Upload
amity-england
View
41
Download
4
Embed Size (px)
DESCRIPTION
Monitoring Beam Intensity in the Tevatron Abort Gap Using Synchrotron Radiation. Randy Thurman-Keup FNAL / AD / Instrumentation. Outline. Motivation for Monitoring the Abort Gap Beam Intensity Synchrotron Radiation Synchrotron Radiation Devices in the Tevatron Some Results - PowerPoint PPT Presentation
Citation preview
f
Monitoring Beam Intensity in the Tevatron Abort Gap Using Synchrotron Radiation
Randy Thurman-KeupFNAL / AD / Instrumentation
f
2/21/2006 R. Thurman-Keup - AD Seminar 2
Outline
Motivation for Monitoring the Abort Gap Beam Intensity
Synchrotron Radiation Synchrotron Radiation Devices in the
Tevatron Some Results Other Odds and Ends
f
2/21/2006 R. Thurman-Keup - AD Seminar 3
Contributors
Tom Meyer – AGI DAQ Eugene Lorman – Synclite Sten Hansen, Heide Schneider – Gating circuit Carl Lundberg, Dale Miller – Technical expertise Sasha Valishev – Synch. Rad., Acc. Phys., etc… Alan Hahn, Harry Cheung, Pat Hurh – Synclite Jim Fast, Ken Schultz, Mark Ruschman, Carl
Lindenmeyer, Ron Miksa – Mech. Mods Morris Binkley – Big giant pulser Stefano de Santis, John Byrd – LBNL Gated PMT Stephen Pordes – Driving force
f
2/21/2006 R. Thurman-Keup - AD Seminar 4
Introduction
The Tevatron operates with 36 bunches in 3 groups called trains
Between each train there is an abort gap that is 139 RF buckets long RF bucket is 18.8 ns Abort gap is 2.6 s
In this talk, abort gap beam intensity measurements are usually normalized to beam around the ring
139 buckets
21 buckets 1113 RF buckets totalTrain
Bunch
Abort Gap
f
2/21/2006 R. Thurman-Keup - AD Seminar 5
Motivation During an abort
Abort kicker magnet ramps up during abort gap Beam in the abort gap is directed towards magnets, CDF,
etc… Quenches (in the past, 10 x 109 caused quenches)
Recent experience……~40 x 109 did not quench (better collimation)
CDF silicon detector damage DØ not impacted as much, protected by CDF collimator
Previous monitors relied on counters external to the beampipe that were timed with abort gap Measured stuff leaving the abort gap, not stuff still in it
Use synchrotron radiation to directly measure abort gap beam Want to be sensitive to a DC beam that is 1 part in 104 of
the total beam
f
2/21/2006 R. Thurman-Keup - AD Seminar 6
Synchrotron Radiation History 100 yrs ago, calculations of radiation from circular
paths Larmour, Liénard, Schott, Schwinger(later), etc…
1947, Observed at 70 MeV e- synchrotron at GE 1944, Could have been betatron radiation
if not for the shielding around the tube 1977, R. Coïsson calculates radiation from
non-uniform magnetic fields such as magnet edges Significant radiation beyond the cutoff
frequency which makes it possible to seeat high-energy proton machines
1979, First observation of protonsynchrotron radiation at CERN
Early 90s, A. Hahn and P. Hurh produceprototype synchrotron radiation detector for Tevatron
f
2/21/2006 R. Thurman-Keup - AD Seminar 7
Synchrotron Radiation
Radiated intensity γ4
Radiated intensity has a peak near the “critical frequency” and drops exponentially beyond Critical frequency γ3
Usually too infrared in proton machines Magnet edge enhances higher frequencies
ret
ret
3
),(
)1(
}){(),(
EnxB
nβ
ββnnxE
t
Rc
et
Light is emitted in cone½ angle ~ 1/
2
ret
2
dteRdd
Id ti
E
magnet~ Bβ
Dipole
f
2/21/2006 R. Thurman-Keup - AD Seminar 8
Synchrotron Radiation @ FNAL
TeV Dipole Critical wavelength ~ 2700 nm
TeV Dipole Edge Critical wavelength ~ 220 nm
Protons
Proton
Antiproton
Antiprotons
Proton light source
DipoleDipole
