Login

116th Impedance Working Group Meeting

→ Europe/Zurich
774/1-079 (CERN)

774/1-079

CERN

20
Show room on map
Carlo Zannini (CERN)
116th Impedance Working Group Meeting
Zoom Meeting ID
67615471072
Host
Elena Macchia
Useful links
Join via phone
Zoom URL
Hide

Present: C. Antuono, P. Bampis, E. De La Fuente, G. Favia, K. Lasocha, N. Mounet,  M. Neroni, C. Pasquino, H. Pommerenke, H. Shreiber, P. Trubacova, C. Wimmelmann, C. Zannini

General info

Actions from previous meetings to be found in relative Indico meeting page.

News from high intensity test:

  • Single beam operation target achieved -> TCLD beam 1 worse than TCLD beam 2, correlation with heating observed
  • Two beams -> Main limitation is the 800 mm diameter ALICE chamber, better with lower e-cloud beams. Consolidation of ALICE chamber in LS3 is under consideration, smaller diameter chamber and smoother transition is suggested by the IWG.

Open and upcoming ECRs

  •         PS BFA09 (Hermann): recap from previous meetings (110th and 112th IWG). Two half moon plugs added at the tank flange to remove high-Q resonance at 1.8 GHz. Computation of power induced in the compensation filter gives about 2 W. It is recommended to dimension compensation filter resistor for 5 W.
  •        BGC beam 2 (Carlo): impedance assessment overview presented. Negligible impact on both longitudinal and transverse beam stability. Power loss maximum 10 W expected. The equipment owner should give a confirmation if this power loss is acceptable or need more considerations and power loss maps.

Wakis development (Clara)

Context & Motivation: Wakis is an open-source wakefield solver for beam-coupling impedance simulations. While it works well on simple geometries (cavities, bellows), it fails on complex geometries like the 6L2 LHC accelerator device, which has non-manifold (non-watertight) surfaces. Robust meshing improvements were identified as critical.

 Three New Meshing Methods Implemented: Three open-source Python (PyVista/VTK) algorithms were evaluated and integrated into Wakis: ray-casting (Select_interior_points), signed distance field (Compute_implicit_distance), and scanline rasterisation (Voxelize_rectilinear). The latter is the fastest, exploiting spatial coherence for efficient collision detection.

STL Geometry Cleaning Techniques: Retriangulation methods (decimation, subdivision, warping) were studied to improve STL input quality. Computation time scales significantly with subdivision level (up to ×227 in CPU time), so the right technique must be chosen per geometry type.

Benchmark Results: All three methods were tested on a resonant cavity and 6L2 sub-components. Voxelize_rectilinear performed best overall — it meshed the 6L2 in 38 seconds (vs. 4 hours with CST) and ran the solver in 13 minutes (vs. 10 hours with CST). However, field leakage through the fingers remains an issue requiring further attention.

Sub-Cell Smoothing (SCS) – Work in Progress: A sub-cell smoothing method was implemented to retrieve fractional material volumes within cells, producing smoother, more accurate mask boundaries. Initial results on the cavity and 6L2 are promising; final 6L2 simulation with SCS is still pending.

Next Steps: Complete the final 6L2 simulation using Sub-Cell Smoothing to validate full pipeline performance on complex non-manifold geometries.

There are minutes attached to this event. Show them.