1 / 47

r esults from measurement and simulation

About LW & Astra Simulations of T he Pitz Gun. r esults from measurement and simulation. m ethods ( LW , PIC, Astra ) and setup. c omparison LW  Astra. m ore a nalysis of LW r esults. about Astra results: sensitivity to numerical p arameters.

toan
Download Presentation

r esults from measurement and simulation

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. About LW & Astra Simulations of The Pitz Gun results from measurement and simulation methods (LW, PIC, Astra) and setup comparison LW Astra more analysis of LWresults about Astra results: sensitivity to numerical parameters about approaches: some analytical calculations second inspection summary

  2. results from measurement and simulation see: DESY/TEMF Meeting - Status 2011 http://www.desy.de/xfel-beam/data/talks/files/03-Gjonaj_Erion_DESY_16.12.2011_new.pdf

  3. methods (LW, PIC, Astra) and setup LW = LienardWichert exact solution of Maxwell problem (based on retarded trajectory) no spatial mesh numeric integration of EoM; fixed time step higher order PIC;PIC = particle in cell exact numerical approach for Maxwell problem spatial mesh & time step numeric integration of EoM; fixed time step extensive convergence test in 2010 carefull comparison in 2011  simulation of PITZ injector for z <= 5 cm Astra approximation based on uniform motion rz - Poisson approach; spatial mesh numeric integration of EoM; variable time step

  4. methods (LW, PIC, Astra) and setup setup bunch charge = 1 nC bunch length = 21.5 psec (Lt=0.0215 rt=0.002) rms-radius = (radius/2) = 0.2 … 0.6 mm MaxE = 60.58 MV/m phi = 223.386 deg (or auto phase = -1.404 deg) E0 = 41.6 MV/m estimated SC limitation (DC field and planar diode) Rrms/mm = 0.2 0.3 0.4 0.5 QSC-limit/nC = 0.185 0.417 0.741 1.157

  5. comparison LW  Astra

  6. comparison LW Astra rms laser spot size = 0.6 mm, slice properties LW norm. (hor;vert) emittance / m c0pz / eV rms{c0pz}/ eV Astra see also: s2e-seminar 2011-Feb-07 http://www.desy.de/xfel-beam/data/talks/files/2011_02_s2e_Gun_TEMF_Auswertung.pdf (z-mean(z))/m (z-mean(z))/m (z-mean(z))/m

  7. comparison LW Astra rms laser spot size = 0.6 mm

  8. comparison LW Astra rms laser spot size = 0.6 mm

  9. comparison LW Astra rms laser spot size = 0.4 mm

  10. comparison LW Astra rms laser spot size = 0.3 mm

  11. comparison LW Astra rms laser spot size = 0.2 mm

  12. more analysis of LWresults

  13. more analysis of LWresults rms laser spot size = 0.6 mm r/m z/m

  14. more analysis of LWresults rms laser spot size = 0.5 mm r/m z/m

  15. more analysis of LWresults rms laser spot size = 0.4 mm r/m z/m

  16. more analysis of LWresults rms laser spot size = 0.3 mm (0.5MP; dt=0.1psec) r/m aliasing z/m

  17. more analysis of LWresults rms laser spot size = 0.2 mm r/m z/m

  18. 0.2mm; more analysis of LWresults micro modulation 0.3mm; r/m 0.4mm; 0.5mm; 0.6mm; z/m

  19. more analysis of LWresults rms laser spot size = 0.3 mm simulation with more particles r/m 0.5M 1M z/m

  20. more analysis of LWresults rms laser spot size = 0.3 mm; calc. with different time step 1mm/24=42um; r/m dt = 0.10 psec dt = 0.05 psec 0.2mm/9 =22.2um; pz z/m

  21. 0.2 mm more analysis of LWresults  this bunch needs more compression ! 0.3 mm, half time step 0.3 mm r/m 0.4 mm  dt = 0.05 psec  dt = 0.10 psec “micro bunching” vs. injection time color frequency = 2x 0.1psec 0.5 mm 0.6 mm z/m

  22. 0.2 mm, <z> = 10 mm 0.3 mm, <z> = 10 mm 0.6 mm, <z> = 10 mm r/m <z> = 20 mm <z> = 20 mm more analysis of LWresults <z> = 30 mm <z> = 30 mm different longitudinal position <z> = 40 mm 50 mm <z> = 50 mm z/m

  23. more analysis of LWresults rms laser spot size = 0.3 mm; slice properties 500k, 0.1psec norm. (hor;vert) emittance / m c0pz / eV rms{c0pz}/ eV (z-mean(z))/m (z-mean(z))/m (z-mean(z))/m

  24. more analysis of LWresults rms laser spot size = 0.3 mm; slice properties 1M, 0.1psec norm. (hor;vert) emittance / m c0pz / eV rms{c0pz}/ eV (z-mean(z))/m (z-mean(z))/m (z-mean(z))/m

  25. more analysis of LWresults rms laser spot size = 0.3 mm; slice properties 500k, 0.05psec norm. (hor;vert) emittance / m c0pz / eV rms{c0pz}/ eV (z-mean(z))/m (z-mean(z))/m (z-mean(z))/m

