some lessons from long pulse operation of negative ion ... · r s hemsworth euratom tore supra iaea...

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 1

Some lessons from long pulse operation of negative ion sources and accelerators

R S Hemsworth

(with help from D Boilson, H P L de Esch, L Svensson, A Krylov, C Grand

and P Massmann)

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 2

The KAMABOKO III negative ion source Designed and built by JAERI, Naka, Japan

Model of the ITER reference design of ion source

☺ First tested in Japan: 280 A/m2 of H-

Sent to the DRFC, CEA, Cadarache, France for testing in deuterium.

Short pulse, 5 s operation

☺ Reproduced the 280 A/m2 of H- but:

with deuterium: Ie extracted >10 !! ID-

Increased magnetic filter strength from 450 Gauss.cm to 900 Gauss.cm:

☺ 220 A/m2 of D-, with Ie extracted ≈1 ID-

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 3

-200

0

200

400

600

800

1000

1200

1400

1600

7 8 9 10 11 12 13 14 15 16 17 18 19 20

t (seconde)

I(G2) (electrons, mA)

I(arc) (A)

I(D-), calorimetric, mA equivalent)

KAMABOKO III Negative Ion Source

Arc power = 80 kW, Source filling pressure = 0.35 Pa, Beam Pulse Length = 5 s

Average J(D-) = 21.6 mA/cm2

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 4

Long pulse operation

• operational advantages & difficulties

• reduced negative ion yield

• reduced plasma grid temperature effect

• increased caesium consumption

• caution in the future operations

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 5

Planned changes to the system ☺ New data acquisition system

☺ New timing system

☺ Water cooled arc series resistances

☺ Change cooling water circuit to ensure adequate flows

☺ Remote operation to allow D- operation for long pulses (neutron dose)

☺ New long pulse plasma grid capable of operation at ≈300 °C

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 6

0

50

100

150

200

250

50 100 150 200 250 300

Plasma Grid temperature (°C)

Incr

ease

in n

egtiv

e io

n cu

rren

t (%

)

Effect of plasma grid temperature on the accelerated negative ion current with the KAMABOKO III ion source.Parc ≈ 40 kW, Psrc ≈ 0.3 Pa, Bias ≈ 5 V, with Cs Short (<5 s) pulses, Mo grid

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 7

Cu/Cr/Zr long pulse grid (6 mm thick).A thermal bridge limits the heat flow to the cooling channels along each edge

Long Pulse Grids Provided by JAERI

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 8

Molybdenum long pulse grid (6 mm thick).A set of stainless steel disks act as thermal bridges between the grid and copper cooling tubes traversing the grid.

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 9

0

100

200

300

400

500

600

0 50 100 150 200 250 300

Time (s)

T PG

(°C

)

0

10

20

30

40

50

60

Parc

(kW

)

Tpg

Parc

Frame cooled grid

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 10

Unplanned changes: Remove/change source flanges

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 11

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 12

Extraction grid supports

Originally the extraction grid supported off the plasma grid by small ceramic insulators

These became metallised due to increased number of breakdowns

Replaced by shielded supports mounted on the acceleration grid

Breakdown protection system

Extraction grid feedthrough

……

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 13

shot 5179, pressure variation, 06-2000, Hydrogen with Cs

0

2

4

6

8

10

0 50 100 150 200 250 300 350 400

Time

Sour

ce P

ress

ure

(µb)

,Id

rain

*10

(A)

0

200

400

600

800

1000

Iarc

(A)

press Idrain*10 IG2*50 Iarc

Advantages

☺ Ability to change parameters during a single shot

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 14

0

0.4

0.8

1.2

1.6

2

0 100 200 300 400 500 600 700

Time (s)

IG2

(A)

0

10

20

30

40

50

Parc

(kW

), V

G2

(kV

)

Shot 8978

IG2

VG2

IH-

Parc

Unexpected long time constant effects

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 15

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

0 200 400 600 800 1000 1200

Time (s)

IH- &

2*I

G2

(A)

Shot 8994

1

2

3

4

5

6

7

IH-

2*IG2

1. VG2 ~8 kV2. Decreased VG2 by 2 kV to 6 kV3. Increased VG2 by 1 kV to 7 kV4. Decreased VG2 by 1 kV to 6 kV5. Reduced filament voltage6. Reduced filament voltage7. Decreased VG2 by 1 kV to 5 kV

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 16

0,0

0,2

0,4

0,6

0,8

1,0

1,2

1,4

1,6

1,8

0 200 400 600 800 1000 1200

Time (s)

Idra

in, I

G2,

Ical

(A)

0

50

100

150

200

250

300

350

400

450

Tpg

(°C

), Pa

rc (k

W)

TpgIdrain

5* Parc

Ical

IG2

1000 s extraction of H- beam.

