Heating Systems Issues, Status, and Plans

1
Heating Systems Issues, Status, and Plans

• David Swain
• Oak Ridge National Laboratory, Oak Ridge, TN,
  USA
• FIRE Engineering Meeting
• PPPL
• June 24, 2003

This research was sponsored by the Office of
Fusion Energy, U. S. Department of Energy, under
contract DE-AC05-00OR22725 with Oak Ridge
National Laboratory managed by UT-Battelle.
2
Summary of status and work to do (from Nov. Eng.
mtg) with comments on status at this meeting

• Ion cyclotron
• Reasonable shape for heating
• Current drive marginally OK need to revise
  requirements, obtain self-consistent solutions,
  do design study with 4 straps/port
• Still have requirements issues (see next VGs)
• Looked at 4 straps/port but it doesnt look very
  attractive if antenna is contained in the
  confines of in the port
• However, 4-strap keyhole antenna looks
  promising (discussion later)
• Lower hybrid
• Reasonable shape for heating and current drive
• Need to check scenarios for consistency, review
  CD calculations, compare with more detailed
  computations
• I think were in good shape here primarily a
  physics issue - Kessel?

3
Summary and comments (cont.)

• Electron cyclotron (for NTM stabilization)
• May be large problem in source availability for
  high frequency ( 170 GHz)
• How aggressive do we want to be in our
  assumptions?
• Need definition of requirements, operating
  scenarios
• Determine frequency (operating range in B0, r/a),
  power required
• Chose 170 GHz
• Power requirement is still a question. - Need to
  be realistic (170 GHz system will be mucho dinero
  per watt)
• Begin conceptual design
• Very pre-conceptual design, invoking launcher
  designs for ITER, putting in same port as LH
  launchers
• Need to do thermal and stress analysis (mainly IC
  and LH)
• Need Prad, Pfusion for different operating
  scenarios
• Need B(t) and heat loads for disruption scenarios
• Manpower and funding are limiting factors
• No thermal or stress analysis has been done.
• Reassess port allocations vis a vis H CD system
  needs.
• Port allocation (Ken Young 1/2003) shows 4 ports
  for ICRF (in a row), two more ports for LH/EC.
  Looks OK to me.

4
Issues from Meade outline

• ICRH
• Frequency range (my addition)
• FWCD 4 strap option concept (future work?)
• Generic port design ?
• LHCD
• Availability of sources
• Launcher feasibility
• ECCD
• Frequency, source availability
• Launcher geometry
• Engineering and Physics Issues are tied together

5
ICRH Frequency Wish List is not compatible with
present antenna

• Desired frequencies and modes of operation
• Heat ions
• 10 T gt 100 MHz 2nd Harmonic T
• 6.5 - 7 T gt 100 MHz 2nd harmonic D (with
  maybe H minority)
• Drive current
• 10 T gt 115 MHz just below 2nd harmonic D
  resonance, hits 2nd harm T at r -a/2. Get some
  absorption by T, but CD requirement (lt 300 kA)
  met
• 6.5 T gt 75 MHz same scenario as above
• 7.0 T gt 80 MHz
• ITB Formation (minority ion CD at r/a 0.5, both
  high and low-field side)
• 10 T gt 115 MHz and 88 MHz He3 or 2T resonance
• 6.5 T gt 75 MHz (He3 or 2T resonance) and 88 MHz
  (H resonance)
• 7.0 T gt 80 MHz (He3 or 2T resonance) and 94 MHz
  (H resonance)
• Bottom line Need 75 to 115 MHz frequency range
  to cover all bases.

6
Present antenna design is simple, reliable but
has a limited tuning range

• Antenna characteristics
• Two current straps
• Straps grounded at each end (violin antenna)
• good mechanical strength
• Each strap fed by 2 coax feeders
• Driven out of phase
• Optimized for 100 MHz
• Low-voltage operation near optimal frequency
• Limited frequency range

Antenna seen from plasma
Side cut
7
Cant fulfill all the desirables with present
antenna design
Antenna power for 35 kV max vs. freq.

• Extending frequency range will require different
  antenna design
• Can we operate with less frequency range?
• Do we need 115 MHz operation?
• Can we use 7.0 T instead of 6.5 T?

Resonances (R) for 6.5 and 10 T
Max. V on rf system vs. freq
8
ICRF status for PVR

• We need to EITHER
• Reduce requirements
• Forget about 115 MHz CD OR
• Operate at 7 T instead of 6.5 T for AT modes
• OR start on new antenna design.
• Will do some quick looks at different electrical
  configurations with broader frequency response
  between now and PVR.
• Aside from that, electrical operating
  characteristics are fairly well-defined and
  should meet the mission.
• Disruption effects not analyzed
• No disruption analysis done, and none likely
  before PVR with limited resources
• However, antenna with current straps grounded at
  each end is intrinsically strong.
• Earlier analyses for high-field machines (CIT,
  BPX) indicated that antennas could be built to
  withstand disruptions, so well rely on existence
  proof at PVR.

