Breakdown Rate Dependence on Gradient and Pulse Heating in Single

Cell Cavities and TD18

Faya Wang, Chris Nantista and

Chris Adolphsen

May 1, 2010

LC Breakdown Limits• For NLC, the breakdown rate limit was chosen so that the 2% pool of spare rf units

would rarely (once a year) be depleted assuming a 10 second recovery period after each breakdown. Operation with a 10 second recovery period has been demonstrated but longer times (30-100 sec) are typically used since the structure monitoring system has a 30 second sampling period. For 60 Hz operation at NLCTA, the breakdown rate limit translates to 0.1 per hour with the nominal 60 cm long structure design. This choice is somewhat soft in that a four-times higher rate would still provide 99% full-energy availability assuming a 5 second recovery time, which has also been demonstrated.

• For reference, at the 0.1 per hour limit, a breakdown would occur in one of the ~ 20,000 NLC X-Band structures once every 120 pulses. Such breakdowns will degrade the luminosity from that pulse, but the beam kicks from the breakdown fields should not inhibit beam operation.

• To 'translate' this limit, for one bkd in 10 hours in a 0.6 m structure at 60 Hz, the rate is 4.6e-7/pulse/structure or 7.7e-7/pulse/m. For CLIC 3-TeV, this same limit gives 1 bkd per 40 pulses, but because of the smaller beam emittance (and hence a smaller collimator aperture), the kicks are more likely to hit the collimators. The bkd kick distribution measured at NLCTA has values as large as 30 keV/c - this level is < 1/10 of the amount needed to hit the collimators at NLC.

98 100 102 104 106 108 110 11210

-7

10-6

10-5

10-4

10-3

Gradient (MV/m)

BD

R (

1/p

uls

e/m

)

500 hr

250 hr

1400 hr

900 hr

1200 hr

TD18 700 hr

Breakdown Rates: T18 and TD18

Pulse width 230ns, Green line for TD18, Others for T18

Single Cell SW Cavity1C-SW-A3.75-T2.60-Cu6N-KEK

Parameters Unit Value

FrequencyGHz

11.427 (Nitrogen, 20 oC)

Cells 1+matching cell + mode launcher

Q (loaded) 4660

Coupling 0.97

Iris Thickness T mm 2.6

Iris Dia. a / % 14.4

Phase Advance Per Cell deg 180

Es/Ea 2.03

Maximum surface electric field for 10 MW

MV/m 399

Maximum surface magnetic field for 10 MW

A/m 6.7e5

Peak pulse heating for 1 μs pulse with flat field of 100 MV/m

oC 24

0 500 1000 1500 20000

5

10

15

20

25

30

35

Time (ns)

Inp

ut p

ow

er

of s

tru

ctu

re (

arb

.u.)

Pre Pulse After Pulse

Main Pulse

Input RF Pulse Maximized initial pulse to reduce fill time

30 35 40 45 50 55 60 65 70110

120

130

140

150

160

170

180

Pulse heating (oC)

Gra

die

nt f

or

flat t

op

of 1

50

ns

(MV

/m)

Constant pulse heating test (45oC)

Constant gradient test (135 MV/m)

Constant gradient test (145 MV/m)

Constant gradient test (151MV/m)

Gradient and pulse heating for normal runing

Measurement Points: Vary Either Pulse Heating or Gradient

0 500 1000 1500 20000

5

10

15

20

25

30

Time (ns)

Ca

vity

Inp

ut P

ow

er

(arb

.u.)

0 500 1000 1500 20000

5

10

15

20

25

30

Time (ns)

Ca

vity

Re

flect

ed

Po

we

r (a

rb.u

.)

0 500 1000 1500 20000

10

20

30

40

50

60

70

Time (ns)

Pu

lse

He

atin

g (

oC

)

Low Pulse HeatingMiddle Pulse HeatingHigh Pulse Heating

Input Power

Reflected Power

Peak Pulse Heating

Breakdown Study with Constant Gradient but Different Pulse Heating from the Pre-Fill ‘Warm-up’

35 40 45 50 55 60 6510

-1

100

101

102

Peak Pulse Heating ( oC)

Bre

akd

ow

n R

ate

(1

/hr)

135 MV/m151 MV/m

1st Test of 145 MV/m

2nd Test of 145 MV/m

Breakdown Rate for Fixed Gradient

0 500 1000 1500 20000

10

20

30

40

50

Time (ns)

Pu

lse

He

atin

g (o

C)

Breakdown Study with Constant Pulse Heating

0 500 1000 1500 20000

5

10

15

20

25

30

Time (ns)

Inp

ut p

ow

er

(arb

.u.)

High gradientMiddle gradientLow gradient

0 500 1000 1500 20000

2

4

6

8

10

12

14

Time (ns)

Re

flect

ed

po

we

r (a

rb.u

.)

