Managed by UT-Battelle for the Department of Energy SNS MEBT : Beam Dynamics, Diagnostics, Performance. Alexander Aleksandrov Oak Ridge National Laboratory
Managed by UT-Battellefor the Department of Energy
SNS MEBT : Beam Dynamics, Diagnostics, Performance.
Alexander Aleksandrov
Oak Ridge National Laboratory
2 Managed by UT-Battellefor the Department of Energy
Introduction: the SNS Project
Parameters:
P beam on target 1.44MW
I beam aver. 1.44mA
Beam energy 1 GeV
Duty factor 6%
Rep. rate 60Hz
Pulse width 1ms
3 Managed by UT-Battellefor the Department of Energy
Introduction: the SNS Front End Systems
IS/LEBT
RFQ
MEBT
65kV; 48mA; .2 π mm mrad
2.5MeV; 38mA; .21 π mm mrad
2.5MeV; 38mA; .27 π mm mrad
3.8m
3.6m
12 cm
No beam diagnostics
No beam diagnostics
2 BCMs6 BPM5 WS
Slit/collect
Designed and build by Berkeley N L
1ms
60Hz
4 Managed by UT-Battellefor the Department of Energy
MEBT functions
Matching beam from RFQ to DTL– 3 quads and 1 buncher would be sufficient
Place for chopper – Requires relatively long drifts
– Increases compexity: match RFQ to chopper – chopper – match to DTL
Place for diagnostics– Try utilizing free space as much as possible.
5 Managed by UT-Battellefor the Department of Energy
SNS chopper functions
Ion source transient
Chopper window
single mini-pulse
Providing gaps for ring extraction kicker rise time – 700ns ON, 300ns OFF pattern at ~1MHz
Macro-pulse shaping– cut off beam during the ion source transient
– create single mini-pulse for turn-by-turn measurements in the ring
– decimate mini-pulses (N-on-M-off ) to reduce beam current
– ramp up current at the beginning of the macro-pulse to help LLRF system in compensating beam loading effects
6 Managed by UT-Battellefor the Department of Energy
SNS Front End chopping system
time
MEBT chopper (10ns, 1e-4)
curr
ent
LEBT chopper
(50ns, 1e-2)
Chopper LEBT MEBT
Ion energy ~25 keV 2.5 MeV
= v/c ~ 0.0073 .073
Max Voltage 3 kV 2.5 kV
gap ~ 14 mm 18 mm
Effective length
~ 27 mm ~350 mm
Max deflection 14 o 1.07 o
Time of flight ~ 12 ns ~ 17 ns
Two stage: LEBT + MEBT
7 Managed by UT-Battellefor the Department of Energy
2.5MeV MEBT
Schematic MEBT Layout
RMS beam envelope in MEBT. Simulation (solid) and wire scanner measurements (dots).