☺ 18 mA/cm2 H- calculated from the Idrain.

Poor transmission: only ~50% accelerated current collected on the calorimeter ≈ 9 mA/cm2

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 17

1000 s extraction of D- beam.Shot 9069, 1000sec D- extraction, 45kW

0

0,5

1

1,5

2

2,5

0 200 400 600 800 1000 1200 1400

Time (sec)

Idra

in, I

G2

(A)

0

50

100

150

200

250

300

350

400

450

Tpg

(°C

), P

arc

(kW

)

Idrain IG2 Ical Parc Tpg (°C)

~60% of accelerated current arrives at the calorimeter ≈ 7 mA/cm2

☺ 11.5 mA/cm2 D- calculated from the Idrain.

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 18

Inertial calorimetry determination of the beam transmission

25

35

45

55

65

75

0 20 40 60 80 100 120 140 160 180 200

Beam on time (s)

Tran

smiss

ion

(%)

Low transmission of accelerated current to the calorimeter ≈55%

Wednesday 11 May 2005 9:25-9:50 D. Boilson: Power transmission from the ITER model negative ion source

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 19

Shot 11074, 20% Temp Effect, 0.3 Pa, 40 kW H2

0

50

100

150

200

250

0 50 100 150 200 250 300 350

Time (s)

Tpg

, Tso

urce

, Par

c

0

0.2

0.4

0.6

0.8

1

Idra

in, I

G2(

A)

Parc Tpg Tsource IG2 Idrain

No significant change in IH- as the PG temperature increased

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 20

0

50

100

150

200

250

50 100 150 200 250 300

Plasma Grid temperature (°C)

Incr

ease

in n

egtiv

e io

n cu

rren

t (%

)

Short pulse

Long pulse

Long pulse grid, cold source walls

Effect of plasma grid temperature on the accelerated negative ion current with the KAMABOKO III ion source.Parc ≈ 40 kW, Psrc ≈ 0.3 Pa, Bias ≈ 5 V, with Cs

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 21

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04

0.045

8000 9000 10000 11000

shot number

Idra

in/P

arc

FCG ACG

No obvious effect of grid material

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 22

Cs “consumption”

Long pulse operation: 4.6*10-6 g/s/aperture.

Estimate for the ITER ion source 2.8*10-9 g/s/aperture) (Short pulse or low power operation)

The ratio is >1600 Source cleaned with water and the “polluted” water was chemically analysed.

The polluted water contained ≈4.5 g ±20% of Cs & a similar amount of tungsten.

Tuesday 10 May 2005: 9:50-10:15 A. Krylov: Caesium and Tungsten behaviour in filamented arc driven KAMABOKO-III beam source

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 23

Suggestions:

1 Cs consumption: The Cs effect disappears not because the Cs has left the ion source, but because the Cs is somehow “blocked” on the walls and the PG surface in loosely bound tungsten – caesium mixture, the tungsten being evaporated from the filaments during the arc operation.

2 Reduced/no PG temperature effect: The composition of the surface of the PG will change dynamically during a long pulse due to the tungsten evaporation from the filaments. This could mask any effect of the PG temperature on the negative ion yield.

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 24

Other long pulse effects: Dark current This appear when HV is applied on the Cadarache 1 test bed. HV only - no gas

- no filaments operating, - no discharge in the source

Reduced by conditioning, i.e. apply HV in series of shots

wait for dark current to reduce

increase HV & repeat 50 – 100 mA at V >250 kV

Seemingly spread generally over the “anode” and cathode” surfaces (measured in all places that have been checked)

But: We increased the (conditioning) pulse length ability and….

We melted a hole in an electrostatic screen!!!

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 25

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 26

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 27

Other long pulse effects: Future????

With the ITER system we foresaw a possible problem with power deposited on the cryopumps by X-rays created by accelerated electrons (stripping).

The UKAEA have started to investigate this using the MCNP package.

Because they (M Kovari) had to input data on electron back scattering, a potential new problem appeared – 40 kW of electron power to the 80 K cryopump screens.

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 28

This time we have been lucky, the problem has been identified and can be solved.

The future will be interesting.

R S Hemsworth

Euratom TORE SUPRA

IAEA TM on Negative ions Long pulse operation 29

Conclusion

Long pulse operation of the MANTIS test bed has thrown up several important potential problems with the ITER negative ion source.

The longer conditioning pulses on the Cadarache 1 MV test bed have reminded us that small powers can be very important over long periods.

Studies in the EU on the ITER NBI test bed have reminded us that other potential problems are to be expected