LCLS Linac Tuning Simulations P Emma LCLS Week

LCLS Linac Tuning Simulations P. Emma LCLS Week April 5, 2005 LCLS Thanks to M. Borland for Elegant code changes in support of these studies 5 Apr. 2005 LCLS Week 1 P. Emma@SLAC. Stanford. edu

Description of the Study Start with low-charge configuration (0. 2 n. C, no CSR) Use elegant to automate tuning; use only ‘real’ diagnostics Add large random errors to linac systems (e. g. , magnets, RF, beam) Assume rough corrections already made (see LCLS Commissioning Workshop, Sep. 2004) Track through linac many times, each step simulating one particular correction (e. g. , b-matching or RF phasing) Use correction devices already built into design (e. g. , BC 2 correction quads, trajectory controls) Evaluate final beam quality, correction convergence, dynamic range, problem areas, etc. 5 Apr. 2005 LCLS Week 2 P. Emma@SLAC. Stanford. edu

Linac System Errors rms errors (Gaussian, 3 -s cutoff) x and y misalignments z misalignments Quads relative gradient errors roll angle errors relative field errors Bends anomalous field gradients (BC’s) BPMs x and y misalignments RF strucs. phase errors (static) relative voltage errors (static) e- beam random charge error initial beta mismatch in x and y Element 5 Apr. 2005 LCLS Week 3 value 300 5 0. 5 2 0. 5 0. 3 3 300 2 1 10 2. 0 unit mm mm % mrad % tol. mm mm deg % % P. Emma@SLAC. Stanford. edu z

Sequential Corrections Applied in Simulations Steer entire machine (initially ignore BPMs in DL 1, BC 2) Set energy and minimize spread in DL 1 (OTR 4, BPM 13 and L 0 b phase/voltage) Set energy and bunch length after BC 1 (BPMS 11, transverse deflector. BC 2 -off, L 1 phase/voltage) Set energy and bunch length after BC 2 (BPMS 21, transverse deflector, L 2 phase/voltage) Beta-match injector (OTR 1 -3, QA 01 -2, QE 01 -4) Correct trajectory again to compensate quad steering Minimize gey after BC 1 with trajectory in QM 12 (WS 11 -13, YC 21201) Minimize gey after BC 2 with L 2 trajectory (WS 21 -24, YCM 12, YC 21601) Minimize gex after BC 2 with L 2 trajectory (WS 21 -24, XCM 11, XCM 13) Correct hx and hx after BC 2 (WS 044 -544, CQ 21, CQ 22) Minimize gex at end of LTU with ‘bumps’ (WS 31 -4, XC 460026, XC 6) Match b after BC 2 & LTU (WS 044 -544, Q 24701, WS 31 -4, Q 6, QEM 3) 5 Apr. 2005 LCLS Week 4 P. Emma@SLAC. Stanford. edu

Trajectory After First Steering DL 1 BC 2 5 Apr. 2005 LCLS Week 5 P. Emma@SLAC. Stanford. edu

Projected Emittance After First Steering Q = 0. 2 n. C 5 Apr. 2005 LCLS Week 6 P. Emma@SLAC. Stanford. edu

b-mismatch amplitude after first steering zx = 3. 3 ! GOAL = 1 5 Apr. 2005 LCLS Week 7 P. Emma@SLAC. Stanford. edu

Final Long. Phase Space After First Steering sz = 8. 6 mm (0. 2 -n. C design is 8. 0 mm) (NOT BAD) DE/E 0 = -0. 3% 5 Apr. 2005 LCLS Week 8 P. Emma@SLAC. Stanford. edu

Set Energy and Minimize Spread in DL 1 (OTR 4) minimize OTR 4 x-spot size while scanning L 0 b RF phase long. DL 1 phase space before long. DL 1 phase space after L 0 L 1 X L 2 OTR 4 after OTR 4 before 5 Apr. 2005 LCLS Week L 3 9 P. Emma@SLAC. Stanford. edu

Set energy and bunch length after BC 1 post-BC 1 long. phase space before post-BC 1 long. phase space after sz = 43 mm sz = 55 mm L 0 L 1 X L 2 L 3 design: 60 mm Sec-25 transverse deflector (BC 2 off) meas. post-BC 1 bunch length BC 1 BPM meas. BC 1 rel. energy L 1 RF phase adjust bunch length to ~60 mm L 1 RF amplitude adjust BC 1 BPM to x = 0 0. 25 mm 5 Apr. 2005 LCLS Week 10 P. Emma@SLAC. Stanford. edu