8 Managed by UT-Battellefor the Department of Energy
SNS MEBT chopper/anti-chopper arrangement
Theoretically can eliminate partially chopped beam
Requires identical choppers, power supplies and anti-symmetric transverse optic
9 Managed by UT-Battellefor the Department of Energy
SNS MEBT Layout
3.7 m of 14 quadrupole magnets; 4 buncher cavities; chopper system; diagnostics
Design beam envelope in the MEBT
10 Managed by UT-Battellefor the Department of Energy
Front End Systems in the FE building
11 Managed by UT-Battellefor the Department of Energy
MEBT BPM and Beam Box
12 Managed by UT-Battellefor the Department of Energy
Wire Scanner and MEBT Beam Box
13 Managed by UT-Battellefor the Department of Energy
New diagnostics
Anti-chopper proved to be unnecessary
Additional diagnostics is added into MEBT “d-box” (former “anti-chopper” box):– In-line beam stop/Faraday cup
– Horizontal and vertical in-line emittance scanners
– Pencil beam apertures (2mm ,1.5mm, 1mm)
– Laser based longitudinal profile measurement (temporary decommissioned)
– View screen (never was used)
– In line Fast Faraday (tested prototype only)
– Several ports for future developments
14 Managed by UT-Battellefor the Department of Energy
Beam diagnostics box replaced the anti-chopper
15 Managed by UT-Battellefor the Department of Energy
MEBT D-box
Slits
Harps
LaserPorts
16 Managed by UT-Battellefor the Department of Energy
MEBT diagnostics scrapers (low power)
collimator position
collimator design
Manually actuated
Observe collimation effect on:
1. MEBT emiitance device
2. Linac wire scanners
3. Loss monitors
17 Managed by UT-Battellefor the Department of Energy
MEBT Performance
MEBT was commissioned with 6% duty factor beam
50mA peak current transport with no measurable losses demonstrated
Anti-chopper has been removed as unnecessary (based on simulations)
X-ray radiation from the rebunchers exceeds the expected levels – Established Radiation Area around the MEBT. No
access allowed during operation
RF amplifiers do not provide design power for Rebunchers #1 and #4– Considering new design based on IOT or tapping RF
power from the RFQ klystron (more in T. Hardek’s talk)
MEBT chopper doesn’t perform to specifications yet
18 Managed by UT-Battellefor the Department of Energy
How we tune MEBT
RFQ tuning
MEBT rebunchers tuning
MEBT quads tuning
19 Managed by UT-Battellefor the Department of Energy
RFQ tuning
No diagnostics to measure absolute transmission No diagnostics to know what is going on inside Can characterize resulting beam in the MEBT
422 RF cells, strongest space charge Defines longitudinal and ,in large part, transverse emittances
Only one parameter to set : RF amplitude– Fit output current vs. RF power curve to PARMTEQ model to find the set point
RF field amplitude [%]
Tra
nsm
issi
on
[%
]
20 Managed by UT-Battellefor the Department of Energy
Characterizing RFQ output beam using MEBT wire scanners
Measure transverse profiles using 5 wire scanners
Search for input Twiss parameters to best fit model to measured data
Repeat several times with different quad settings
Twiss parameters have been stable since 2003
21 Managed by UT-Battellefor the Department of Energy
MEBT tuning Set strengths of 14 quadrupole magnets
– Establishing proper beam profile for chopper operation– Matching beam to DTL– Use design quad values usually (~5% PS calibration
accuracy)– Verify with wire profile measurements
Set strengths of 6 hor. and 6 ver. dipole steerers – for minimum beam centroid offset at 6 BPM locations
Set amplitude and phase of 4 rebuncher cavities– Set amplitude to design or max available (~80% of nominal)– Set phase to non-accelerating bunching phase using phase
scan– Verified once in 2004 with laser longitudinal profile scanner
22 Managed by UT-Battellefor the Department of Energy
MEBT transmission is always good
Beam current after RFQ (red), MEBT (blue)
23 Managed by UT-Battellefor the Department of Energy
MEBT rebuncher phase scan
24 Managed by UT-Battellefor the Department of Energy
Ambiguities during the set up
MEBT rebunchers phase– Resulting set point depends on BPMs selected for
scan – (5º - 10º uncertainty)
BPM05
BPM10
Example of MEBT rebuncher scan producing different results with different BPMs
25 Managed by UT-Battellefor the Department of Energy
Beam envelope measurements
First wire scanner in the MEBT
Last wire scanner in the MEBT
Gaussian core + ‘tail’ dependent on MEBT tuning
26 Managed by UT-Battellefor the Department of Energy
MEBT to DTL transverse matching
Vertical Beam profile measured at DTL3 exit for different matching quadrupole settings
Approaches Gaussian shape when best match is achieved
FPAE057, TPAT030
27 Managed by UT-Battellefor the Department of Energy
Beam profile measurements after DTL tank3 for various Q13,Q14 settings
28 Managed by UT-Battellefor the Department of Energy
Measurement of MEBT chopper kick strength
Kick strength is measured by measuring beam deflection at chopper target location
Kick strength of the prototype is ~15% below design value
New design will meet design specs with ~15% margin
29 Managed by UT-Battellefor the Department of Energy
MEBT chopper efficiency
Simple strip line of solid copper (~16 ns TOF)
30 Managed by UT-Battellefor the Department of Energy
Vertical beam position along the pulse. Chopper is OFF
Vertical beam position along the pulse. Chopper is ON
Chopper OFF
Chopper ON
LEBT chopper performance (II)
Measured effective beam size increase due to large fraction of partially chopped beam. With 750 Ohm resistor.
