sign in sign up
<<
Home , Gyrotron

28 GHz Gyrotron and 4.6 GHz Heating Options for LDX Columbia University P. P. Woskov1, A. C. Boxer1, J. L. Ellsworth1, D. T. Garnier2, J. Kesner1, M. E. Mauel2, P. C. Michael1 1Plasma Science and Fusion Center, MIT, Cambridge, MA, 02139 USA, 2Department of Applied Physics, Columbia University, New York, NY 10027 USA

ABSTRACT - The success of fully levitated LDX 28 GHz Gyrotron Heating 4.6 GHz Klystron Heating operations is now motivating increasing the plasma heating to explore the full potential of high beta plasma stability 100 meter TE11 Circular Aluminum Transmission Line A 100 m in a dipole magnetic configuration. Adding 28 GHz gyrotron B B A and 4.6 GHz klystron heating using sources available at MIT is being evaluated. A Varian model VGA- 10 cm i.d. 15 m WR-187 ~15 m 8050M 28 GHz gyrotron that is capable of 200 kW in 0.1 s WR-187 A: WR-187 to Circular Transition pulses or 10 kW CW is on site, and the output of one 250 kW, B: Conical Taper 5.7 to 10.0 cm i.d. 0.5 s klystron at the adjacent C-Mod site could be transmitted 4.6 GHz LDX Kystrons to LDX. Adding these sources would significantly increase One 250 kW klystron maximum heating power above the present 15 kW maximum would be used to access higher beta plasmas. Heating at 28 GHz in for LDX experiments particular, would be above cutoff for plasma densities up to the 1013 cm-3 range which are presently not accessible with the current low power 2.45 to 10.5 GHz heating sources. Overmoded would be used for both 28 and 4.6 4.6 GHz X-mode launched GHz to efficiently transmit the 100+ kW power levels into LDX perpendicular to the to heat at the electron cyclotron fundamental and harmonics. from the outside mid-plane would be directed at the electron Introduction cyclotron resonance at the fundamental and High-power heating sources available at MIT harmonics. could be implemented with modest effort to increase LDX 3 MW, 12 Klystron Bank at C-Mod One possible geometry for 28 GHz X-mode launched heating plasma density and pressure • Overmoded circular TE for main 100 m run to reduce loss perpendicular to the magnetic field. The heating beam would be directed Rectangular TE / Circular TE 11 01 11 • Fundamental rectangular WR-187 to negotiate bends at input and output So far successful levitated plasmas have been achieved at the fundamental and 2nd harmonic electron cyclotron resonances. Transmission Line Attenuation • Aluminum tubing is proposed for the 100 m run to reduce costs with a maximum of 15 kW of microwave heating power Circular TE Transmission Line Attenuation 02 7 - 2.45 GHz magnetron, 2.5 kW → Al conductivity 2.50 x 10 mhos/m The gyrotron output is mainly in the circular TE02 mode. A mode filter or converter - 6.4 GHz klystron, 2.5 kW will be used (not shown) if necessary to achieve a pure mode for transmission. Simple Rectangular to Circular Transition 100 mm dia. Al wg -0.004 dB/m 100 m -0.4 dB - 10.5 GHz klystron, 10 kW http://www.ham-radio.com/sbms/techpapers/K6BLG/circ_rect.html WR-187 -0.02 dB/m 30 m -0.6 dB Assuming copper waveguide with a conductivity of 5.88 x 107 mhos/m, the TE b r • If another 0.5 dB loss is allowed for a total loss of 1.5 dB, then it may 02 TE01 TE11 A 200 kW, 28 GHz gyrotron is currently located adjacent to mode loss at 28 GHz for the transmission line components shown above are: a be possible to transmit up to 175 kW from one klystron to LDX LDX as shown above right, but needs to be made Homogenous Transition, Cutoff Wavelengths Equal operational 63 mm diameter -0.004 dB/m 3.5 m -0.014 dB 2π r 1.8114 λ ==2a ra= Parabolic Circular Taper - Requires high voltage power for the 38 mm diameter -0.021 dB/m 5.0 m -0.105 dB c Two bends 38 mm dia. -0.042 dB/m 3.2 m -0.134 dB 1.8114 π J. L. Doane, Inter. J. Infrared & Millimeter Waves, 5, 737-751, 1984 • 80 kV, 8.0 A for 200 kW, 0.1s pulses WR-187, 47.55 x 22.15 mm 55.73 mm dia. • 40 kV, 0.87 A for 10 kW, CW Taper Lengths for which Mode Conversion Vanishes The total ohmic transmission losses are calculated to be about 5% (-0.25 dB). It - A total of 1960 A current for 4 main coil and 2 gun Quarter Wavelength Matching Section 1 is reasonable to expect a transmission efficiency to the plasma of 90% allowing 22 2 2 coil magnets Ka()−λ a⎡⎛⎞⎤ another 5% losses for the window and taper. ' Power-Voltage Impedance o 21 nπ a1, a2 input and output radii, - Cooling water connections for magnets, gyrotron Z =⋅ZZin out Ln =−⎢⎜⎟1⎥ Matching Condition ⎜⎟a1 = 27.8 mm, a2 = 50 mm collector, and body PKaaoo⎢⎝⎠ln ()21 ⎥ Many of the 28 GHz transmission components are available or easily fabricated. 2b 2r 2b' ⎣ ⎦ Z ==0.932 Z = = 2 Z ' ==1.365 in a out r a' for TE −TM conversion Ko = -0.65, Po = 0.90 A 3 MW, 4.6 GHz 12 klystron system is located a little over Valsov Launcher, Circular Guide 11 11 A taper length as short a 16 cm (n=1) 100 m away at the C-Mod site ⎛⎞k 28 GHz ' r a’ = 51.5 mm α = sin−1 cn bbr=⋅ b’ 3 4.6 GHz, 57 - 100 mm Dia. could be used but a length of 32 cm ⎜⎟ TE02 b’ = 35.2 mm -An efficient waveguide could bring power from one ⎝⎠k Parabolic Taper (n= 2) would be better because TE a’ 2 12 250 kW klystron to LDX ν ' k = 0 n k = 2π mode conversion loses (not shown) 28 GHz TE cn a λ 02 1 would be lower at longer length. Other ' Transition Length, 4.6 GHz n=1 38 mm dia. ν 0n root of J’ (ν)=0 possible modes for conversion from The possible implementation of these additional sources of 0( n=2 a, waveguide radius L λ 95% of quarter 1 (%) Conversion Mode 0 TE at 4.6 GHz do not propagate in corrugated λ λ 11 microwave heating to significantly increase LDX plasma TE ν’ = 7.016 oc λg = 22.37 mm 10 20 30 tapered 02, 02 = guide wavelength 100 mm diameter g 4 Length (cm) performance are described and details of the waveguide radius bend a =17.5 mm λ 22− λ to account for 38° co L = 21.25 mm 22 α shunt capacitance tran 2 F =−aa2 L transmission lines given L = 2acosα 30 mm Parabolic Profile az()=+ a1 2 Fz ( 21)

