Design overview of the low-field side reflectometer for ITER

Design overview of the low-field side reflectometer for ITER

C.M.Muscatello1, C.Anderson1, J.Anderson1, A.Basile2, R.L.Boivin1, M.Duco2, D.K.Finkenthal3, A.Gattuso1, J.Klabacha2, G.J.Kramer2, M.LeSher1, L.Meneses4, G.H.Neilson2, R.O’Neill1, W.A.Peebles5, R.Seraydarian5, M.Sibilia2, P.Sichta2, D.Sieving1, A.Sirinelli4, G.Wang5, W.Wang2, A.Zolfaghari2

1General Atomics, 3550 General Atomics Crt, San Diego, CA 92121-11222Princeton Plasma Physics Laboratory, P.O. Box 451, Princeton, NJ 08543-04513Palomar Scientific Instruments, San Marcos, CA 920694ITER Organization, Route de Vinon sur Verdon, 13115, St Paul Lez Durance, France5University of California, Los Angeles, 475 Portola Plaza, Los Angeles, CA 90095-1547

ABSTRACT

The low-field side reflectometer (LFSR) will supply ITER with three important plasma measurements: 1) electron density profile, 2) electron density fluctuations, and 3) poloidal rotation. Simultaneous measurements of the three quantities are enabled by an array of six monostatic antennas which inject from an equatorial port on the outboard side of the ITER vessel. Low-loss transmission lines, consisting of corrugated, overmoded waveguide and miter bends, transmit the 30 – 165 GHz, O- and X-mode signals to and from the ITER plasma. Integrated transmission-line components serve a range of purposes, such as protection from high-power stray radiofrequency radiation, accommodation of transmission-line displacement, and simultaneous measurement of reference and plasma phases during the discharge. Broadband transmission signals are realized by full-band microwave transceivers combined with quasi-optical multiplexing. A field-programmable gate array (FPGA) processor demodulates the profile reflectometer signals, enabling real-time density profile measurements for Plasma Control System feedback. A full-scale transmission line test facility provides an integrated environment to assess the performance of critical LFSR components.

I. INTRODUCTION

Reflectometry relies on wave cutoff surfaces to collect information on characteristics of the density. The electron density profile, the electron temperature profile, and, for certain wave polarizations, the magnetic field, dictate the location of the cutoff for a probing wave with certain frequency. Reflectometry, in its many forms, is a common diagnostic technique in magnetic confinement devices used routinely to measure the electron density profile, electron density fluctuations, and plasma rotation velocity.

The ITER low-field side reflectometer is the result of the evolutionary development of reflectometry for magnetically-confined fusion devices and shares many functional features with its contemporary counterparts 1,2,3–10,11–17,. Some of the key features are listed in Table 1, denoting the presence of that feature on the existing tokamak. ITER LFSR is truly unique; not one of the existing systems incorporates all of the key features, and none currently have the capability of real-time phase processing.

Besides the functional requirements, the ITER environment presents unique challenges that affect the design and implementation of LFSR. Despite the commonplace of reflectometry in existing magnetic fusion experiments, LFSR must be compatible with the requirements of the nuclear facility and withstand the nuclear and thermal power loads during high-performance plasmas with high reliability. These requirements influence important design aspects and must be compatible with the measurement requirements of the diagnostic.

II. PRELIMINARY DESIGN

LFSR will provide measurements of three basic plasma parameters: 1) electron density profile, 2) electron density fluctuations and, 3) poloidal plasma rotation. The overall design solution is a broadband (30 – 165 GHz) reflectometer system supporting both O-mode and X-mode

1

14th Intl. Reflectometry Workshop - IRW14 (Lausanne) 22-24 May 2019 O.113___________________________________________________________________________

polarizations. Both frequency-modulated continuous wave (FMCW) operation for profile measurements and fixed-frequency operation for fluctuation and rotation (Doppler) measurements are incorporated. The fluctuation and Doppler transceivers are still in the early phases of development, and detailed discussion of those is deferred to a later phase. Figure 1(a) shows ITER baseline profiles for electron density (ne), electron temperature (Te), and magnetic field (B). Corresponding O-mode and X-mode cutoffs and cyclotron resonances are plotted along the midplane in Fig. 1(b). The cutoffs and resonances include a relativistic correction18. Since LFSR probes near the midplane from the low-field side of the torus, the upper X-mode cutoff is used for the primary measurements. LFSR is optimized to be an edge diagnostic, but core measurements are accessible under certain conditions (e.g. with reduced magnetic field or with full field and sufficiently low density). Contour plots of constant cutoff frequency versus magnetic field and electron density are shown in Figs. 1(c) and 1(d) for upper X-mode and O-mode polarizations. The relativistic correction has been applied with an assumed electron temperature of 10 keV. For a particular value of ne (and B for upper X-mode), the cutoff frequency is lower(higher) for higher(lower) values of Te. With upper X-mode, the outboard density profile can be probed from the far scrape-off layer (SOL) to the top of the pedestal with ne up to about 1x1020 m-3. For lower ne at the top of the pedestal, access into the core is possible but limited by the maximum probe frequency. Since O-mode measurements in ITER provide limited edge coverage, they provide a partially redundant dataset to compare with the upper X-mode measurements.

