TandemMirror ExperimentUpgrade neutral pressure measurement diagnostic systemsW. L. Pickles, S. L. Allen, D. N. Hill, A. L. Hunt, and T. C. Simonen Citation: Review of Scientific Instruments 56, 978 (1985); doi: 10.1063/1.1138010 View online: http://dx.doi.org/10.1063/1.1138010 View Table of Contents: http://scitation.aip.org/content/aip/journal/rsi/56/5?ver=pdfcov Published by the AIP Publishing Articles you may be interested in EB endlossion analyzer for TandemMirror ExperimentUpgrade Rev. Sci. Instrum. 56, 1117 (1985); 10.1063/1.1138242 Xray spectroscopy diagnostics for the Tandem Mirror ExperimentUpgrade (abstract) Rev. Sci. Instrum. 56, 847 (1985); 10.1063/1.1138066 Design of a high energy xray imaging system for the Tandem Mirror ExperimentUpgrade (abstract) Rev. Sci. Instrum. 56, 845 (1985); 10.1063/1.1138063 Characteristics of the neutral beam injectors for the Tandem Mirror Experiment Upgrade Rev. Sci. Instrum. 54, 1615 (1983); 10.1063/1.1137321 Measurement of sloshingion spatial profiles in end cell of tandem mirror experimentupgrade (TMXU) Phys. Fluids 26, 2335 (1983); 10.1063/1.864434

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:

130.88.90.110 On: Sun, 21 Dec 2014 14:34:46

Tandem-Mirror Experiment-Upgrade neutral pressure mea.surement diagnostic systems

w. L. Pickles, S. L. Allen, D. N. Hill, A. L. Hunt, and T. C. Simonen

Lawrence Livermore National Laboratory, University a/California, Livermore, California 94550

(Presented on 18 September 1984)

The Tandem-Mirror Experiment-Upgrade (TMX-V) has a large and complex system of Bayard­Alpert, magnetron, and Penning gauges, in addition to mass spectrometers (RGA), all of which measure neutral pressures in the many internal regions ofTMX-U. These pressure measurements are used as part of the confinement physics data base as well as for management of the TMX-V vacuum system. Dynamic pressures are modeled by a coupled-volumes simulation code, which includes wall reflux, getter pumping, and plasma pumping.

iNTRODUCTION

Data acquired from the Bayard-Alpert, magnetron, and Penning gauges and the RGA mass spectrometers are pro­cessed and stored in the Tandem-Mirror Experiment-Up­grade (TMX-U) physics data base at three separate rates (from once every 10 min to once every 200 Ils during a plas­ma shot). In general, the magnetically shielded and unshield­ed Bayard-Alpert gauges have worked well in up to 3-kG magnetic fields. In contrast to the Bayard-Alpert gauges, the Penning and magnetron gauges are much more difficult to operate successfully. Neutral pressure measurements within 30 cm of the plasma are routinely measured by the unshielded Bayard-Alpert gauges with a 20% accuracy at a large number of locations along the TMX-U central-cell plug plasma.

To evaluate the data gathered from the above gauges, we have developed a coupled-volumes computer code to simulate the complexities of the gas dynamics in TMX-U. The code has 61 coupled volumes that include the ability to

z = 210 FMWS 90° Z·o

calculate plasma pumping, wall reflux, and getter pumping. We also use the code to calculate the axial and radial pres­sure variations, and to predict the pressures in regions where there are no measurements. By using the RGA's, we found that the species present in the neutral gas between the first wall and the plasma are H 2, HO, and D2 in varying ratios. We are attempting to connect this large body of pressure data to more direct spectroscopic measurements of neutral density in the plasma in order to develop halo shielding mod­els.

