Development of non-intersecting transverse and longitudinal Profile Monitors

L. Groening, Sept. 15th, 2003GSI-Palaver, Dec. 10th, 2003, A dedicated proton accelerator for p-physics at the future GSI facilities

Peter Forck (GSI), HIPPI meeting, Oxford Non-intersecting transverse and longitudinal Profile Monitors

Development of non-intersecting

transverse and longitudinal Profile Monitors

P. Forck, A. Bank, W. Barth, C. Dorn, A. Peters, H. Reeg

Gesellschaft für Schwerionenforschung, Darmstadt

HIPPI Meeting 2005, Oxford Non intersecting methods for:

• Preventing destruction of intersecting material• Parallel observation at different locations• Monitoring of possible time-varying processes

• Goal: Same precision as intersecting methods

Outline:• Transverse profile monitor by Beam Induced Fluorescence BIF

• Bunch Structure Monitor BSM based on residual gas electron spectroscopy

• Transmission control by transformers



High Current Injector (RFQ&IH) Alvarez DTLSingle Gap Resonators

The UNILAC Facility at GSI

Achieved current for U-beam (tpulse = 200 μs)

U4+: 16 emA U28+: 5 emA U73+: 2 emA

1.4 MeV/u11.4 MeV/u



Basics of Beam Induced Fluorescence

Physics of fluorescence for N2 residual gas: p + N2 p + (N2+)*+ e- p + N2

+ + + e- • Excitation of residual gas molecules by beam’s energy loss• Decay of N2

+ levels generate light, blue light 390 nm < < 470 nm, lifetime = 60 ns.

Realizations at Los Alamos, CERN, Orsay/Saclay, Uni-Frankfurt, GSI, COSY ….Fluorescence of 200 keV p in N2 (1961)

Spectrum confirmed at CERN-PS/SPS from 1 to 450 GeV.

LANL (D. Gilpatrick et al.)

p at MeV in 5*10-5 mbar N2

-20

20

60

100

140

180

320 360 400 440 480

Wavelength (nm)



Image Intensifier used at GSI-LINAC

Technical realization of image intensifier at GSI:

• Photo cathode S20 UV: γ-e- conversion, 15 to 25 % efficiency, 200 nm < λ < 650 nm

• Two step MCP (25 mm diameter): 106 fold amplification

• P 46 phosphor: e- -γ conversion, 300 ns decay, 500 nm < λ < 600 nm

• Minifying taper coupling to CCD chip (1/2’’): 7% transmission• Digital camera (Basler A311f): Firewire interface



Test Setup at GSI-LINAC

Compact chamber with 150 mm insertion:

Installation behind Alvarez at 11 MeV/u



Typical Result at GSI-LINAC

Features:

• Single photon counting

• High resolution (here 0.3 mm/pixel),

can easily be matched to application

• Low background (sometime larger

contribution by neutrons and )

Beam parameters at GSI-LINAC:4.7 MeV/u Ar10+ beam

I=2.5 mA equals to 1011 particles

One single macro pulse of 200 s

Vacuum pressure: p=10-5 mbar (N2)

bump restricted ~1 m,

no influence to beam detected



Application of Beam Induced Fluorescence

Special application

Variation during the macro pulse detectable:

Switching of image intensifier

Exposure window during macro-pulse

Signal treatment

Statistics offers ‘offline’ optimization

statistics integration time resolution

Beam parameter:

Ar10+ at 11 MeV/u with 8 mA



In Preparation: Digital Interface for Firewire

Digital camera offers: no loss of data-quality, versatile trigger, variable exposure timeCCD-camera: Basler A311f featuring 649x494 pixels, 12 bit, 50 frames/s, IEEE 1394bIris/MCP-gain variation: Remote controlled iris by local, ethernet based DACReadout: HUB optical fiber real-time controller running RT-LabVIEW (NI)Status: DAQ in preliminary design phase

LabVIEW Software:

DAQ System:



Novel Device for non-intersecting Bunch Shape Measurement

Bunch-Shape seldom measured !

Scheme for novel device:

• Secondary electrons for residual gas

• Acceleration by electric field

• Target localization by apertures

and electro-static analyzer

(Δy = 0.2 to 2 mm, Δz=0.2 to 1 mm)

• rf-resonator as ‘time-to-space’ converter

same as intersecting method (INR-Moscow)

• Readout Ø70 mm MCP + Phosphor + CCD

• Measurement done within one macro-pulse

(not yet achieved due to back-ground)



Realization for Bunch Shape Monitor at UNILAC

E-field and the energy-analyzer:Installation for beam based tests:



First Results from Bunch Shape Measurement at UNILAC

Features:

Single electron detection

Recorded within few

macro-pulses

Resolution better

50 ps = 20@108MHz

Pressure bump required

• Back-ground should be

suppressed

Beam parameters:

Ni14+ at 11.4 MeV/u

I=1.5 mA, 200 μs macro pulse

Average: 8 macro pulses

Pressure p=2*10-6 mbar

Deflector power P=15 W

Time information carried by the residual gas e- is transferred to spatial differences:



First Application from Bunch Shape Measurement

Variation of buncher:

• Bunch shape was determined,

influeneced by buncher

• Pick-up: No measurable influence

• Emittance determination possible

Beam parameters:

