1. High Gradient tests at the KEK Nextef Facility
WE08
XB-10
High Gradient tests at the KEK Nextef Facility
Shuji Matsumoto
On behalf of Nextef Group
Accelerator Laboratory, KEK
2018/9/20
XB-10 Cockcroft Institute

2. XB-10 Cockcroft Institute
Contents
Three topics I will report today are as follows
1) Review of two X-band high power stations in KEK; Nextef, the 100MW X-Band test station and KT-1, 50MW.
2) Recent high power test results of TD18_Disk#2 (a CLIC prototype structure with HOM damping slots).
3) A overview of our on-going programs such as the pulse compression system of Nextef as well as the future plans of X-band study.
2018/9/20
XB-10 Cockcroft Institute

3. XB-10 Cockcroft Institute
Nextef facility
Mission
Nextef stands for NEw X-band TEst Facility.
Nextef was proposed in 2006 and was constructed in 2007 as a reassembled facility of GLCTA to be a 100MW high power station for the X-band accelerating structure tests.
Since 2008, the facility runs for the development of CLIC prototype high gradient accelerator structures (of 100MV/m or beyond). The collaboration program is ongoing.
KT-1
A 50MW facility originally started as an X-band klystron test station in 2006.
Small experiments utilizing the klystron power are performed.
2018/9/20
XB-10 Cockcroft Institute

4. Nextef 100MW X-Band station in KEK
Nextef facility
Nextef 100MW X-Band station in KEK
y/m
2007/8 Start operation
2007/12 Commissioning Power production went upto 100MW
2008/1 Start Retest of KX03
2008/9 Start test of T18
/7 Start TD18Quad#5 test.
2009/12 Starts TD18Disk2 test.
2010/11 T24Disk installed. Ready to start testing
Features
Combined power of klystrons to produce GHz RF of peak power above 100MW (maximum power available depends on the pulse width).
Use of PPM klystrons (originally developed for GLC project).
Associated with a concrete shield (Shield-A) which is fully equipped for the structure high power tests.
No beam available.
2018/9/20
XB-10 Cockcroft Institute

5. Nextef station configuration
Nextef facility
Nextef station configuration
A: Modulator
B: Klystrons
C: Low Loss Power Transfer Line
D: Accelerating structure under test in Shield-A (5m W x 3m D x 2m H)
The modulator drives two 50MW PPM klystrons.
The combined power is transferred into Shield-A.
Local control room (not shown in this picture).
2018/9/20
XB-10 Cockcroft Institute

6. Plan View inside Shield-A observables along beam axis
Nextef facility
Plan View inside Shield-A observables along beam axis
ACC str: Structure under test
FC: Faraday Cup, PM: Profile Monitor,
AM: Analyzer magnet, GV: Gate Valve.
Comment: Some of the details are incorrect. The drawing should be updated.
2018/9/20
XB-10 Cockcroft Institute

7. XB-10 Cockcroft Institute
Nextef facility
Performance
　24Hr operation is possible due to the linkage with KEKB Linac operation. (The linac operation=6000Hr/year.)
　Local control as well as a remote control through Linac control system.
　Possible to store / handle the operation data as well as experimental data through Linac control system (LINUX and EPICS).
Specifications
Frequency
11.424GHz
Max power production
100MW
Max power for test *
75MW
Pulse width
400ns
Repetition rate
50pps
* Measured Transmission Loss is 25%.
2018/9/20
XB-10 Cockcroft Institute

8. What we have achieved : Nextef
Nextef facility
What we have achieved : Nextef
Continuous run established (Total RFON time is more than 20000hrs from 2006) .
Maximum power 70MW, 250ns, 50pps for recent structure tests.
The maximum power production is about 100MW. It is practically determined by the performance of one of the klystrons: Frequent Klystron gun breakdowns occur in one of the klystrons at the cathode voltage above 460kV. Below this voltage klystrons are stable.
2018/9/20
XB-10 Cockcroft Institute