AUG DIII-D JET WEST

Monostatic Antenna

(bistatic TL)

Corrugated waveguide TL

Vacuum window in corrugated waveguide

Routine, automated data analysis

Simultaneous fluctuation measurement

Real-time analysis

Table 1. List of key ITER LFSR design features currently incorporated into the AUG, DIII-D, JET and WEST profile reflectometer systems.

The LFSR system is comprised of a number of sub-systems, including: antenna assembly, transmission line, back-end instrumentation, stray radiation protection, and data acquisition / system control. Each sub-system is designed to enable the diagnostic to meet its required measurement functions and to meet or exceed the safety requirements of the ITER plant. Figure 2 shows a diagram of the LFSR TL network used to route microwave signals between the tokamak and the diagnostic hall. The boxed regions are subject to different environmental conditions and therefore the components and instrumentation permitted for use within each region have different requirements. The four primary regions are: in-vessel, interspace and port cell, gallery, and diagnostic hall. The LFSR design is discussed within the context of these regions.

The in-vessel portion of LFSR, from the vacuum windows to the antennas, is shown in Fig. 3. A monostatic antenna array provides the required measurement functions. The open end of circular, corrugated waveguide forms each of the antennas. The Doppler antenna at the top of the array is tens of centimeters above the midplane, lies in a constant z-plane, and points slightly away from the centerline of the torus. Due to this geometry, it injects a beam

2


perpendicularly to the magnetic field with a nonzero poloidal angle-of-incidence at the cutoff. The monostatic Doppler antenna collects a portion of the back-scattered signal. Antennas for profile and fluctuation measurements are arranged near the midplane, lie in constant z-planes, and also point slightly away from the centerline of the torus to compensate for toroidal field ripple. This geometry enables near-perpendicular angle-of-incidence to the outboard cutoff surfaces. To accommodate height variation of the ITER plasma, the profile/fluctuation antennas are arranged as an array of five. This provides the ability to select the antenna nearest the plasma midplane for optimal coupling.

Figure 1. (a) Full-field ITER baseline profiles of ne, Te, and B. (b) Corresponding cutoff and cyclotron frequency profiles with relativistic correction. Contour plots of constant frequency (in

GHz) versus ne and B for (c) upper X-mode and (d) O-mode polarizations. The Greenwald density is plotted as the vertical dashed line.

The TL between the antennas and vacuum windows forms a labyrinth; a pair of miter bends creates a periscope, which limits neutron streaming through the front end of the TL. The majority of the TL consists of 63.5-mm diameter corrugated, overmoded waveguide, capable of supporting the HE11 waveguide mode. The HE11 mode couples very efficiently to a free-space Gaussian beam19. Because of this, the open end of the corrugated waveguide acts as an excellent antenna for LFSR.

Because the LFSR TLs are long (~ 50 m) and ITER pulse lengths will be hundreds of seconds, thermal expansion and mechanically-induced displacements of the TLs on the order of centimeters can make non-negligible contributions to the measured phase. Simultaneous

3


measurements of the plasma phase and a reference phase, containing information on the TL expansion, are necessary for accurate density profile measurements. Within each of the lower five TLs, the mirror in the first miter bend closest to the antenna contains an embossed feature on its surface that retroreflects a small fraction of the transmitted signal to the receiver. This mirror provides in-situ reference phase calibration, enabling simultaneous phase measurements of the reference point close to the plasma and phase measurements of the cutoff layers. A similar technique for measuring the reference phase utilizing a free-standing wire has been demonstrated with success on DIII-D20. The free-standing wire is not possible for ITER LFSR due to the potential for large stray RF power loads entering the antennas.

The mirror in the second miter bend of each TL has a periodic grated surface which functions as a frequency filter to reduce the amount of stray electron cyclotron heating (ECH) power at 170 GHz propagating through the LFSR TLs. A grated mirror was fabricated and tested at the TL test facility, and the frequency response shows strong, narrow rejection against 170 GHz, with transmission less than -20 dB of the incident power. Two grated mirrors are installed in each TL, providing rejection of about 20 dB of 170 GHz ECH in both X- and O-mode polarizations.

Figure 2. End-to-end functional diagram of LFSR transmission line.

A double-window assembly is attached at the closure plate to form a vacuum barrier separating the evacuated in-vessel TL components from the ex-vessel region of the TL. The vacuum-barrier assembly of each TL consists of two disks for redundancy and increased protection against the possibility of tritium leaks. In order to reduce the interference between and the reflection from the window pair, the two disks are made of slightly different thickness and are tilted, each by 10° in the same plane and with opposite sense. Insertion loss and reflectivity tests with a pair of quartz discs were conducted at the TL test facility. Results indicate that, with the 10° tilt, the reflected power measured by the receiver is -30 dB of the transmitted power. The 10° tilt has negligible effect on the transmission response.

During machine warm-up, operation, and bake-out, displacements up to 3 cm are expected at the equatorial port-plug flange. The TL components installed within the interspace and port cell must allow for displacement relative to the vessel and maintain alignment during all operation states. Displacements of the TL are enabled by two gaps in the corrugated waveguide near the vacuum vessel. One gap is located on the air side of the vacuum windows where the Gaussian telescope is located (Fig. 3). The Gaussian telescope is quasi-optical solution for maintaining transmission of the signal during transverse displacement of the TL. Another gap exists in the corrugated waveguide at the vacuum window assembly. The gap between the waveguide and vacuum window on the air side is intended to permit axial displacement of the TL. A section of

4


waveguide is held in-line with the vacuum window, and it can translate along its axis through linear bearings. Three-dimensional displacement of this section of waveguide follows vessel movement through linkage arms attached to the Gaussian telescope.

The TLs penetrate a lintel between the port cell and equatorial gallery, where a secondary confinement barrier is located. The confinement barrier fulfills several requirements in terms of microwave transmission, leak rate, pressure differential, and fire sectorization.

The longest portion of the LFSR TL is located in the equatorial gallery, the space between the diagnostic hall and the port cell. The majority of this portion of the TL consists of straight sections of waveguide with miter bends. One set of miter bends is orientated 30° out-of-plane, such that the TLs extending radially from the vessel turn towards the diagnostic hall. The TLs penetrate the lintel between the equatorial gallery and diagnostic hall, where another secondary confinement barrier is located, similar to the first one discussed.

Figure 3. In-vessel and interspace segments of the LFSR transmission lines.

The diagnostic hall contains microwave transceivers, microwave signal conditioning equipment, a switching array for TL routing, data acquisition and processing equipment, and control system. A diagram of the LFSR signal routing in the diagnostic hall is depicted in Fig. 4.

The back-end instrumentation of LFSR is defined as the parts of the system in the diagnostic hall that include the electronic hardware and microwave components for generating and detecting the reflectometer signals. The millimeter-wave signals couple between the transceivers and multiplexer with corrugated scalar horns. For the profile reflectometer, quasi-optical multiplexers combine the signals from separate transceivers onto one common corrugated waveguide on transmit, and split the bands to their separate transceivers on receive. X- and O-mode transceivers are incorporated for FMCW measurements of the electron density profile. The fluctuation and Doppler reflectometers will be capable of producing multiple, fixed-frequency signals that are arbitrarily tunable. The design allows for the profile and fluctuation signals to occupy any of the TLs that feed the lower five antennas. The TL that feeds the upper antenna is dedicated for Doppler signals.

The microwave source/receivers for the profile reflectometers are heterodyne FMCW transceivers. The profile reflectometers consist of V-, W-, and D-bands (50 – 165 GHz) for the X-mode transceiver, and Q-, V-, and W-bands (30 – 110 GHz) for the O-mode transceiver. Due to the wide frequency ranges of W- and D-bands and the bandwidth availability of microwave components, the W- and D-bands are split into two sub-bands and are diplexed within fundamental waveguide. A representative schematic of one band (or sub-band) of a FMCW

5


transceiver is shown in Fig. 5. The voltage-controlled oscillator (VCO) voltage is ramped over time s to produce a frequency sweep in the range of a few GHz to tens of GHz. The signal is split into probe and local oscillator (LO) paths. A delay line within the LO path increases the propagation time of the LO signal to the mixer. The time delay of the LO is somewhat mismatched compared to that of the probe to generate positive intermediate frequency (IF) signals from the reference phase mirror and plasma. The probe and LO signals are amplified and frequency-multiplied. The probe RF is transmitted to the quasi-optical multiplexer with a scaler horn and returns through the same scaler horn after roundtrip propagation through the plasma and the TL. The LO and probe RF are combined at the mixer to produce the IF signal.

