Electron Lens Simulation Updates Valentina Previtali Vince Moens, - - PowerPoint PPT Presentation

▶
electron lens simulation updates
SMART_READER_LITE
LIVE PREVIEW

Electron Lens Simulation Updates Valentina Previtali Vince Moens, - - PowerPoint PPT Presentation

Aug 20, 2022 18 likes •382 views

Electron Lens Simulation Updates Valentina Previtali Vince Moens, Giulio Stancari, Alexander Valishev (FNAL, Chicago IL) Stefano Redaelli (CERN, Geneva CH) the e-lens: what is it and why is it interesting? The e-elens is a device generating

slide-1
SLIDE 1

Electron Lens Simulation Updates

Valentina Previtali Vince Moens, Giulio Stancari, Alexander Valishev (FNAL, Chicago IL) Stefano Redaelli (CERN, Geneva CH)

slide-2
SLIDE 2

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

the e-lens: what is it and why is it interesting?

The e-elens is a device generating an electron beam which travels parallel (overlapped) with the p-beam It has be proven (Tevatron experience) to be an effective scraping device (data from Giulio’ s presentations). Main advantages w.r.t. conventional scraping:

  • no “hard” material close to the beam
  • size can be easily/

quickly changed acting on magnetic fields

  • great flexibility:
  • scraping efficiency can be tuned by acting on e-beam current /
  • peration mode
  • device can control single bunches individually

2

slide-3
SLIDE 3

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

The e-elens is a device generating an electron beam which travels parallel (overlapped) with the p-beam A hollow electron beam has been proven (Tevatron experience) to be an effective scraping device (see Giulio’ s presentations). Main advantages w.r.t. conventional scraping:

  • no “hard” material close to the beam
  • size can be easily/

quickly changed acting on magnetic fields

  • great flexibility:
  • scraping efficiency can be tuned by acting on e-beam current /
  • peration mode
  • device can control groups of bunches individually

3

the e-lens: what is it and why is it interesting?

slide-4
SLIDE 4

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

The e-elens is a device generating an electron beam which travels parallel (overlapped) with the p-beam A hollow electron beam has been proven (Tevatron experience) to be an effective scraping device (see Giulio’ s presentations). Main advantages w.r.t. conventional scraping:

  • no “hard” material close to the beam (MP, impedance)
  • size can be easily/

quickly changed acting on magnetic fields

  • great flexibility:
  • scraping efficiency can be tuned by acting on e-beam current /
  • peration mode
  • device can control groups of bunches individually

4

the e-lens: what is it and why is it interesting?

slide-5
SLIDE 5

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

The “e-lens for LHC” project so far

E-lens review in November 2012 @ CERN: outcomes

first integration studies: identified possible installation locations in LHC strong experimental evidences from Tevatron measurements first simulation results assessing beneficial effects of the electron lens

Positive feedbacks, however installation in LS1 is not feasible FNAL and CERN have agreed on the roadmap:

the device will be responsibility of BI (CERN) integration studies will be carried out by ME group (CERN) the Physics study is to be lead by FNAL -> delivery of a conceptual design by next collaboration meeting (Oct-Nov 2013)

5

slide-6
SLIDE 6

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

plan for conceptual design study

identify expected scraping efficiency implication on beam dynamics integration (not covered here)

Scraping efficiency: First answers by SixTrack simulations Realistic e.m. field computation & non linear map for proton beam core

6

slide-7
SLIDE 7

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

WARP calculations

possible e-lens issues

Vince Moens, CERN technical student, now @FNAL Realistic e.m. field computation & non linear map for proton beam

7

slide-8
SLIDE 8

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

WARP: particle in cell plasma 3D code which computes the e-beam evolution in space and the associated e.m. fields. Detailed geometry & electron beam evolutions included experimental data on measured profiles on the new high- current 1 in gun are used as benchmark Evaluation of e.m. magnetic fields along the p trajectory and creation of a non linear map to be inserted in a tracking code. Detailed evaluation of the e-lens effect on the beam core. Identify possible issues related to the e-lens use in the LHC code setup is done, results coming soon!

