguide, there is the much more complex issue of mode purity. It is possible to design an effective mode selective branch guide directional coupler in smooth wall overmoded waveguide. In the typical highly overmoded corrugated waveguide propagating the HE{sub 11} mode, however, obtaining an adequate coupling factor can be difficult, and branch guide attenuation and phase velocity matching over several meters become concerns. A more practical approach for large diameter corrugated waveguide is to obtain a sample of the propagating beam at a miter bend mirror. At low power, the mirror could be a thin metal screen. At the megawatt level, however, heat removal must be considered. For example, at 110 GHz at 1 MW, taking the surface resistance of copper to be 0.10 {Omega}, the dissipation on a 45{sup o} copper mirror would be 750 W or 1500 W for H or E plane reflection, respectively. With a peak to average power ratio of 3.7 for the circular HE{sub 11} mode, in 31.75 cm diameter corrugated waveguide the peak dissipation can be as high as 500 W/cm{sup 2} at the center of the mirror. An edge cooled thin metal screen is not therefore practical, but a thick plate containing a single narrow channel, at the bottom of which is a row of holes in the remaining thin wall, can be adequately water-cooled on its face. To maintain vacuum and focus the radiation from the holes, the narrow channel is filled by a fused quartz plate, the shape of which is a 45{sup o} sector of a circle having a truncated apex at the coupling holes. These are being used as power monitors on the DIII-D ECH system and on other systems. Since this single row of holes samples only part of the wave field, however, interference among higher order modes, even when they are of low amplitude, can produce noticeable variations in received power, depending on the exact gyrotron frequency and temperature of the waveguides. The only certain way to avoid this problem is to use a true beam splitter that provides a sample of the fields in the entire cross section of the waveguide. Our proposed means of overcoming the heat removal problem is to make the mirror an edge cooled CVD diamond disk having a very thin copper coating on the high power side of the mirror. A two dimensional array of holes is etched through the copper. An image of the incident wave free of aliasing is launched on the opposite side of the copper film. This process is easily scaled to frequencies greater than 170 GHz. We have analyzed the thermal response of such a plate bonded to water-cooled heat sinks on the periphery of both faces. In our case, the heat sinks have 32 mm by 45 mm apertures. The gray diamond we are using is 60 mm diameter with two flats 48 mm apart. For a diamond 1.1 mm thick having an assumed thermal conductivity of 1300 W/m-K and heated with a total of 1500 W (times a correction for the resistivity increase with temperature) on one face with the circular HE11 distribution, the steady state temperature at the center is 210 C, starting at 35 C. The temperature reaches steady state by 1 s, while the 1/e time is about 160 ms. This temperature is slightly higher than the steady state gyrotron window temperature. If necessary we can reduce the peak temperature, at greater cost, by using a thicker gray diamond, say 1.5 mm, or by using a white diamond 1.1 mm thick having 1800 W/m-K conductivity. Either would bring the final central temperature down to <160 C. Since we are using edge cooling, this result is independent of waveguide diameter, so this device can be made for any diameter.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2002},
month = {9}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share: