Today's Progress 5. May. 2007
T0 TOF analysis by &pi+/- beam-triggered data : E570
T0 slewing correction by pi^- beam run
Here, we try the T0 slewing corection by &pi+ beam run, run 222/404, for each cycles. The results are confirmed by &pi- beam run, 246/515. By using the T0-T0 data, we cannot determine the xi-dependence. Therefore, we just perform the slewing correction by deltaE term. The xi term can be studied by T0-PA analysis, although, it is expected to be almost negligible.
Results
1st cycle results of T0 1st stage slewing correction. 1st stage correction is done by polynomial(up to 3rd) of 1/sqrt(phL*phR) of 2 layers for mean time difference.
| ROW ID | Layer 1 order | Layer 2 order | Layer 1 - fit regin | Layer 2 fit region | Gaussian &sigma for &pi+/- beam (psec) (run 222/246) | Gaussian &sigma for K+/- beam (psec) (run 225/220 ) |
|---|
| 1 | 3 | 3 | 0.051-0.104 | 0.031-0.109 | 79.83/79.43 | 46.23/42.08 |
| 2 | 3 | 2 | 0.051-0.119 | 0.031-0.094 | 78.58/78.71 | 46.43/47.45 |
| 3 | 3 | 3 | 0.051-0.109 | 0.036-0.179 | 98.86/100.41 | 68.07/68.91 |
| 4 | 3 | 2 | 0.056-0.129 | 0.031-0.094 | 77.87/79.98 | 46.56/47.61 |
| 5 | 3 | 3 | 0.056-0.104 | 0.031-0.099 | 79.89/80.41 | 52.67/58.22 |
2nd cycle results of T0 1st stage slewing correction.
| ROW ID | Layer 1 order | Layer 2 order | Layer 1 - fit regin | Layer 2 fit region | Gaussian &sigma for &pi+/- beam (psec) (run 404/515) | Gaussian &sigma for K+/- beam (psec) (run 403/417) |
|---|
| 1 | 2 | 3 | 0.046-0.094 | 0.031-0.119 | 80.77/89.56 | 46.02/44.84 |
| 2 | 3 | 3 | 0.051-0.119 | 0.031-0.089 | 78.79/83.63 | 45.69/47.67 |
| 3 | 3 | 3 | 0.056-0.109 | 0.036-0.164 | 93.23/106.9 | 56.08/63.70 |
| 4 | 3 | 2 | 0.051-0.124 | 0.031-0.099 | 81.87/86.42 | 48.42/52.73 |
| 5 | 3 | 3 | 0.056-0.104 | 0.031-0.099 | 78.08/79.92 | 61.87/64.51 |
Row-by-row 1/sqrt(phL*phR) VS T0time(2) - T0time(1) correlation after 1st stage slewing correction done above. A systematic tendency
that T0 mean times are small if 1/sqrt(phL*phR) being small (i.e. for large pulse height), is clearly seen even after the 1st stage correction, for all counters. This must be removed by adding additional term to the correction function, which are locally determined, and locally effective by 50~100 psec.
cycle1 results from run 222
cycle2 results from run 404
Second stage correction is activated only for 1/sqrt(phL*phR) < 0.1. The second stage correction function is prepared to correct the larger PH region, and defined as a parabola with its differential and value at 0.1 being 0 and 0 . The parabora is obtained by fitting the higher PH region for pi-beam events AFTER 1st stage global correction. The resulting behaviour of the time difference with respect to the 1/sqrt(phL*phR) is shown below, row-by-row. Now, we hardly see the deviation of the time difference from the constant.
1st cycle results of T0 2nd stage slewing correction. 2nd stage correction is done by parabora of (1/sqrt(phL*phR)-0.1) on 2 layers for mean time difference.
| ROW ID | Gaussian &sigma for &pi+/- beam (psec) (run 222/246) | Gaussian &sigma for K+/- beam (psec) (run 225/220) |
|---|
| 1 | 79.33 /78.55 | 47.97/42.01 |
| 2 | 78.60 /78.43 | 47.74/50.53 |
| 3 | 98.81 /100.4 | 67.19/69.76 |
| 4 | 77.19 /79.22 | 47.98/48.36 |
| 5 | 79.98 /81.05 | 58.22/65.32 |
2nd cycle results of T0 2nd stage slewing correction.
| ROW ID | Gaussian &sigma for &pi+/- beam (psec) (run 404/515) | Gaussian &sigma for K+/- beam (psec) (run 403/417) |
|---|
| 1 | 82.36 /88.17 | 49.21/46.39 |
| 2 | 78.18 /82.41 | 48.03/50.21 |
| 3 | 92.85 /107.4 | 58.88/65.20 |
| 4 | 81.49 /86.37 | 49.65/52.48 |
| 5 | 77.81 /79.60 | 63.36/65.96 |
Note that it is very difficult to see the improvement due to the 2nd stage correction by the width/mean of T0 2-1 time difference, because the effects kill each other for Kaon, for which two pulse heights are strongly correlated! But this correction would be meaningful for T0-PA/T0-NC analysis, to avoid expected deviation of the time origin and/or the aggravation of the resolution.