High Gradient tests at the KEK Nextef Facility Shuji Matsumoto On behalf of Nextef Group Accelerator Laboratory, KEK WE08 XB-10 22/05/13 1 XB-10 Cockcroft Institute
Feb 06, 2016
High Gradient tests at the KEK Nextef Facility
Shuji MatsumotoOn behalf of Nextef Group
Accelerator Laboratory, KEK
WE08XB-10
23/04/22 1XB-10 Cockcroft Institute
ContentsThree 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.
23/04/22 2XB-10 Cockcroft Institute
Mission
Nextef 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.
Nextef stands for NEw X-band TEst Facility.
KT-1
•A 50MW facility originally started as an X-band klystron test station in 2006.•Small experiments utilizing the klystron power are performed.
23/04/22 3XB-10 Cockcroft Institute
Nextef100MW X-Band station in KEK
y/m2007/8 Start operation2007/12 Commissioning Power production went upto 100MW2008/1 Start Retest of KX032008/9 Start test of T182009/7 Start TD18Quad#5 test.2009/12 Starts TD18Disk2 test.2010/11 T24Disk installed. Ready to start testing
Features•Combined power of klystrons to produce 11.424GHz 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.
Nextef facility
23/04/22 4XB-10 Cockcroft Institute
Nextef station configuration
A: ModulatorB: KlystronsC: 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).
Nextef facility
23/04/22 5XB-10 Cockcroft Institute
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.
Nextef facility
23/04/22 6XB-10 Cockcroft Institute
• 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%.
PerformanceNextef facility
23/04/22 7XB-10 Cockcroft Institute
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.
Nextef facility
23/04/22 8XB-10 Cockcroft Institute
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, 11.424GHz 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.
Nextef facility
23/04/22 9XB-10 Cockcroft Institute
Test done at KT-1, an example.Test of CERN Cinel load and KEK Waveguide valve
CERN CINEL High power RF Load
KEK Waveguide valve
RF
2009.10.28
Lead Box(It is opened at this moment)
Nextef facility
23/04/22 10XB-10 Cockcroft Institute
We are operating two X-band high power stationsKT-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
Nextef facility
23/04/22 11XB-10 Cockcroft Institute
Klystrons
Spec of Nextef klystron: originally designed as a 75MW klystron for GLC project.
Frequency 11.424GHz
Peak Power 75MW
Pulse width 1.s
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
We have two(at Nextef)+one(at KT-1) ppm klystrons in operation and one spare.(“ppm” stands for “periodic permanent magnets”.)
Nextef facility
23/04/22 12XB-10 Cockcroft Institute
X-band klystrons
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.
Nextef facility
0.010.020.030.040.050.060.070.080.090.0
350.0 400.0 450.0 500.0 550.0Cathode Voltage (kV)
Pow
er
(MW
) /
Eff
. (%
)
PowerEfficiency
Result of high power testing. PPM4 (2003).
-50
-40
-30
-20
-10
0
10
20
30
-0.5
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0 1 2 3 4 5
Pulse Waveforms
Volts
Volts (RF Power Signal)
time (micro sec)
Cathode Voltage
RF Power
Cathode Current
Collector Current
38.4MW @ 439kV
23/04/22 13XB-10 Cockcroft Institute
Structure Test Procedure:How the research Collaboration works
Design for CLIC
(CERN)
Fabrication of parts (KEK)
Bonding (SLAC)
CP (SLA
C)
VAC bake (SLAC)
High power test
(NLCTA-SLAC)
High power test (Nextef-
KEK)
Two structures are made and treated as the twin for the evaluation.