Half-Dipole
C11 warm sectionPickoff Mirrors
proton pbar
System known as Synclite
f
2/21/2006 R. Thurman-Keup - AD Seminar 9
Synchrotron Radiation @ FNAL
40
-4
mm
840-4-8
mm
40
-4
mm
840-4-8
mm
40
-4
mm
840-4-8
mm
40
-4
mm
840-4-8
mm
40
-4
mm
840-4-8
mm
40
-4
mm
840-4-8
mm
40
-4
mm
840-4-8
mm
40
-4
mm
840-4-8
mm
40
-4
mm
840-4-8
mm
104
2
4
6
810
5
2
4
6
810
6
phot
ons
/ 25n
m /
60 x
109 p
roto
ns
1000800600400200Wavelength (nm)
200 nm 300 nm 400 nm
500 nm 600 nm 700 nm
800 nm 900 nm 1000 nm
Illumination at the pickoff mirror
200 400 600 800 100010
0
102
104
106
Wavelength (nm)
phot
ons
/ 25
nm
/ 6
0 x
109 p
roto
ns
1000 GeV
900 GeV
800 GeV
700 GeV
600 GeV
500 GeV
400 GeV
980 GeV
This data generated by Synchrotron Radiation Workshop (SRW) calculation
Typical Proton Bunch
Half million γ / 20 nm
f
2/21/2006 R. Thurman-Keup - AD Seminar 10
Expected # of photons
Wanted to measure DC beam of 1 part in 104
Total beam is 1013 want to measure 109 1.2 x 108 / abort gap
Synchrotron radiation calculation # of 400 nm photons / 25nm / 6x1010 protons ~ 2 x 105
Optical losses ~ 35% efficiency (50% from beam splitter) # of photons / 109 DC beam / 25 nm / rf bucket ~ 1
Wavelength acceptance ~ 200nm Gating duration ~ 30 buckets PMT Quantum Efficiency ~ 15% Typical # of photoelectrons ~ 40
f
2/21/2006 R. Thurman-Keup - AD Seminar 11
Abort Gap Intensity Monitor Made use of existing synchrotron light system
Measures beam profile, including abort gap, using lens and camera
Added beam splitter and gated photomultiplier tube
f
2/21/2006 R. Thurman-Keup - AD Seminar 12
Synclite Device
X-Y Mirror
X-Y Mirror
Pickoff Mirror
Beampipe
Beam Splitter
25:1 Filter
CID Camera
Filter Wheel
Synchrotron Light
Quartz Window
PMT Module*
Lens
Proton Optics Box
PMT Module*Contains optional 100:1 Filter
25:1 and 100:1 filters are Neutral Density Filters
f
2/21/2006 R. Thurman-Keup - AD Seminar 13
Synclite Device
Future home of PMT
Proton Synclite Box
Antiproton Synclite Box
Picture taken beforePMT installation
f
2/21/2006 R. Thurman-Keup - AD Seminar 14
Synclite Device
Synchrotron light spot
~ 9 inBeam splitter
Picture taken with small CCD camera
The specks are radiation-damaged pixels
f
2/21/2006 R. Thurman-Keup - AD Seminar 15
Abort Gap Intensity Monitor
Synchrotron light box
Abort Gap PMT
f
2/21/2006 R. Thurman-Keup - AD Seminar 16
Abort Gap Intensity Monitor Made use of existing synchrotron light system
Measures beam profile, including abort gap, using lens and camera
Added beam splitter and gated photomultiplier tube Photomultiplier had to be gateable and insensitive to light
present just before the gate (bunch intensity is several thousand times brighter than DC beam) Rules out just gating the output of the PMT
f
2/21/2006 R. Thurman-Keup - AD Seminar 17
Abort Gap Intensity Monitor Made use of existing synchrotron light system
Measures beam profile, including abort gap, using lens and camera
Added beam splitter and gated photomultiplier tube Photomultiplier had to be gateable and insensitive to light
present just before the gate (bunch intensity is several thousand times brighter than DC beam) Rules out just gating the output of the PMT Custom gating circuit for generic photomultiplier
Pulse two dynodes
f
2/21/2006 R. Thurman-Keup - AD Seminar 18
Abort Gap Intensity Monitor
Photocathode
Dynode 1
Photocathode
HV below photocathode
Nominal PMT Behavior (Gated On)
Gated Off PMT Behavior
Dynode 2
Dynode 3
Dynode 4
Dynode 1
Dynode 2
Dynode 3
Dynode 4
HV below dynode 3
Two dynodes are capacitively coupled to pulsed voltage source.
