Minutes March 5th, 2008
Present: Elias Métral, Giovanni Rumolo, Benoit Salvant, Elena Shaposhnikova.
Agenda
- Benoit Salvant (slides, pdf)
"Update on TMCI measurement"
Minutes
- Benoit started a web page which will be linked to Gianluigi’s section and Elena’s upgrade studies web page. This webpage should contain the slides and minutes of the informal SPS impedance meetings, which started last year.
- B. Spataro wants to refine his simulations. This is what he is still doing, sending me all the infos.
- I asked for the drawings of the 800 MHz RF cavities (which are in the process to be scanned => I still have to wait about 2 weeks) and wire scanners (I have them and I will send them to Bruno asap).
Elena would like to know the way the incoherent band is shifting with intensity when emittance Blow up is made with some band limited noise (double RF bucket in SPS and single RF bucket in the LHC).
Reminder: There is no longitudinal space charge yet in HEADTAIL.
Summary:
- Measurements of tune shift with intensity below and above the instability threshold (TMCI MD Nov 4th 2007) show a discontinuity in the shift of the main spectral line when the instability appears. Comparing these measurements with HeadTail simulations, this shift can be understood as the consequence of the coupling of transverse modes -2 and -3.
- Observing the -2 and -3 transverse modes shift before the coupling and the instability seem very difficult in the conditions of the experiment. Maybe a more careful correction of the chromaticity and the amplitude detuning would help enhancing the signal.
- The issues that came up in comparing the simulations with the measurements are currently a work in progress:
1) Initial longitudinal parameters for the simulations are not obvious to choose, as the experiment shows that the bunch is not matched to the bucket.
2) With current parameters the instability occurs at 4e10 p/b in the simulations instead of 8e10 in the measurements. Transverse space charge is not included in the simulations, and may explain this difference.
3) Fitting the tune shift with current in HeadTail and the measurements means introducing a transverse shunt impedance of 13.44 MOhm/m, instead of an expected 19 MOhm/m from Helmut’s transverse impedance assessment of the SPS using Sacherer’s equation with a parabolic bunch distribution. Assuming Gaussian or Parabolic distribution yield different values of the effective impedance.
Next steps:
1) Continue the comparison between parabolic and Gaussian bunches for Sacherer’s formula (mode 0 only) => Effective and real impedance.
2) Produce a table of effective impedance deduced from Helmut’s measurements over the years (Gaussian bunch, Parabolic bunch and HEADTAIL simulations).
3) Simulate the TMCI with the MKE kickers only (with the real wakefield) => Was done already in the past by Giovanni to see the threshold. Now one wants to look at the evolution of the different modes, which modes are coupling etc., and what is missing to explain the measurements.
4) Start HEADTAIL simulation for the longitudinal plane => Quadrupolar oscillation vs. intensity with a broad-band impedance first and then we will use the wakefield that we already have.
Fritz made an apparition and said that due to the granularity the parabolic bunch (whose spectrum is going to higher frequencies than the Gaussian one) is better than Gaussian => This is one of the conclusions of the workshop on Shottky noise held at Chamonix in 2007.
Author: Benoit Salvant CERN
AB/ABP-LIS