sc-13 – consequences for us?

1

SC-13 – Consequences for us?

• Overview of the SC-13 Workshop• Some Highlights• Main Outcome• Questions and Responses

• Our Week with the Code Experts

5/31/2013 FS Outcome of SC-13

22

SC-13 Workshop Organization The scientific program by IAC:

Y.Alexahin (FNAL), O. Boine‐Frankenheim(TUD/GSI), Hofmann (GSI/HI-Jena), J. Holmes (SNS), S. Machida (RAL), E. Metral (CERN), K. Ohmi (KEK), J. Qiang (LBL),F. Zimmermann (CERN)

The program, the presentations and session and workshop summaries are available at the indico page:

https://indico.cern.ch/conferenceDisplay.py?confId=221441

Approved/Sponsored:

ICFA,EUCARD, ACCNET, HIC4FAIR, LIU


3

SC-13 Workshop Participants


There have been 82 registrants distributed over the following countries, laboratories and companies: Switzerland (CERN) ------------------------------------------------------------------ 38 Germany (GSI, Frankfurt University) ------------------------------------------- 13 UK (STFC) ------------------------------------------------------------------------------- 12USA (Fermilab, SNS, LANL, LBNL, MIT, Tech-X Cor., Muon Inc.) ---- 11 Japan (KEK) ---------------------------------------------------------------------------- 2Austria (MedAustron) --------------------------------------------------------------- 1China (IHEP) ---------------------------------------------------------------------------- 1France (Saclay) ------------------------------------------------------------------------ 1 Mexico (universidad de sonora) ------------------------------------------------- 1Russia (ITEP) --------------------------------------------------------------------------- 1Sweden (ESS) -------------------------------------------------------------------------- 1


SC-13 The Group Photo

5

SC-13 Workshop Sessions


The workshop lasted for 3 and half days and the program has been grouped in the following sessions:

16th of April Stimulus/Project Modeling/Theory/Overview Schmidt, BartosikSpace Charge Studies Ohmi, HuschauerDiscussion Hofmann, Fitterer 17th of AprilMachine model (Theory and Instruments) Prior, Forte Code development Alexahin, WagnerDiscussion Holmes, Aumon

18th of AprilHigh intensity effects Machida, Noll Synergies Zimmermann, HeinDiscussion Metral, Wasef

19th of AprilMitigation and advanced techniques Boine-Frankenheim, Hernalsteens

Joint Discussion/Outlook Hofmann, Holmes, Metral, Bartosik, Rijoff

1. The LIU goal: “Double the number of particles per bunch” ➔ Space Charge has become very relevant for LHC injectors

2. Generational Problem for the Injectors We have started to create a ➔ new team of expertise on space charge covering all 3 rings

3. Strengthen Collaboration with other Laboratories (resurrected or new):a) GSIb) KEKc) Fermilab/SNSd) Others?

4. We are at a crossroad for our LIU studies: a) Huge effort made to do experiments until this springb) ➔ This workshop c) Simulations to understand and improve our machines to reach LIU goal

Why this SC-13 Workshop? General CERN Interest


This can only be a small sample with emphasis and direct relevance in view of LIU

Apologies for having to leave out lots of relevant material!

Looking for catchy statements that reflect what the space charge community is presently thinking…

Some Hightlights



Chris PriorSpaceChargeModel

Status of the space charge studies and measurements in the CERN PSB Vincenzo Forte – Space charge workshop – CERN - 16/04/2012

Measurements: half integer resonance crossing… (static w. point above 2Qy=9)

300 350 400 450 500 550 600 6500

50

100

150

200

C time [ms]

Inte

nsity

[e10

]

@ C450 (switching off

QNO correctors)

Nice case for code benchmarking (simulations are on-going)

Transverse emittances constant

300 350 400 450 500 550 600 6504

4.2

4.4

4.5

4.6

4.8

5

Ctime [ms]

prog

ram

med

tune

s

HorizontalVertical

Vincenzo PSB


III. 4th order Resonance

The 4th order resonance seems to be excited by space charge• Raymond WASEF, Space Charge Workshop, 16/04/13,