When pulsed, the dynodes are pushed to their nominal voltage level.
f
2/21/2006 R. Thurman-Keup - AD Seminar 19
Abort Gap Intensity Monitor Made use of existing synchrotron light system
Measures beam profile, including abort gap, using lens and camera
Added beam splitter and gated photomultiplier tube Photomultiplier had to be gateable and insensitive to light
present just before the gate (bunch intensity is several thousand times brighter than DC beam) Rules out just gating the output of the PMT Custom gating circuit for generic photomultiplier
Pulse two dynodes ~200 ns settling time after gate application Relatively inexpensive Not all PMT’s work correctly
End window tube had 10 μs transient after application of gate Side window tube worked better, installed for several months
Some sensitivity to pre-gate light
f
2/21/2006 R. Thurman-Keup - AD Seminar 20
Abort Gap Intensity Monitor
Anode output of PMT into 50 ohms
HV Gate start
2 usec
Note turn-on transient
LED Pulses
Test stand used a pulsed blue LED to simulate the beam bunches and a constant low level green LED to simulate the DC beam
Output of PMT into 50 ohms
HV Gate start
20 usec
Note turn-on transient
End-window tube gating transient
Pre-gate light sensitivity
f
2/21/2006 R. Thurman-Keup - AD Seminar 21
Abort Gap Intensity Monitor Made use of existing synchrotron light system
Measures beam profile, including abort gap, using lens and camera
Added beam splitter and gated photomultiplier tube Photomultiplier had to be gateable and insensitive to light
present just before the gate (bunch intensity is several thousand times brighter than DC beam) Rules out just gating the output of the PMT Custom gating circuit for generic photomultiplier Hamamatsu gated regular PMT (inexpensive, $5K)
PMT module complex circuitry would be in tunnel Exhibited same sensitivity to light as FNAL gating circuit
f
2/21/2006 R. Thurman-Keup - AD Seminar 22
Abort Gap Intensity Monitor Made use of existing synchrotron light system
Measures beam profile, including abort gap, using lens and camera
Added beam splitter and gated photomultiplier tube Photomultiplier had to be gateable and insensitive to light
present just before the gate (bunch intensity is several thousand times brighter than DC beam) Rules out just gating the output of the PMT Custom gating circuit for generic photomultiplier Hamamatsu gated regular PMT (inexpensive, $5K) Hamamatsu gated MCP style PMT on loan from LBNL
Expensive! (~$20K /tube) 2-stage Micro Channel Plate PMT – Gain of <= 106
5ns minimum gating time w/no noticeable settling time No sensitivity to pre-gate light Very large extinction ratio
f
2/21/2006 R. Thurman-Keup - AD Seminar 23
DAQ Systems
Gated PMT in tunnel
Linear Amplifier in service building
Fast Integrator
DigitizerMVME
Processor
VME Crate
Accelerator Network
CID Camera in tunnel Framegrabber Rackmount
PC in service building
RS-170 Video
Accelerator Network
Synclite DAQ System
Abort Gap DAQ System
Average 1000 turns of data in each of the 3 abort gaps.
For abort gap:
Average 200 camera frames of data, each containing ~75 turns
f
2/21/2006 R. Thurman-Keup - AD Seminar 24
Abort Gap Intensity (Synclite)
Peak is ~7 x 109
CID Camera Image
Raw Background Subtracted
f
2/21/2006 R. Thurman-Keup - AD Seminar 25
Abort Gap Intensity Monitor
TEL On
Tevatron Energy (980 GeV)
Gate 1
Gate 2
Gate 3
TEL Off
0 1 2 3 4 5Time (hrs)
Ab
ort
Gap
Bea
m In
ten
sity
(10
9 p
arti
cles
)
-1
0
1
2
3
4
TEL On
Gate 3
Abort Gap 3
TEL On
Abort Gap 2
Gate 2
TEL On
Abort Gap 1
Gate 1
Abort Kicker Turn-on
f
2/21/2006 R. Thurman-Keup - AD Seminar 26
Calibration
Two options:Calibrate by gating on a bunch (with attenuators in place) and comparing to FBI intensity
Calibrate by turning the TEL off and then back on and comparing to the DCCT measurement of the accumulated beam intensity
f
2/21/2006 R. Thurman-Keup - AD Seminar 27
Calibration
Periodically check this using TEL trips. Seems to vary by 40-50%
f
2/21/2006 R. Thurman-Keup - AD Seminar 28
Abort Gap Intensity Monitor
0
20
40
60
80
0 45 90 135 180Time (min)
0
20
40
60
80
Ab
ort
Gap
Inte
nsi
ty (
109
par
ticl
es)
Beam
Inten
sity (1012 p
articles)
Abort gap beam intensity after longitudinal damper went berserk and started shaking beam everywhere (AGI saturates at ~70E9)
f
2/21/2006 R. Thurman-Keup - AD Seminar 29
Microbunches
0 50 100 150 200 250 300 350 4000
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
Time (2ns)
Bunch 1 Bunch 36
End of Abort Gap 3
Bunch 36 is 2 s to the left of this plot
The red crosses show the location of the center of the RF buckets.