0 500 1000 1500 20000

50

100

150

Time (ns)

Gra

die

nt (

MV

/m)

115 120 125 130 135 1400

0.5

1

1.5

2

2.5

3

3.5

4

Flat top gradient (MV/m)

Bre

akd

ow

n r

ate

(1

/hr)

First TestSecond TestThird Test

Flat top = 160 ns

Increase much less than expectation of (139/119)^25 ~ 50

Breakdown Rate for Fixed Peak Pulse Heating

130 135 140 145 150 155 16010

-1

100

101

102

Gradient (MV/m)

Bre

akd

wo

n R

ate

(1

/hr)

40 45 50 55 6010

-1

100

101

102

Peak Pulse Heating ( oC)

Bre

akd

wo

n R

ate

(1

/hr)

Breakdown Rate with Varying Gradient and Pulse Heating

Blue for the 1st test and Red for the 2nd test.

0 500 1000 1500 20000

10

20

30

40

Time (ns)

Pow

er (

arb.

u.)

Damage to Single Cell Irises

V Dolgashev, L Laurent

a [mm] 4.06 3.36 2.66

a/(%) 15.4 12.8 10.1

d [mm] 2.794 2.054 1.314

e 1.21 1.18 1.15

f [GHz] 11.424 11.423 11.424

Q(Cu) 5098 5364 5458

vg/c [%] 2.25 1.47 0.87

r’/Q [LinacΩ/m] 10195 12560 15034

Es/Ea 1.97 1.88 1.88

Hs/Ea [mA/V] 5.85 5.2 4.85

Sc/Ea2 [mA/V] 0.52 0.39 0.3

For regular cellsP = 59.8 MW for <G> = 100

MV/mActive Length = 15.8 cm

First Middle Last Cell

TW Structure with Damping (TD18)

0 100 200 300 400 500 600 700 800 900 10000

5

10

15

20

25

n

|Ez|

[kV

/m]

Last Cell Surface Magnetic Field

T18 and TD18 BDR Pulse Heating Dependence

20 30 40 50 60 70 80 90 10010

-7

10-6

10-5

10-4

10-3

Peak Pulse Heating at Last Cell (K)

BD

R (

1/p

uls

e/m

)

T18, 900 hrsTD18, 100 MV/m @ 100ns,870hrsTD18, 100MV/m@ 150ns, 960hrs TD18, 100MV/m@200ns, 750hrsTD18, 115MV/m@150ns, 550hrsTD18, 120MV/m@150ns, 570hrsTD18, 100MV/m@230ns, 700hrsTD18, 105MV/m@230ns, 680hrsTD18, 100MV/m@50ns

110 MV/m

Pulse Heating BDR Test

0 500 1000 1500 2000-10

0

10

20

30

40

50

60

70

Time (ns)

Po

we

r to

Str

uct

ure

(M

W)

150ns - normal150ns - pre heating

0 500 1000 1500 20000

20

40

60

80

Time (ns)

Pe

ak

Pu

lse

He

atin

g in

La

st C

ell

(K)

100 MV/m, 200 ns100 MV/m, 150 ns100 MV/m, 150 ns pre-heating

60 65 70 750.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

2.2

2.4

Peak Pulse Heating in Last Cell (K)

BD

R (

1/h

ou

r)

100MV/m@150ns100MV/m@150ns + pre-heating100MV/m@200ns

A Coaxial Two Mode Cavity is Being Designed to Study E and B Effects Somewhat Orthogonally

TEM3 TE011

|Hs||Es|

A coaxial cavity resonant with 11.424 GHz TEM3 and TE011 would be excited by two rf sources, one coupling to each mode.

The high E field on the center conductor is determined solely by the TEM3 excitation, with the peaks at the zero points of the H field.

Adding TE011 increases the H field, preferentially around the central E field lobe.

|Hs||Es|Center conductor:r = 0.635 cm

C. Nantista

Latest Design

TEM3

TE01

WR90

0.592”

11.42375 GHz

11.4243 GHz

Es Bs

Pulse Heating Summary• All NLC structures and undamped CLIC structures operate with

pulse heating below 50 degC – do not believe it impacts BDR at this level as damped and undamped NLC structures performed the same despite the 20-40 degC variation in heating (should recheck with T18)

• TD18 and single cell BDR show clear pulse heating dependence above 50 degC

– Since TD18 differs from T18 mainly in the outer cell wall, indicates that heating (or B field) ~ 0.5 cm from irises has an influence on breakdown at the irises (as did the sharp-edge coupler heating in earlier NLC structures)

• Pre-heating tests at constant gradient confirm the pulse heating sensitivity

– However, longer exposure to moderate gradient (half of max) during the pre-heating period could be a factor