Proton Calorimetry/Meetings/2021/02/17: Difference between revisions

From PBTWiki
Jump to navigation Jump to search
Line 10: Line 10:
*FTDI script now writes raw data to single file (opening and closing file once) and performs 50Hz rolling average, updating single file .csv file with averaged values in pC.
*FTDI script now writes raw data to single file (opening and closing file once) and performs 50Hz rolling average, updating single file .csv file with averaged values in pC.
**Plotting and fitting routines now much simpler as they just need to access .csv file as fast as possible, with no hex to decimal conversion required.
**Plotting and fitting routines now much simpler as they just need to access .csv file as fast as possible, with no hex to decimal conversion required.
**Slowest part of FTDI scipt is the conversion of each acquisition to decimal for rolling average (roughly 185us instead of 170us). Will investigate if this can be parallelised, though performance gains are small considering that 5 modules will require minimum integration time of roughly 185us (more data processing will be required, however).
**Slowest part of FTDI scipt is the conversion of each acquisition to decimal for rolling average (roughly 185us instead of 170us). Will investigate if this can be parallelised, though performance gains are small considering that 5 modules will require minimum integration time of roughly 181us. The processing of the extra module's data will increase the conversion time.
**Photodiode charge uncertainty could be calculated from the standard deviation of the 50Hz average.
*Installed Bootcamp on personal laptop with Vivado etc.
*Installed Bootcamp on personal laptop with Vivado etc.
*Next major step is to investigate non-ROOT-based fit.
*Next major step is to investigate non-ROOT-based fit.

Revision as of 15:35, 17 February 2021

Minutes for UCL Proton Calorimetry Meetings, 17th February

Present

Simon Jolly, Raffaella Radogna, Saad Shaikh, Fern Pannell

Raffaella Radogna

Saad Shaikh

  • FTDI script now writes raw data to single file (opening and closing file once) and performs 50Hz rolling average, updating single file .csv file with averaged values in pC.
    • Plotting and fitting routines now much simpler as they just need to access .csv file as fast as possible, with no hex to decimal conversion required.
    • Slowest part of FTDI scipt is the conversion of each acquisition to decimal for rolling average (roughly 185us instead of 170us). Will investigate if this can be parallelised, though performance gains are small considering that 5 modules will require minimum integration time of roughly 181us. The processing of the extra module's data will increase the conversion time.
    • Photodiode charge uncertainty could be calculated from the standard deviation of the 50Hz average.
  • Installed Bootcamp on personal laptop with Vivado etc.
  • Next major step is to investigate non-ROOT-based fit.
  • Will work on HEP summer school poster and problem sheet.

Fern Pannell