8

slide-9
SLIDE 9

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

new 1-in cathode measurements used for benchmark

9

slide-10
SLIDE 10

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

1 2 3 −0.2 0.0 0.2 −

0.00 0.05 0.10 0.15 −0.02 0.00 0.02 −

WARP setup new 1-in cathode measurements used for benchmark

10

slide-11
SLIDE 11

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

Latest SixTrack results

Scraping efficiency: answers by SixTrack simulations

e-lens advantages

Valentina Previtali, TOOHIG fellow@FNAL

11

slide-12
SLIDE 12

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

Simulation inputs

collimation version of SixTrack + elens particle number: 6400 turns: up to 200K

  • ptics: nominal squeezed LHC 7 TeV

e-lens inputs: 1,2 A - inner radius 4σx collimation inputs: only primary collimators at 6σx initial distribution: horizontal distribution between 4 and 6σx, no off momentum

12

slide-13
SLIDE 13

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

Latest SixTrack results

3 operation modes have been identified so far: DC mode: e-lens is ON (scraping relies on the strong non linearity of lens field) AC mode: e-beam is modulated with particle tune (improved from last presentation, see details later) diffusive mode: e-beam current is randomly switched off or on on turn-by-turn basis

13

slide-14
SLIDE 14

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

DC mode

0.311 0.3115 0.312 0.3125 0.313 4.5 5 5.5 6 frac Qx Ax [σx] no elens DC

e

  • l

e n s c

  • n

t r i b u t i

  • n

:

  • v

e r a l l ~ l i n e a r b e h a v i

  • r

14

slide-15
SLIDE 15

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

DC mode

0.311 0.3115 0.312 0.3125 0.313 4.5 5 5.5 6 frac Qx Ax [σx] no elens DC “suspect region”

e

  • l

e n s c

  • n

t r i b u t i

  • n

:

  • v

e r a l l ~ l i n e a r b e h a v i

  • r

15

slide-16
SLIDE 16

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

0.019 0.02 0.021 0.022 0.023 0.024 0.025

  • 1.65 -1.6 -1.55 -1.5 -1.45 -1.4 -1.35 -1.3 -1.25

xp [mrad] x [mm] no elens DC radial+jitter DC radial

resonance at about 5.7 σx

notice the contribution from e-beam current jitter

16 islands along the phase-space ellipse: high order resonance

Still no appreciable scraping effect

16

slide-17
SLIDE 17

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

AC mode

trying to attack the problem analytically: Harmonic oscillator with a driving force which depends on time and position. The problem is solvable for non-hollow elens, uniform e-beam (landau, parametric resonance) for this case attempt to solve analytically the “hollow” e-lens case not straightforward

Resonance frequency is double of the system “natural” oscillation frequency

n=2

17

slide-18
SLIDE 18

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

finding the resonance frequency by “brute force”

0.2 0.4 0.6 0.8 1 2 4 6 8 10 Nfinal/N0

  • mult. factor n

ODD mult. factors EVEN mult factor

2 “families”: ODD and EVEN n factors. Even factors are generally more efficient. n=2 is, so far, the most efficient

18

I simulated the scraping effect of an el-lens driven by different multiples of the natural frequency nω0, with the multiplying factor n in the range {1, 2 ... 10}

slide-19
SLIDE 19

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

finding the resonance frequency by “brute force”

0.2 0.4 0.6 0.8 1 2 4 6 8 10 Nfinal/N0

  • mult. factor n

ODD mult. factors EVEN mult factor

2 “families”: ODD and EVEN n factors. Even factors are generally more efficient. n=2 is, so far, the most efficient

19

about 75% cleaning in 20s

I simulated the scraping effect of an el-lens driven by different multiples of the natural frequency nω0, with the multiplying factor n in the range {1, 2 ... 10}

slide-20
SLIDE 20

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

finding the resonance frequency by “brute force”

0.2 0.4 0.6 0.8 1 2 4 6 8 10 Nfinal/N0

  • mult. factor n

ODD mult. factors EVEN mult factor

2 “families”: ODD and EVEN n factors. Even factors are generally more efficient. n=2 is, so far, the most efficient about 75% cleaning in 20s

20

I simulated the scraping effect of an el-lens driven by different multiples of the natural frequency nω0, with the multiplying factor n in the range {1, 2 ... 10}

what is ω0

slide-21
SLIDE 21

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

We have seen from the study of the tune that there is no a single “natural frequency” of the system

from .3110

  • s

c i l l a t i

  • n

f r e q u e n c y w i t h

  • c

t u p

  • l

e s

0.311 0.3115 0.312 0.3125 0.313 4.5 5 5.5 6 frac Qx Ax [σx] no elens DC

to .3125

the resonance condition must cover all the particles = all the tunes repeat cycles over all the tunes (optimization in progress)

21

what is ω0

slide-22
SLIDE 22

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

We have seen from the study of the tune that there is no a single “natural frequency” of the system

from .3110

  • s

c i l l a t i

  • n

f r e q u e n c y w i t h

  • c

t u p

  • l

e s

0.311 0.3115 0.312 0.3125 0.313 4.5 5 5.5 6 frac Qx Ax [σx] no elens DC

to .3125

the resonance condition must cover all the particles = all the tunes repeat cycles over all the tunes (optimization in progress)

22

what is ω0

~5 s

Frequencies of the order of 6.8 KHz, precision of 1 Hz required, total Δf∼25Hz in ∼5 s.

