BCAL Reconstruction Meeting 2013-03-14
Teleconference Time: 11:00 a.m. EST
- ESNET (Number is 8542553) and EVO session (GlueX Calorimetry meeting room)
- Phone connection only upon request.
- +1-866-740-1260 : US+Canada
- +1-303-248-0285 : International
- then enter participant code: 3421244# (remember the "#").
- or www.readytalk.com (and code without the #)
Participant Direct Lines
- JLab Phone: in CC F326 is 757-269-6460 (usual room)
- JLab Phone in CC L207 is 757-269-7084
- Phone in the Regina Videoconference Suite is 306-585-4204
- Mar 7 Meeting
Do we need time-walk on FADC timing? (David)
- Do we have 12 bits or more in the FADC? (David)
- Show plots with timewalk correction applied. (Will)
- Will's Slide 8 spikes in spectrum?
- Will's yellow bumps in Slide 4?
- Mar 14 Meeting
- Slide 2 expand scale.
- Slide 3 add error bars
- Slide 4 Project on y-axis to compare fits (peak vs mean on Slide 5)
Time-walk (t-w) correction methodology. (Andrei)
- BCAL t_res and z_res info. (Zisis)
- Check PoissonD generator for discreteness and will examine quantization part of MCSMEAR. (Andrei)
- Moyal fit to Slide 2? (Zisis)
- Any other business
Attendees: David Lawrence (JLab), Will Levine (CMU), Andrei Semenov, Zisis Papandreou (Regina)
- Updates: Will showed updated results.
- Fitting to the peak vs mean can have a 0.5ns difference which is significant.
- Slide 2. Action Item: expand range to make sure tails at high end are not cut off. What is a good fit function to use? Gaussian over part of the range or Moyal over full range?
- Slide 3. The y-axis reflects the peak position. Will will assign errors to these points in the future.
- Slide 4. Action Item: project this plot on the y-axis and compare fit
- t-w correction needs to be done iteratively. Andrei will provide methodology by email. We need to look at t_diff between the two ends as well as t_mt (meantimer).
- David asks: 'What level of t-w correction to we need to achieve first order cleanup of extra clusters due to bad timing?' I.e., how good does t_res need to be? Can we get the MC in good enough form to help answer this?
- David: Chris Cuevas says that the Matt's algorithm (see his NIM article) has been implemented in the FPGA of the fADC and that the t-w will be mostly gone. This is in a readout mode on a few flash samples and we should investigate whether it can be tuned further. The integral in pulse mode gives the charge and time.
- Slide 6. Picket fence disappears once quantization step in MCSMEAR is turned off. Andrei will check PoissonD generator to see if its is indeed discrete and will examine that part of MCSMEAR.