Recent 30 GHz results @ CTF2

M. Gerbaux, S. Döbert, Jan Kovermann, R. Zennaro

• Introduction

• 30CNSDsbCu_speed-bump Motivation

High power RF test results

Effect of the speed bump on the damages

• TM020 3.5 mm structure Motivation

High power RF test preliminary results

• Conclusion and future plans

Outline

CR

30 GHz production (PETS line)and test stand

Thermionic gun

Linac

D FFD

DF

F

D F D

D F D D F D

D F D

DF DF DF DF DF DF DF DF DF

D F D

F DF D

D FFFDD

DF

F

DDFF

FF

D F DD F D

D F DD F D D F DD F D

D F DD F D

DF DF DF DF DF DF DF DF DF DFDF DF DF DF DF DF DF DF DF DF DF DF DF DF DF DF

D F DD F D

F DF DF DF D

CTF2

CLEX

To set the stage…

30 GHz activities : • Reminiscence of the time when CLIC operating frequency was not yet 12 GHz.

• Now used to test scientific hypothesis.

• The speed bump and TM02 structures are based on the design of the older 3.5 mm diameter structure.

Why speed bump?

From Igor’s presentation at the X band workshop:

Very often we do observe, that after accelerating structure processing the most of the surface modifications take place in a few first cells. Also the number of cells involved is correlated with the group velocity, the less the Vg the fewer cells modified.

What we do certainly know, the breakdown ignition is a very fast process: 0.1 -10 ns. If so, one can propose the main difference between the “first” and “second” cell is accessible bandwidth.And the lower group velocity the more the difference.The first cell, if breakdown occurs is loaded by the input coupler/waveguide and is very specific in terms of bandwidth. In other words, the first cell can accept “more” energy during breakdown initiation than the following ones. We do not know the exact transient behavior of the breakdown and the structure bandwidth could play important role.

HDS 60 L

PINC

HDS 60 S

PINC

Courtesy of

Riccardo Zennaro

Speed bump (TM03)

R=14.398 mm

R_iris= 2.428 mm

Iris_thickness= 1mm0

0.01

0.02

0.03

0.04

0.05

0.06

0 1 2 3 4 5

cell #

vg

/c

2nd mode speed bump

regular cell nominal value

3rd mode speed bump

3rd mode,final version

Goal : « protect » the structure by lowering Vg in the first cell (usually the most damaged).

Tested in both direction :

• RF fed from the input (4.1×106 pulses, 2186 breakdown) → the speed bump plays its role• RF from the output (1.7×106 pulses, 501 breakdown) → the speed bump has no effect

Test structure in disks: 30 cell and identical mode launcher of the conventional “3.5 mm structure”

Courtesy ofRiccardo Zennaro

→ Equivalent to 19 + 8 « SLAC hours » @ 60 Hz…

6

1606 1608 1610 1612 1614 1616 1618 1620 1622 1624 162630

40

50

60

70

80

90

100

Time (h)

Acc

eler

atin

g g

rad

ien

t

1606 1608 1610 1612 1614 1616 1618 1620 1622 1624 1626

20

40

60

80

100

Time (h)

Pu

lse

len

gth

@ 50%@ 75%

OK

Breakdown rate calculation

Breakdown rate vs gradient – speed bump structure

65 70 75 80 85 90 95 100

10-6

10-5

10-4

10-3

10-2

Peak gradient (MV/m)

Bre

akd

ow

n r

ate

C30-sbUpper limits

-- Old 3.5mm structure

3.5 mm structures comparison

• Vertical error bars calculated assuming an error of 1 on the breakdown number and 500 on the number of pulses.

• Horizontal error bars : standard deviation of the measured gradient.

• Red circles are upper limits (no breakdown recorded).

• Fit doesn’t take into account these points.

No effect observed on the breakdown rate : similar results in both directions and for the 3.5 mm structure (same design without speed bump).

