monitoring beam intensity in the tevatron abort gap using synchrotron radiation

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


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


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


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


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


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


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


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


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


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


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


Synclite Device

Future home of PMT

Proton Synclite Box

Antiproton Synclite Box

Picture taken beforePMT installation

f


Synclite Device

Synchrotron light spot

~ 9 inBeam splitter

Picture taken with small CCD camera

The specks are radiation-damaged pixels

f


Abort Gap Intensity Monitor

Synchrotron light box

Abort Gap PMT

f




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





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



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





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



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





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





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


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


Abort Gap Intensity (Synclite)

Peak is ~7 x 109

CID Camera Image

Raw Background Subtracted

f



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


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


Calibration

Periodically check this using TEL trips. Seems to vary by 40-50%

f



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


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


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


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


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


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


Pedestal behavior

f


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


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


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

monitoring beam intensity in the tevatron abort gap using synchrotron radiation

Documents