Much better now with 50Ohm resistor
31 Managed by UT-Battellefor the Department of Energy
Beam emittance after MEBT. Vertical (top) and horizontal (bottom)
Measured emittance after MEBT
Dependance of RMS emittance upon peak beam current
Design value = .3 mm mrad(RMS normalized)
32 Managed by UT-Battellefor the Department of Energy
Emittance along the pulse
Emittance at the MEBT exit is not sensitive to beam current
33 Managed by UT-Battellefor the Department of Energy
mode-locked laser diagnostics for longitudinal profile measurement
Longitudinal bunch profile in MEBT
Longitudinal bunch profile (blue dots) and Gaussian fit (red line).
Bottom right: rms bunch length vs. the rebuncher phase
(measurements – squares, simulation – stars, solid line – quadratic fit).
34 Managed by UT-Battellefor the Department of Energy
MEBT Collimation Experiments
Scrapers
downstream BCM
upstream BCM30
25
Emittance scan without scraper Emittance scan with scraper
35 Managed by UT-Battellefor the Department of Energy
Transverse tails in DTL
horizontal a = 8.6 sigma
Asymmetric tail in horizontal plane
Clear effect of MEBT scraper
Do not see effect in vertical plane from horizontal scraping – weak coupling
DTL wire scanners usable to 10-3 level (~ 4 sigma)
Significant portion of transverse tail originate in Front End
vertical a = 16 sigma
36 Managed by UT-Battellefor the Department of Energy
Symmetric tails in both plane
No measurable effect of MEBT scraper
CCL wire scanners usable to 10-2 level (~ 3 sigma)
Significant portion of transverse tail in CCL does not originate in Front End
or wire scanner measurements not reliable yet
Transverse tails in CCL
horizontal a = 13 sigma
vertical a = 12 sigma
37 Managed by UT-Battellefor the Department of Energy
PARMILA vs measurements
38 Managed by UT-Battellefor the Department of Energy
Do we understand beam dynamics in our linac well enough?
good model requires– Accurate computation algorithm– Accurate representation of beam line elements (strength, position, edge
fields, etc. ) More complicated in longitudinal plane as element parameters (RF settings)
are not given by found from the model itself
– Reliable verification Good beam diagnostics tools
Good model should be able – To provide good agreement with measured beam parameters – To predict change of beam parameters in response to changes in beam
line elements– To provide above capabilities from end to end, not only piecewise – Accuracy of agreement should be well within RMS size
We use different codes for centroid motion and RMS simulations– Online Model, PARMILA, IMPACT
39 Managed by UT-Battellefor the Department of Energy
Accuracy of beam modeling in different parts of the linac
Transv.
centroid
Transv.
RMS
Long.
centroid
Long.
RMSRFQ ? not so good ? ?
MEBT good good not so good not so good
DTL good not so good ? ?
CCL very good not so good very good not so good
SCL not so good ? good not so good
40 Managed by UT-Battellefor the Department of Energy
Do we understand halo?
Expected in transverse plane– Much studied during design process
– Observe non-Gaussian transverse profiles Do not have diagnostics to measure halo extent below ~ 1%
– Does not seem to be a source of losses so far DTL serves as a collimator ?
– Do not have reliable information on losses inside DTL Probably responsible for some ring injection losses
Did not expect in longitudinal plane– Very low level
– Have only limited direct diagnostics tools ( CCL BSMs )
Problem of initial distribution for simulations– “reference” distribution used at design stage seems not
adequate