Set energy and bunch length after BC 2 post-BC 2 long. phase space before post-BC 2 long. phase space after sz = 12 mm sz = 8. 7 mm DE/E 0 = +0. 8% L 0 L 1 L 2 X L 3 design: 8. 0 mm sec-25 transverse deflector (BC 2 on) meas. post-BC 2 bunch length BC 2 BPM meas. BC 2 energy L 2 RF phase adjust bunch length to 8. 0 mm L 2 RF amplitude adjust BC 2 BPM to x = 0 0. 25 mm 5 Apr. 2005 LCLS Week 11 P. Emma@SLAC. Stanford. edu

Trajectory After BC 1/BC 2 Energy Corrections 5 Apr. 2005 LCLS Week 12 P. Emma@SLAC. Stanford. edu

b-match at OTR 2 in the Injector L 0 b QA 01 -02, QE 01 -04 OTR 3 OTR 2 OTR 1 L 0 L 1 5 Apr. 2005 LCLS Week X L 2 L 3 13 P. Emma@SLAC. Stanford. edu

b-match at OTR 2 with quads: QA 01 -2, QE 01 -4 sx = 67. 5 mm sy = after correction L 0 before correction 57. 9 mm L 1 X L 2 76. 1 mm 5 Apr. 2005 LCLS Week L 3 sy = 61. 2 mm 14 P. Emma@SLAC. Stanford. edu

Measure Emittance on OTR 11 -13 after BC 1 OTR 13 OTR 12 OTR 11 OTR 2 L 0 L 1 5 Apr. 2005 LCLS Week X L 2 L 3 15 P. Emma@SLAC. Stanford. edu

Vertical offset in QM 12 can be trouble y OTR 12 x gey gex L 0 L 1 5 Apr. 2005 LCLS Week X L 2 L 3 16 P. Emma@SLAC. Stanford. edu

Minimize gey on OTR 12 using YC 21201 x y gex gey L 0 L 1 5 Apr. 2005 LCLS Week X L 2 L 3 17 P. Emma@SLAC. Stanford. edu

BC 2 Minimize gey at WS 21 with YCM 12 & YC 21601 gex gey L 0 L 1 5 Apr. 2005 LCLS Week X L 2 L 3 18 P. Emma@SLAC. Stanford. edu

BC 2 Now Repeat for gex on WS 21 (XCM 11 & XCM 13) gex gey L 0 L 1 5 Apr. 2005 LCLS Week X L 2 L 3 19 P. Emma@SLAC. Stanford. edu

BC 2 Correct Dispersion in BC 2 (WS 044 and CQ 21, CQ 22) gey gex L 0 L 1 5 Apr. 2005 LCLS Week X L 2 L 3 20 P. Emma@SLAC. Stanford. edu

Tweak LTU traj. to minimize gex on WS 31 -4 BC 2 XC 460026 XC 6 gey gex L 0 L 1 5 Apr. 2005 LCLS Week X L 2 L 3 21 P. Emma@SLAC. Stanford. edu

b-match needs correction after BC 2 and in LTU Wow! (zx = 3. 3) GOAL = 1 5 Apr. 2005 LCLS Week 22 P. Emma@SLAC. Stanford. edu

After b-matching at WS 044 and WS 31 Q 24701 Q 6 QEM 3 zx= 4 ? z 1 hand tweaked, much like real machine 5 Apr. 2005 LCLS Week 23 P. Emma@SLAC. Stanford. edu

Emittance After Final b-matching (no CSR) gey= 0. 91 mm gex = 0. 88 mm 5 Apr. 2005 LCLS Week 24 P. Emma@SLAC. Stanford. edu

Final Longitudinal Phase Space (no CSR) DESIGN TUNEUP 2 k. A 5 Apr. 2005 LCLS Week 25 P. Emma@SLAC. Stanford. edu

Final Slice Emittance (no CSR) limit at 0. 2 n. C, 2 k. A Slice emittance originates from Parmela run at 0. 2 n. C, 35 A (C. Limborg) 5 Apr. 2005 LCLS Week 26 P. Emma@SLAC. Stanford. edu

Final Slice Energy Spread (no CSR) limit at 0. 2 n. C, 2 k. A Slice energy spread originates from laser-heater (set a bit low here) 5 Apr. 2005 LCLS Week 27 P. Emma@SLAC. Stanford. edu

Summary Tuning results look encouraging (at 0. 2 n. C) Steering (Dy) after BC 1 can be very important Setting energy accurately in BC 2 needs work Beta-matching through LTU may benefit from additional e-diagnostics between DL 2 bend pairs Limited resolution of diagnostics not included (they must meet specifications) BC 2 off to meas. BC 1 sz is clumsy (3 rd T-cav? ) Should add CSR (but very slow) LCLS 5 Apr. 2005 LCLS Week 28 P. Emma@SLAC. Stanford. edu