Ni14+ at 11.4 MeV/u

I=2 emA, 200 μs macro pulse

Average: 4 macro pulses

Pressure p=10-5 mbar



Beam Space Charge Contribution

Simulation method:

e- trajectory calc. inside beam pipe

& linear optics for energy analyzer

Simulation parameter:

Ekin= 11.4 MeV/u

Parabolic bunch shape

∓0.5 ns longitudinal root points

∓5 mm transversal root points

Variation of current (as for Ni14+ )

Simulation result: stronger influence as for

standard method,

but still acceptable

The residual gas e- are influenced by beam’s E-field in addition to the monitor E-field Simulation of influence for different currents:



Dynamic Transmission control at UNILAC

ACCT clamping

Integration window

ACCT signal

40 μs/div

Variation of maximal loss

via software input:

8 different input thresholds

8 different macro-pulse duration

by electric chopper in front of RFQ

Save protection of equipment.FPGA-electronics:ACCT V/f-converter Up/down-counter: 1st ACCT ↑, 2nd ↓

Digital comparator chopper



Conclusion and Outlook

Beam Induced Fluorescence BIF: First prototype in operation for ‘single photon counting’, usable during UNILAC operation

Data acquisition in design phase (responsible engineer just hired)

More investigation with high current required

possible problems: broadening by space charge field, two-step excitation….

Non-intersecting Bunch Shape Monitor:

Prove-of-principle performed, resolution lower than 50 ps = 20 @ 108 MHz

Improvements for back-ground suppression in preparation beam test necessary

Calculations and measurements of signal deformation due to beam space charge required

Device in experimental condition engineering design for operation required

Dynamic Transmission control: System design finished

Hardware in operation

Improvements of operation control required



Comparison for different Gases at p Source (Saclay)

Choice of fluorescence gas:

• High fluorescence yield at optical wave-length

• Short lifetime of excited level

• Good vacuum pumping

Results:

Profile is independent of gas

Care:

• Long lifetime (N2+: 60 ns)

broadening by beam space charge

• Light emitted by primary ions

e.g. p + N2 H* + N2+

(only important for Ekin<1 MeV)

• At large N2 density (p>10-3 mbar): Two-step processes e.g. N2 + e- N2* + e- possible

Profiles from different gasses

Example: Ion source 100 keV, 100 mA protonsP. Ausset et al. (Orsay/Saclay)

N2, Ne

Ar, Kr

Xe



Non-intercepting Profile Measurement based on Energy Loss

Target e--density ~ 1/Ekin (for Ekin> 1GeV nearly constant)Strong dependence on projectile charge

Profile determination from ionization and excitation of residual gas.

M. Plum et al.:

p in N2 at CERN-PS

Standard monitors: SEM-Grid, Wire-Scanner, Scintillation Screen, OTR-Screen… Disadvantage: intercepting, problems for time-varying processes

Non-intercepting profile measurement:

• Large beam power can destroy the material

• Synchrotron: Monitoring during full cycle • LINAC: Monitoring at different locations,

variation during the macro-pulse

Physics: electronic stopping power

Bethe-Bloch formula:

- dE/dx = const · Zt ρt /At · Zp

2 · 1/β2 · [ ln(const ·γ2β2/I) –

β2]

cr

oss

sect

ion

α dE

/dx

pc [GeV]



Technical Realization Possibilities for BIF

Double MCP: + single photon, 106-fold amp.

- resolution limited (MCP-channels)

Example: GSI-LINAC (300 μm/pixel)

Single MCP: - lower 103-fold amp.

+ higher resolution

Example: CERN-SPS (160 μm/pix), R. Jung

et al.

Photo-cathode: Only for required wavelength interval to avoid dark currents, e.g.

S20UV: 200<λ<650 nm dark rate 500 e-/cm2/s, S25red: 300<λ<900 nm 30000 e-/cm2/s

Phosphor: Fast decay ↔ lower sensitivity e.g. P47: τ = 0.1 μs, P43: τ = 1000 μs IP43 ~ 4 · IP47

Problem: Radiation hardness of CCD camera



BIF at Synchrotrons

Example: CERN SPS and PSB,PS (R. Jung, M. Plum et al.)

Photon yield scales like Bethe-Bloch energy loss ΔE

d for p with 100 MeV < E kin < 450 GeV

Comparison to wire scanner at SPS

Gas N 2 Xe

ΔE /photon 3.6 keV 46 keV

lifetime 58 ± 0.3 ns 59 ± 1 nsMethod: fluorescence decay by ~5 ns long bunches



Realization of Bunch Shape Monitor at UNILAC



Standard Bunch Shape Determination

Standard intersecting method developed by INR-Moscow (A. Feschenko, P. Ostrumov et al.):

Insertion of a 0. 1 mm wire at 10 kV

Emission of e- within < 0.1 ps

Acceleration toward 1mm slit

Rf-deflector as time-to-space converter

Detection with Slit+Cup or MCP

Resolution better 1o or 10 ps



Dynamic Transmission control at UNILAC

40 μs/div

Verification for transmission control:

Artificial beam loss by

quadrupole variation

chopper window decrease



High Current Transmission control at UNILAC

FPGA

Development of non-intersecting transverse and longitudinal Profile Monitors

Documents