137 GHz ECE Measurements • At 137 GHz 500 eV equivalent black Radiometer calibrated to window 137 GHz ECE LDX Shot # 810010018 137 GHz ECE LDX Shot # 810010014 body signals are observed at the LDX • A 137 GHz radiometer was installed to view ECE directed Ceramic waveguide with liquid nitrogen black body 600 450 bottom window vertically downward from the high field region of LDX 400 Non linearity of microwave IF 500 500 eV equivalent • This is a lower limit for the actual - max. LDX field is 3.25 Tesla on inside F-Coil 350 350 eV equivalent detector taken into account black body signal emission at the plasma since the plasma - only harmonic emission observed at 137 GHz 400 300 black body signal 137 GHz radiometer by: does not completely fill the view of the 250 radiometer • Radiometer noise temp. 3000 K with 3 GHz double side band 300 Levitated KPin=+ V out(1 α V out ) Supported 35 mm dia. 1.1 m long ceramic waveguide from radiometer to 200 • 200 • In F-coil supported discharges the 2.45 fused quartz window for HE mode launched field-of-view 150 and 6.4 GHz sources do not produce 11 K = 59.72 mV/eV 100 10 kW 100 10 kW significant hot in view of the

- 1.9 dB insertion loss Effective Window Temperature (eV) from LN calibration at window 10.5 GHz Effective Window Temperature (eV) 10.5 GHz 137 GHz radiometer View of 137 GHz radiometer setup on 2 50 - another 7.0 dB attenuator inserted to prevent 0 floor below LDX. A short run of Al 2.5 kW 0 2.5 kW radiometer saturation after initial observations α= 0.01 mV/mV 6.4 GHz 6.4 GHz • The presence of high energy electrons corrugated waveguide with a miter bend -100 -50 2 for typical HP-33330 zero bias -5 0 5 10 15 20 25 -5 0 5 10 15 20 25 would make possible harmonic ECE • Field of view at plasma ~0.3 m 1/e dia. encompassing >12 connects to the ceramic waveguide Time (sec) Time (sec) detector used here heating at above 100 GHz harm. partially obstructed by F-Coil catcher (not shown) 2.5 kW, 2.45 GHz 2.5 kW, 2.45 GHz