------------------------------------------------------- Minutes of the ABP-RLC team meeting of 22.04.2005 present: EB, Ulrich Dorda (UD), AG, WH, EM, TP, FR special guests: Fritz Caspers (FC), Mario Deile (MD), Marek Gasior, Paul Proudlock web site: http://ab-abp-rlc.web.cern.ch/ab%2Dabp%2Drlc/ ------------------------------------------------------- (1) Minutes of the last meeting and pending actions ---------------------------------------------------- The discussion is postponed to the next RLC-LHC team meeting. (2) Impedance measurements of LHC Roman pots (Mario Deile) ---------------------------------------------------------- MD presents measurements of beam induced signals in the TOTEM Roman pots installed last year in the SPS and bench measurements of field attenuation and impedance with the wire method. The Roman pots in the SPS were instrumented with a wire loop and a pin probe to measure E and B fields, respectively, and two Al windows of thickness 140 and 210 micron were tested. EMI measurements were repeated with open and with terminated coax cable outside the pot near the beam pipe, using both a CNGS beam and an LHC-type beam with 25 ns bunch spacing: the corresponding signals at 200 MHz and at 40 MHz were clearly visible, together with several harmonics of the bunch frequency, both with the loop mounted inside the Roman pots and with a reference loop mounted outside. The conclusion is that most of the signal was picked up through the cables in the SPS tunnel. Transmission measurements in the lab as a function of frequency for different distances between Roman pot windows and wire show a reduced transmission with increasing distance, an E-field attenuation by the windows >= 60 dB @ 40 MHz, and a B-field attenuation of 50-60 dB @ 40 MHz. Further lab measurements will be needed to assess the effect of the LHC beam signals on the Roman pot silicon detector electronics. There are plans to mimic the LHC bunches using a spark generator with ~20 kV pulses and ~20 A peak current. Bench measurements of the transmission coefficient S_21 with the wire method show two resonances at ~650 and ~1250 MHz which move and become stronger when closing the Roman pots (there is also a stronger pill-box resonance at ~1800 MHz due to end flanges, not part of the Roman pots). Using the retracted position of the Roman pots as a reference yields a longitudinal coupling impedance Z~1.2 kOhm around 700 MHz with Q~114, for a 2 mm distance of both pots to the wire, corresponding to Z/n~19 mOhm and to a dissipated power of 220 W with LHC beam. These somewhat pessimistic results are obtained adopting a so-called "improved distributed model" and do not properly take into account the 25 ns bunch spacing. The Q-value can be reduced by a factor ~5 using little pieces of ferrite glued around the moving part of the pots in their cavity. The geometry of these ferrites has to be optimised. WH remarks that a 25 ns spacing is pessimistic for TOTEM operation, originally foreseen with only 43 bunches. MD replies that operation with a luminosity of up to 10^33 is now contemplated. (3) Follow-up of collimator impedance measurements (EM) ------------------------------------------------------- The predicted tune shifts for a lower graphite resistivity of 10 microOhm m are closer to the tune measurements obtained independently by Marek Gasior (BBQ method) and by Fritz Caspers&Tom Kroyer (using a 245 MHz pick-up). These two independent experimental results are in fairly good agreement and are consistently larger than the estimates based on the linear theory by a factor ~two for a jaw half-gap below 1 mm. The theoretical tune shifts predicted by the linear theory are essentially the same as those obtained by the classic thick-wall resistive wall impedance, showing that the so-called "inductive by-pass" effect is not tested by the SPS measurements. Predictions based on the Piwinski approach and including the finite beam size are somewhat smaller than the measured tune shifts. The spectra obtained for collimator-in conditions show a reproducible reduction of the peak betatron oscillation amplitude compared to those for collimator-out conditions. According to FC this may indicate that, for larger impedance, there is less coherence of the self-excited beam oscillations. FC underlines also the existance of a time structure of the measured signals, ranging from the scale of msec to sec. This time structure has still to be understood. (4) Zotter's resistive-wall impedance with AC conductivity (EM) --------------------------------------------------------------- Using the “general” Zotter formula (see RLC of 04/02/05) with AC conductivity yields Burov & Lebedev results in the low and intermediate frequency regime (kHz -> GHz) and the results reported by AG (see RLC meeting of 18/03/05) using Bane’s formula in the very high frequency range (THz). Therefore Zotter's formalism unifies the two approaches and is also valid for any beam velocity. FC remarks that the effect of the surface roughness and of the an-isotropy of the graphite micro-crystals probably becomes overwhelming in the THz range. (5) Simulation of trapped modes in the TDI absorber (AG) -------------------------------------------------------- Preliminary HFFS results indicate two trapped modes around 770 MHz, with Q~60000 and R/Q ~ 1 LinacOhm, leading to a power deposition of some 5 kW for the LHC beam. These trapped modes can be suppressed by avoiding a (useless) transition to very large pipe apertures near the two ends of the TDI vacuum chamber. Possible further trapped modes will be computed by AG using a second-iteration geometry with a modified transition region and, depending on the results, ferrites absorbers or a vacuum chamber made out of SS instead of Cu may be envisaged. (6) Reports from other meetings and AOB --------------------------------------- FR reports that FZ's presentation at the last LTC meeting on the new results of LHC e-cloud simulations had generated some misunderstanding, including the wrong impression by the AB department leader that the last few years of e-cloud studies may have been a waste of time. FR and FZ will write a short paper to clarify the situation and state that beam scrubbing may be needed in the LHC, even if the heat load for nominal LHC beam intensity seems now compatible with the available cryogenic power. EM will represent the RLC-LHC team at the LHC Collimation WG and, in particular, will attend the meeting of next Monday. EM will also report about collective effects aspects discussed at the WG on Beam Losses and Radiation, chaired by G. Arduini. FR will be absent from 25 to 27 April. Posted on the web: Slides by MD, EM, and AG Web site: http://ab-abp-rlc.web.cern.ch/ab%2Dabp%2Drlc/