Design of multipulse Thomson scattering diagnostic for SST-1 tokamak

Design of Multi-Pulse Thomson Scattering Diagnostic for SST-1 Tokamak

Ajai Kumar, C. Ramdas, Chhaya Chavada, Y.C. Saxena, R. Rajesh,B. R.Kumar, S. Sunil, A. Thaker, R. Singh, V. Aravind, & V. Chaudhari

Institute For Plasma ResearchBhat, Gandhinagar 382 428, India

ABSTRACT

This paper describes the design and subsystem testing of vertical, horizontal and divertor

Thomson scattering systems for SST-1 tokamak. These Thomson scattering systems will use

multiple 30 Hz Nd:YAG lasers to measure the electron temperature and density profile

periodically throughout, the 1000 sec plasma discharge of SST-1. The system is designed for

high spatial resolution (1.0 cm) and wide dynamic range (20ev –3 keV).

The modular design of the vertical Thomson scattering diagnostic system on SST-1 has

allowed simultaneous measurements in the main plasma and in the divertor region using

same laser. Three different optical imaging lens systems, with magnification of 0.2, are

designed to image scattering volume on to the linear array of optical fibers to provide spatial

resolution of 1.0 cm. The scattered photons are dispersed by two different sets of five channel

interference filter polychromators to cover the dynamic range of divertor, SOL, and main

plasma. The scattered light will be detected by cooled Si-APD.

Laser control and data acquisition will be performed in real time a by PXI based system

throughout the plasma discharge. Initially data analysis will be done after the plasma shot, but

later on it will be done in real time.

Results of subsystem testing like, laser beam-clustering concept, laser beam transport,

imaging parameters of filter polychromators, detector electronics are discussed.

1. INTRODUCTION

SST-1 tokamak is a large aspect ratio tokamak with elongation of 1.7-1.9, triangularity of

0.4-0.7. It is designed for double null configuration. The machine is configured with super

conducting magnets (both poloidal and toroidal) to achieve plasma of 1000 sec duration. The

major and minor radii of the machine are 1.1 m and 0.2 m respectively. Other details of SST-

1 tokamak are given in ref. [1].

Thomson scattering has long been a standard diagnostic for measuring the electron

temperature and density in tokamak [2-7]. For temporal evolution of the plasma, it is

important to understand the processes such as L-H transition, ELM’s, beta limits, disruption,

activity of other modes, and rapid density increase due to injection of pellets etc. To analyze

these plasma phenomena, a high time resolution is required. It is cost effective to use a multi-

pulse laser system to study all the events, when shot to shot reproducibility is difficult. Due to

dynamic behavior of the plasma, during the heat up phase, the nested magnetic flux surface

moves outwards to compensate for the increasing thermal pressure. Then the hot magnetic

plasma center moves out of the range of the vertical laser beam and hence will no longer be

covered by the measuring system, whose geometry remains fixed. It is not possible to send

laser beam through the central vertical plasma chord of SST-1. To analyze the entire plasma

cross-section, one has to enlarge the number of observation points.

For the temporal evolution of the electron temperature (Te) and density (ne) profiles over

1000 sec of SST-1 plasma duration, a multi-point, and multi-pulse 1. Vertical Thomson

scattering System (VTS), 2. Horizontal tangential Thomson scattering System (HTS) and 3.

Divertor Thomson scattering System (DTS) were designed. These systems together will cover

a parameter range of, ne ≥ 1 x 1012 cm-3 and Te: 20eV to 3.0 keV. Practical issues like space

availability, machine interface, accessibility, and simplicity of alignment, cost effectiveness,

long-term stability and reliability etc were important considerations for the design.

2. Laser Source And Beam Transport Line

2.1 Laser Source

As per signal estimation and analysis, for the measurements of low electron density, ~ 9J of

Nd:YAG laser energy is required. The present design uses multiple lasers. These laser beams

are combined into a common beam path. Low divergence lasers are required to minimize the

laser beam waist inside the plasma to maximize the signal to noise ratio. We have selected a

Nd:YAG laser ( 6 total, Continuum Model Powerlite 9030) which produces; 1.6 J, 10-ns, 30

Hz laser pulse with divergence of <500 µrad. The laser beams are packed together along a

common beam path. These beams will partially overlap in the far field in side SST-1 vacuum

vessel. Configuration of six Nd:YAG lasers permit the following mode of laser firing:

1. 1.6J per pulse at 180 Hz (Low energy high repetition rate mode)

2. 9.6J per pulse at 30 Hz (High energy low repetition rate mode)

3. 1.6J per pulse at ~3KHz triggered at given plasma event. (Burst mode)

For measuring the plasma events, laser-firing rate will be changed from 30Hz to 180 or 3 kHz

before the event occurs. After the event, laser firing rate will changed to 30 Hz rate. A PXI

based, in-house developed, time sequencing module will be used for this purpose.