9
Lower-frequency range option

• Giving up 115 MHz operation will allow operation
  at 75 MHz
• For this case, can deliver 20 MW over 72 to 100
  MHz frequency range, assuming that everything in
  rf system can handle 35 kV

Antenna power for 35 kV max vs. freq.
Max. V on rf system vs. freq
10
Capacitor-tuned option

• Wider frequency band, tunable over 75-115 MHz
• Capacitors in vacuum will require tuning remotely
  between shots - how?
• End of current strap is free, transmits
  disruption forces to capacitor - can it stand it?
• More real estate taken up by HV rf in vacuum,
  more chance for breakdown
• Less shielding
• Cost more

Bottom line About the only advantage is it will
operate over wider freq. range
11
For the future (FY04?)...

• Look at 4-strap keyhole antenna?
• More power per port
• Factor of 2 better CD efficiency
• Requires recess in vessel
• Mechanical design (vacuum seal, installation) may
  be difficult
• Look at broadband antenna in more detail?

12
30 MW lower hybrid system

• 5 GHz needed to drive current in AT modes (must
  keep f gt 2 fLH).

• Waveguide array for 5 GHz
• Choose RF power flux 53 MW/m2,
• Need 0.57 m2 of WG area for 30 MW
• Each waveguide opening is 5.7 cm high x 0.65 cm
  toroidally
• Need 1500 waveguides

Thanks for help from Stefano Bernabei
13
LH launchers in two ports deliver 30 MW

• For each port, put array as shown.
• 96 toroidally x 8 poloidally.
• LH launcher contour must conform very closely to
  the plasma contour for good coupling. The taller
  the coupler, the greater the constraint on the
  plasma outer separatrix shape.
• Geometry flexibility precludes filling entire
  port with LH launcher, so need 2 ports for LH
  system.
• Good news
• Can put EC launchers above and below LH antenna

14
Source availability at 5 GHz is an issue, but
solvable

• Is a 1 MW, long-pulse, source available at 5 GHz?
• Not exactly.
• But A 3.7 GHz, 750 kW, 1000 s klystron (TH2103
  C) is presently available from Thales Electron
  Devices (in EU was Thomson-CSF)
• Thales did a cost estimate for ITER in 1998
• 64 to 72 klystrons at 5 GHz, 1 MW CW
• 16 to 18 High-voltage power supplies, each 80
  kV/100A
• 64 to 72 RF amplifiers
• The result was an estimate of 1.15 /W, for the
  sources and HV supplies only. (I dont know if
  any RD costs would have to be added).
• Requested info from CPI (US vendor), but no
  response so far.

15
Electron cyclotron system

• Needs to be absorbed at particular r/a to
  stabilize NTMs
• Assume midplane launcher with steerable optics to
  aim launched power at particular location on
  (almost) vertical resonance surface
• r/a range that can be reached depends on location
  of resonance surface

Need steerable launcher that can change direction
in vertical and toroidal direction launch waves
so toroidal when reaching resonance region if
good CD is needed.
16
EC launcher off-midplane, use similar to ITER
upper launcher?

• ITER upper launcher
• Steerable mirrors. Beams reflected by four
  mirrors through a vertical slot at the front
  shield.
• Poloidal RF beam steering capability ( 5)
• Accurate focussing of the RF power on the m 2
  and m 3/2 plasma flux rational surfaces is
  obtained

ITER midplane EC launcher
ITER upper EC launcher
17
Frequency of EC sources is significant issue

• Requirement
• Be able to reach r/a 0.6 to stabilize NTMs at
  rational q-surface locations
• Questions
• How much power is needed?
• How much current must be driven?
• For what magnetic fields will NTM stabilization
  be needed?
• As B0 increases, required operating frequency
  increases.

• Gyrotron availability
• US and Russian tubes at 1 MW level for f 120
  GHz
• Proposed development of 170 GHz tubes for ITER
  (JA, US, RU)
• will require significant RD effort
• 200 GHz? Fuggedaboutit! (major development
  effort would be needed)

NTM control range
18
Proposal for EC system design frequency (fm. Nov.
mtg).

• Pick f 170 GHz for design study
• Allows penetration to r/a 0.6 at B0 7 T
• Allows penetration to r/a 0.25 at B0 6.5 T
• Compatible with ongoing EC source development
  work (we get it for free)
• Consider higher-frequency operation (and
  consequent development of higher-frequency
  source) as an upgrade that could be done
• after initial operation with 170 GHz
• or if it becomes evident it will be needed

19
ECH power requirements and cost

• Cost info from ITER cost estimate
• US has just completed a cost estimate for the
  ITER EC system
• 170 GHz
• 20 MW
• Long-pulse/CW system
• Cost estimate for sources HV supplies is gt x/W
  (not counting RD)
• Cost estimate for launchers, transmission lines,
  etc. is (0.6 x)/W
• (I could tell you what x is, but then Id have to
  kill you)
• (But its a lot.)
• Issue What is a realistic power?
• How much do we need to stabilize NTMs?
• How much can we afford?

20
Stress and thermal analysis needed by
PVR?(from November Engineering Meeting)

• We need information to do the analyses
• Normal operation scenarios
• Prad(t) - radiated power (UV, soft x-rays) to
  midplane port
• Pfusion(t) - neutron flux at outer midplane
• Pulse length
• Disruptions
• B(t) in front of midplane port
• Power to wall from disruption
• Time is of the essence
• We have funding and manpower limitations
• If we want serious engineering analysesby the
  PVR, then input needed by Jan.(?)
• Probably dont have the resources(i.e., funding)
  to do much will need to prioritize.