Figure 4. Signal routing and conditioning in diagnostic hall.

Transceivers based on this design were built and tested at the LFSR TL test facility. They achieve a minimum sweep time of 4 s, and a dwell time of about 1.25 s is sufficient to produce highly repeatable sweeps. Further development and upgrades of the FMCW transceivers is ongoing, and one high-priority area of improvement involves increasing the maximum sweep rate. Results from Tore Supra with solid-state sources capable of full-band sweeps down to 2 s and repetition times of 3 s were reported8. More recently, results from profile reflectometers with 1-s sweeps and 1.25-s repetition times have been obtained21. Assuming technological advances continue in this area, sources capable of full-band sweeps and repetition times approaching 1 s or shorter could expand the capabilities of the LFSR profile reflectometer to diagnose density fluctuations with characteristic frequencies in the hundreds of kHz.

The LFSR Instrumentation and Control (I&C) system provides local data acquisition, control, monitoring, and alarm/event handling and logging. The LFSR data acquisition system is composed of two output data streams. One data stream is processed in real time using a field-programmable gate array (FPGA). The FPGA employs direct IF conversion to process the phase from the IF output using a quadrature demodulation algorithm. A prototype FPGA phase processor of this kind has been built and tested as part of the ITER Toroidal Interferometer/Polarimeter (TIP) program22,23. The data product from the LFSR FPGA is the accumulated plasma phase as a function of probe frequency. The phase data are sent to the PCS where the inversion algorithms are applied and the density profile is calculated. A separate

6


data stream of unprocessed IF data is continuously archived during normal operations for post-pulse and offline analyses. Another aspect of the LFSR I&C is the system of sensors which monitors the health of the system. In extreme conditions, the system may be exposed to harmful stray RF radiation, excess heating, over-pressurization, or even tritium leaks. Various types of sensors and instrumentation must be implemented to guard against such anomalous events.

III. SUMMARY AND PROSPECTUS

The ITER LFSR will provide important measurements of electron density profile, fluctuations, and plasma rotation. The design largely represents a traditional reflectometer in terms of basic functionality, but its combination of low-loss waveguide components, broadband-enabling technologies, and ultrafast, real-time process capabilities sets it apart as a truly innovative system. In addition to performing its measurement functions, the design also conforms to the environmental factors of the ITER facility in terms of mechanical, heat, and nuclear loads. Data from the full-scale transmission-line test facility and modeling results from full-wave simulations suggest that high signal-to-noise is achievable with the current design, enabling sub-centimeter spatial resolution and microsecond temporal resolution of the edge density profile.

Figure 5. FMCW transceiver schematic of one (sub)band of the profile reflectometer.

ACKNOWLEDGMENTS

We are grateful and indebted to Edward Doyle for the major contributions he made to this project. We would also like to thank Xuan Nguyen for his support of many different engineering aspects of the project. This work is supported by US DOE Contract No. DE-AC02-09CH11466. All US activities are managed by the US ITER Project Office, hosted by Oak Ridge National Laboratory with partner labs Princeton Plasma Physics Laboratory and Savannah River National Laboratory. The project is being accomplished through a collaboration of DOE Laboratories, universities and industry. The views and opinions expressed herein do not necessarily reflect those of the ITER Organization.

DISCLAIMER

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency

7


thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

VII. REFERENCES1 A.Silva, M.E.Manso, L.Cupido, M.Albrecht, F.Serra, P.Varela, J.Santos, S.Vergamota,

F.Eusébio, J. Fernandes, T.Grossmann, A.Kallenbach, B.Kurzan, C.Loureiro, L.Meneses, I.Nunes, F.Silva, and W. Suttrop, Rev. Sci. Instrum. 67, 4138 (1996).

2 A.Silva, M.Manso, S.Vergamota, L.Cupido, L.Meneses and I.Nunes, Rev. Sci. Instrum. 72, 307 (2001).

3 A.Silva, L.Cupido, M.E.Manso, F.Serra, I.Nunes, J.Santos, P.Varela, S.Vergamota, L.Meneses, V. Grossman, F.Silva, C.Loureiro, F.Nunes, B.Kurzan and W.Suttrop, Rev. Sci. Instrum. 70, 1072 (1999).