TD18 Test
23/04/22 14XB-10 Cockcroft Institute
T18_Disk_#2 TD18_Disk_#2
Test structures made as twinsTD18 Test
23/04/22 15XB-10 Cockcroft Institute
Unloaded 100MV/mT18 TD18
0 2 4 6 8 10 12 14 16 180
50
100
150
200
250
iris number
P [
MW
] (b
lack
), E
s (
gre
en),
Ea (
red
) [M
V/m
],
T
[K
] (b
lue)
, S
c*50
[MW
/mm
2 ] (m
agen
ta)
8.1 12.5
148
232
2.7
4.4
76
126
53.0
37.4
Pinload = 53.0 MW, P
outload = 37.4 MW
Eff = 0.0 % tr = 0.0 ns, t
f = 0.0 ns, t
p = 100.0 ns
P (M
W),
Es
(MV
/m),
Ea
(MV
/m),
T
(C),
Sc*
50 (
MW
/mm
2)
Iris number
P
Ea
Sc
Es
T
T18 unloaded 100MV/m
0 2 4 6 8 10 12 14 16 180
50
100
150
200
250
iris number
P [
MW
] (b
lack
), E
s (
gre
en),
Ea (
red
) [M
V/m
],
T
[K
] (b
lue)
, S
c*50
[MW
/mm
2 ] (m
agen
ta)
29.1
47.0
155
226
3.2
4.4
79
120
57.5
34.3
Pinload = 57.5 MW, Pout
load = 34.3 MW
Eff = 0.0 % tr = 0.0 ns, tf = 0.0 ns, tp = 100.0 ns
P (M
W),
Es
(MV
/m),
Ea
(MV
/m),
T
(C),
Sc*
50 (
MW
/mm
2)
Iris number
P
Ea
Sc
Es
T
TD18 unloaded 100MV/m
TD18 Test
23/04/22 16XB-10 Cockcroft Institute
Processing of damped structuresTD18_Disk_#3 tested at SLAC
0 100 200 300 400 500 600 700 800 900 10000
20
40
60
80
100
120
Accumulated rf process time (hr)
BDR (1/hr)
<G> for regular cell (MV/m)
Pulse Width (divided by 10 ns)
0
1000
2000
3000
Total # BD
RF-ON period (hrs)
Pul
se w
idrh
(ns
)/10
E
acc
(MV
/m)
TD18_Disk_#2 tested at KEK
Big difference in ramping speed.(may be partially due to the difference in
processing procedure)Both reached 100MV/m 240ns.
TD18 Test
23/04/22 17XB-10 Cockcroft Institute
Breakdown rate T18 vs TD18
Undamped vs DampedSLAC vs KEK
Sensitive to Eacc (Es, Hs) ~ X10 / 5 (MV/m)BDR(damped) ~ 100 X BDR (undamped)
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.
TD18 Test
23/04/22 18XB-10 Cockcroft Institute
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!?
TD18 Test
23/04/22 19XB-10 Cockcroft Institute
Breakdown rate versus width or T
T18_Disk_#2BDR versus pulse width
Tight correlation with pulse width.
TD18 Test
TD18_Disk_#3 BDR versus T (pulse temperature rise)
Undamped
Damped
Tight correlation with pulse temperature rise. Same at 100MV/m but pulse length T Same T 83~84 degC but diff. Eacc
23/04/22 20XB-10 Cockcroft Institute
TD18_#2 Evolution of breakdown rateTD18 Test
23/04/22 21XB-10 Cockcroft Institute
TD18_#2 BDR versus widthat 100MV/m around 2800hr
and at 90MV/m around 3500hr
TD18 Test
The power of the width depends on Eacc. 23/04/22 22XB-10 Cockcroft Institute
Evolution of dark current till early April in TD18_Disk
Dark current reduced by three order of magnitude. It followed roughly modified F-N formula. reduced from 70 to 40. * Es
max ~ 5~7 GV/m
IPAC10TD18 Test
23/04/22 23XB-10 Cockcroft Institute
Comparison of dark current
0.01
0.1
1
10
100
50 60 70 80 90100 200
T18_#2 Dark Current evolution081128-081224-090224-090414-090515
FC-Mid [microA] (081128)
FC-Mid [microA] (081224)
FC-Mid microA (253ns, 090225)
FC-Mid microA (253ns, 090414)
FC-Mid microA 090515
FC-Mid microA
Eacc [MV/m]
T18_Disk
TD18_Disk
TD18_Quad
Eacc for peak dark current of 10 A90MV/m 70MV/m 40MV/m
TD18_Quad
Presented to IPAC10TD18 Test
23/04/22 24XB-10 Cockcroft Institute
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.
TD18 Test
23/04/22 25XB-10 Cockcroft Institute
•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.
TD18 Test
23/04/22 26XB-10 Cockcroft Institute
Run 98 switching in three power levels for 46.2 hours
Switching every second among three power levels, controlled by Ushimoto program.
90 – 95 – 100 MV/mRF pulse counts
Inpu
t po
wer
(M
W)
Histogram of pulses in terms of Eacc
Nicely separated with each other.
Input power (MW)
TD18 Test
23/04/22 27XB-10 Cockcroft Institute
Pulse width=252ns
Run 98
81
43 Higher power pulse makes bigger BDR.
We compare this result to the usual BDR plot…
3.3*10-6 / 2.6*10-5 / 1.4*10-4 [BD/pulse/m]
TD18 Test
23/04/22 28XB-10 Cockcroft Institute
Relevant data points of BDR vs Eacc
Data points from jumping among three levels
TD18 Test
23/04/22 29XB-10 Cockcroft Institute
Run 100 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.