I. PRESSURE MEASUREMENT

We measure pressure (neutral density) in many of the regions of the TMX-U vacuum vessel using 23 unshielded and five magnetically shielded Bayard-Alpert gauges. The basic gauge system is described in Ref. 1, and the complex­ities of the X-U vessel regions are described in Ref. 2. The electronics system and data-acquisition system are described in Ref. 3. The new capabilities of this gauge system result

Z;. 558 FWPS190 D SWPS2900D RGA9000

FCCS1 135° OS SCCS2135 os RGA 135"OS

Z - S88 FEPS19000 SEPS2 90°0 RGA90 D

z + 1040 FWFSS60 0

W-+---

Z + 738 W01105 W

W02100 SU

Z + 588 WPPW 135"S

Z + 239 ~90oD

Z - 588 EPPE 45°N

Z - 528 EP245'W EPPW 45"N EP1 45 W

Z=44BFME~ z· 418 FMETO

Z -738 E0175 E E0280 Nu

o g Opening down

U B Opening up

= Along Z opening west W

= Along Z opening east E

@ Opening north N

o Opening south S

FIG. \. TMX-U vacuum vessel regions with all gauge locations shown. The new unshielded high-field Bayard-Alpert gauges are labeled WCP. ECP, ETP, WTP, CCE, and CCW.

978 Rev. Sci.lnstrum. 56 (5), May 1985 oo34-6748/85/050978"()3$01.30 © 1985 American Institute of Physics 978 This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:

130.88.90.110 On: Sun, 21 Dec 2014 14:34:46

~~!Y

1 Stuelded 'J ~=en~l= coileetOl' coline ... wIth B field

2 Umh.elded [J Fat m..anatron gM.I9Iah1lO8d to TMXU 8 fMtld

J Seli held 0 Cold Cilthode PannIng G....,.

4 EII,.,.".lheld :=: M""~trometer ~'

Dlnstde

FIG. 2. Isometric drawings of the TMX-U first or warm wall showing the location of all the gauges used for fast pressure measurements.

from the successful operation of the following three gauge types between the first wall and the plasma during a shot: (I) unshielded Bayard-Alpert gauges in B fields of up to 3 kG; (2) magnetron-type cold-cathode gauges (EW); and (3) Pen­ning-type cold-cathode gauges (E liB). Figure 1 shows the locations of all gauges in the TMX-U vessel including the new gauges.

U. UNSHIELDED SA YARD-ALPERT GAUGES

Our experiments indicate that unshielded Bayard-Al­pert gauges can operate reliably in up to 3·kG magnetic fields. The gauge is aligned so that the collector, the filament, and the local TMX-U B field are colinear plasma with B perpendicular to the axis of the filament and collector. In this orientation, we achieve a sensitivity that is ~ 20% low in a 3-kG field. This use of Bayard-Alpert gauges in strong magnetic fields is discussed in detail in Ref. 4. These gauges are located just inside the first or warm wan and donot view the plasma directly. Figure 2 is an isometric drawing of the first or warm wall of TMX-U showing the location of the gauges and the complexity of the warm-wall shape.

10-4.---------------------::j 6/8/84 Shot 29

WCP

N 10-5

,~/ ::I: \, e !: f " ::! f

11. 10-6 ""--

10 20 30 40 50 60 70 80 90 100

Time (mo)

FIG. 3. Pressure during TMX-U plasma operation in six regions. The WCP is 30 em from the plasma near the gas box in the central cell. Note pressure drop results from sealing and unsealing at plasma to gas box limiter.

979 Rev. Sci. Instrum., Vol. 56, No.5, May 1985

FIG. 4. TMX-U magnetron gauge with housing to prevent plasma-induced signals in the gauge.

Figure 3 shows a typical data record from five shielded and one unshielded (WCP) Bayard-Alpert gauges. The WCP gauge is located near the gas box on the west side of the TMX-U central cell about 30 em from the plasma. The pres­sure drops are due to the increasing size of the plasma radius until it "seals" the plasma to the gas-box limiter preventing escape of neutral gas. The ratio of the pressure near the plas­ma (WCP) to the pressure at the outer edge of the central-cell (FCCS 1) is 20 or 30.

III. MAGNETRON GAUGES

Commercially available magnetron-type gauges were packaged for TMX-U to prevent plasma reduced signals from appearing in the gauge. The housing is a Faraday shield with an 80-mesh screen and step baffle. Figure 4 shows all the components disassembled. Three of these gauges were mounted near the plasma in TMX-U and used the TMX-U B field to operate the gauge. The B field strength was 3-5 kG and was on for 2 s. The gauges were able to ignite and mea­sure pressure. Figure 5 shows the pressure measured by a magnetron gauge and nearby Bayard-Alpert gauge during plasma operation. The magnetron gauge is noisy and does not track fast pressure changes nearly as well as the Bayard­Alpert gauge. The magnetron-gauge conductance-time con-

5.0

4.5 8/2/84 Shot 10

4.0 Bios - 3900 V G' 3,58 10-8

Gain 25 3.5

£ <; 3.0 ...