CERN8

Wasef PS

12

Compensation of resonancesbefore after

Compensation of 2qx+qy=1

3qy=1 is clearly enhanced

vertical tuneconstant

horizontal tuneconstant

Huschauer PS

13

Compensation of resonancesbefore after

Compensation of 3qy=1

additionally: reduction of

2qx+qy=1

vertical tuneconstant

horizontal tuneconstant

Huschauer PS


Bartosik SPS

Outcome of SC-13

Slow extraction in Debuncher using Orbit• 3-order resonance with variable tune Qx and sext. str. K2• First “strange” results for extraction: “intensity drop”

intensity vs turns

• “Intensity drop” was resolved simply at the beginning: Valery advised to make mesh refinements

PIC: “Total Beam size increases at slow extraction => mesh number should be increased to keep the cell size”

Kapin Fermilab

5/31/2013 FS

Outcome of SC-135/31/2013 FS

Simulations with ORBIT by V. Nagaslaev

Ramps are given in tables; Npart in bunch ~ 2.5e12

Kapin Fermilab


N_macro_surv vs Turn Number for the Debuncher

Timing on CERN computer• Macro version

~20-24h • MADX-SC ~2-4h

MAD-X V3 with MacrosMADX-SC V5

Outcome of SC-13

3Qy=16 – uncorrected

Phase and amplitude data obtained from

Fourier spectra

Beam intensity Vertical beam position Normalised phase space

Fourier spectrum

|)ˆsin(|31|| 2

0

10030

y

y

aa

h

resonance strength and phase:

|)ˆ|2

)(ˆsgn(2

2 010030 yy

5/31/2013 FS

Urschütz Driving terms (PSB)

3Qy=16 – compensated

Beam intensity Vertical beam position Normalised phase space

Fourier spectrum

Calculated compensation currents (for two independent skew sextupoles):

IXSK2L4 = -29.3 A, IXSK9L1 = +15.3 A

|h0030| = 9.0±0.6*10-3 mm-1/2

ψ0030 = -21º±14º

Results from the measurements:

Urschütz Driving terms (PSB)


Machida PIC


Outcome of SC-13

Machida PIC

5/31/2013 FS


Machida PIC


Ohmi J-PARC MR

Holmes pyORBIT



Some History

Micro-scale unphysical effects: H-detuning (left), fake tune evolutions, artificial dipole oscillations,…May jeopardize the validity of beam modeling

Without SC – NO Micro-scale unphysical effects.

WANTED:Consistency between ~Exact Trackers and Approximate SC Solvers

Search for remedies

Recent findings, when using ORBIT:

Frank Schmidt, et al.Micro-scale effects

APC Seminar at Fermilab, November 2012“Benchmarking of PTC_ORBIT & SYNERGIA ”, https://indico.fnal.gov/conferenceDisplay.py?confId=6115


Library of TemplatesPre-Assigned densities

Ring Disc Template

• General • Halo• Hollow Beam

EXP

ARB

-r_m 0 +r_m

3.A Hybrid Solvers, SC Templates

Private Communications:

A. Friedman, D. Grote, I. Hofmann, M. Reiser, J. Struckmeier and R. York


Nghiem


Hofmann GSI

38

SC-13 Main Outcome I


1) The workshop has shown a rich activity on space charge related topics at CERN (LIU), GSI (FAIR), and RAL (ISIS upgrade).Several studies have been presented on experimental work and in particular concerning the LIU studies and ISIS upgrade. 2) In the session of code developments, the issue of code benchmarking has found a lot of attention with regards to long-term tracking. The presentation of the status of code benchmarking has been discussed and the issue about the noise created by PIC codes has evoked intense discussions. This topic has raised broader interest, in particular by J. Amudson, Eric Stern, J. Holmes, Ji Qiang, and Jean-Luc Vay. The decision has been taken to use the GSI test suite for benchmarking of frozen space charge models also for the benchmarking of PIC codes, both 2.5D and 3D. Firm plans have been made to benchmark Synergia and Orbit and hopefully also IMPACT and WARP.

39

SC-13 Main Outcome II


3) Some discussion has focused on role of GPU in the high intensity beam dynamics. Effort reported by GSI, RAL, FNAL were discussed at a deep technical level, and the present difficulties were highlighted.

4) Interesting from a theoretical standpoint has been a discussion about possibly equating PIC noise with intrabeam-scattering. To this end, previous work by J. Struckmeier has been reviewed on the effect of PIC noise on emittance growth. The audience has expressed the interest to study this topic in some detail.

