Minutes of the Linac4 Diagnostics Working Group Meeting held on 12 March 2009

Present: U.Raich, J.Tan, G.Bellodi, L.Soby, G.Tranquille, C.Dutriat, B.Cheymol, E.Bravin, W.Weterings, B.Holzer, V.Vlachoudis, C.Carli, T.Hermanns, K.Hanke, B.Mikulec.


1. Beam losses along Linac4 and transfer lines (G.Bellodi)

2. Status of the BLM design for Linac4 (B.Holzer)

3. Status of FLUKA simulations for Linac4 (V.Vlachoudis)

1. Beam losses along Linac4 and transfer lines

G.Bellodi presented beam loss studies along Linac4 and the transfer lines (see BeamLossesL4_March2009.pdf). Plotting aperture/rms beam size along the line for the nominal case, this factor always stays >8 in Linac4 (100% transmission); along the transfer lines there is one bottleneck visible just before PSB injection (septum?) where this factor decreases to only ~2.5 (99.97% transmission).

Including initial beam jitter and machine errors, transmission can get as low as 68.57% in Linac4 for the worst case (50k input particles, ~2k runs). Losses are concentrated in DTL tank 3, but loss patterns differ between runs. These numbers do not include any steering to improve the situation. If steering is used for the worst case (68.57% transmission), transmission can be improved to 99.986%. Losses stay below the limit of 1 W/m even for SPL duty cycle.

Concerning the transfer line simulations including errors (1750 runs), zero transmission corresponds to the worst case. Also this case can recover a transmission of 99.72% if the beam is steered adequately. 92.26% is the average transmission. Preferred loss locations are high dispersion and transition regions.

It was concluded that the cause of the losses is not a statistical effect, but the sum of machine imperfections. Steering should always lead to acceptable beam loss levels. In any case for Linac4 the installed beam loss monitors should have a certain flexibility to be moved along the line (enough cable length etc.). As mentioned already in a previous meeting there should also be some additional mobile BLMs available for the transfer lines.

It was agreed that BLMs represent devices to protect the machine against high activation (limit: 1 W/m) and for machine tuning (not to guarantee personnel safety). The question was raised how to assure the compliance of this limit with only approximately 1 BLM per 8 m in the Linac (10 BLMs in total; see https://twiki.cern.ch/twiki/bin/view/SPL/Minutes04February2008) and even less (~half) coverage along the transfer lines (~10-14 BLMs; see https://twiki.cern.ch/twiki/bin/view/SPL/Minutes09June2008).
To assure machine protection the BLMs have to be connected to the beam interlock controller.
In particular in the PSB injection region an appropriate number of BLMs (and other redundant loss measurement equipment) at optimised locations has to be foreseen to protect distributor, septum and the H0/H- dumps. In addition, ~8 BLMs might be required around the injection bump region in the PSB ring (see https://twiki.cern.ch/twiki/bin/view/SPL/Minutes07July2008). Final numbers and positions still need to be clarified there.

It is not clear if in particular at low energies the BLMs will be sensitive enough to detect energy losses. FLUKA simulations with the final geometry are needed to evaluate the loss spectrum. Specifications need to be set accordingly for energy thresholds. Also resolution, accuracy and time response should be discussed.

Finally G.Bellodi showed the BLM distribution used at J-PARC. They use >50 BLMs in their linac of different types and response times (between 1.6-40 us). Up to 50 MeV they rely on the detection of neutrons and gammas. The accuracy for all types is 1%.

2. Status of the BLM design for Linac4

An overview of the BLM design status was given by B.Holzer (LINAC4_BLM_10_3_09.pdf). She started with a table detailing the specifications for LHC, SPS, PS, PSB and Linac4. K.Hanke clarified that the specifications noted down for Linac4 should not be taken as such - this was just a first step for further discussions that never took place, but should be realised now.

The design of a high dynamic range digitizer with bipolar input is currently underway. Several architectures are being considered. Assembly and testing of an analog test board is currently in progress. The digital data treatment unit based on the LHC version still needs to be designed.

BI needs information on the particle spectrum in critical locations as well as precise specifications and information on data treatment, logging etc. are needed. Even though the Linac4 startup has been delayed by one year (2014), the original installation schedule should be maintained (cable installation starting mid-2011; equipment installation from 2012 with hardware tests starting mid-2012).

3. Status of FLUKA simulations for Linac4

Up to now no FLUKA simulations have been made to describe the expected energy spectrum from Linac4 beam losses. Only beam dump activation studies are underway. Already some time ago E.Mauro worked on a geometry model, which would have to be updated. V.Vlachoudis is hoping to have E.Mauro as fellow in his section from July 2009 in which case he could start working on this project. It is clear that no result can be expected before end 2009.

In addition studies need to be performed on the activation around the head and tail dump area in the PSB injection region (F. Cerutti?).

-- BettinaMikulec - 16 Mar 2009

Topic attachments
I Attachment History Action Size Date Who Comment
PDFpdf BeamLossesL4_March2009.pdf r1 manage 405.6 K 2009-03-16 - 11:28 BettinaMikulec Beam losses along Linac4 and transfer lines for nominal case and including errors.
PDFpdf LINAC4_BLM_10_3_09.pdf r1 manage 212.7 K 2009-03-16 - 11:29 BettinaMikulec Status BLM design for Linac4.
Edit | Attach | Watch | Print version | History: r2 < r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r2 - 2009-03-16 - BettinaMikulec
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    SPL All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright