GlueX Start Counter Meeting, July 2, 2015
GlueX Start Counter Meeting
Thursday, July 2, 2015
11:00 am EDT
CEBAF Center F326/327
- Review minutes from the May 21 meeting
- DAQ Tasks for Spring 2015 Commissioning Run
- Plots for upcoming reviews
- Action Item Review
- The BlueJeans Meeting ID is 556 286 544 .
- Connect to the Meeting
Talks can be deposited in the directory
/group/halld/www/halldweb/html/talks/2015 on the JLab CUE. This directory is accessible from the web at https://halldweb.jlab.org/talks/2015/ .
A recording of the meeting is available here.
- FIU: Werner Boeglin
- JLab: Mahmoud Kamel, Mark Ito (chair), Eric Pooser, Simon Taylor, Beni Zihlmann
- There are two methods for measuring the light propagation delay:
- Take events with two hits in different start counters and measure their time difference and compare it to the difference in light propagation distance. Eric has a new fitting procedure: first cut the data into slices in z, fit each of those and use the results to fit as a function of z. He gets 11 cm/ns in the straight section, 16 cm/ns at the bend, and 20 cm/ns in the nose. On the bench at FIU he measured 19 cm/ns in the nose (method 2, below).
- At FIU delays were measured as a function of position along the counters at fixed locations. Eric has redone the fitting procedure, using a method similar to that of method 1, automating several steps which were done by hand before.
- To compare the two methods calibration constants were generated for both. These have been put into a private CCDB and the PID library was modified to access the constants. Right now there is a problem with the beam-data-generated constants. Eric is looking into this now.
- When Eric looks at start counter time corrected for light propagation (using Simon's 15 cm/ns), for particle time-of-flight, and for time of interaction in target he gets widths of around 300 ps over most of the start counter, consistent with our design goal.
- We discussed a scheme where we simply measure time delay as a function of position directly to get a correction function or correction table to be interpolated and applied, without explicit reference to nose, bend, or straight. Although this is different way of parametrizing the correction, Eric pointed out that the framework for analyzing the data is the same and is already developed.
- From the FIU measurements there were measurable differences in the propagation between paddles. At present we do not have enough statistics to do a paddle-by-paddle calibration using beam data.
- Time-walk corrections have been revisited using the fitting procedure he described for the propagation delay. The parameters he gets are stable and uniform counter-to-counter. The constants are in CCDB now.
- Attenuation length measurement suffers from low statistics, not enough even to do an all-counters-together measurement. For now we will use data from the bench data measured at FIU.
Plots to Show at Reviews
We discussed some possibilities for plots illustrative of the status of the calibration. We settled on a representative histogram of start counter time vs. RF time, corrected for time-of-flight, light propagation in scintillator, and the position of the interaction in the target. Also a plot showing the resolution, obtained from the aforementioned histogram, as a function of position in the start counter.