• During the Joint Discussion Session 7 Question have been presented to the community. These questions have been formulated beforehand both at CERN & GSI

• I have modified them for this talk

• The full set can be found on the web:➔http://indico.cern.ch/internalPage.py?pageId=9&confId=221441


Question & Responses

41

Question 1


1) What about the modelling of our machines?

Responses: • What we really need is a better description of the machines• Our simulations codes are more or less fine benchmarking!➔•The question is how we can improve the nonlinear description of our machines? What are the possible methods?

Modeling of linear and nonlinear chromaticity Resonance driving terms (correct chromaticity, kick in both planes, )

osome data in PS, PSB: difficult since only 1 family of sextupoles

Experience at other labs: o Chromaticity, DTAo ISIS: turn-by-turno Octupolar components by local bumps

•There has been a significant amount of work being done for the PS in particular. But we still need a reliable magnet by magnet model which would require a more structured effort.

42

Question 2


2) What are the relevant mechanisms for beam loss and emittance growth?

Responses:

• There seems to be consensus that resonances due to non-linear fields in conjunction with the tune spread due to space charge seems to lead to both problems: just approaching a resonance leads to emittance blow-up and resonance crossings may lead to losses.

• On a deeper level the community is split about the questions if these incoherent or also coherent effects are important, as being discussed concerning the half integer resonance crossing in the PSB.

43

Question 3


3) Which instrumentation do we need to measure our space charge effects?

Responses:

•Concerning instrumentation there has mainly been proposals to provide 1000 turn BPM systems with good resolution • Further the developments concerning wire scanners to allow for instance a better resolution of beam halo, i.e. improving existing instrumentation. •Transverse tomography and collimators to reconstruct the transverse profile have been proposed as possible new tools.

•It was also mentioned that kickers in both planes are important to study coupled resonances.

•There has been a proposal to study coherent effects: A quadrupolar pickup should allow to measure coherent response of the beam to the half integer resonance. In fact, this pickup would allow for the first time the measurement of the real space charge tune spreads in a machine.

44

Question 4


4) PIC versus Frozen model: Which difference is "normal" and which is not?

Responses:

• This issue has been central in the discussion during the workshop.

•On the one hand, benchmarking of several PIC codes have now been pursued and compared with the results of frozen space charge codes. ➔This effort will take the better part of this year!

•On the other hand, renewed interest has been expressed to look into the noise issue of PIC codes in long-term storage ring simulations in more detail by several teams.

•It has been mentioned by PIC code developers that the single particle behavior allows a lot of in-sight into the understanding the working of the PIC codes.

45

Question 5-7


5) There seem to be some beams in the PS with most of the beam in the PS below the integer resonance not too ➔many losses nor emittance growth. Can we explain this?

6) What is the maximum allowed space charge tune shift? 7) What can we do from the optics point of view to reduce space charge effects?

Responses 5) No clear position from the audience except that the experimental data would have to be well documented to allow for a clarification.

Responses 6-7) Both of the last 2 questions simply would need more studies. In fact, our simulation tools have to be better understood and benchmarked with experiments before a reliable answer can be given.

46

Code Developer Meeting 22-26.04.13


• Invitation of code developers after SC-13 workshop • Jeff Holmes(SNS) ORBIT proper• Leonid Vorobiev ORBIT expert of Fermilab • Eric Stern for Synergia• Giuliano Franchetti stayed on for a few days

• Main goals• Learn from the experts about the 2 PIC codes• Demonstrate issues we had with un-physical behavior• Plan benchmarking effort for both codes• Mitigation efforts for the noise problem• Alternative approaches

• Lively discussions and presentations all week long

47

Code Developer Meeting II

• The support for the ORBIT code as we know it will be discontinued this year!

• It will be replaced by pyORBIT• Mainly replacing Interface• Hopefully they will introduce a hook to PTC• There might even be a manual!

• Jeff gave a lecture about the most relevant ORBIT modules

• Benchmarking is agreed to proceed


48

Code Developer Meeting III

• Eric explained MPI as used in Synergia and gave a detailed report on the RUN facilities in their code.

• Jeff raised the question if PIC codes are the proper tool for long-term studies in storage rings! He expressed hope that the so-called FMM technique might be the right approach for the future.

