-----------------------------------------------------------
Minutes of the ABP-RLC team meeting of 17.06.2005
present: EB, HBU, UD, AG, WH, JJ, EM, TP, FR, FZ
excused: DS
web site: http://ab-abp-rlc.web.cern.ch/ab-abp-rlc/
------------------------------------------------------------
(1) Minutes of last meeting, pending actions, announcements
-----------------------------------------------------------
=> ACTION (EM+FZ): compare and understand the different predictions
of the Burov-Lebedev theory calculated by FZ and EM.
STATUS: EM had sent his notebook to several people, and
presented a summary report (see below)
There is no further action needed from EM on this comparison.
The main reasons for the discrepancy were a missing factor
beta/(4 pi) in FZ's notebook, the different bunch shapes
(parabolic vs. Gaussian), and the different degrees of
approximation in the Burov-Lebedev impedance formula.
The last difference was a small effect compared with the
other two.
(2) Comparison with FZ and AK for the transverse RW impedance (EM)
------------------------------------------------------------------
EM compared three formalisms: Vos, Burov-Lebedev, and Zotter.
Evaluation by EM, AK, & FZ for a test example showed some
differences at low frequencies when either approximations
in the Burov-Lebedev paper or the Vos formalism were used.
The exact impedance is the lowest. At high frequencies, the
three formalisms and different evaluations agree.
The advantage of Burov-Lebedev, implemented by AK, is that
it can treat any number of layers. The Zotter formalism is at
the moment limited to two layers, but it is valid to arbitrarily
high frequencies, above ~1 MHz, for any velocity, and even for
different velocities of beam and wave.
It still remains to reconcile the Piwinski-based solution
with the Burov-Lebedev one.
(3) Follow-up of beam-beam related optics (WH)
-----------------------------------------------
WH reported five different optics requests from CMS, with
about four different values of beta-star and different
crossing angles. One is nominal and one is for TOTEM.
The other three are for a new letter of intent. Two of
the new optics can be adapted from existing ones, but
the third new wish with a beta* of 100-200 m and 300
microrad crossing angle, is excluded by present
magnet-strength limits for the crossing angle.
WH is studying in detail the various aspects of the
beam-beam interaction for the different optics versions.
He will summarize his findings in a report. Many
constraints and experiments requests need to be considered,
such as a high-dispersion value at the detectors,
the exact phase relationship between the IP and the
detector, the absence of obstacles for forward particles
(off momentum p*), or reversal of spectrometer polarity.
FR pointed out that beam parameters like beta-star
and beam emittance should be determined by the
accelerator physicists and not by the experiments.
He also recommended to discuss conflicts between
TOTEM requirements and new collimators near the IPs
at the LTC. E. Tsesmelis is presently following up
this issue.
HBU mentioned that some optics have different integer
tunes.
JJ described that Alice has an optics with the smallest
beta-* of all (0.5 m). JJ is taking care of the ALICE
optics.
FR reported that the LHC emittance budget may be too
tight in view of the measured injection kicker ripple.
HBU and FZ suggested that some margin might be gained by
reducing the emittance in the SPS and allowing for
larger blow up in the LHC.
(4) Head-tail growth rates vs Q' for the LHC at injection (EM+FZ)
-----------------------------------------------------------------
EM and FZ followed up a request by Lyn Evans to compute head-tail
instability rise times as a function of chromaticity.
JJ explained that this question came up in the LHC MAC, triggered
by a question of F. Willeke, who referred to experience
with unstable beams on the ramp in HERA ('Batman effect').
EM mentioned that in HERA the interplay of coupling and
chromaticity is important.
FZ considered an impedance model consisting of beam screen,
warm chambers, 41 phase-1 collimators, and a broadband resonator
with Q=1 and resonant frequency of 5 GHz. He computed single-bunch
growth rates as a function of Q' for m=0 and m=3, assuming
Gaussian bunches, and coupled rigid-bunch (m=0) growth rates
for Q'=0 and Q'=10. FZ found that he could not reproduce
several of the effective impedance quoted in the design report.
EM considered a similar but more refined impedance model,
including separate contributions for MQ, BW and locally correct
treatment of TDI.
For the beam screen, he found the same effective impedances as
FZ, which at 20 MHz differ from the design report. For the
m=0 single-bunch mode his results was similar to FZ's, but the
rise times were widely different for higher-order modes.
With parabolic bunches, Landau damping can be guaranteed
up to a chromaticity of 7. For Gaussian bunches the maximum
stable chromaticity value is 4. The Landau damping limit is
computed by assuming a tune shift of 2e-3 at 6 sigma.
EM's minimum coupled bunch rise time for Q'=0 of 50 ms was
the same as that found by FZ, but the most unstable modes
were not the same.
FZ pointed out that in the SPS no unstable head-tail modes
are observed up to chromaticities of Q'=30.
(5) Follow-up of beam-beam simulations (TP)
-------------------------------------------
TP performed strong-strong beam-beam simulations for different
numbers of IPs for RHIC, using COMBI in a rigid and in a soft
Gaussian approximation. The results for the tune shift for the
rigid case agree with a matrix calculation. A factor 2 error
in the tune shift was found in Ji Qiang's calculation presented
at PAC'05.
(6) Longitudinal impedance of a transverse slot (TCDS) (EM)
-----------------------------------------------------------
EM quoted the slot impedance computed by AG as Z/n~30 microOhm.
There is an option to reduce the slot width (longitudinal extent)
from 1 mm to 0.5 mm for all 11 slots per tank (2 tanks in total).
Using a formula for the slot impedance of Zotter, EM found an
impedance of Z/n~15 microOhm. The formula does not strictly
apply to slot length comparable to the chamber radius, as is the
case here, which could explain the factor 2 discrepancy with
AG's calculation. Regardless using this formula to determine
the dependence on the slot width, EM showed that a marginal
gain of about 15% could be expected. It was thought that this
was not worth the effort of changing the design.
[After the meeting is was pointed out by AG and EM that this value
referred to TCDQ and rough scaling to TCDS would yield ~12 microOhm.]
(7) AOB
-------
FR announced the creation of two new working groups
PAF - proton accelerators for the future
POF - physics opportunities with future proton accelerators
Posted on the web: Slides by EM, TP, FZ
Web site: http://ab-abp-rlc.web.cern.ch/ab-abp-rlc/