These data were collected using a PMT pulse-over-threshold counting technique
These small bunches are ~105 smaller than a normal bunch
Longitudinal profile in the end of an abort gap
Captured beam should be bunched in the center of RF buckets
f
2/21/2006 R. Thurman-Keup - AD Seminar 30
0
500 50 100 150 200
-5
0
5
10
15
20
25
Inte
nsity
(E
9 ar
ound
the
rin
g)
DC Beam Elsewhere
Take 1 6-bucket sample between each bunch and ~10 in each abort gap
~30 hour store duration (sample # = time)
0 10 20 30 40 50 60 0
100
200
-5
0
5
10
15
20
25
Inte
nsity
(E
9 ar
ound
the
rin
g)
Sample #
Position #
Position # Sample #
f
2/21/2006 R. Thurman-Keup - AD Seminar 31
DC Beam Elsewhere
0 50 100 150 200 2500
7
14
21
28
-10
0
10
20
30
40
50
60
70
80
Time (hrs)
Bucket #
Measure every bucket in 1 train except the 5 centered on each bunch
Compare change from beginning to end of store as a function of bucket
Inte
nsity
(E
9)
f
2/21/2006 R. Thurman-Keup - AD Seminar 32
DC Beam Diffusion
0
5
10
15
05
1015
20-2
0
2
4
6
8
10
12
14
Interbunch Gap #Bucket # mod 21
In
tens
ity (
EO
S -
BO
S;
E9
arou
nd t
he r
ing)
0 2 4 6 8 10 12 14 16 18 20 15
6
7
8
9
10
11
12
Bucket mod 21
In
tens
ity (
End
of
stor
e -
Beg
in o
f st
ore;
E
9 pa
rtic
les
arou
nd r
ing)
Leading Bunch Trailing Bunch
Averaged over11 interbunchgaps
Expected Diffusion Direction
f
2/21/2006 R. Thurman-Keup - AD Seminar 33
Issues
No automatic pedestal subtraction Pedestal varies by ± 1 E9 over time Pedestal very sensitive to beam synchronous EM
noise Will be helped by new PMT (see Future)
No automatic gain measurement Checked periodically (TEL trips) Correlation between TEL and bunch calibrations
Cross-talk from Synclite Need synchronization for any automatic solutions
OAC?
f
2/21/2006 R. Thurman-Keup - AD Seminar 34
Pedestal behavior
f
2/21/2006 R. Thurman-Keup - AD Seminar 35
Pedestal behavior
f
2/21/2006 R. Thurman-Keup - AD Seminar 36
Summary
Abort Gap Monitoring PMT has been in use for nearly 2 years
Integral part of TeV operations Checked by the sequencer before normal
beam aborts Watched by CDF and MCR
i.e. I get paged when it hiccups
f
2/21/2006 R. Thurman-Keup - AD Seminar 37
Future
2006 Shutdown Replace LBNL PMT with newly purchased PMT
Hamamatsu R5916U-50 ++ 3 stage MCP vs. 2 stage (better S/N; N being EM pickup) 10% duty cycle vs. 1% duty cycle for gating
Install second one in antiproton box Typical proton bunch intensity is 250-300 E9 Pbar bunch intensities now approaching 100 E9
Efforts to see pbar abort gap beam in Synclite have not seen anything large (maybe possibly hints of < 1E9 at beginning of store, but complicated by pickoff mirror movement)
Possibly attempt to automate gain and pedestal measurements OAC?
f
2/21/2006 R. Thurman-Keup - AD Seminar 38
Stuff
R. Coïsson, Opt. Commun. 22, (1977) 135
On Synchrotron Radiation in Non-Uniform Magnetic Fields
Phys. Rev. A20 (1979) 2Angular-Spectral Distribution and Polarization of Synchrotron Radiation from a “Short” Magnet
R. Bossart, et. al., Nucl. Instr. and Meth. 184 (1981) 349
Observation of Visible Synchrotron Radiation Emitted by a High-Energy Proton Beam at the Edge of a Magnetic Field