Managed by UT-Battelle for the Department of Energy SNS Injection and Extraction Systems Issues and Solutions by M. Plum for the SNS team and our BNL collaborators

Managed by UT-Battellefor the Department of Energy

SNS Injection and Extraction Systems

Issues and Solutions

by M. Plum

for the SNS team and our BNL collaborators

HB2008Nashville, TN

Aug. 24-29, 2008

2 Managed by UT-Battellefor the Department of Energy M. Plum, HB2008 Workshop

SNS injection

Ring injection area

4 magnet chicane

Gradient septum

Quadrupole magnet and dual-plane dipole corrector

Injection septum

Secondary foil

Primary foil

150 kW Injection dump

Injection dump beam

line

HEBT beam line

Ring

Injection painting kicker magnets

Injection painting kicker magnets


Functions of chicane magnets Closed orbit bump of about 100 mm

Merge H- and circulating beams with zero relative angle

Place foil in 2.5 kG field and keep chicane #3 peak field <2.4 kG for H0 excited states

Field tilt [arctan(By/Bz)] >65 mrad to keep electrons off foil

Funnel stripped electrons down to electron catcher

Direct H− and H0 waste beams to IDmp beam line

H- beam from Linac

ThinStripping Foil

To InjectionDump

ThickSecondary Foil

p

H0

H-Dipole magnets

H- beam from Linac

ThinStripping Foil

To InjectionDump

ThickSecondary Foil

p

H0

H-Dipole magnets


SNS injection issues

Chicane magnets do not function as designed– Bend angles have been adjusted to give good injection into

ring, but this causes problems in the injection dump beam line

Original design did not allow individual control over the H0 and H− waste beams– We’ve since added a C-magnet just downstream of the septum

magnet

High beam loss in injection dump beam line– Beam halo

– Scattering in the secondary stripper foil

Beam profile and position info at the vacuum window / dump difficult to determine – We plan to add a view screen at the vacuum window


Inj. dump beam line modifications to date

Oversize & thicker primary stripper foil

Thinner, widersecondary stripper foil

Increase septum magnet gap by 2 cm

New C-magnet

New WS, view screen,BPM, NCD (ridicules)

Shift 8 cm beam left

Electron catcher IR video

Radiation monitor on vacuum window water cooling return pipe

beam line drawing from J. Error


Beam loss due to scattering

For a given aperture, the probability of Rutherford (large angle Coulomb) scattering outside the aperture separately depends on the target and the apertures

By replacing the secondary foil with a thicker material we can estimate the fraction of the loss due to scattering

2

yl1 1e e xl02 2 2

xl yl xl xl yl yl

2Zm r t 1 1 1P N tan tan

M A

(R. Macek, 2004)

target angles

0

10000

20000

30000

40000

50000

60000

70000

80000

90000

100000

0 5 10 15 20

Scattering Angle [milliradians]

Nu

mb

er

ex

ce

ed

ing

an

gle

0

5

10

15

20

25

30

35

40

45

Ra

tio

Foil

View screen

Ratio

Foil total

Foil

VS

ratio

(Plots from J. Holmes)

Scattering angle (mrad) Scattering angle (mrad)

Num

ber

exce

edin

g an

gle

Num

ber

exce

edin

g an

gle

Rat

io


0

0.005

0.01

0.015

0.02

0.025

Ring_

Diag:B

LM_A

Mov

02

HEBT_Diag

:BLM

_Mov

08

IDm

p_Dia

g:BLM

00a

IDm

p_Dia

g:BLM

00b

IDm

p_Dia

g:BLM

01a

IDm

p_Dia

g:BLM

01b

IDm

p_Dia

g:BLM

01c

IDm

p_Dia

g:BLM

01d

IDm

p_Dia

g:BLM

01e

IDm

p_Dia

g:BLM

_Mov

01

BLM

rad

/pu

lse

0

10

20

30

40

50

60

rati

o in

/ o

ut

VS in

VS out x 10

in/out

16/Apr/08 data, 10 mp, Sim. H0 steered to minimize beam loss

Single beam species tuned to minimize beam loss

One well-tuned beam Simulated H0 beam, production tune

Foil: (total loss) = a x (scattering) + b x (base loss)

View screen: (total loss) = 50 x a x (scattering) + b x (base loss)

Conclude that for simulated H0 beam, 30 to 90% of beam loss is due to foil scattering. We need a thinner foil!