TD18 Test
23/04/22 30XB-10 Cockcroft Institute
Vaccum issuesVacuum at low power level
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.
Acc Input power
Pressure @IN-WG
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 …
TD18 Test
23/04/22 31XB-10 Cockcroft Institute
6.3E-5
2.5E-7
4.0E-6
1.0E-6
T18
TD18
T24
It does not change so much now. We will watch the behaviour.
opeartion started
TD18 Test
23/04/22 32XB-10 Cockcroft Institute
Demands for system upgradeFuture plans
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.
23/04/22 33XB-10 Cockcroft Institute
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 750nsPCS 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.
Future plans
23/04/22 34XB-10 Cockcroft Institute
Nextef Delay Line Pulse Compressor
Kly
str
on
Acc
eler
ato
r
TE11R
TE11L
750ns → 150nsGain = 3.6 @ 3dB
22m - One ChannelK
lys
tro
nA
ccel
erat
or
1500ns → 300nsGain = 3.3 @ 3dB
TE11R
TE11L
TE12LTE12R
TE11 ModeConverter
TE11-TE12
Reflector (Mode exchanger)
Upgrade possible if we use a higher mode.
25 MW×2 Klyston×gain=3→ 150MW
Phase1
Phase2
Future plans
Tunable reflector
23/04/22 35XB-10 Cockcroft Institute
C-band RF Source(will be removed.)
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.
Nextef pulse compressor LayoutThe reflector will be put.
X-bandRF Source
PCS header(3dBHybrid+Mode Launcher)
Future plans
A
23/04/22 36XB-10 Cockcroft Institute
Production of RF components:Kazakov Mode Converter
(Rect TE10→Circ TE01)
Noboru Kudoh
Future plans
23/04/22 37XB-10 Cockcroft Institute
01020304050607080
0 200 400 600 800 1000 1200 1400 1600
ns
MW
PPM4 2003)(PPM4G2A(2006)
25MW,0.75us pulse compression
50MW,0.25us: current operation point
Parameter choice / klystron stability
The points reflect the limit of the RF instability
Future plans
pulse compressionPhase 2
23/04/22 38XB-10 Cockcroft Institute
Improvement Plan for KT-1Demand: 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.
Future plans
23/04/22 39XB-10 Cockcroft Institute
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.
Future plans
23/04/22 40XB-10 Cockcroft Institute
KT-1X-band KT-2
Reflector
Delay line
Shield-B
NextefX-band
Lead Box (now workingfor small experiments)
Future plans
23/04/22 41XB-10 Cockcroft Institute
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.
Future plans
23/04/22 42XB-10 Cockcroft Institute
Status Future plans
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.
23/04/22 43XB-10 Cockcroft Institute
Nextef and KT-1 as of 2011
A
KT-1toB
B
Nextef with PCS
Future plans
23/04/22 44XB-10 Cockcroft Institute
Summary: Future plansMost 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.
23/04/22 45XB-10 Cockcroft Institute
Future plans
Nextef near future plan revised as of IWLC10
9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2
2010 2011 2012
PPM6B, Component tests, NWG test or serving shield-B, …….
A
B
KT-1
C-band Structure Test
Components fabrication
KT1 B connection
PC commissioning
TD24_Disk #3
X-band basic studies Construction in shield-B
T24_Disk #3TD18_Disk #3
TD24_Disk_R05 ?? ??
PC Construction??
23/04/22 46XB-10 Cockcroft Institute
ConlusionsNextef, 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.
23/04/22 47XB-10 Cockcroft Institute
5th Collaboration Meeting on X-band Accelerator Structure Design and Test Program
Date: May 16th to 18th , 2011Place: Seminar hall of 4th building, KEK, Tsukuba, Japan (KEK Homepage: http://www.kek.jp/intra-e/)
Chaired by K.Ueno (KEK).Organized by T.Higo(KEK) S.Tantawi(SLAC) W. Wuensch(CERN).
23/04/22 48XB-10 Cockcroft Institute
Now we suppose those issues to be covered:
Structure design: electrical and mechanical Manufacturing of parts; disks, couplers and loadsState-of-the-art precision machiningFabrication of quadrant/half structureExperimental study on high gradient performanceBreakdown physics and simulationsOn-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!
23/04/22 49XB-10 Cockcroft Institute