6.853 KHz 6.875 KHz

slide-23
SLIDE 23

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

https:/ /dl.dropbox.com/u/7248203/AC_movie.gif

23

effect on the halo while we cycle the resonant frequency over tune range (1 full cycle about 5 s, total of 4 cycles)

is 75% enough?

“hard” edge at ~4.5 sigma

slide-24
SLIDE 24

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

https:/ /dl.dropbox.com/u/7248203/AC_movie.gif

24

effect on the halo while we cycle the resonant frequency over tune range (1 full cycle about 5 s, total of 4 cycles)

is 75% enough?

“hard” edge at ~4.5 sigma

slide-25
SLIDE 25

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

Diffusive mode

2 different random mode tested: ON-OFF mode: then electron beam was either reduced to zero (OFF) or at its full power (ON) randomly on a turn-by-turn basis. Random mode: the electron beam current was modulated on turn by turn basis by a random multiplier in the range [0,1]; Smoother-slow scraping PROS completely uncorrelated with the particle state (both amplitude and tune) works simultaneously for Vertical and Horizontal plane the scraping efficiency can be easily increased by in- creasing the beam current. CONS: not “hard-edge” as AC mode

25

slide-26
SLIDE 26

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

0.2 0.4 0.6 0.8 1 50000 100000 150000 200000 N/N0 [-] turns [-] ON-OFF RANDOM

Random vs ON-OFF

26

slide-27
SLIDE 27

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

scaling with the current

0.2 0.4 0.6 0.8 1 50000 100000 150000 200000 N/N0 [-] turns [-] 1.2 A 2.4 A

almost reaching the performances of AC mode

2X current

27

slide-28
SLIDE 28

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

28

not as “hard edge” as AC mode

slide-29
SLIDE 29

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

29

not as “hard edge” as AC mode

slide-30
SLIDE 30

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

Conclusions

preliminary results with 6track are encouraging (N/N0 < 25% in 20 s) identified diff op modes which allow us to go from smooth to fast cleaning Diffusive mode seems to be preferable for: easy operation mode flexibility independent of particle tune characterized performances of new e-gun designed for the LHC (4A, 10 KV) simulation of details of e-beam with WARP, studies of impact of e-lens on beam core working of conceptual design for LHC in collaboration with LHC collimation working group

30

slide-31
SLIDE 31

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

Thanks :)

31

slide-32
SLIDE 32

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

Reserve Slides

32

slide-33
SLIDE 33

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

The “e-lens for LHC” project so far

E-lens review in @ CERN last year...

identified possible installation locations strong experimental evidences from Tevatron measurements first simulation results assessing beneficial effects of the electron lens

Positive feedbacks, however installation in LS1 is not feasible More work is required

the device will be responsibility of BI (CERN) integration studies will be carried out by ME group (CERN) the Physics study is to be lead by FNAL -> delivery of a conceptual design by next collaboration meeting (Oct-Nov 2013)

  • 0.3
  • 0.2
  • 0.1

0.1 0.2 0.3 9900 9950 10000 10050 10100 Dx [m] s [m] Dx Dy 50 100 150 200 250 300 350 400 β [m] IP4 β x β y

identified an optimal location downstream IP4. Space is available (for the moment). First integration studies are promising (Adriana, Belen)

33

slide-34
SLIDE 34

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

The “e-lens for LHC” project so far

E-lens review in @ CERN last year...

identified possible installation locations strong experimental evidences from Tevatron measurements first simulation results assessing beneficial effects of the electron lens

Positive feedbacks, however installation in LS1 is not feasible More work is required

the device will be responsibility of BI (CERN) integration studies will be carried out by ME group (CERN) the Physics study is to be lead by FNAL -> delivery of a conceptual design by next collaboration meeting (Oct-Nov 2013)

34

slide-35
SLIDE 35

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

initial Ax final Ax

not as “hard edge” as AC mode

35

slide-36
SLIDE 36

Hi-Lumi / LARP CM20 - NAPA, April 2013

  • V. Previtali

initial Ax final Ax

36