We expect similar numbers for production case with both H− and H0 waste beams

0.000

0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

0.045

0.050

BLM

0.00E+00

5.00E+00

1.00E+01

1.50E+01

2.00E+01

2.50E+01

3.00E+01

3.50E+01

4.00E+01

4.50E+01

5.00E+01

rati

o in

/ o

ut

VS out x 10

VS in

in / out #1

in / out #2

1/Jun/08 data, 2 mp

90% of loss is due to scattering

50%30%

0.000

0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

0.045

0.050

BLM

0.00E+00

5.00E+00

1.00E+01

1.50E+01

2.00E+01

2.50E+01

3.00E+01

3.50E+01

4.00E+01

4.50E+01

5.00E+01

rati

o in

/ o

ut

VS out x 10

VS in

in / out #1

in / out #2

1/Jun/08 data, 2 mp

90% of loss is due to scattering

50%30%

ratio

VSfoil x10

ratio VS

foil x10


0

0.005

0.01

0.015

0.02

0.025

Ring_

Diag:B

LM_A

Mov

02

HEBT_Diag

:BLM

_Mov

08

IDm

p_Dia

g:BLM

00a

IDm

p_Dia

g:BLM

00b

IDm

p_Dia

g:BLM

01a

IDm

p_Dia

g:BLM

01b

IDm

p_Dia

g:BLM

01c

IDm

p_Dia

g:BLM

01d

IDm

p_Dia

g:BLM

01e

IDm

p_Dia

g:BLM

_Mov

01

BLM

rad

/pu

lse

0

10

20

30

40

50

60

rati

o in

/ o

ut

VS in

VS out x 10

in/out

16/Apr/08 data, 10 mp, Sim. H0 steered to minimize beam loss

0

0.005

0.01

0.015

0.02

0.025

Ring_

Diag:B

LM_A

Mov

02

HEBT_Diag

:BLM

_Mov

08

IDm

p_Dia

g:BLM

00a

IDm

p_Dia

g:BLM

00b

IDm

p_Dia

g:BLM

01a

IDm

p_Dia

g:BLM

01b

IDm

p_Dia

g:BLM

01c

IDm

p_Dia

g:BLM

01d

IDm

p_Dia

g:BLM

01e

IDm

p_Dia

g:BLM

_Mov

01

BLM

rad

/pu

lse

0

50

100

150

200

250

300

350

400

rati

o in

/ o

ut

VS in

VS out x 100

in/out

14/Aug/08 data, 10 mp, Sim. H0 steered to minimize beam loss

Foil scattering losses with thinner sec. foil

One well-tuned beam – old foil One well-tuned beam – new foil

Replaced secondary stripper foil August 2008

Old foil 18 mg/cm2 carbon-carbon (Allcomp)

New foil 3.2 mg/cm2 polycrystalline graphite (ACF Metals)

Ratio of losses view screen / foil increased from 50 to 300

Conclude that beam loss due to scattering is now ~6x less

ratio

VSfoil x10

ratio

VS

foil x100


HB2008 injection Q&A

Does the system perform as expected? Did the simulations/calculations performed during the design stage accurately predict the actual performance? – No. Design bend angles of chicane set points were not correct.

Beam loss in injection dump beam line was much higher than expected. Vertical deflection in chicane #4 was not expected.

What are the major limitations in performance? Were they known in the design stage? – Beam loss in the injection dump beam line. Not known in the

design stage.

If someone were to begin now designing the same type of system for a similar machine, what is the one piece of advice that you would give them? – 3-D field simulations and tracking in complex regions such as

injection area. Map magnets well enough to determine higher order multipoles, for a wide range of currents. Allow independent control over multiple beams.


Ring extraction

Extraction kickers

Lambertson septum magnet

RTBT beam line


Extraction system issues

Tilted beam (cross plane coupling)– Due to large skew quad component in the extraction septum

magnet

Lack of diagnostics to measure beam path in ring and first 27 m of the RTBT– Have not yet found set points that give a good launch into

the RTBT

Lack of beam profile and position info at the vacuum window and target– Diagnostic closest to target is 9.5 m away

– Still have a discrepancy between halo thermocouple monitor and the BPM extrapolation method


Tilted beam caused by skew quad component in extraction septum magnet

Tilted beam on the target view screen

horizontal vertical

RTBT20 wire scanner for 3 different horizontal injection kicker amplitudes

Beam distribution at BPM25 in the extraction line, reconstructed using single minipulse injection and varying extraction time(S. Cousineau & T. Pelaia)

(S. Cousineau)

X (

mm

)

Y (mm)


Harmonics calculation (see J.G. Wang HB2008 poster)

Integrated skew quad component 0.26 – 0.28 T at 1 GeV beam energy

~75% due end effects

~5% due to proximity of quad


End of RTBT

BPM

Wire scanner or harp

Wire scanners

Harp

9.52 m from last BPM & profile monitor to face of target

Thermocouple halo monitor


How we determine position and profile at the target

Thermocouple halo monitor used to center beam on target

Physics application “RTBT Wizard” – Determines beam position based on upstream beam position

monitors – ~4 - 8 mm different than halo monitor

– Determines beam density and rms beam size based on on-line model and fitted profiles

S. Cousineau and T. Pelaia


HB2008 extraction Q&A

Does the system perform as expected? Did the simulations/calculations performed during the design stage accurately predict the actual performance? – Except for cross plane coupling, as near as we can tell, it is

working as expected. We knew there were not as many diagnostics as we’d like.

What are the major limitations in performance? Were they known in the design stage? – Difficult to determine extraction kicker set points due to lack of

beam position information. Difficult to determine beam size, density, and position on target. We knew this in the design stage.

If someone were to begin now designing the same type of system for a similar machine, what is the one piece of advice that you would give them? – Map magnets well enough to determine higher order multipoles,

and take into account field distortion due to nearby magnets. Especially important for large beams.

– Install adequate diagnostics to allow easy determination of critical beam parameters


Summary & future work

SNS injection issues are fairly well understood– Beam loss is still too high

– We are working on a view screen for the injection dump

– We are now considering another increase in the aperture of the injection dump beam line

SNS extraction issues are well understood– We are working to modify the extraction septum

magnet to reduce skew quad component

– We are working on a view screen for the target

Documents

Managed by UT-Battelle for the Department of Energy SNS Injection and Extraction Systems Issues and Solutions by M. Plum for the SNS team and our BNL collaborators