9. KT-1 : 50MW X-band test station
Nextef facility
KT-1 : 50MW X-band test station
(y/m)
2006/ 5 Start operation (as an X-band klystron test stand)
2006/6 Klystron test (PPM6A)
2007/1 Started to run for “small experiments”
2008/5 Narrow waveguides (SUS003 followed by CU005) testing.
2009/9 Testing RF Loads
Features
Use of a single PPM klystron, GHz RF of peak power of 50MW, 400ns is available. This station runs for small-size experiments such as a narrow waveguide and RF component test (e.g. RF load).
The current klystron sitting at KT-1 is old (constructed in 2003, repaired 2004) but it has been healthy so far.
24Hr operation is possible.
This station is rather crucial for Nextef, since sometimes we need to check the performance of the high power devices before their installation.
2018/9/20
XB-10 Cockcroft Institute

10. XB-10 Cockcroft Institute
Nextef facility
Test done at KT-1, an example. Test of CERN Cinel load and KEK Waveguide valve
Lead Box
(It is opened at this moment) RF
CERN CINEL
High power RF Load
KEK Waveguide valve
2018/9/20
XB-10 Cockcroft Institute

11. We are operating two X-band high power stations
Nextef facility
We are operating two X-band high power stations
KT-1 (50MW)
Klystron test, small experiments (in Lead shield box) or component tests
Test area in the bunker(Sheild-A)
Nextef (100MW)
Modulator and twin ppm klystrons
2018/9/20
XB-10 Cockcroft Institute

12. XB-10 Cockcroft Institute
Nextef facility
Klystrons
We have two(at Nextef)+one(at KT-1) ppm klystrons in operation and one spare.
(“ppm” stands for “periodic permanent magnets”.)
Frequency
11.424GHz
Peak Power
75MW
Pulse width
1.6ms
Repetition
150Hz
Cathode Voltage
480kV
Cathode Current
266A
Perveance
0.8uK
Efficiency
>55%
Main Focus
PPM
Max B / period
0.32T / 30mm
Magnet Material
NdFeB
Spec of Nextef klystron: originally designed as a 75MW klystron for GLC project.
2018/9/20
XB-10 Cockcroft Institute

13. XB-10 Cockcroft Institute
Nextef facility
X-band klystrons
Result of high power testing. PPM4 (2003).
We usually operate it at 50MW 400ns 50pps maximum. One tube has frequent gun breakdowns when it is operated more than at ~460kV. The performance of our X-band ppm focus klystron has a limit in available peak
power and pulse width by RF instability.
So far the tubes are good enough for our purposes.
2018/9/20
XB-10 Cockcroft Institute

14. Structure Test Procedure: How the research Collaboration works
TD18 Test
Structure Test Procedure: How the research Collaboration works
Design for CLIC (CERN)
High power test (NLCTA-SLAC)
High power test (Nextef-KEK)
Fabrication of parts (KEK)
CP (SLAC)
VAC bake (SLAC)
Bonding (SLAC)
Two structures are made and treated as the twin for the evaluation.
2018/9/20
XB-10 Cockcroft Institute

15. Test structures made as twins
TD18 Test
Test structures made as twins
T18_Disk_#2
TD18_Disk_#2
2018/9/20
XB-10 Cockcroft Institute

16. Unloaded 100MV/m T18 TD18 TD18 Test
P (MW), Es (MV/m), Ea (MV/m), DT(C), Sc*50 (MW/mm2)
Iris number P Ea Sc Es DT
T18 unloaded
100MV/m
P (MW), Es (MV/m), Ea (MV/m), DT(C), Sc*50 (MW/mm2)
Iris number P Ea Sc Es DT
TD18 unloaded
100MV/m
2018/9/20
XB-10 Cockcroft Institute

17. Processing of damped structures
TD18 Test
Processing of damped structures
TD18_Disk_#3
tested at SLAC
TD18_Disk_#2
tested at KEK
1000
2000
3000
Total # BD
RF-ON period (hrs)
Pulse widrh (ns)/10 Eacc (MV/m)
Big difference in ramping speed.
(may be partially due to the difference in processing procedure)
Both reached 100MV/m 240ns.
2018/9/20
XB-10 Cockcroft Institute

18. XB-10 Cockcroft Institute
TD18 Test
Breakdown rate
T vs TD18
Undamped vs Damped
SLAC vs KEK
Damped structure:
High pulse temperature rise?
Any other mechanism?
Single-cell SW test also shows tight correlation of BDR to pulse temperature rise.
Helps if reducing temperature rise?
See the result of TD24 to be tested in several months.
Sensitive to Eacc (Es, Hs)
~ X10 / 5 (MV/m)
BDR(damped) ~ 100 X BDR (undamped)
2018/9/20
XB-10 Cockcroft Institute