• Leonid explained in detail both his mitigation proposal for minimizing the noise in ORBIT and his space charge templates as a possible long-term solution.

• Summary: http://frs.web.cern.ch/frs/Source/space_charge/SC-13/Code_developer_meeting_22-26.04.13/


49

Reserve



Ohmi J-PARC MR

53

Question 1


1) What we are really missing at the moment is a better description of our machines WRT to nonlinearities and in particular the variation from magnet to magnet. How can we determine this true non-linear model of the machine and how shall we treat our limited knowledge about it (keywords: sigma of the individual multipolar component, how many seeds). What are the different methods, which can be used to make some progress in this aspect (and time scale)? Which instrumentation do we need to measure "properly" the machine nonlinearities?

1.1)What is the meaning of “properly” in a space charge dominated regime that require 105 turns of storage time? In particular, the requirements are significantly affected by the physics: there are regimes in which 10% error in the knowledge of nonlinear components is enough for a good description of the machine, but there are other regimes where the concept of “good” or “properly” is difficult to be defined.

•Responses: What we really need is a better description of the machines; our simulations codes are more or less fine; The question is how we can improve the nonlinear description of our machines? What are the possible methods?

linear and nonlinear chromaticityresonance driving terms (correct chromaticity, kick in both planes, )

osome data in PS, PSB: difficult since only 1 family of sextupoles

experience at other labs: ochromaticity, DTAoISIS: turn-by-turnooctupolar components by local bumps

•There has been a significant amount of work being done for the PS in particular. But we still need a reliable magnet by magnet model which would require a more structured effort.

54

Question 2


2) For the actual necessity of projects what are the mechanisms for beam loss and emittance growth that are most important => Review of the relevant issue of each project and review of their relevant mechanisms including a comparison with some simple formulae Do we have good agreement?

Responses: There seems to be consensus that resonances due to non-linear fields in conjunction with the tune spread due to space charge seems to lead to both problems: just approaching a resonance leads to emittance blow-up and resonance crossings may lead to losses. On a deeper level the community is split about the questions if these incoherent or also coherent effects are important, as being discussed concerning the half integer resonance crossing in the PSB.

55

Question 3


3) Which instrumentation do we need to measure our space charge effects? What has been the progress in instrumentation in the past years and what are the plans for the future?

•Responses: Concerning instrumentation there has mainly been proposals to provide 1000 turn BPM systems with good resolution and further the developments concerning wire scanners to allow for instance a better resolution of beam halo, i.e. improving existing instrumentation. •Transverse tomography and collimators to reconstruct the transverse profile have been proposed as possible new tools.•It was also mentioned that kickers in both planes are important to study coupled resonances. •There has been a proposal to study coherent effects: A quadrupolar pickup should allow to measure coherent response of the beam to the half integer resonance. In fact, this pickup would allow for the first time the measurement of the real space charge tune spreads in a machine.

56

Question 4


4) The numerical aspects of beam physics codes with space charge are very important => In particular when we benchmark codes of either type frozen model or PIC type: which difference is "normal" and which is not.

•Responses: This issue has been central in the discussion during the workshop. •On the one hand benchmarking of several PIC codes have now been pursued and compared with the results of frozen space charge codes. This effort will take the better part of this year. •On the other hand, renewed interest has been expressed to look into the noise issue of PIC codes in long-term storage ring simulations in more detail by several teams.•It has been mentioned by PIC code developers that the single particle behavior allows a lot of in-sight into the understanding the working of PIC codes.

57

Question 5-7


5) There seem to be some beams in the PS with most of the beam below the integer resonance and with not too many loses and emittance growth. Can we explain this?

6) What is the maximum space charge tune shift evolution over the past years / decade in the different machines. Did we make progress? Can we dream to reach more than 0.4? What are all the possibilities to push forward the limit? Is the only possibility to fight space charge to increase the injection energy? 7) What can we do from the optics point of view to reduce space charge effects?

Responses 5) No clear position from the audience except that the experimental data would have to be well documented to allow for a clarification.

Responses 6-7) Both of the last 2 questions simply would need more studies. In fact, our simulation tools have to be better understood and benchmarked with experiments before a reliable answer can be given.

sc-13 – consequences for us?

Documents

liu5312013 fsoutcome

workshop summaries

workshop organization

workshop simulations

metral cern

zimmermann cern

liu studies

bunch space charge