1. Polarized Proton Acceleration in the AGS with Two Helical Partial Snakes October 2, 2006 SPIN 2006 H. Huang, L.A. Ahrens, M. Bai, A. Bravar, K. Brown, E. D. Courant, C. Gardner, J.W. Glenn, A.U. Luccio, W.W. MacKay, V. Ptitsyn, T. Roser, S. Tepikian, N. Tsoupas, J. Wood, K. Yip, A. Zelenski, K. Zeno

2. Depolarizing Resonances in the AGS Imperfection Resonances arising from sampling of error fields, fields due to closed orbit errors, etc. G  =n (integer) G  =5,6,…45partial snake(s) Spin tune is given by cos  s =cos (  /2) cos G  s can not be an integer, avoided all imperfection resonances Vertical Intrinsic Resonances arise from sampling of focusing fields due to finite beam emittance. G  =kP  y Strong ones: G  =0+ y,12+ y,36  y AC dipole, strong partial snakes Weak ones: G  =24  y,48- y fast crossing speed, strong partial snakes Horizontal Intrinsic Resonances 1. horizontal non-vertical stable spin direction due to strong partial snake interaction with horizontal motion. 2. betatron motion coupled to the vertical betatron motion by coupling elements: solenoid, helical magnet. G  =k  x fast crossing speed, strong partial snakes Partial Snake Resonances strength proportional to nearby intrinsic resonance strength. G  =kP  m y, m>1 avoid the resonance tunes

3. Spin Tune and Fractional Vertical Tune With two helical magnets installed, the lattice is largely distorted at low energies. It took quite a lot efforts to set the vertical tune close to integer. Vertical tune is higher than 8.98 at all major intrinsic resonances. It is even as high as 8.99 at 36+.

4. Ramp Measurement RHIC injection energy was lowered by 1 unit of G  for better spin transmission. Note that there is not much data before G  =7.

5. Polarization as Function of Vertical Tunes Snake resonance effect is clearly seen.

6. 9 th Harmonic at 36+ Estimated imperfection resonance strengths with large sin 9 th harmonics. As can be seen, with 8mm sin 9 th harmonic amplitude, the resonance strength is comparable to the 16% partial snake in the AGS. With certain phase between them, the effect of partial snakes is canceled by the large orbit distortion at G  =45 (near 36+ y ). Measured polarization as function of the sine 9 th harmonic amplitude at 36+ y. The dashed line is to guide the eyes. The location of the polarization dip agrees with model prediction. (Mei)

7. Horizontal Polarization Profile Polarizatio n Target Position (mm) 15% Cold Snake 10% Cold Snake Target Position (mm) Normalize d Rate 15% Cold Snake 10% Cold Snake The horizontal resonance effect is measurable over the whole ramp. The following snake setup gave the best polarization: 10% cold snake, 5.9% warm snake.

8. 20022003200420052006 Intensity (10 11 ) 0.7 1.01.5 Polarization30%40%50% 65% AGS Polarization Evolution Slow ramp rate Normal ramp rate New warm snake Cold snake Booster Scraping Source polarization and intensity gradually improved over this period.

9. Horizontal Tune Near Integer The idea is to put horizontal tune near 9 (~8.95) while maintain vertical tune close to 9 (~8.98). Both tunes are within the spin tune gap. With the fractional part of the two tunes are so close, the coupling has to be corrected very well. The old solenoid snake will be ramped to overcome the coupling. Since the horizontal resonance strength are very weak, the horizontal tune does not need to be so as close to integer as vertical tune. The idea has been tested with beam. With added strength from nearby quad strings, we should be able to put the tunes into the gap as needed.

10. Horizontal Tune Near Integer (2)

11. Horizontal Tune Near Integer (3)

12. Summary 65% polarization with 1.5  10 11 intensity achieved with two partial snakes in the AGS. The following snake setup gave the best polarization: 10% cold snake, 5.9% warm snake. Added four compensation quads for warm snake worked very well. The lattice is easier to handle than last year without them. The intensity dependence is very benign with this setup. There is polarization loss of 5% due to each of vertical and horizontal resonances. The injection and extraction mismatch is about 1%. We expect 73% polarization with 82% input. There is still an relative 10% polarization loss unexplained. Move horizontal tune into the spin tune gap next run. Spin tracking effort is underway to understand the polarization loss. More polarization ramp measurements during next run.

13. Spare slides

14. AGS Stability RHIC injection measurements with very long measuring time. Averaged over 420 million events. Error bar for each bunch is 4.7%. Trying to look for AGS variation from shot to shot. Need more statistics.

15. pC CNI Polarimeter in the AGS ultra-thin Carbon ribbon (target) 5  g/cm 2 250  m wide ~32 cm Si strip detectors 12 vertical strips read-out with waveform digitizers beam direction beam 24 mm 12 mm pp pp left right

16. Ramp Measurement (2) With total of eight runs, the error bars are already quite small. If data from 45 degree detectors are included, they can be further reduced.