ILC 07 May 30-June 3, 2007 N.Solyak 1 MULTIPACTORING SIMULATIONS IN CAVITIES AND HOM COUPLERS Nikolay Solyak, Ivan Gonin, Jiajian Li Fermilab

ILC 07 May 30-June 3, 2007 N.Solyak 2 Analytical Model MatCad simulations. Simple 1-D RF fields. High order resonances. Impact energy and SYE file for simulation of the enhanced function Quasi 2D model for cavity field near cell equator (V.Shemelin) MultiPac 2.1 Package Axis-symmetric RF components, such as coupler, cavity, windows 2D electromagnetic field distribution in cavities and couplers Graphical MATLAB user interface, easy to use Analyst (Omega3P) Powerful tool for MP phenomena investigation for complex RF devices in real 3D fields High order FE for electromagnetic field simulation Paralleling algorithm allows to run a few million FE at the reasonable time Simulation Tools

ILC 07 May 30-June 3, 2007 N.Solyak 3 Codes benchmarking - 3 rd harm. cavity Transmitted /Reflected power. Temperature sensors on HOM couplers cavity #2: Vertical test Results Strong MP in both HOM couplers Both HOMs coupler fractured after high power vertical tests. No MP in cavity (0-mode) Surface magnetic field in “0”-mode.

ILC 07 May 30-June 3, 2007 N.Solyak 4 MP in notch gap 0.6 mm 3D simulation Omega 3P(Analyst ) MP in 2 mm Leg-wall gap 3 rd Harmonic Cavity: MP in old HOM design Cavity No.3. Results of vertical test: MP observed at Eacc~0.7MV/m (Q drop). Quench at Eacc~14MV/m. Second resonance frequency of HOM was tuned higher than designed value.

ILC 07 May 30-June 3, 2007 N.Solyak 5 Cavity No.3: Results of the Cold Test #3 QvsT at Pi-mode. F= MHz. Cavity baked 150C 48 h. Rres=25 nOhms. At Eacc=12 MV/m quench in the bottom HOM. Q vs. max. surface H-field for some modes. Quench in HOM for Pi and 8/9Pi modes and in the cavity cells for lower modes. T-sensor between welds of the bottom HOM go up to maximum 6 K. RF fall time 20 msec. MP 12MV/m

ILC 07 May 30-June 3, 2007 N.Solyak 6 New Designs of the HOM coupler F 2 =4400 MHz 1-post design F 2 =4080 MHz F 2 =4280 MHz Trimmed Initial design -1a Initial design No.1 HOM-1post HOM1-original 1 1 HOM1-modif cavity Hpk/H Epk/E HOM field - Comparison No MP up-to 23 MV/m in modified design

ILC 07 May 30-June 3, 2007 N.Solyak 7 Egap ~ 3.8 V/m; MP in TESLA cavity and HOM coupler FE mesh, ~1 million second order elements E max ~ 48.1 V/m  = E gap / E acc ~ mm Enhancement function for gap 1.73mm and 1.3GHz E MP_analytical ~ 200 kV/m E acc = E MP /  ~ 1.25 MV/m Analyst results E max ~ 3.8 V/m HOM coupler 1-D 3-D Test: Q vs. E acc MP in the gap

ILC 07 May 30-June 3, 2007 N.Solyak 8 MP in TESLA cavity HOM coupler (cont.) E gap ~ ( ) V/m;  = E gap / E acc ~ ( ) E acc = E MP_analytical /  ~ 28 – 45 MV/m E>2V/m Enhancement Counter Function for gap 8.2mm and F=1.3GHz in Analytical MP model MP_analytical ~ kV/m MP in gap 8.2mm

ILC 07 May 30-June 3, 2007 N.Solyak 9 MP in TESLA cavity HOM coupler (cont.) Analyst results

ILC 07 May 30-June 3, 2007 N.Solyak 10 MP in TESLA cavity CF, Impact Energy and ECF. No MP. Small activity at E acc ~24 MV/m. Resonance trajectories near equator Eacc~24 MV/m

ILC 07 May 30-June 3, 2007 N.Solyak 11 FE mesh and E&H Fields distribution in LL mid-cell MP in Low Loss cavity

ILC 07 May 30-June 3, 2007 N.Solyak 12 MP in LL cavity (cont.) Start points for MP calculations (top) and secondary yield plot (below) Counter (CF) (top) and Enhanced Counter Function (ECF) (below). No MP because ECF < 0.001, E peak /E acc ~2.36 ~30 MV/m

ILC 07 May 30-June 3, 2007 N.Solyak 13 MP in LL cavity (cont.) At the equator resonance trajectories was found, but impact energy is too small to create enough secondary electrons Distance map