  26. about Astraresults

  27. about Astra results: sensitivity to numerical parameters rms laser spot size = 0.6 mm no SC limitation is it possible to provoke numerical microbunching? (fine mesh, big time step) no: Astra is too clever; during emission it does not use the user defined time step rms laser spot size = 0.2 mm strong SC limitation: late emission is pulsing Hammersley random mesh lines time step

  28. about Astra results: sensitivity to numerical parameters rms laser spot size = 0.2 mm strong SC limitation: late emission is pulsing r/m pitzTT 100x40 pitzTT 100x40 Nz Nr emitted particles 40.5% z/m

  29. about Astra results: sensitivity to numerical parameters rms laser spot size = 0.2 mm strong SC limitation: late emission is pulsing r/m pitzTT 100x40 pitzN 100x40 Hammersley emitted particles 40.5% random emitted particles 40.5% pulsing (multiple fronts) is no artifact of pseudo random gen. z/m

  30. about Astra results: sensitivity to numerical parameters rms laser spot size = 0.2 mm strong SC limitation: late emission is pulsing r/m pitzTF 200x40 pitzTFF 1000x40 Nz Nr Nz Nr Hammersley emitted particles 41.9% Hammersley emitted particles 45.4% z/m

  31. about Astra results: sensitivity to numerical parameters rms laser spot size = 0.2 mm strong SC limitation: late emission is pulsing r/m not quite the same time pitzTT 100x40 pitzTTfixed 100x40 Nz Nr Nz Nr Hammersley emitted particles 40.5% Hammersley emitted particles 39.1% z/m

  32. about Astra results: sensitivity to numerical parameters rms laser spot size = 0.2 mm strong SC limitation: late emission is pulsing r/m pitzTFFF still running: pitzTFFF, 10000x40 pitzTFFR, 1000x200 pitzTFF 1000x40 43.8%  45.4% pitzTFFR 1000x200 43.2% z/m

  33. about Astra results: sensitivity to numerical parameters pitzTFFF10000x40 r/m q/Q q/Q = 0.05 10 mesh lines z/m z/m

  34. about Astra results: sensitivity to numerical parameters result is extremely sensitive on initialconditions and bothmesh settings no initial energy spread caused by r-mesh 0.55eV initial energy spread

  35. about approaches

  36. about approaches: some analytical calculations the planar diode exact solution of Maxwell problem no explicit appearance of retarded time exact solution of EoM with  = slice index, t= ejection time of slice , , external field a three dimensional driven problem in the following: three different approaches to calculate comparison for

  37. about approaches: some analytical calculations three approaches “M” exact Maxwell solution “SUM” solution: slices in individual uniform motion with and z(v,t) the slice velocity “BUM” solution: bunch in uniform motion with and “com” = center of mass

  38. about approaches: some analytical calculations injection (z = 0) bunch charge accelerating field time of injection (rectangular) (rms = 0.5 mm) (rms = 0.3 mm) … planar diode __ SUM __ BUM x_ M space charge limitation

  39. about approaches: some analytical calculations injection (z = 0) (rms = 0.2 mm) … planar diode __ SUM __ BUM x_ M space charge limitation M calculates stronger field reduction than UM & BUM !

  40. about approaches: some analytical calculations after injection (t>Ti) bunch charge accelerating field time of injection (rectangular) (rms = 0.5 mm) … planar diode __ SUM __ BUM __ M

  41. about approaches: some analytical calculations after injection (t>Ti) … planar diode __ SUM __ BUM __ M

  42. second inspection

  43. second inspection lost particles vs. accelerated particles (@ injection) 0.2mm: ASTRA, lost particles = blue pitzTT 100x40  injection radius / m  injection radius / m injection time / sec  injection time / sec  integratedcharge accelerated particles lost particles integratedcharge accelerated particles lost particles injection time / sec  injection time / sec 

  44. second inspection lost particles vs. accelerated particles (@ injection) 0.2mm: ASTRA, lost particles = blue pitzTT10000x40 pitzTT 100x40  injection radius / m not calc. injection time / sec  integratedcharge accelerated particles lost particles injection time / sec 

  45. second inspection lost particles vs. accelerated particles (@ injection) pitzTT 100x40  injection radius / m  injection radius / m injection time / sec  injection time / sec  integratedcharge accelerated particles lost particles injection time / sec 

  46. second inspection correct numbers: q = 1nC Ti = 21.5 psecEacc=41.61 MV/m 0.2mm: ASTRA, lost particles = blue (rms = 0.2 mm) pitzTT 100x40  injection radius / m  injection radius / m injection time / sec  … planar diode __ SUM __ BUM x_ M M-shielding should be stronger than BUM-shielding!

  47. summary SC limited emission: differences between measurement and ASTRA ASTRA incomplete model predicts strong SC limited emission SC limited emissions needs extreme mesh resolution (usually not fulfilled at end of emission) very sensitive to initial conditions (is it real physics?) SC shielding seems to agree with analytical model LW should include all effects has been benchmarked with PIC or vice versa (but: same tracker) less SC shielding: seems to agree with measurement problems with boundary condition at injection (mirror-point-charge) micro modulation: related to time step and time of injection analytical model (Maxwell  Poisson) predicts stronger shielding measurement …?!

More Related