Vertical Emittance Tuning at the Australian Synchrotron Light Source Rohan Dowd Presented by Eugene Tan
Overview • Machine overview • LOCO method and minimisation technique • Minimisation results and Measurements • Sextupole alignment measurements • Conclusion ICFA - Low Emittance Ring Workshop 3-5 October 2011
Australian Synchrotron Light Source Overview • Double bend lattice • 14 fold symmetric • Gradient dipoles • Corrector and skew quad coils on sextupoles. • Independent power supplies for all multipoles. • Storage Ring Parameters Energy 3 GeV Circumference 216 m RF Frequency 499.654 MHz Peak RF Voltage 3.0 MV Current 200 mA Betatron Tune (h/v) 13.3/5.2 Momentum Compaction 0.002 ε x (nominal) 10.4 nm∙rad ICFA - Low Emittance Ring Workshop 3-5 October 2011
LOCO method • LOCO – Linear Optics from Closed Orbits. • Adjusts the linear optics in the model to fit the real machine data • Model response matrix – Machine response matrix = Error • Minimise error by adjusting the model ‘fit parameters’ • Fit Parameters normally include: – BPM/Corrector gains and coupling – Corrector gains and coupling – Quadrupole strengths – Skew Quadrupole strengths • We fit skew quadrupole components in all multipole magnets in our model lattice to represent rolls and misalignments. ICFA - Low Emittance Ring Workshop 3-5 October 2011
LOCO Outputs Verified from attenuation measurements for the BPM. Outputs: • BPM Gains + Couplings • Skew components • Quad Strengths • Corrector gains/tilts From these outputs we can calibrate the model and calculate skew corrections needed to Fits to varying machine coupling states (0.01- adjust coupling 17% coupling) Consistent BPM/corrector coupling results for differing machine coupling settings – LOCO is not attributing the machine coupling to BPM/corrector coupling Fits to identical machine coupling (10 samples) ICFA - Low Emittance Ring Workshop 3-5 October 2011
Emittance Coupling minimisation • Emittance coupling ( ε y / ε x ) calculated from LOCO Calibrated model. • Minimisation algorithm used to adjust skew quads to desired emittance coupling. • Emittance coupling can be adjusted to arbitrary amounts with this method. Set Coupling LOCO Measured Calculated ε y (pm) Coupling 0.0% 0.01% 1.0 0.1% 0.12% 12.2 0.2% 0.23% 23.5 0.3% 0.33% 33.7 0.4% 0.43% 43.9 0.5% 0.54% 55.1 0.6% 0.64% 65.3 0.7% 0.74% 75.5 0.8% 0.84% 85.7 0.9% 0.92% 93.8 1.0% 1.04% 106.1 Phys. Rev. ST Accel Beams, 14, 012804 (2011)
Touschek Lifetime vs RF • By taking single bunch lifetime over extended period the Touschek component of the lifetime can be extracted. 2 1 Nr c D ( ) e max x , , 2 3 8 z max x y x u u 3 e ln( u ) 1 e D ( ) e du ( 3 ln( ) 2 ) du 2 2 u 2 u 2 di i i dt a b 1 a i be 0 i ( t ) 1 a b i a ( 1 e ) 0 ICFA - Low Emittance Ring Workshop 3-5 October 2011
Touschek Lifetime vs RF • Touschek component will also change with RF voltage. • 2.1% energy acceptance (measured) • Curve fit by varying ε y / ε x , other values fixed. • Blue curve fit corresponds to ε y = 1.24 pm Set ε y / ε x Fitted ε y / ε x ε y (pm) (%) (%) 0.01 0.012 ± 0.003 1.2 ± 0.3 0.06 0.043 ± 0.013 4.5 ± 1.3 0.10 0.093 ± 0.025 9.4 ± 2.6 ICFA - Low Emittance Ring Workshop 3-5 October 2011
Tune Crossing Results LOCO Model ε y (pm) η y Component (pm) Betatron Coupling Total Measured ε y Model – Measured Component (pm) (pm) (pm) 92.31 ± 2.77 2.24 ± 0.09 86.90 ± 2.65 89.55 ± 2.65 2.76 ± 3.83 76.30 ± 2.23 2.96 ± 0.12 76.50 ± 2.14 78.64 ± 2.14 2.34 ± 3.09 41.41 ± 1.24 1.73 ± 0.07 40.39 ± 1.84 42.23 ± 1.84 0.82 ± 2.22 22.75 ± 0.68 2.04 ± 0.08 19.69 ± 1.33 21.72 ± 1.33 1.03 ± 1.49 12.75 ± 0.38 2.14 ± 0.09 10.71± 1.02 12.85 ± 1.02 0.10 ± 1.09 7.55 ± 0.23 1.73 ± 0.07 5.41 ± 0.71 7.14 ± 0.71 0.41 ± 0.75 1.05 ± 0.03 0.73 ± 0.09 0.59 ± 0.30 1.32 ± 0.31 0.27 ± 0.31 ICFA - Low Emittance Ring Workshop 3-5 October 2011
How to improve? • With current correction scheme we are stuck at 1-2 pm. • Instead of correcting for misalignments, lets eliminate them. • Recent ring alignments have made the coupling worse! • Need to measure them independently – Beam based method ICFA - Low Emittance Ring Workshop 3-5 October 2011
Sextupole offsets • Shunt each sextupole magnet family to different strengths and take a response matrix at each point • Perform LOCO analysis and fit skew quadrupole terms to each sextupole. • Gradient of skew field vs sextupole field gives vertical offset. ICFA - Low Emittance Ring Workshop 3-5 October 2011
Sextupole Offset Results • Average offset: 70 microns • Systematic alignment effect? ICFA - Low Emittance Ring Workshop 3-5 October 2011
Cross Checks Applied BPM Offset Measure Mean Beam Difference from zero Offset +125 234.6 ± 10.6 128.7 ± 18.8 +75 167.7 ± 16.5 61.8 ± 22.6 0 105.9 ± 15.5 0 -75 33.1 ± 17.8 -72.8 ± 23.6 -125 -16.1 ± 21.6 -122 ± 26.6 Magnet Original Offset Applied Shim New Offset Delta offset ( μ m) ( μ m) ( μ m) ( μ m) Sector 9 SFB -108.4 ± 44.6 150 -249.3 ± 7.2 140.9 ± 45.2 Sector 11 SFB -56.7 ± 10.0 100 -120.4 ± 56.0 -63.4 ± 57.4 Sector 9 SDA -14.6 ± 9.9 100 -118.3 ± 8.3 -103.7 ± 14.1 Cross checks show that amplitude of offset is correct and individual magnets can be adjusted accurately. ICFA - Low Emittance Ring Workshop 3-5 October 2011
Sextupole Realignment – latest results Re-aligned section (girder #16-21) now has much lower offsets. Simulated minimal vertical emittance has reduced from 1pm to 0.7 pm. ICFA - Low Emittance Ring Workshop 3-5 October 2011
Conclusions • ε y of 1.3 pm.rad achieved through LOCO based minimisation. • We have developed tools to accurately measure sextupole misalignments and eliminate them. • Aim to have all sextupoles realigned by end of year and expect to reach < 1 pm emittance • Also working on an inexpensive way to increase available skew correctors ICFA - Low Emittance Ring Workshop 3-5 October 2011
Thank you ICFA - Low Emittance Ring Workshop 3-5 October 2011
Recommend
More recommend