ILC 07 May 30-June 3, 2007 N.Solyak 14 FE mesh and E&M Fields distribution in RE mid-cell MP in Re-entrant cavity

ILC 07 May 30-June 3, 2007 N.Solyak 15 MP in Re-entrant cavity (cont) Start points for MP calculations (top) and secondary yield plot (below) CF (top) and ECF (below). No MP because ECF < 0.001, E peak /E acc ~ MV/m

ILC 07 May 30-June 3, 2007 N.Solyak 16 MP in Re-entrant cavity (cont) At the equator resonance trajectories was found, but impact energy is too small to create enough secondary electrons Distance map

ILC 07 May 30-June 3, 2007 N.Solyak 17 MP in ICHIRO cavity (2D) Start points for MP calculations (top right), Counter Function/Impact Energy and Enhancement CF (bottom) MP levels (E peak /E acc ~2.36): Resonance order: E-surface pk field: MV/m E acc : 16; 18.6; 19.5; 24; 30; 33 MV/m E&H fields in ICHIRO cavity

ILC 07 May 30-June 3, 2007 N.Solyak 18 MP in ICHIRO cavity (2D)

Global Design Effort 19 E max ~ 285 V/m E ~ 5V/m or E max FE mesh ~200K elements Resonance trajectories: E acc ~28.5MV/m MP in ICHIRO cavity (3D) Electric Fields

ILC 07 May 30-June 3, 2007 N.Solyak 20 Ichiro#1 w/ HOM MP in ICHIRO cavity (3D) cont. Ichiro#0 (w/o HOM)

ILC 07 May 30-June 3, 2007 N.Solyak 21 Ez vs z for β=0.61 cavity in 2D (Omega2P) SC SNS cavities β=0.61 and β=0.81 have end tubes with different diameters. β= mm and 130mm β= mm and 140mm 2D simulations of cavities show the different E-field strength in planes of HOM coupler: β = % and 11.3% of E max β = % and 9.2% of E max Therefore MP activity in HOM coupler can be predicted in coupler placed at the larger diameter end-tube, where the field is higher. SNS β=0.81 cavity

ILC 07 May 30-June 3, 2007 N.Solyak 22 Analytical simulations show the possibility of multipacting in 4 and 5.3mm gaps between HOM coupler cylindrical body and flat surface of antenna leg. Analytical simulations for MP and Enhanced Counter Function (ECF) after 20 impacts in 4mm (left) and 5.3mm (right) gaps for F=805MHz and n=1. Calculation shows: MP_analytical =~( ) kV/m for 4 mm and ~( ) kV/m for 5.3 mm 4mm 5.3mm Gap = 4mm Gap = 5.3 mm SNS β=0.81 cavity (cont.)

ILC 07 May 30-June 3, 2007 N.Solyak 23 Electric field distribution FE mesh, ~1 million second order elements E max ~ 26.6 V/m E gap ~ ( ) V/m;  = E gap / E acc ~( ) E acc = E MP_analytical /  ~ mV/m zoom E-field in HOM Ratios: E max / E acc ~ 2.19 E gap ~ 0.5 V/m; E acc = E MP_analytical /  ~ 5.1MV/m 5.3mm  = E gap / E acc ~ E>0.75V/m E>0.25V/m SNS β=0.81 cavity (cont.) Fields in Cavity and HOM coupler

ILC 07 May 30-June 3, 2007 N.Solyak 24 Enhanced Counter Function and plots of resonance trajectories after 20 impacts in HOM coupler. ANALYST shows two ranges of multipacting, first one around 5-6MV/m is strong and narrow and second have one order less enhanced counter values but is very wide. SNS β=0.81 cavity (cont.) E acc = 5.1 MV/m - analytical model E acc =(10-26) MV/m - analytical model

ILC 07 May 30-June 3, 2007 N.Solyak 25 Experimental results on a high-β cavity showing typical multipacting behavior. Black circles show the Q and the blue triangles shown the radiation level measured outside the dewar. “The analysis of vertical test data on high-β cavities shows that the onset occurred from 9.5 to 18.6 MV/m with a mean of 11.7 MV/m, and a standard deviation of 1.8 MV/m. Onset of multipacting was defined as the gradient at which a sudden drop in Q-value was first observed. Typical results are shown in Fig. below.” SNS β=0.81 cavity (cont.)

ILC 07 May 30-June 3, 2007 N.Solyak 26 Summary Develop Tools for 1-D, 2-D and 3-D MP simulations in complex SRF cavities and couplers Programs are benchmarked using experimental data for 3 rd harmonic cavity MP was simulated for TESLA-ILC, LL, Reentrant, ICHIRO and SNS cavities.