5 15 25 350

50

100

150

200

250

300

350

400

450

500

550

Power (MW)

Num

ber

of o

ccur

renc

esC30-SB

All breakdowns

5 15 25 350

50

100

150

200

250

300

350

400

450

500

550

Power (MW)

Num

ber

of o

ccur

renc

es

Reversed C30-SBAll breakdowns

Number of breakdowns in the two experiments

Iris 1 Iris 2 Iris -2 Iris -1

Speed bump

General view of the 30CNSDsbCu_speed-bump afer cutting

SEM inspection was performed on these irises

Iris 1

Iris -1

Damages

Iris 2

Iris -2

Damages

• Same phase advance• Same P/c• Same aperture and iris shape• Same field configuration in the iris region

TM02 structure

TM02 regular cellTM01 regular cell “reference”

Test structure in disks : 30 cells, same mode launcher as the “conventional” 2π/3, Ø 3.5 mm.

• Different group velocity (4.7% vs 2%)• Different R/Q (29 kΩ/m vs 12 kΩ/m)

but

Is it possible to change some global parameter without changing local field distribution?

Only by changing the propagating mode Courtesy ofRiccardo Zennaro

50 55 60 65 70 75 80 85 90 95 10010

-6

10-5

10-4

10-3

10-2

Peak gradient (MV/m)

Bre

akdo

wn

rate

3.5 mm C30-SB Reversed C30-SB TM02 upper limits only TM02 last "real" points

TM020 structure – preliminary results

1

2

3

5

4

6

7

8

9

10

1112

13

1415

16

17

18

19

20

21

22

23

24

25

26

27

28

2930

At the moment, equivalent to ≈26 « SLAC hours »

50 55 60 65 70 75 80 85 90 95 10010

-6

10-5

10-4

10-3

10-2

Peak gradient (MV/m)

Bre

akdo

wn

rate

TM02 upper limits only TM02 last "real" points TM02 fit 3.5 mm C30-SB Reversed C30-SB

TM020 structure – preliminary results

Conclusion and future plans

• The 30CNSDsbCu_speed-bump worked well and the speed bump seems to reduce the damages due to breakdowns.

• This suggests to test a « speed bump » structure at 12 GHz.

• The TM020 structure is still under test (still conditioning ?) but the results are not very promising.

• If vg was the key parameter, the achievable gradient at a given BD rate should be higher than for the other 3.5 mm structures.

• If it was rather surface field, the results should be more or less the same.

• The experiment confirms neither one nor the other. Is this only due to fabrication issues or is there another key parameter?

• It underlines the difficulty to draw conclusions with a single structure !

La ré

serv

e du chef

as ds vgs( ) a1 d1 vg1( )

TM01: 2π/3 Vg=4.7%

TM02: 2π/3 Vg=2.0%a

Vg

d

Direct comparison of VgCourtesy of

Riccardo Zennaro

18

0 1 2 3 4 5 6 7 8

x 104

0

500

1000

1500

2000

2500

Pulse number

En

erg

y

IncTransRefl

C30-sb

In total : • 4,101,250 pulses, mainly at 1 Hz

corresponding to 18.99 SLAC hours at 60 Hz• 2186 breakdowns

Weird things due to calibration problems (now solved)

Mathias Gerbaux - CTF3 Collaboration Technical meeting - 27/01/2009

19

0 0.5 1 1.5 2 2.5 3 3.5

x 104

0

500

1000

1500

2000

2500

3000

Pulse number

En

erg

y (

mJ

)

IncTransRefl

1 point every 50 pulses

C30-sb-reversed

In total : • 1,704,650 pulses, mainly at 1 Hz

corresponding to 7.89 SLAC hours at 60 Hz• 501 breakdowns

Weird things due to calibration problems(now solved)

Mathias Gerbaux - CTF3 Collaboration Technical meeting - 27/01/2009

500 1500 2500 35000

100

200

300

400

500

Energy (mJ)

Num

ber

of o

ccur

renc

es

C30-SBAll breakdowns

500 1500 2500 35000

100

200

300

400

500

Energy (mJ)

Num

ber

of o

ccur

renc

es

Reversed C30-SBAll breakdowns