'f g

2.5

~ 2.0

~ 1.5 ct

1.0

0.5

30 40 50 60 70 80

Time (msl

FIG. 5. Overlay of the pressure measured by a magnetron gauge and a near­by Bayard-Alpert gauge during plasma operation ofTMX-U,

Diagnostics systems and other applications 979

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:

130.88.90.110 On: Sun, 21 Dec 2014 14:34:46

10-7r-________________________ ~--------~

9/8/84 Shot 38 cc

"~'l

CC I-D2 b-,

o 20 40 60 80 laO 120 140 160 180 200

Time (msl

FIG. 6. TMX-U RGA diagnostic shows Ho partial pressure in each plug and Ho, HD, and Do partial pressure in the central cell from 20 ms before to 200 ms after a shot.

trol was 1 ms and should not have affected the gauge re­sponse. We also experienced difficulty with magnetron gauge igniting and mode changing when we used different bias voltages. We find an approximate calibration of 0.2-1 A/Torr for these gauges. Currently, we are attempting to increase our use of this gauge type.

IV. PENNING GAUGES

We use a commercially available I-Torr I/s ion pump as a Penning gauge. This gauge was intended for the Mirror Fusion Test Facility (MFTF-B) operation. The sensitivity of this gauge is ~ 10 AITorr,5 which is better than the magne­tron gauge, but we have had difficulty getting these gauges to tum on in the TMX-U B field. At present, we have only preliminary data from these gauges.

V. MASS SPECTROMETERS

We use three magnetically shielded commercial RGA mass spectrometers to obtain species as a function of time from very slow rates to 1 ms during a plasma shot. Figures 1-3 show the location of these units and their design is dis-

980 Rev. Sci. 'nstrum., Vo •. 56, No.5, May 1985

cussed in Ref. 3. Figure 6 shows the H2 partial pressure as a function oftime in each end plug, in addition to the H2, HD, and D2 partial pressures in the central cell during a plasma shot, where the walls have already desorbed most of their Hz inventory.

Vi. PLASMA NEUTRAL DENSITY

We are attempting to install in the short term a new diagnostic that measures neutral density inside the plasma as a function of radius and time. This technique, which involves observation of the H or D Bal.hmer alpha lines, is discussed by Allen in these proceedings. 6

Vii. CONCLUSiONS

We are finding that Bayard-Alpert gauges are by far the most effective diagnostic gauge system for fusion plas­ma-pressure measurements. We are continuing to develop the potential of the magnetron and Penning gauge types, and are attempting to generate reliable, easy to use diagnostic methods for measuring neutral atomic densities inside the plasma.

ACKNOWLEDGMENT

This work was performed under the auspices of the U. S. Department of Energy by the Lawrence Livermore Na­tional Laboratory under contract No. W-7405-ENG-48.

'w. L. Pickles, M. O. Calderon, M. R. Carter, C. A. Clower, R. P. Drake, A. L. Hunt, D. Lang, T. C. Simonen, and W. C. Turner, 1. Vac. Sci. Tech­noL A 1. 1288 (1983).

2W. L. Pickles, A. K. Chargin. R. P. Drake. A. L. Hunt, D. D. Lang. J. J. Murphy, P. Poulsen, T. C. Simonen, T. H. Batzer, T. P. Stack. and R. L. Wong, J. Vac. Sci. Techno!. 20. 1177 (1982).

'A. L. HUnt, F. E. Coffield. and W. L. Pickles, J. Vac. Sci. TechnoL A I, 1293 (1983).

'G. D. Martin, Jr., General Observations on Hot Filament Ion Gauge Oper­ation in Homogeneous B fields MA IT-80, Princeton. Plasma Physics, Princeton, NJ, Report No. MATI-80, 1961. ~H. F. Dylla, J. Vac. Sci. TechnoL 20, 119 (1982). oS. L. Allen et al. (these proceedings).

Diagnostics systems and other applications 980 This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:

130.88.90.110 On: Sun, 21 Dec 2014 14:34:46