Inter-Surf Calibration
The best way to do this would be from timing comparisons between pulses seen in multiple channels on different SURFs. This can be done using some fo the testing data that was taken at Willy field, but the real test will be from reconstructing calibration pulser events.
For now it has been tested using the clock channels (which is somewhat less than robust as the clock channels are also used to ccalibrate the event by event jitter). Using the interpolation method it is possible to check the calibration, by looking at transitiosn other than the first negative transition (which is used in the interpolation method to remove jitter).
Results
Summary
Although it is hard to draw conclusions from this data the fit method appears to perform marginally better with seemingly narrower chip distributions that are mor closely clustered around zero. Until this test is repeated using external data, all conclusions can only be preliminary. The following table comes from the 2nd positive data.
| Calibration Name | RMS of Mean | Mean of RMS |
|---|---|---|
| kVTFullJWPlusFastClockZero | 196 ps | 111 ps |
| kVTFullJWPlusClockZero | 107 ps | 100 ps |
The RMS of Mean column is intended as an estimator on systematic error (i.e this is an estimate of the fixed offset not taken out by the calibration), the Mean of RMS is intended as an estimator of statistical error (i.e the event to event stability of the calibration).
Copyright © 2006 Ryan Nichol, (last modified 07 Jun 2007)