For the safe operation of these lasers, against the EMI and RF fields of the SST-1 machine,

these lasers are kept in a separate laser diagnostics room. All the six lasers are placed on a

vibration free table.

2.2 Laser Beam Packing

The laser beam-packing [8] scheme is shown schematically in fig. 1. In the present scheme,

a right-angle prism with high energetic dielectric mirror coating on their sides for packing

Fig. 1. Top inside view showing six-channel laser beam-packing scheme and its cross-section

at the end of laser table

laser beams are used. All the six lasers are arranged on a vibration free table such that each

pair of lasers will face each other and with a difference in height. The prisms are mounted on

precision tilt and rotation stages. The beams from each pair of lasers will be directed towards

the common path by the 45o surfaces of the prism such that the beam pack at the end of the

table will be a 2x3 matrix. The beam pack passes from the laser table and traverses

approximately 28 m to the focusing lens. To reduce the size of beam transport optics, top and

bottom pairs of beams are tilted such that these pairs of beams coincide with the middle pair

at the lens position. By overlapping the beams at focusing lens, the focus point of the

individual beams will be positioned behind each other. This way collection lens will see three

sets of overlapping beams. This arrangement also simplifies the alignment.

A proto-type experimental set (fig. 2) was assembled to test and develop the alignment

procedure for packing and transporting laser beams over 30 meter of distance. This set up

makes use of 4 He-Ne lasers of divergence 500 ìrad. The center-to-center distance of 10 mm

Fig. 2 : A proto-type experimental setup for the laser beam packing scheme

between all the 4 beams were maintained. At a distance of 30m, a 4m focal length lens was

mounted in the beam path to measure the beam position and their overlap at different distance

from lens position. These measurements were made by Spiricon beam profiler and visible on

paper.

2.3 Beam Transport Line

The beam transport line for all the three Thomson scattering system is common upto the

machine location. The beam pack from the laser table will enter the SST-1 hall through a

window, then the beam will be folded towards the machine with the help of beam folding

mirrors. These folding mirrors, along with their gymbal mount and translation stages are

mounted on sub-hertz vibration isolation platforms. To avoid human hazards and loss due to

scattering from water molecules and dust during the free air transportation, the beam passes

through a beam transportation line maintained at 10-2 Torr of pressure.

To test laser beam transport over 30m of distance with the use of 5 beam folding mirrors,

beam stability, beam profile, beam size at different locations in the focal plane of lens, loss of

beam energy upto 30 meter and to set up the alignment procedure for beam transport, a test

stand was designed and fabricated (fig. 3). The measurements were made by beam profiler,

Fig. 3. A test stand for laser beam transport, optimization of beam defining aperture size and

to measure focused beam parameters

burn paper and energy meter. It is observed that for beam size in the focal plane simple

expression, wo ~ F tanè ~ Fè, holds good.

3. Vertical Thomson Scattering System (VTS)

The vertical Thomson scattering system provides the plasma parameter profiles along the

vertical chord from Z = - 24 cm to +24 cm through R=107 cm, which is 3 cm off to plasma

center. Measurements will be made from 19 spatial points, each corresponds to 1.0 cm in

height at ~ 1.5 cm interval. In the present scheme, the pack of beam enters the vessel through

the bottom vertical port (Port BV9) and exits through the top vertical port (Port TV9). The

near 90o scattered photons will be collected through the radial port (Port R9) at the same

toroidal plane.

The beam pack transported by the common beam line optics up to machine foundation will

be diverted for the vertical Thomson scattering system by a set of beam steering mirrors. The

beam pack will be focused to the chord center by 4 m focal length plano-convex lens. The

beam pack will pass through a 150 mm dia AR coated fused silica window located at the end

of a drift tube attached to the port flange (fig. 4). To reduce stray light level, beam passes

through a set of four concentric light baffles in side drift tube. The expected size of the beam

pack cross section at the focal plane will be 2x 8 mm2 (for the 1/e2 energy) and 5x13 mm2

(for the 2.5/e2 energy).

Fig. 4. Schematic layout of vertical Thomson scattering diagnostics showing the laser beam

propagation from the table through vacuum window, set of baffles, SST vacuum

vessel and its exit to beam dump

After passing through the desired chord of the plasma, the laser beam pack exit through a

Brewster window attached at the end of a similar drift tube used at the entrance side. After

exit from the drift tube, beam is dumped on V shaped carbon block with the help of 450

mirror. This mirror is used to avoid the dust accumulation on optical components. To

withstand the average power, this carbon block will be water-cooled. The ablated particles

from this block will be removed by forced air.

