steve kahn mucool/mice meeting berkeley october 21, 2002 · oct 21, 2002 s. kahn -- hard edge geant...

13
Oct 21, 2002 S. Kahn -- Hard Edge Geant 1 Hard Edge Simulation of TETRA Ring in GEANT (again) Steve Kahn Mucool/Mice Meeting Berkeley October 21, 2002

Upload: trankhue

Post on 29-Jun-2018

215 views

Category:

Documents


0 download

TRANSCRIPT

Oct 21, 2002 S. Kahn -- Hard Edge Geant 1

Hard Edge Simulation of TETRA Ring in GEANT (again)

Steve KahnMucool/Mice Meeting

Berkeley October 21, 2002

Oct 21, 2002 S. Kahn -- Hard Edge Geant 2

Why No Progress With Realistic Fields

Need to Understand the GEANT with Hardedge Fields before Attempting Realistic Fields.

My attempts (and others) have reported large losses in using GEANT with this simulation.

Valeri Balbekov has shown that one can achieve a reasonable amount of cooling with reasonably good transmission through this TETRA ring.

Rick Fernow has similarly achieved similarly good results with a high FoM for the TETRA ring in ICOOL.

•Field maps of the solenoids have been supplied to Makino and Berz for use in their COSY model.

Oct 21, 2002 S. Kahn -- Hard Edge Geant 3

Concern whether RF is Properly Phased

Energy Gain

Es at cavity center

Correlation between Es and Energy Gain implies that the RF cavities are not correctly phased.

Oct 21, 2002 S. Kahn -- Hard Edge Geant 4

Reference Track to Phase RF Cavities

Figure shows a reference muon circulating in the ring for 20 turns.

Reference track is launched on axis with ideal Pµ and no PT.RF cavities are active.Absorbers are present.

dE/dx is turned on but no random processes are.

Track shows very little deviation from axis (shown on next transparency).

Oct 21, 2002 S. Kahn -- Hard Edge Geant 5

Phasing the RF — Looking at Path Time

The top figure shows the path time that the reference particle traverses one quadrant.

Only the gradient is varied.RED line indicates what is expected from frequency.

The lower figure shows the RMS variation of the path time for

Cavity 1 in the four quadrants of turn 1.The quadrant path time for all cavities in turn 1.

These figures indicate that the correct gradient is ~15.1 MV/m.

Quadrant Time vs Gradient

34

34.01

34.02

34.03

34.04

34.05

34.06

34.07

34.08

34.09

34.1

10 11 12 13 14 15 16 17

Gradient, MV/m

Tim

e, n

s

Cavity 1All CavitiesExpected

RMS Time Differences

0

0.05

0.1

0.15

0.2

0.25

10 11 12 13 14 15 16 17

Gradient, MV/mTi

me

devi

atio

ns, n

s

Cavity 1All Cavities

Oct 21, 2002 S. Kahn -- Hard Edge Geant 6

Phasing the RF — Looking at Energy Gain

The top figure shows the mean difference in energy gain between corresponding cavities in adjacent quadrants for the 1st turn.

Ideally this difference should be zero. All cavities in all should have the same energy gain difference.

The lower figure shows the RMS variation of the energy gain differences between corresponding cavities in adjacent quadrants.Both figures also imply that the ideal gradient should be ~15.1 MV/m.

Energy Gain vs Gradient

-0.001

0

0.001

0.002

0.003

0.004

0.005

10 11 12 13 14 15 16 17

Gradient, MV/m

Mea

n En

ergy

Gai

n, G

eV

Cavity 1All Cavities

Energy Gain vs. Gradient

0

0.001

0.002

0.003

0.004

0.005

0.006

10 11 12 13 14 15 16 17Gradie nt, MV/m

RM

S En

ergy

Dev

iatio

n, G

eV

Cavity 1All Cavities

Oct 21, 2002 S. Kahn -- Hard Edge Geant 7

Using the Optimized Gradient

Es alongReference Path

Energy along Reference Path

•The reference particle sees a more stable Esand energy variation along its path at least during the early turns.

•There is still room for improvement.

Oct 21, 2002 S. Kahn -- Hard Edge Geant 8

Deviations of Reference Particle from Ideal Orbit

Oct 21, 2002 S. Kahn -- Hard Edge Geant 9

Typical Run

Figure shows 20 particles passed through ring with RF on and wedges in place:

σx=σy= 4 cm, σct=8 cmσPT=32 MeV/c, σE=18 MeVCorrelation between E, PT, BNo decaysNo random processes

dE/dx is mean value

Figure illustrates losses that typically occur in corner regions.

Oct 21, 2002 S. Kahn -- Hard Edge Geant 10

Emittance Calculations

The figure shows the transmission, transverse emittance. (Ignore the 6D emittance, there is a problem with it).

The transmission drops to ~10% in 12 turns.Emittance drops significantly but that is due to losses more than cooling.

Transmission

4D Emittance

Oct 21, 2002 S. Kahn -- Hard Edge Geant 11

Dispersion Plots

The figure shows the dispersion along the straight solenoid. Dx

Dy

Oct 21, 2002 S. Kahn -- Hard Edge Geant 12

Transverse Phase Space Plots

Px vs x at same position for 1st

nine turns

Oct 21, 2002 S. Kahn -- Hard Edge Geant 13

Longitudinal Phase Plot

E vs. ct at same position for 1st nine turns