Choose timezone
Your profile timezone:
BGI Regular Meeting
→
Europe/Zurich
865/1-B03 (CERN)
BGI Regular Meeting
Participants
Present: Chiara, Gunn, Hampus, James, Juri, Lauren, Lukas, Mark, Ray, Wilfried.
HL-LHC BGI
Follow-up from previous meeting
Outstanding points reviewed from the previous meeting. A rough updated estimate for the mechanics has now been made and was presented as part of the updated budget review. The minimum bending radius for the high-voltage cable is still to be confirmed, and this remains an input needed to converge on the magnet-side integration. It was agreed that a right-angle connection is likely to be required because a direct bend appears too tight for the available space.
The possibility of integrating shielding directly into such a right-angle connection was discussed. A custom interface mounted on the flange, potentially incorporating a shielded connector or cable gland, now seems the most realistic route. The exact available space around the flange and feedthrough remains to be checked in detail with Lauren before a final concept is selected.
HV feedthrough and connector integration
The HV feedthrough remains an open design point. The current preference is for a 45 Ω controlled-impedance solution, but the previously considered feedthrough does not satisfy this requirement, so alternative candidates must be identified. One possible baseline is to proceed with a suboptimal but practical feedthrough solution in the short term, and then ask the RF experts to help assess the impedance impact.
It was noted that approximately 75 mm are available between the front flange and the end of the threaded part of the electrical feedthrough. This provides a concrete geometrical constraint for the coming design iterations. The external cable connection will also need careful shaping to avoid sharp points and local electric-field enhancement. Possible mitigations such as rounded terminations or local insulating covers were mentioned.
Timepix mounting, brazing and glue tests
Glue outgassing tests are still ongoing. The PCB samples are currently under test, and additional glue preparation remains to be completed before the final vacuum-related checks can be closed. In parallel, the proposed brazing route has been sent to the CERN brazing specialists for feedback, and gold-plating options still need to be followed up with Jerome.
The main unresolved issue is no longer the ceramic-to-copper-tungsten brazing itself, which is considered relatively standard, but rather the later attachment of the Timepix assembly and the compatibility of this sequence with bump-bonding constraints. A key concern is whether the required temperatures and applied pressure during soldering or later process steps could damage the already bonded chip-sensor assembly.
Because of the number of coupled steps and interfaces, it was agreed that a dedicated meeting should be organized with the relevant specialists to map the full assembly workflow, identify where each option could fail, and clarify which steps must happen before or after bump bonding. The glue route remains an important backup solution and should continue in parallel.
Simulations and mechanical tolerances
Updated VIPM+CST simulations were presented to quantify the effect of angular misalignment between the instrument and the magnetic field. For the horizontal BGI, misalignment around the relevant axis leads to ExB drift and mainly shifts the measured profile; the associated sigma error remains acceptable over the cases studied. Misalignments around the other axes were shown to be less critical for profile-width reconstruction in the present configuration.
The conclusion from the simulations is that the most important point is to keep the angle between magnet and instrument as small as possible. From a mechanical standpoint, the standard ±200 µm linear tolerance is considered sufficient. This was explicitly stated in the presentation and was accepted as the working tolerance to pass to mechanics. It was also noted that this tolerance should apply not only between instrument and magnet, but also internally to the parallelism of cathode and anode during assembly.
Alignment strategy and fiducialization
A broader discussion followed on how the relevant references will be established during installation: beam axis, magnetic axis, mechanical axis of the magnet, vacuum chamber references, and the instrument flange reference. The current thinking is that the magnet will first be installed and aligned relative to the beam, after which the vacuum chamber and instrument will be aligned to the magnet. However, the exact relation between magnetic and mechanical axes still needs to be formally captured and made usable for installation.
It was emphasized that the future maintenance scenario must also be considered. The team would like to avoid a concept in which every intervention requires a full surface pre-assembly campaign. Repeatability after removal and reinstallation therefore needs to be checked. It was suggested that flange tightening procedures and the use of alignment shims or pins will be important to guarantee the required positional reproducibility.
Budget and EVM structure
An updated HL-LHC BGI budget breakdown was presented, structured by work unit: beamline/vacuum, mechanics, electronics, magnets, power converters and infrastructure. The beamline contribution from the vacuum group is about 120 kCHF. Mechanics for five vacuum chambers and five instruments, including one spare, are currently estimated at about 300 kCHF. Electronics are estimated at about 173 kCHF.