19. TD18_#2 all BDR versus Eacc
TD18 Test
TD18_#2 all BDR versus Eacc
It seems difficult to get a smooth curve as function of Eacc by collecting all data points scattered in time and in operation parameter space.
Usually we get a smooth curve by measuring intentionally with focusing to take data of BDR vs Eacc!?
2018/9/20
XB-10 Cockcroft Institute

20. Breakdown rate versus width or DT
TD18 Test
Breakdown rate versus width or DT
TD18_Disk_#3
BDR versus DT
(pulse temperature rise)
Undamped
Damped
Tight correlation with pulse temperature rise.
Same at 100MV/m but pulse length  DT
Same DT 83~84 degC but diff. Eacc
T18_Disk_#2
BDR versus pulse width
Tight correlation with pulse width.
2018/9/20
XB-10 Cockcroft Institute

21. TD18_#2 Evolution of breakdown rate
TD18 Test
TD18_#2 Evolution of breakdown rate
2018/9/20
XB-10 Cockcroft Institute

22. XB-10 Cockcroft Institute
TD18 Test
TD18_#2 BDR versus width at 100MV/m around 2800hr and at 90MV/m around 3500hr
The power of the width depends on Eacc.
2018/9/20
XB-10 Cockcroft Institute

23. Evolution of dark current till early April in TD18_Disk
TD18 Test
Evolution of dark current till early April
in TD18_Disk
IPAC10
Dark current reduced by three order of magnitude.
It followed roughly modified F-N formula.
b reduced from 70 to 40.
b * Esmax ~ 5~7 GV/m
2018/9/20
XB-10 Cockcroft Institute

24. Comparison of dark current
TD18 Test
Presented to IPAC10
Comparison of dark current
T18_Disk
TD18_Disk
TD18_Quad
TD18_Quad
Eacc for peak dark current of 10 mA
90MV/m MV/m MV/m
2018/9/20
XB-10 Cockcroft Institute

25. Summary of TD18_Disk#2 testing
TD18 Test
Summary of TD18_Disk#2 testing
Testing of TD18_Disk#2 structure is finished successfully.
Breakdown rates were and is being measured. It is 10-4/pulse/m at 100MV/m at 252nsec. It is fair to conclude that BDR of TD18_Disk#2 is greater than that of T18 by about one-two order of magnitude (depends on Eacc).
The amount of dark current has been decreasing by a few order of magnitude during the test (over some 2000hrs) and became very similar to that of T18.
The field enhancement factor, beta, has decreased by factor of 2. The final value is some 40.
Dark current was initially reduced but recently has been saturated.
2018/9/20
XB-10 Cockcroft Institute

26. XB-10 Cockcroft Institute
TD18 Test
A few attempts to understand the properties of RF breakdowns in the structure were done: operation with RF double pulses, operation with additional eight pulses after an RF reflection interlock and operation with switching RF input power periodically.
I quickly show the latter two results.
2018/9/20
XB-10 Cockcroft Institute

27. Run 98 switching in three power levels for 46.2 hours
TD18 Test
Run 98 switching in three power levels for 46.2 hours
Histogram of pulses in terms of Eacc
Input power (MW)
Input power (MW)
90 – – MV/m
RF pulse counts
Nicely separated with each other.
Pulse width=252ns
Switching every second among three power levels, controlled by Ushimoto program.
2018/9/20
XB-10 Cockcroft Institute

28. XB-10 Cockcroft Institute
TD18 Test
Run 98 43
Higher power pulse makes bigger BDR.
We compare this result to the usual BDR plot… 8 1
3.3*10-6 / 2.6*10-5 / 1.4* [BD/pulse/m]
2018/9/20
XB-10 Cockcroft Institute

29. Relevant data points of BDR vs Eacc
TD18 Test
Relevant data points of BDR vs Eacc
Data points from jumping among three levels
2018/9/20
XB-10 Cockcroft Institute

30. Run 100 at 100 MV/m level Behavior of eight pulses after BD
TD18 Test
Run at 100 MV/m level Behavior of eight pulses after BD
More than 50% of breakdowns are not accompanied by any consecutive breakdowns.
Some are followed by breakdowns.
Correlation between number of consecutive breakdowns and the hardness of the initial breakdown should be evaluated.
2018/9/20
XB-10 Cockcroft Institute

