Notes of Meeting Arc Half Cell Mock-up #7
11th May 2023
List of people attending the meeting:
F. Carra (chairman), J. Bauche, L. Baudin, J. Coupard, M. Garlasche, M. Guinchard, K. Hanke, R. Kersevan, H. Mainaud Durand, G. Papazoglou, A. Piccini, S. Rorison, C. Tetrault, M. Timmins, F. Valchkova-Georgieva, L. Von Freeden.
Slides of the presentations are available in the Indico page related to the meeting (ArcHalfCellMock-up Meeting #7 (11 May 2023) · Indico (cern.ch)).
Agenda:
1)
F. Carra introduced the meeting by reviewing the actions of the meeting #6 minutes. It would be interesting when we have more information to share the link of the Tuning workshop (28th of June) on this Indico Page (action A. Piccini).
The action on the magnet assembly sequence is still opened: all stakeholders should provide their inputs in the table available in the Indico event https://indico.cern.ch/event/1278881/. Once the table is completed, it will be shared with colleagues from Technical Infrastructure Working Group responsible for the coordination of the assembly sequence at the scale of the facility.
F. Carra also asked M. Timmins to share the graphs sent by A. Unnervik about the evolution of the price as a function of the quantity of elements produced (action M. Timmins). M. Timmins mentioned that there is a 2-3 factor between the prototype price and the series price of the elements.
2)
F. Valchkova-Georgieva presented updates on the tunnel cross section integration for regular arc. The idea is to cut a section of concrete over a height of 500 mm and a width of 223 mm. Hence it is no longer necessary to reduce the height of the beamline of the collider from 1030 mm to 650 mm. The beamline of the collider can remain at a height of 1030 mm without compromising the integration in the RFs sections.
M. Timmins indicated that this configuration can be problematic to actuate the jacks of the collider. K. Hanke noted that this configuration is certainly not optimal and needs to be analysed and optimised.
J. Bauche asked how the magnets will be installed, because in this configuration there is an important cantilever. F. Valchkova-Georgieva indicated that Fraunhofer have already worked on this configuration, and they seem to have no issues installing the magnets. She reminded that it is in the case of scenario 4 of the RF/cryogenic layout point L that it would be necessary either to increase the diameter of the tunnel or to lower the beamline of the collider to avoid integration problems (cf. integration presentation Reunion Integration FCC (17 March 2023) · Indico (cern.ch)). J. Bauche commented that it is surprising that a small section regulates the design of the whole tunnel layout: why not consider this scenario 4?
J. Bauche and M. Timmins added that this layout does not seem to be optimal in terms of maintenance, operation and even safety.
F. Carra asked if this configuration is compatible with FCC-hh? F. Valchkova-Georgieva confirmed that it is also compatible.
F. Carra asked if this option will be presented at the FCC week. F. Valchkova-Georgieva confirmed. F. Carra observed that presentations at the FCC week should be coordinated so one baseline is presented.
3)
H. Mainaud Durand presented the PACMAN project, its outcomes and the possible applications for FCC-ee. The slides were shown at the International Workshop on Future Linear Colliders (LCWS2017) are available on the Indico page of the event. in the work frame of CLIC project the aim of the PACMAN project was to develop a new semi-automated method of fiducialisation and alignment of the quadrupoles and accelerating cavities on the 40000 girders foreseen for CLIC and at the same time to develop a 5-axis adjustable platform. H. Mainaud Durand explained that this project was focusing on a 2-m-long girder, one of the difficulties will be to extrapolate to 6-m-long girder needed for FCC-ee SSS.
Slide 17 presents in a short video the work achieved by a team of students in the context of the Euspen Challenge. They propose a strategy to align the elements on the girder and to use the Digital Twin to readjust the alignment after a perturbation.
F. Carra asked if the PACMAN configuration is compatible with the use of a single vacuum chamber? H. Mainaud Durand confirmed that it was already the case.
M. Timmins asked if the position was held after alignment during transport and handling, how was it held i.e., what was the maintenance strategy? H. Mainaud Durand answered that they did not have time to assess the maintenance. A mobile alignment bench, transportable in the tunnel, would have been developed to realign a girder if needed. In DESY, the components are glued on the girder to mitigate the consequences of the vibrations during the transport on the alignment of the components.
M. Timmins asked what were the outcomes in terms of stability when introducing adjustment platforms in the assembly? H. Mainaud Durand explained that it is difficult to assess because all the components were glued after being aligned.
F. Carra asked how does the alignment team plan to use the mock-up? H. Mainaud Durand explained that the mock-up will be interesting to test alignment strategies. And to go further, it would be interesting to test the impact of parameters such as temperature on the alignment. Indeed, the elements will be aligned at 20°C but in the tunnel, there are important temperature variations which can impact the alignment. It would therefore be interesting to monitor this impact with the Digital Twins and to develop a method of correction by using the mock-up.
4)
A. Piccini presented the updates concerning the analyses of the dynamic stability of supporting systems: girder and booster’s support. The slides are available on the Indico page of the event. The methodology consists of modelling the girder, the magnets (quadrupole and sextupoles) and the girder supports using springs. The model is excited by the Power Spectral Density (PSD) of the ground displacement (measurements performed in LHC tunnel) and the PSD of the beam displacement is computed as an output. The integrated Root Mean Squared (RMS) displacements at the level of the beamline are then computed and compared to the specifications.
M. Guinchard indicated that it would be interesting to study the response to an impulse excitation, because it is happening for the LHC.
F. Carra asked if it is possible to optimise the model by considering a girder with filler in it? Did Framag send data about the girder study? C. Tetrault answered that Framag team thinks they can meet the given specifications but ther have not done any FEM yet, they will send a cost estimate. L. Baudin added that some simulation trials have been performed to evaluate the impact of the filer. By filling the studied steel girder with Framag filer (data from their website), we can gain a factor of 1.8 in terms of stability. And Framag had indicated in a meeting that simplifying simulations like this are about 80% of accuracy.
F. Carra noted that the specifications are not met for low frequencies below 1 Hz. A. Piccini answered that in fact these specifications are not met, but what is problematic is that even considering only ground motion they are not fulfilled. This is why it is important to review these specifications with T. Raubenheimer and K. Oide (action A. Piccini). M. Guinchard noted that for frequencies below 1 Hz this can be corrected by beam-based alignment. He indicated also that it is important to understand if the specifications stand for the magnetic axis or the beamline, because a factor of 10 can exist between them (this was exhibit for the LHC main dipoles for instance). In addition, it is essential to pay attention to the dynamic structure of the magnet and their fixations, because even if the girder is optimised in terms of stability, it will be useless if the magnets damage the stability.
M. Timmins indicated that these simulations these simulations are currently a rough approximation and need to be refined. L. Baudin reminded that the aim is to benchmark these simulations to optimise them and to have an accurate model at the end. M. Timmins proposed to set up a meeting with PSI to get their inputs (action M. Timmins).
M. Guinchard proposed to benchmark the simulations for some of the elements like the magnets, or even do the inverse, run simulations on elements already tested and compare the results, using for instance the results gathered on the Sesame project.
F. Carra commented that the results are promising, but we cannot confirm at this stage that the specs are met, since, when adding the other elements to the simulation process (flexibility of the magnet, transfer function mechanical axis to beam axis, vibration cross-talk between booster and collider, etc.), the vibration amplitude will increase. Anyhow, at least in terms of order of magnitude, it seems that we are in the good direction.
Actions follow up
Opened actions
Minutes written by: Audrey Piccini and Lucie Baudin.