Project

General

Profile

Feature #4300

Investigate why events come in bursts (on a short time scale)

Added by Kurt Biery over 6 years ago. Updated over 5 years ago.

Status:
New
Priority:
Normal
Assignee:
-
Target version:
Start date:
07/07/2013
Due date:
% Done:

0%

Estimated time:
Duration:

Description

Boris noticed some time ago that events in the WH14N teststand come in bursts.

Alden Fan has also noticed this in the data taken with the full DAQ. Here is an email excerpt from Alden:

I noticed something in the DS50 data while developing DarkArt that I hope you can explain. I am reading event times using the trigger_time member of the V1495 fragment metadata. Histogramming the event times, I see that there are intervals of up to several seconds long where apparently no events were recorded. See attached plots. The x axes are in microseconds, and I plotted only the first 1000 events, so the horizontal range is on the order of tens of seconds. Do you know why the data comes in intervals? Luca suggested this is maybe the buffer unfilling when it is full.

Recently, Boris and I noticed that the bursty behavior seemed less pronounced at the WH14N teststand when we decreased the number of 1720 buffers from 64 to 16, but this was not a deliberate study of this effect.

We should look into why this is happening, and understand if a smoother pattern of accepting events is possible and/or desirable.

run3481_event_times.png (14.7 KB) run3481_event_times.png Kurt Biery, 07/07/2013 03:13 PM
run3481_event_times_zoom.png (13.6 KB) run3481_event_times_zoom.png Kurt Biery, 07/07/2013 03:13 PM

History

#2 Updated by Kurt Biery over 5 years ago

  • Target version set to KnownIssues


Also available in: Atom PDF