Progress on the C Mod FIR PolarimeterProgress on the C-Mod FIR PolarimeterSystem

P. Xu1, J.H. Irby1, W. F. Bergerson2, D.L. Brower2, W.X. Ding2, S Shi i 1 S W lf 1S. Shiraiwa1, S. Wolfe1

1 MIT Plasma Science and Fusion Center, Cambridge, MA2 University of California, Los Angeles, CAy , g ,

APS DPP Nov.8-12,2010 1*Supported by USDoE award DE-FC02-99ER54512

AbstractAbstract

• A poloidally viewing FIR polarimetry diagnostic is being developed for the Alcator C-Mod Tokamak. The primary diagnostic components are a two-wave FIR laser at 117.73 microns and newly developed detectors whose performance characteristics will be described. Faraday rotation will be used both to refine the q-profile measurement by adding constraints to EFIT , and to study density and magnetic field fluctuations. A three-chord system y y g yhas been installed, one chord of which is being tested during the FY10 C-Mod campaign. The FIR laser source is affected by both stray magnetic fields and mechanical vibrations present in the experimental cell therebyfields and mechanical vibrations present in the experimental cell thereby impacting the measurement. Methods developed to mitigate and correct for these effects will be discussed. Initial Faraday data will be compared with expectations from numerical simulationexpectations from numerical simulation.

APS DPP Nov.8-12,2010 2

OutlineOutline

• Motivation• Polarimetry Theory• System Geometry and Critical Components• Initial Measurement Results• Summary and Plans

APS DPP Nov.8-12,2010 3

MotivationMotivation

• Detailed information about the plasma current profile is of great importance to C-Mod lower hybrid current drive program

• A Far-Infrared (FIR), multi-chord polarimetry system will provide direct measurement of the internal magnetic field structure and its temporal evolution and will compliment MSE measurements on C Modevolution and will compliment MSE measurements on C-Mod

• Diagnostic geometry and plasma parameters similar to those expected on g g y p p pITER will provide an excellent environment for system optimization

L IF b d id h f d i ill ll f• Large IF bandwidth of detection system will allow measurements of magnetic and density fluctuations

APS DPP Nov.8-12,2010 4

Polarimetry TheoryPolarimetry TheoryNon-perturbative diagnostic to measure the internal magnetic field of hot plasmaThe polarization of a laser beam passing through a magnetized plasma will be changed by both the Faraday rotation and Cotton-Mouton (CM) effects.

∫∫With density profile from Interferometer or Thomson scattering, the Faraday rotation integral can be either directly inverted to obtain magnetic field

11 3 2[deg] 4.84 10 [ ]en B dl SIε λ−⊥= × ∫13 2

//[deg] 5.24 10 [ ]en B dl SIα λ−= × ∫

y g y gor used to constrain EFIT, thereby improving the accuracy of magnetic field and current density profile reconstruction.

CM ff t th F d t ti t C M d ( JET dCM affects the Faraday rotation measurement on C-Mod (same as on JET and ITER). Numerical simulation for C-Mod polarimter system shows similar level of Faraday rotation and CM, around 10 degrees depending on the plasma parameters and measurement location.

Standard Counter Rotation Beam method (Dodel/Kunz technique) is used to measure the Faraday rotation. This method is not susceptible to CM effect.

5APS DPP Nov.8-12,2010

GeometryRetros-reflectors in inner-wall

Upper table

Poloidally viewing geometry: laserPoloidally viewing geometry: laser beams pass through the K horizontal port window

FIR In vessel retro-reflectors in inner-wall provide a double pass measurement

6APS DPP Nov.8-12,2010

GeometryGeometry

Three poloidally viewing chords installed, one of which is working at this time Retro-reflector positions onthis time. Retro-reflector positions on inner wall : z1=3.9cm, z2=17.5cm, z3=39.2cm.Upper chord is well outside the separatrix to provide zero rotationseparatrix to provide zero rotation boundary condition and estimate of noise levelLower chord is around the maximum Farada rotation angle position th sFaraday rotation angle position, thus gives us best signal.6 retro-reflectors installed now, and will have 10 after the next up to air,

f hi h ill h th ti

7

some of which will reach the magnetic axis

APS DPP Nov.8-12,2010

FIR Lasers and Detectors

Two CO2 pumped FIR lasers THz Schottky barrier diode detectorsTwo CO2 pumped FIR laserswavelength is 117.73μm

about 150 mW power output per laser (CW)

Corner cube mixers from FarranTechnology (left) and UCLA (middle)Waveguide coupling mixer from Radiometer-Physics (right)Radiometer-Physics (right)New VDI detectors with planar diode technology have been developed with higher sensitivity, which will allow for more chords (ARRA)

8

more chords (ARRA)

APS DPP Nov.8-12,2010

Air-tight Enclosures

Air-tight enclosures seal the optics and significantly reduce the laser power loss from water vapor absorptionwater vapor absorption.

Sound proofing material to reduce the acoustic noise

P l d ibilit f ti l9

Panels - good accessibility for optical adjustments, feedthroughs for power, control and signals

Magnetic ShieldingDuring run 2010, the lasers were not running stably during plasma discharges. The laser instability clearly correlates with the amplitude of magnetic field around the lasers probably caused by the Invar plate and rods inside the laser cavity.Shielding part of the lasers with iron plate improved but did not solve the problem.A shielding structure covering the whole lasers has been installed and will be tested over the next few weeks

We are working with the laser company to replace Invar with carbon fiber as an ultimate solution (Jan 2011).

