LIFETRAC: Beam-Beam Simulation for Tevatron
project home  |  goals  |  results  |  meetings  |  documentation  |  team |  references

Project Scope and Goals
Beam-Beam phenomena in Tevatron are summarized in detail in V.Shiltsev's talk [1]. Our project is aimed mainly at creating a credible model of luminosity evolution when the beams are put into collision at 980 GeV. This implies first, taking into account all diffusion mechanisms and their interference with beam-beam interaction and second, realistic representation of the machine optics.
Results

Weekly Meetings
take place on Fridays, at 3 pm in the Penthouse, WBT.

If you are interested to be notified of the upcoming meetings and their agenda, send email to Valery Lebedev.


Past Meetings

5/21/2004

Sasha: continued with scallops. Varied initial emittance for the antiproton bunch number 2. The blow up changes with emittance but the amplitude of change is approx. twice smaller than observed between bunches 2-11. Changing machine chromaticity in order to adjust the difference between bunches due to beam-beam did not have any effect on the situation. There was very strong dependence on the emittance of the proton bunch, but the simulation should be repeated conserving Np rather than xi. We propose to test a possibility of changing the working point and moving it closer to half-integer resonance. Namely, Valeri produced a Tevatron optics with the tunes 0.53, 0.52. The tune pace in this region is free from strong resonances. The main problem encountered are chromatic aberrations of the beta function which limit available dynamical aperture (longit.)

Alvin: Analysis of longitudinal oscillation spectra.
Plan:

  • Test new bunch spacing
  • Implement Shottki spectrum calculation
  • Study dependence of the blow up during the first step on the length of the step more thoroughly
  • Small xi - proton helix, proton lifetime

5/14/2004

Sasha: simulation of scallops with lifetrac. For some working points a scallop-like distribution of emittances is seen after first 10,000 turns. However, theory predicts only bunches #1 and #12 to be different because of different tunes. Idea of explanation of difference between bunches 2-11: a) they have different emittances. b) different chromaticities.

Plan:

  • Vary initial emittance and chromaticity e.g. for bunch #2
  • Vary dp/p inside train (was not implemented yet)
  • Study Separation of 24 or 23 buckets instead of 21
  • Pursue study of proton life time (small xi)

5/7/2004

Yuri: Resonance crossing driven by synchrotron phase fluctuations caused by RF noise. The conclusion: RF noise of the order of 1E-10 results in huge (~0.4 rad) betatron phase fluctuation which randomizes the resonance completely.
Yuri: Coherent oscillations excited by dipole (local) can be significant because of different phase advances for protons and antiprotons between excitation location and IP. Conclusion: excited coupled oscillations with large pbar amplitudes and small proton.
Valeri: after optics correction faster decay of particles with large dp/p is observed. In future, beam-beam will limit sigma dp/p and hence the life time. A strong call for improvement of beam-beam effect to cure this. Proposals: 1. consider the CERN working point. 2. Change bunch pattern to eliminate the nearest parasitic IP's.
Sasha: Investigation of scallops. First stage - analysis of experimental data. The emittance measurements in the beginning of store are done first before and second after! scraping. This leads to stronger 'shaving' of first bunches in a train (those with larger emittances). Will try fitting with some simple model.

Plan:

  • Acquire experimental data for horizontal scallops
  • Model of fitting the beam with a gaussian after scraping
  • Simulate scallops with lifetrac: dependence on w.p.

4/30/2004

VS: parametrization of luminosity lifetime

Sasha: short description of simulation being in progress (xi=0)

4/23/2004

Alvin about his progress in simulating diffusion mechanism. Panagiotis on Strong-Strong status: first efforts were successful, the new code runs. Now ready to specify realistic physics parameters. Sasha: implemented calculation of beam size fitted with a Gaussian.

The plan:

  • make a run with xi=0
  • study dependence on initial antiproton emittance

4/16/2004

Yuri on noise levels: the additional simulations were finished and they confirm previous conclusions. Sasha about results of the week: testing of 12 and 19 longitudinal slices at the main IP's coincide well with predictions of analytical calculation. Calculation for 1st and 6ths bunch (will we see the 'scallops') is underway.

We decided to use a different characteristic instead of the rms emittance to display beam properties, since large-amplitude particles contribute strongly to this value. For example, a gaussian fit can be used.

The plan:

  • Add 12th bunch to comparison of 1st and 6ths
  • New optics (starting from Wednesday)

4/09/2004

Elliott McCrory on Fitting the Luminosity Decay. Sasha on results of the past week. Yuri on phase averaging for different slicing schemes (12,15,23). He evaluated the resonance driving terms:
  • Ideal optics (good averaging) -> 12 slices differ from 'infinite' number by a factor of 2.
  • If the averaging is 'broken', the difference becomes only 20%.
Yuri: dependence of model on amplitude of noise. The conclusion is: the noise level we have chosen seems very reasonable in terms of ratio with beam-beam.

Plan for the week:
  • Exclude LR in design optics and make a run with 19 slices.
  • Change bunch length without varying the RF voltage.
  • Start 'real' simulations?

4/02/2004

Alvin presented analysis of SBD data of the longitudinal phase space density evolution during store #3261. He compared the experimental data and an analytical model involving solution of the diffusion equation. There is a discrepancy in the results: the model suggests that the tails should be more populated than it is observed.

Alex showed simulation results for the past week (see the results section).
Yuri gave his understanding of the present picture: Comparison of the full design optics and the one with excluded Long Range interactions shows that:

  • the emittance is increased because of diffusion only
  • losses are caused by LR interactions
With "realistic" optics the main IP's contribute to the emittance growth but again, the losses are due to LR's. Probably, a different fill pattern can be considered to avoid this effect of parasitics.
Plan for the next week: survey what number of slices is required (with 55 cm longitudinal beam size) - 12, 15 or 23.

3/26/2004
We discussed results of the finished runs. Yury presented his understanding of scallops. Plans for the week:

  1. Alex: run design optics with 6 sigma hard aperture without noise.
  2. Alex: run 'real' optics, 6 sigma aperture + noise
  3. Yury: working point #0, design optics, 20sigma aperture, noise x 2
  4. Yury: number of slices + longer bunch
  5. Study feasibility of another working point.

3/19/2004
Valery summarized our current aims and plans. Alex about porting the code to GP Farms: our accounts there are set up, we have to install the supporting software (compiler, libraries, etc.) to compile lifetrac. Intermediate results of current simulation run: we are comparing two kinds of machine optics - the design and the measured. They have different helices, optics of main IP's, and the latter has crossing angle. The measured optics exhibits higher losses and bigger emittance growth. To identify the source of problems we vary simulation conditions: exclude parasitic collisions, eliminate crossing angle and 'repair' main IP's optics. Currently, almost every variation is done, the results will follow. Plans for the next week:

  1. Yury will work on analytical analysis of finite crossing angle and shifted beam 'waists'.
  2. Two more runs with the noise switched off must be done to complete the study of dynamical aperture due to beam-beam.
  3. Run design optics with excluded parasitic collisions, compile the results of comparison
  4. Decrease the hard aperture limit from 20 to 6 beam sigma and repeat comparison of the design and measured optics.

Team
  • Yury Alexahin
  • Valery Lebedev
  • Dmitry Shatilov
  • Alexander Valishev

References
  1. V. Shiltsev, Beam-Beam Phenomena in Tevatron
  2. D. Shatilov, Part. Acc. 52, 65, 1996
  3. Meeting on Beam-Beam Simulations for Tevatron, March 5, 2004, Fermilab

last modified 5/24/2004    A. Valishev
Security, Privacy, Legal Fermi National Accelerator Laboratory