The largest change remains in magnets. The magnet estimate has increased significantly relative to earlier iterations, reaching roughly 700 kCHF, including the new magnets, refurbishment of trim magnets, and associated design-office effort. By contrast, savings have been made on the power-converter side because existing converters can be recycled and only electronics upgrades are needed.
The resulting total project cost is now about 1.4 MCHF, compared with an available envelope of about 1.35 MCHF. It was agreed that this should be treated carefully in the EVM: the present subsystem budgets are effectively zero-contingency estimates, and the magnet overrun should be clearly declared as a specific cost pressure rather than silently absorbed into the rest of the project. James will circulate the updated structure so that the EVM and deliverables can be cleaned up and aligned with the present final-production scope.
Actions
-
Gunn to provide the minimum HV cable bending radius and cooling-water parameters.
-
Lauren and Wilfried to confirm the detailed space envelope around the flange/feedthrough region.
-
Wilfried/Mark to identify candidate HV feedthrough solutions, including a practical baseline for RF review.
-
James to organize a dedicated meeting on the Timepix/ceramic/brazing/bump-bonding workflow.
-
James to circulate the updated budget/EVM structure and reflect the magnet cost increase explicitly.
PS & SPS BGI
Software and commissioning status
Work is continuing on stabilizing the PS/SPS systems and bringing the software chain into routine use. The immediate goal is to obtain one profile at one moment in the machine cycle; the next milestone after that will be bunch-by-bunch acquisition at a fixed cycle point.
The current schedule is a week behind the original plan, but the team considered the goals still achievable. Next week is expected to be important for closing the remaining software issues, including automatic calibration adjustments and overall system stability. OP has already been consulted on which cycle points should be prioritized first, so the initial tests can focus on the most relevant measurement windows.
Measurements and GUI integration
For the first deliverable, the intention is to target a single useful cycle point and demonstrate a reliable profile measurement there. Once that is established, the same framework can be extended to bunch-by-bunch operation. This will require the usual time-offset correction, but the team expects that previous work from the existing detectors can largely be reused.
The question of integration into the operational GUI was also raised. The expert tools are already usable through the navigator/software framework, but the operator-facing display is still largely in the hands of the OP teams. The team would prefer not to push too early, but it was noted that it would be useful to understand when a fixed operational display might realistically become available.
Temperature monitoring and shielding
A discussion was held on whether additional temperature monitoring should be added, especially in the SPS, where beam-induced heating and wakefield-related effects remain a concern. At present the system reads temperatures on cooling-water lines, but this is not very sensitive at the low dissipated power levels involved. Moving or adding a Pt100 directly onto the vacuum chamber or flange was considered more useful.
This would provide at least a simple indication of whether beam-induced heating is occurring during scrubbing or high-intensity operation. Similar external Pt100 measurements have been used successfully on SPS wire scanners in the past to observe chamber temperature increases. Since spare PLC channels and infrastructure are already available, this looks like a straightforward improvement and should be pursued.
The team also discussed whether, during a future technical stop, the SPS horizontal instrument should be equipped with the newer higher-open-fraction shielding that was recently installed in the vertical instrument. This will depend partly on the operational results obtained with the present hardware, but the feeling was that bunch-by-bunch measurements may be difficult with the current horizontal shielding and that the comparison could be useful.
PS HV upgrade preparation
For the PS vertical BGI, the team confirmed the plan to replace the present 20 kV-rated HV cable with the LHC-type cable already used elsewhere. This is being done now so that the infrastructure is in place for a future HV upgrade of the instrument. The corresponding EN-EL ticket has been accepted and planned.
However, the cable replacement alone is not enough to raise the operating voltage. To do that, the PS instrument itself would still need design changes, including a revised feedthrough and an updated ceramic geometry with larger creepage distance. The team estimated that the required hardware changes are of order 10 kCHF and could most likely be covered from operations budget when the timing is right.
This upgrade remains attractive because the present limitation on cathode voltage is one of the reasons the PS vertical instrument struggles at injection when the beam is close to the detector. The team agreed to revisit the detailed implementation after the LHC priority period, while taking advantage now to order or prepare any long-lead parts.
Actions
-
Wilfried/James - Add or relocate Pt100 temperature monitoring to the SPS vacuum chamber/flange.
-
James - Keep under review whether the SPS horizontal shield should be replaced with the newer high-open-fraction version.
-
Gunn - Follow through the PS HV-cable installation and prepare the later instrument-side HV upgrade concept.
There are minutes attached to this event.
Show them.