The scattered photons from 480 mm long plasma chord are collected and imaged by a

(designed) three lens imaging system. The designed parameters of this lens system are; image

size: 96mm, image surface radius of curvature: 120mm, magnification: 0.2, solid angle: 2x

10-2 sr, scattering angle: 760 to 116o. The fig. 5 shows the schematic of ray-tracing diagram

Fig. 5. Ray tracing diagram for VTS imaging lens system for 0.2 magnification

for this lens system. The collection lens images the plasma chord on linear array of 19 optical

fibers (2mm core diameter and NA 0.22). These fibers are low OH all silica 35-40 meter

long, having transmission >98% between 700-1100 nm range. These fibers are coupled to

fiber holder block having 120mm. radius of curvature. Lemo connectors are connected to the

other end of fiber, which will be coupled to five channel filter polychromators

A viewing dump made of graphite tiles, having 15o serrations of 10mm deep will be

mounted at the in-board side of the vessel wall to view against black background. To avoid

the coating during wall conditioning, a mechanical shutter is provided in front of the

collection windows. For an in-situ alignment of the collection optics, a cylindrical piece with

a semi-spherical projection at its center will be dropped collinear to the laser chord, from the

exit side of drift tube.

4. Divertor Thomson Scattering System (DTS)

The VTS laser beam [9,10] after it passes through the plasma will be reflected back for X-

point measurement in divertor region, before it is dumped. To avoid the interference with the

vertical Thomson scattering system, the exit beam from VTS will be shifted in toroidal plane

by 20. This provides the simultaneous measurements of plasma parameters from VTS and top

divertor region. The graphite beam dump of VTS will be removed for the operation of DTS

operation. The laser beam will enter the vessel through a drift tube, having a set of 4

concentric beam define baffles, fixed on side of trapezium box (fig. 6). The beam will pass

through the divertor plasma almost parallel to the field line from null point to the divertor

plate. The laser beam will be dumped inside the vessel on a beam dump machined in

Rayleigh horn configuration

Fig. 6. Schematic layout of DTS diagnostics showing reflection of VTS beam for DTS

purpose and its passage from window, apertures, to beam dump inside vacuum vessel

The scattered photons from the diverter region will be imaged by a three lens system on a

linear array of 8 all silica optical fibers (core dia: 2 mm). The magnification for this lens

system is 0.2. The scattered photons from 8 spatial points distributed over a plasma chord of

~ 15cm long with 1.0 cm resolution will be collected through the same viewing window of

VTS. The other end of the fiber will be coupled to polychromator with filter set for low

temperature measurement.

5. Horizontal Thomson scattering System (HTS)

The horizontal tangential Thomson scattering (HTS) system [11] will give the plasma

parameter profiles along the radial chord from a = -20 cm to + 6 cm at Z = 0 plane.

Measurements will be made from 10 spatial points from the core with 1cm resolution and 6

spatial points from the edge with 0.5 cm resolution.

The beam pack is transported from the common beam line optics to the drift tube with the

help of beam steering mirrors (fig.7). A set of 6 concentric beam-defining baffles is mounted

Fig. 7. Schematic layout of HTS diagnostics showing tangential laser beam propagation,

through vacuum window, set of apertures and by beam deviating optics to dump. The

collection view is also illustrated.

in this drift tube, which is mounted at a tangential angle to plasma. In order to take the beam

out of the machine, a beam deviating optics is designed with the help of two right-angled

prisms (fig.8) For the cleaning purpose, these prisms can be taken out by a translator system.

Laser beam pack is focused by 4m lens near to a = - 12 cm, in similar fashion as for VTS.

Fig. 8. Schematic lay-out of HTS beam deviating optics

The scattered photons from the HTS region will be imaged by two different imaging lens

systems. One of the lens systems has a magnification of 0.4 where as other one has 0.1. The

scattered photons from 10 (from core plasma) and 6 (from edge plasma) spatial positions

will be imaged on the 2mm core dia all silica optical fiber liner array. The other end of the

fiber will be coupled onto a filter polychromator.

6. Laser Beam And Collection Fiber Optics Alignment

For accurate electron density profile measurements, an alignment stability between laser

beam and object fields of collection fiber optics is of essential importance in Thomson

scattering systems [3,12]. An arrangement of narrow beam width and minimum object field

width as narrow as beam width for improving the signal / noise ratio of Thomson scattering

light to plasma light is one of the key issues to maximize the precision.

Three different alignment systems are worked out and are in testing stage. 1. A He-Ne laser

is made collinear with the Nd:YAG laser beam. The beam alignment will be made by using

three computer-controlled mirror gimbals located at floor pit and at the exit of Brewster

window. The alignment will be achieved by adjusting the center of leakage laser light to the

optimized position of each target screen by monitoring the digitized images from the three

corresponding CCD cameras. 2. A portion of focused light, diverted by beam splitter, will be

sampled. It will be monitored by linear array photodiode camera working at 1.5 MHz rate to

get the information about the change in laser divergence/ pointing stability. This information

can identify the number of poor quality laser pulses during plasma discharge. 3. The

alignment between object field of collection fiber optics and the laser beam will be monitored

by four optical fiber bundles (size 3x3 mm, single fiber core dia 250µm) mounted at image

plane of collection lens system. The other end of this fiber bundle will be coupled to

detectors. With the help of four different outputs, the parallel and rotational displacement of

the object field of the collection fiber with respect to laser beam in side the vacuum vessel

will be calculated. From the calculated value, misalignment of collection fiber optics holder

will be corrected by servo-motor controlled feedback system to the misalignment tolerance of

< ± 1.0 mm.

7. Filter Polychromators

The scattered light carried by the fibers is fed to a five channel interference filter

polychromators for the dispersion into different wavelength bands [2,5,13]. In the present

polychromators design, optimization is made between size and filter performance in terms of

angle of incidence and the cone angle of light. Stability of alignment, maintenance, optical

component size, good laser line rejection, and required spectral resolution are other important

design criteria

The scattered signal enters the polychromators via all silica optical fiber (2mm core dia,

NA: 0.22). The scattered light is imaged on interference filter by an aspheric input lens, by

placing fiber at the focal length. The focal length of this lens (35mm) is chosen such that the

beam dia at the fifth filter is less than the clear diameter (36 mm). The design has an angle of

incidence of 5 o and cone angle of 4 o at the filter. This permits the use of filter with required

lowest bandwidths 2.5nm.

As mentioned above, the temperature range to be studied is in the range of 20 eV to 3 keV

with the measurement error of < 5%. To select the filter set, which will cover this temperature

range, a code was developed. The reported values of collection lens efficiency, fiber optics,

filters transmission, detector quantum efficiency as a function of wavelength and estimated

values for laser energy, observation volume, total detector noise and optical transmission

were used as input parameters to the code. The wavelength transmission of different

polychromators channel is shown in fig. 9. Stray light rejection of the filters at the laser

880 900 920 940 960 980 1000 1020 1040 1060

0

20

40

60

80

100

Wavelength (nm)

Tra

nsm

issi

on

Fig. 9. Optical transmission curve of five-channel polychromator

wavelength are > 105. The band pass for each filter is optimized to provide good temperature

resolution. A, 2.5nm wide filter at laser wavelength will be used for Rayleigh calibration.

Filters are mounted in two parallel precision aluminum blocks (fig.10). There is a hole in

Fig. 10. Schematic of five-channel polychromator using 2 inch dia optics

one of the blocks at 50 to the surface for light signal to enter the polychromators. The input

fiber chuck can be adjusted in X, Y, Z, and è plane of input lens. The fiber, with lemo

connector, can be removed and reinstalled without loss of alignment. Coupling of the

transmitted light from filter to light detector is done by a short focal length (38.5mm)

aspheric condenser lens mounted, perpendicular to beam axis i.e. at 5 0 to block surface,

behind the filter. No extra adjustments are required for aligning different filters and coupling

lens of polychromators as it is done automatically by design itself. The image size, ~2.4mm,

is smaller than the 3mm diameter detector. The 2.4mm dia output image of transmitted band

of the polychromators will be directly coupled to thermo-electrically cooled 3mm dia silicon

avalanche photodiode (EG&G LLAM Head CD1787). The adjustment in X, Y and Z

direction is provided for detector box.

To measure the image size at the different filters and at the detector plane, a proto-type

five-channel polychromators was fabricated (fig. 11). The measured values are ~ 36mm at the

last filter and 2.4mm in detector plane. The non-transmitted part of signal was dumped on

Fig. 11. A proto-type five channel polychromator for testing image size of transmitted band

and alignment

graphite block. For the alignment of filters and lenses (by removing filter at that location), the

light dump was removed in-situ and a diode laser mounted from rear side was switched on.

The detectors along with the power supply will be mounted inside the polychromators box.