10APS DPP Nov.8-12,2010

Faraday Rotation Measurement w/o Plasma

Dodel/Kunz technique is used to measure the Faraday rotation anglemeasure the Faraday rotation angleThe standard deviation of Faraday rotation measurements are as low as 0.10 without plasma dischargeDominant vibration induced noise appears at ~30 Hz and ~240 Hz.The FIR lasers are very sensitive to the vibration, and contribute strongly to the , g y240 Hz noise.We suspect a mirror mount defining the laser cavity length is causing the vibration in this frequency range (also fixvibration in this frequency range (also fix Jan 2011)Cables and pumping lines, directly connecting to the laser housing, have been wrapped with vibration dampingbeen wrapped with vibration damping pads and attached to the table to reduce the low frequency vibrations.

11APS DPP Nov.8-12,2010

CalibrationCalibrationA half-wave plate was placedA half wave plate was placed before the wire-grid beam splitter and rotated by stepping motor to simulate the Faradaymeasured motor to simulate the Faraday rotation expected from plasmaThe difference between the measured and theoretical Faraday rotation data is about 20%, which is possibly due to theoretical p ythe polarization dependence of the reflection coefficient of the copper mesh beam splitter orcopper mesh beam splitter or other optical components.

12

Faraday Rotation Measurement during Plasma Di hDischarge

FIR lasers were unstable during discharges during our first operation of the FIR system. Large variations in both the laser power and wavelength were observed.

Laser Intermediate

Frequency(IF)

DetectorP

Efforts to improve the laser stability

Power

Efforts to improve the laser stabilitybreak the grounding loops between laser controller, PLC and digitizer, by adding

isolation amplifiers.Covered the FIR lasers at critical locations with up to 0.5’’ thick iron platesp pbuilt electronics (frequency meter and integrator ) to provide feedback control of the laser

difference frequency. 13

Measurement & Simulation Following System Improvements(chord z1=3.9cm)

Laser lF

Detector signal level

Radial magnetic field

Measured FaradayRotation(black)Rotation(black)

Simulation(red)

D itDensity

14

Plasma Current

Measurement & Simulation (chord z1=3 9cm)Measurement & Simulation (chord z1=3.9cm) Laser IF and detector power signals are much more stable with iron plate shielding and feedback electronics while there are extra vibrations causedshielding and feedback electronics, while there are extra vibrations caused by iron plates.Detector power is observed highly correlated with the radial magnetic field strength measured around the lasers.

egre

es)

Measurement

FRO

T(de Measurement

Simulation

Time(s)

F

APS DPP Nov.8-12,2010 15

Measurement & Simulation (chord z1=3 9cm)Measurement & Simulation (chord z1=3.9cm)

The measured Faraday rotation signal shows ~2-3 degrees of error from -1s, y g gwhen there is no plasma but a rapidly increasing magnetic field.

E t d F d t ti i i l t d b EFIT d TS l t d itExpected Faraday rotation is simulated by EFIT and TS electron density. Measurement and simulation agree at the 10% level.

Error Sources need to be eliminated before attempting any physics studies.Phase error caused by IF instability

∆α~ π/c *∆IF*(Lsig-Lref)The calibration must still be applied to the system, which may affect the discrepancy between simulation and measured data. p y1. Phase calibration to remove the affect of polarization-sensitive optics2. Alignment calibration to improve the colinearity of two FIR laser bbeams.

APS DPP Nov.8-12,2010 16

Measurement & Simulation (chord z=17.5cm) Laser lF

Detector signal level

Measured FaradayRotation(black)

Simulation(red)

Density

Plasma Current

17APS DPP Nov.8-12,2010

Measurement & Simulation (chord z=17 5cm)Measurement & Simulation (chord z=17.5cm) es

)T(

degr

ee

Measurement

Simulation

FRO

T

Time(s)

APS DPP Nov.8-12,2010 18

Simulation Results with low density LHCD (1080320013)y ( )

Data from 2008 when significant current was driven by the Lower Hybridcurrent was driven by the Lower Hybrid system has been used to simulate what would have been seen by the polarimeterpolarimeter

Large Faraday rotation angle variation are indicated during low electron density LHCD discharge fromelectron density LHCD discharge from simulation based on MSE constrained EFIT and TS

Small change in densityg yLow density discharges in FY2010

indicate sustained LH current drive that should be easily measured with the ypolarimeter

No obvious rotation angle variation seen during high density LHCD during

t/s19

FY2010 campaign, from both measurement and simulation.APS DPP Nov.8-12,2010

SummarySummary

A three chord polarimeter system has been installed on C-Modp yWell designed enclosures keep laser beam path in low humidity environment to avoid large water vapor absorption. Sound proofing material reduces the acoustic noise in the system.p g yNew magnetic shielding structure designed and installed for FIR lasers.

A mock up test w/o plasma has shown ~0 10 uncertainty in the rotationA mock-up test w/o plasma has shown ~0.1 uncertainty in the rotation angle (1ms temporal resolution)

I i i l i h l di h h d F dInitial measurements with plasma discharge show expected Faraday rotation with errors at the 10% level. We will continue to investigate the sources of noise in the system following improvements to the FIR laser t bilit bl l dstability problems are resolved.

20APS DPP Nov.8-12,2010

PlansPlans

Improve laser stability during plasma dischargesTest polarimeter measurement for low density LHCD plasma discharges, in which significant Faraday rotation angle variation is expectedwhich significant Faraday rotation angle variation is expectedUpgrade the system to run 3 chords together with the higher sensitivity VDI detectorsIntegrate Faraday rotation data into EFIT to provide additional constraint.Continue strong collaboration with UCLA on development of the polarimeter and fluctuation measurementsand fluctuation measurements

21APS DPP Nov.8-12,2010