What is in here: Figs Fit results by the detector (2000 data, average over 10 random seeds)

Transcription

1 What is in here: next page (just a reminder): the fitting function I used (the same as in 99: no pile-up subtraction, some parameters are fixed for later fit start times) Figs.1-12 R and other parameters vs fit start time (2000 data, one random sequence) for all data summed, first and second halves of the ring summed, odd and even detectors summed (by summed i mean the data were summed, not the results for each detector) Figs Results for R from 5-par fit (unacceptable until very late times, of course, but the results may be of interest) for both 1999 and 2000 data Figs Fit results by the detector (2000 data, average over 10 random seeds) Figs.20. Comparison of CBO parameters 1999 vs 2000

2 @ > F A > > > > Fitting Procedure minimization with MINUIT "! Fitting Function with 13 free parameters: $# %'& ( )+*-,/ :9;.=<?>'@ >BA pile-up: < $#C% & (D) EGF. 0 F H6 7D.I8J. KL8 F B>,M. N F % &PO Q QSR'T UWV X coherent betatron oscillation:,y. 0IZ Z. 8[Z % & Q R U]\ T X muon losses:,/. N_^M` % & O Q R Q Uba+c T X 3 parameters fixed after studies pile-up phase KL8 F = 0 pile-up enhancement due to fast rotation: N F = 0.72, d = 12.7 e s (simulation, PU-Sub) Data from 22 detectors fitted separately and combined

3 j " 2000 data: plots through page 12 are from 13-parameter fit, for one random sequence only 1. (y) R in ppm with an offset; (x) fit start time in e s dets 1-12 summed in red, dets summed in blue, all data in green dets 4,5 are gated on after 45 e s the difference at 80 fhg s is 10 ppm (i )

4 data: (y) R in ppm with an offset; (x) fit start time in e s odd dets summed in blue, even dets summed in red, all data in green

5 3. Asymmetry vs fit start time in e s (top) all data summed; (middle) first and second halves of the ring fitted separately; (bottom) odd and even detectors summed and fitted separately

6 4. Muon lifetime (e s) vs fit start time in e s (top) all data summed; (middle) first and second halves of the ring fitted separately; (bottom) odd and even detectors summed and fitted separately

7 5. G-2 phase (rad) vs fit start time in e s (top) all data summed; (middle) first and second halves of the ring fitted separately; (bottom) odd and even detectors summed and fitted separately

8 6. Pile-up fraction vs fit start time in e s (top) all data summed; (middle) first and second halves of the ring fitted separately; (bottom) odd and even detectors summed and fitted separately Pile-up phase (cut 2 GeV) from PUS in 2000 is (-77 k 21) mrad, in a good agreement with (-68 k 31) mrad in 1999

9 7. Pile-up asymmetry vs fit start time in e s (top) all data summed; (middle) first and second halves of the ring fitted separately; (bottom) odd and even detectors summed and fitted separately

10 8. The amplitude of CBO vs fit start time in e s (top) all data summed; (middle) first and second halves of the ring fitted; separately; (bottom) odd and even detectors summed and fitted separately

11 9. The phase of CBO (rad) vs fit start time in e s (top) all data summed; (middle) first and second halves of the ring fitted; separately; (bottom) odd and even detectors summed and fitted separately

12 10. CBO frequency (khz, fixed after 80 fhg s) vs fit start time in e s (top) all data summed; (middle) first and second halves of the ring fitted; separately; (bottom) odd and even detectors summed and fitted separately The frequency is about 5 khz lower than in 1999.

13 11. Muon loss lifetime (e s, GAUSSIAN form, fixed after 68 fhg s) vs fit start time in e s (top) all data summed; (middle) first and second halves of the ring fitted; separately; (bottom) odd and even detectors summed and fitted separately

14 12. for one random seed only; (top) all data summed; (middle) first and second halves of the ring fitted; separately; (bottom) odd and even detectors summed and fitted separately

15 13. 5-par fit to 2000 data, is not acceptable before e s (!) The jitter at early times is the phase pulling due to neglecting CBO, muon losses (top) R vs time for all data in green; dets 1-12 in blue; dets in red (bottom) all data in green; odd dets in red; even dets in blue

16 14. 5-par fit to 2000 data (top) KLl between the first and the second halves of the ring vs time : almost 20 ppm at 80 e s, much worse than from 13-par fit (fig.1) (bottom) KLl between the odd and even detectors vs time

17 15. 5-par fit to 1999 data, is not acceptable before e s (!) The jitter at early times is the phase pulling due to neglecting CBO, muon losses (top) all data in green; dets 1-12 in blue; dets in red (bottom) R vs time for all data in green; odd dets in red; even dets in blue

18 16. 5-par fit to 1999 data (top) KLl between the first and the second halves of the ring vs time : almost 20 ppm at 80 e s, much worse than from 13-par fit (fig.1) (bottom) KLl between the odd and even detectors vs time

19 data: Results of multi-parameter fit by the detector ( e s), det 25 is the data summed, results averaged over 10 random seeds (top) R (ppm) vs det; (middle) vs det; 9bottom) lifetime (e s) vs det.

20 data: other parameters vs detector (top) asymmetry, g-2 phase (rad); (middle) normalization constant, pile-up fraction; (bottom) pile-up asymmetry

21 data: other parameters vs detector (top) CBO amplitude and phase (rad); (middle) CBO frequency (khz) and lifetime (e s); (bottom) muon loss amplitude and lifetime (fixed at 31.5 e s for all detectors, gaussian form of the fit function)

22 20. CBO parameters 1999 (red) vs 2000 (blue): (top) CBO amplitudes and phases overlayed; (middle) CBO frequency (khz) and lifetime (e s) in 2000; (bottom) CBO frequency (khz) and lifetime (e s) in 1999