The detectors and interference filters are sensitive for the temperature variation. . The cooling

channels are provided above and below the filters in side the aluminum blocks for

temperature stabilization.

8. Detection System

Silicon avalanche photodiode (APD) mounted on Peltier thermo-electric circuit (EG&G,

model CD 1787) is used for the detection of transmitted wavelength band of scattered signal.

A feedback circuit is used to control the current through the Peltier cooler and to stabilize the

temperature of the APD to ±0.10C at the pre-selected value of 13.50C (optimized value from

signal to noise ratio). The output of the detector-preamplifier module [14] is connected to

three stage high frequency amplifiers (Burr-Brown OPA-603, bandwidth 50MHz). The

second stage of amplifier can be addressed via remote control to change the gain. Overall

gain of amplifier will vary from 10 to 100 using eight different selections. Each detector-

amplifier circuit has two outputs; a slow (1MHz bandwidth) direct-coupled output for

calibration and background light measurements, and a high frequency (50 MHz bandwidth)

output, which measures the laser scattered signal. The high frequency circuit is accomplished

by using a 100-ns Bessel pole [15] delay line. The delayed signal is subtracted from the non-

delayed signal to reject the lower frequency background light. A schematic of the amplifier

circuit and its tested frequency response are shown in fig.12, 13. A crowbar circuit is

Fig. 12. 3-stage high frequency amplifier circuit for APD detector

0 20 40 60 80 100

20

40

60

80

100

Gai

n

Frequency (MHz)

Fig. 13. Frequency response curve of 3-stage APD amplifier

provided to cut high voltage of APD in case APD signal exceeds the set safe limit.

9. Data Acquisition

The number of photons scattered into a given spectral channel are measured by integration

of detector current for a time period equal to signal pulse duration synchronized with laser

pulse. Normally, such measurements are done by using fast charge sensitive ADC. A PXI

based 5 channel (correspond to each polychromators) is being developed with 12bit

resolution, < 5µsec conversation time, and a sensitivity of ~ 0.125pc/count. . After

digitization, the data is stored in LIFO storage element of depth of 16 locations. The data

acquisition and transfer of data will be controlled by PXI driver module (PXI model 8335). A

gate pulse triggered by the laser itself will synchronize the data acquisition with the laser

pulse signal. For this purpose, the signal from a laser monitor diode will be fed to a in-

housed developed fast discriminator and adjustable delay line. Two gate pulses (100-ns apart)

will be generated to gate on high and low frequency channel charge ADC.

A real time, PXI based system to operate and control lasers and to acquire data will be used.

The entire system will be on PXI bus operating on ~80 Mbytes/sec.

10. Summary

Engineering and optical designs of most of the subsystems have been completed. These are

at different fabrication stages. The imaging performance of polychromator has been tested.

For collection lens system, these measurements will be done shortly with the help of

simulated scattering geometry. The alignment procedures for laser beam packing, beam

transport to distance of 30m are worked out on a test stand. This work is in progress for

further details.

A proto-type three-stage high frequency amplifier has been developed. Work related to

coupling Bessel pole delay line to amplifier circuit will be starting shortly. Work related to

the development of charge-ADC and interlock control system for laser operation is in

progress.

References

[1] Saxena, Y.C., SST Team, Nuclear Fusion, 40 (2000) 1069

[2] Rohr, H., Report IPP III/121 B (1987)

[3] Carlstrom, T.N., et al, Rev. Sci. Instrum. 63 (1992) 4901

[4] Watterson, R., and Chen, K., Rev. Sci. Instrum. 61 (1990) 2867

[5] Narihara, K., et al, Rev. Sci. Instrum. 66 (1995) 4607

[6] Johnson, D., et al, Rev. Sci. Instrum. 70 (1999) 776

[7] Rajesh, R., et al, Pramana, 55 (2000) 733

[8] Trost, P.K., et al, Rev. Sci. Instrum. 61 (1990) 2864

[9] Carlstrom, T.N., et al, Rev. Sci. Instrum. 68 (1997) 1195

[10] Behm, R., et al, Rev. Sci. Instrum. 70 (1999) 768

[11] Murmann, H., and Huang, M., Plasma Physics and Controlled Fusion, 27 (1985) 103

[12] Yoshida, H., et al, Rev. Sci. Instrum. 68 (1997) 1152

[13] Carlstrom, T.N., et al, Rev. Sci. Instrum. 61 (1990) 2858

[14] Hsieh, C.L., et al, Rev. Sci. Instrum. 61 (1990) 2855

[15] Johnson, D., et al, Rev. Sci. Instrum. 72 (2001) 1129