4 P.Moreau, F.Clairet, J.M.Chareau, M.Paume, and C.Laviron, Rev. Sci. Instrum. 71, 74 (2000).5 F Clairet, R.Sabot, C.Bottereau, J.M.Chareau, M.Paume, S.Heuraux, M.Colin, S.Hacquin, and

G. Leclert, Rev. Sci. Instrum. 72, 340 (2001).6 F.Clairet, C.Bottereau, J.M.Chareau, and R.Sabot, Rev. Sci. Instrum. 74, 1481 (2003).7 R.Sabot, A.Sirinelli, J.M.Chareau, and J.C.Giaccalone, Nucl. Fusion 46, S685 (2006).8 F.Clairet, S.Heuraux, C.Bottereau, D.Molina, L.Ducobu, F.Leroux, and A.Barbuti, Rev. Sci.

Instrum. 81, 10D903 (2010).9 G.Hornung, F.Clairet, G.L.Falchetto, R.Sabot, H.Arnichand, and L.Vermare, Plasma Phys.

Control. Fusion 55, 125013 (2013).10 E.J.Doyle, T.Lehecka, N.C.Luhmann, and W.A.Peebles, Rev. Sci. Instrum. 61, 2896 (1990).11 G.Wang, E.J.Doyle, W.A.Peebles, L.Zeng, T.L.Rhodes, S.Kubota, X.Nguyen, and

N.A.Crocker, Rev. Sci. Instrum. 75, 3800 (2004).12 L.Zeng, W.A.Peebles, E.J.Doyle, T.L.Rhodes, N.Crocker, X.Nguyen, C.W.Wannberg and

G.Wang, Rev. Sci. Instrum. 85, 11D843 (2014).13 L.Zeng, G.Wang, E.J.Doyle, T.L.Rhodes, W.A.Peebles, and Q.Peng, Nucl. Fusion 46, S677

(2006).14 L.Meneses, L.Cupido, A.A.Ferreira, S.Hacquin and M.Manso, Rev. Sci. Instrum. 77, 10E927

(2006).15 L. Meneses, L. Cupido, A. Sirinelli, and M.E. Manso, Rev. Sci. Instrum. 79, 10F108 (2008)16 A.Sirinelli, B.Alper, C.Bottereau, F.Clairet, L.Cupido, J.Fessey, C.Hogben, L.Meneses,

G.Sandford, and M.J.Walsh, Rev. Sci. Instrum. 81, 10D939 (2010).17 L.Cupido, E.De La Luna, C.Antonucci, A.Guigon, F.J.Van Amerongen, W.A.Bongers, A.J.H.

Donné, M.F.Graswinckel, A.Bruschi, S.Cirant, A.Simonetto, C.Sozzi, D.Wagner, E.Manso, L.Meneses, F.Silva, P. Varela, N. Balshaw, J.M. Chareau, G. Conway, J. Fessey, S. Hanks, R.Pearce, V.Ricardo, D.Sands, D.Starky, and T.Tisconia, Fusion Eng. Des. 74, 707 (2005).

18 E.Mazzucato, Phys. Fluids B Plasma Phys. 4, 3460 (1992).19 P.J.B.Clarricoats and A.Olver, Corrugated Horns for Microwave Antennas (The Institution of

Engineering and Technology, 1984).20 L.Zeng, G.Wang, T.L.Rhodes, W.A.Peebles, C.Sung & R.Lantsov, Rev. Sci. Instrum. 89,

10H112 (2018).21 F.Clairet, C.Bottereau, A.Medvedeva, D.Molina, G.D.Conway, A.Silva and U.Stroth, Rev. Sci.

Instrum. 88, 113506 (2017).22 M.A.Van Zeeland, T.N.Carlstrom, D.K.Finkenthal, R.L.Boivin, A.Colio, D.Du, A.Gattuso,

F.Glass, C. M.Muscatello, R.O’Neill, M.Smiley, J.Vasquez, M.Watkins, D.L.Brower, J.Chen, W.X.Ding, D. Johnson, P.Mauzey, M.Perry, C.Watts & R.Wood, Plasma Phys. Control. Fusion 59, 125005 (2017).

23 M.A.Van Zeeland, T.N.Carlstrom, D.K.Finkenthal, T.Akiyama, R.L.Boivin, A.Colio, D.Du, A. Gattuso, F.Glass, C.M.Muscatello, R.O’Neill, M.Smiley, J.Vasquez, M.Watkins, D.L.Brower, J.Chen, W.X.Ding, D.Johnson, P.Mauzey, M.Perry, C.Watts & R.Wood, Rev. Sci. Instrum. 89, 10B102 (2018).

8


Documents

Design overview of the low-field side reflectometer for ITER