31. Vaccum issues Vacuum at low power level
TD18 Test
Vaccum issues Vacuum at low power level
Acc Input power
Outgassing is observed at the monitor located at left arm of Acc output port when the power comes down (about a few MW) .
Exact location of the outgas has not been identified yet.
We inserted a long waveguide (and put an additional pump) between the structure and the load to “separate” the vacuum. We expect better vacuum due to the added pump but …
2018/9/20
XB-10 Cockcroft Institute

32. XB-10 Cockcroft Institute
TD18 Test
T18
opeartion started
TD18
6.3E-5
2.5E-7
4.0E-6
1.0E-6
T24
It does not change so much now. We will watch the behaviour.
2018/9/20
XB-10 Cockcroft Institute

33. Demands for system upgrade
Future plans
Demands for system upgrade
There are always demand to improve the performance of a facility.
Of course this is true for us..
For example, we have a room to improve the peak power of Nextef by introducing a relevant pulse compressor system, which should be useful for accelerating structure studies. Also from a technical point of view, this should be somewhat important for us to actually handle extremely high power over one hundred MW.
We have started this upgrade work within the same framework of collaboration between three labs.
2018/9/20
XB-10 Cockcroft Institute

34. Nextef Pulse Compressor
Future plans
Nextef Pulse Compressor
A pulse compression system (PCS) will be installed in order to improve the peak power of Nextef 100→150MW. 100MW power available for structure tests.
A practical solution to getting higher peak power, necessary for the structure
tests in the future.
Use of single delay line (circular waveguide of phi 80mm). The line partially exists already as a C-band transmission line in Nextef experimental area. The circular polarized TE11 mode in X band is employed to store energy in this line.
The compression gain of 3 expected. Final power of 150MW 150ns pulse width is expected in Phase 1.
PCS INPUT :X-band PPM Klystrons: 25 MW each X2 X 750ns
PCS OUTPUT: Gain 3 → 150MW, 150ns.
In Phase 2 we utilize TE12 mode also to obtain 300ns output pulse while the peak power will be kept.
2018/9/20
XB-10 Cockcroft Institute

35. Nextef Delay Line Pulse Compressor
Future plans
Nextef Delay Line Pulse Compressor
22m - One Channel
25 MW×2 Klyston
×gain=3
→ 150MW
TE11L
Klystron
750ns → 150ns
Gain = 3dB
Phase1
Accelerator
TE11R
Tunable reflector
TE11 Mode
Converter
Upgrade possible if we use a higher mode.
TE11L
Klystron
1500ns → 300ns
Gain = 3dB
Phase2
TE12R
TE12L
Accelerator
TE11R
TE11-TE12
Reflector (Mode exchanger)
2018/9/20
XB-10 Cockcroft Institute

36. Nextef pulse compressor Layout
Future plans
Nextef pulse compressor Layout
The reflector will be put.
22m Delay line
Originally constructed as the C-band transmission line in 2009 and will be converted as the X-band delay line for pulse compressor.
C-band RF Source
(will be removed.)
X-band
RF Source A
PCS header
(3dBHybrid+Mode Launcher)
2018/9/20
XB-10 Cockcroft Institute

37. Production of RF components: Kazakov Mode Converter
Future plans
Production of RF components: Kazakov Mode Converter
(Rect TE10→Circ TE01)
Noboru Kudoh
2018/9/20
XB-10 Cockcroft Institute

38. Parameter choice / klystron stability
Future plans
Parameter choice / klystron stability
25MW,0.75us pulse compression
pulse compression
Phase 2
50MW,0.25us: current operation point　
The points reflect the limit of the RF instability
2018/9/20
XB-10 Cockcroft Institute

39. Improvement Plan for KT-1
Future plans
Improvement Plan for KT-1
Demand:
There has been some requests on the testing of the cavity structures such as single cell structures, short structures, so on.
Those experiments do not require much power and it can be provided by our single klystron.
Problem:
However it is difficult to perform these programs at current KT-1 station where only a small lead box is there for small experiments. It is not practical to make a good shield nearby.
Solution:
A simple solution is to use Shield-B which locates in next door. This can be done by constructing a new power transfer line.
2018/9/20
XB-10 Cockcroft Institute

40. Power delivery from KT-1 to Shield-B
Future plans
Power delivery from KT-1 to Shield-B
We are able to start basic studies with structures using this test setup, though it sacrifices the usual KT-1 klystron activities (=Klystron test) somehow.
This experimental activities should complement the series of high gradient tests on CLIC prototype structures proceeded at Nextef.
2018/9/20
XB-10 Cockcroft Institute

41. XB-10 Cockcroft Institute
Future plans
Nextef
X-band
Shield-B
Lead Box (now working
for small experiments)
Delay line
Reflector
KT-1
X-band
KT-2
2018/9/20
XB-10 Cockcroft Institute

42. Whole Configuration of the Proposed Power Line from KT-1 to Shield-B
Future plans
Whole Configuration of the Proposed Power Line from KT-1 to Shield-B
The power from the klystron is transmitted through the circular waveguides (40mm diameter) with TE01 Mode. Those circular pipes, bends and vacuum ports are common to those used in the pulse compressor.
2018/9/20
XB-10 Cockcroft Institute

43. XB-10 Cockcroft Institute
Future plans
Status
Components: More than 80% fabricated.
A hole was made through a wall between KT1 and shield-B.
Installation of KT1-B in early next year.
Pulse compression setup will follow KT1-B setup.
2018/9/20
XB-10 Cockcroft Institute

44. XB-10 Cockcroft Institute
Future plans
Nextef and KT-1 as of 2011
KT-1toB
Nextef with PCS A B
2018/9/20
XB-10 Cockcroft Institute

45. XB-10 Cockcroft Institute
Future plans
Summary: Future plans
Most of the components of those lines for Nextef Pulse Compressor and KT-1 to B are in the final stage of the fabrication. The whole parts will be ready for construction soon.
The construction depends on the schedule of the testing in Shield-A and –B. The construction work will start in the end of this year.
We expect the power of more than 100MW * with 150ns pulse width in Nextef as well as that of 35MW *, 400ns in Shield-B available.
* Due to the transmission loss.
2018/9/20
XB-10 Cockcroft Institute

46. Nextef near future plan revised as of IWLC10
TD18_Disk #3
T24_Disk #3 A
TD24_Disk_R05 ?? ??
TD24_Disk #3
Components fabrication ??
PC commissioning
PC Construction
C-band Structure Test
Construction in shield-B B
X-band basic studies
KT1  B connection
PPM6B, Component tests, NWG test or serving shield-B, …….
KT-1
2018/9/20
XB-10 Cockcroft Institute

47. XB-10 Cockcroft Institute
Conlusions
Nextef, a 100MW X-band power station in KEK, has been running for the test of high gradient structures in last three years. The facility is dedicated for CLIC prototype structure tests in the framework of an international collaboration.
High power test of the structure TD18_Disk, the structure with damping slots has been over. The basic time trend of the BDR is and it is fair to say that its final value is from ten to hundred times higher (the factor depends on Eacc) than that of T18 structure.
Works for the future plans are ongoing to establish the higher peak power at Nextef and to provide a new experimental area (Shield-B) for KT-1.
2018/9/20
XB-10 Cockcroft Institute

48. XB-10 Cockcroft Institute
5th Collaboration Meeting on X-band Accelerator Structure Design and Test Program
Date: May 16th to 18th , 2011
Place: Seminar hall of 4th building, KEK, Tsukuba, Japan
(KEK Homepage:
Chaired by K.Ueno (KEK).
Organized by T.Higo(KEK) S.Tantawi(SLAC) W. Wuensch(CERN).
2018/9/20
XB-10 Cockcroft Institute

49. XB-10 Cockcroft Institute
Now we suppose those issues to be covered:
Structure design: electrical and mechanical
Manufacturing of parts; disks, couplers and loads
State-of-the-art precision machining
Fabrication of quadrant/half structure
Experimental study on high gradient performance
Breakdown physics and simulations
On-going and possible new collaborations
…. Others
In this 5th collaboration meeting, we set one day dedicated for the effort to understand and suppress the vacuum breakdowns in addition to the usual collaboration meeting. Especially we try to introduce those activities ongoing in Japan for making perfect or special surface based on the state-of-the-art precision machining technology.
Details and registration procedures will be posted at conference web site soon.
See you in Tsukuba in next spring!
2018/9/20
XB-10 Cockcroft Institute