What Does the CRT Do?

The ProtoDUNE -SP Cosmic Ray Tagger (CRT) is a set of 32 scintillator modules mounted just upstream and downstream of the TPC. They were originally designed to veto muons for the Double Chooz experiment. The CRT sends trigger signals to the DAQ that can be used to read out the TPC when there are muon-like hits both upstream and downstream of the detector. Also, the CRT board reader records which strips in each module were hit to deliver something like 2.5cm^2 position resolution information about hits that could have caused readout. Because the CRT trigger signal and strip-by-strip readout are transmitted over different cables, the trigger board can still trigger on the CRT even if the board reader isn't writing out data.

Adding the CRT to Run Control

Just like adding any other component to Run Control. The unusual thing about the CRT is that an expert could be running it without Run Control. That is probably very unlikely now, but you should check with the CrtExperts if there is an unresolved eLog entry that talks about working on the CRT.

Monitoring the CRT's Performance

1. Watch out for alarms from your current partition! The CRT can trigger an alarm if its Average_Fragment_size (~= data rate) gets too small.
2. Keep an eye on the CRT in Monet. Under the CRT entry in the list-tree on the left side on Monet, look at the PerJob page. White channels in this picture are dead, and the one in this screenshot is
   the only known dead channel as of November 14, 2018. At the moment, the rates themselves aren't so meaningful because of how histograms are getting merged.
   
3. From the Main Run Control window for any partition, open the Trends window
   
4. At the top-right corner of the Trends window, open the Selection Tool window
   
5. There is a list-tree of DAQ "nodes" on the left side of the Selection Tool window. Click on the black arrow next to CRT to expand the tree into a list of CRT board readers.
   
6. Click on crt0 and drag down to crt4. All 4 CRT board readers should now be selected.
   
7. You can select a metric to view in the pull-down window to the right of the list tree. The default metric, Average_Fragment_size_double, is good for monitoring the CRT because it is proportional to the number of CRT boards that were included in an average data event. Click the Add button to get the metric you've selected to plot for all 4 CRT board readers.
8. Watch for size-48 and size-0 Fragments on the plot at the bottom of the Selection Tool. Size-48 Fragments seem to generally be empty for all systems I've worked with.
9. If any CRT board reader's Average_Fragment_size_double goes below 48, contact CrtExperts!
10. For a more thorough test of whether we are triggering on the CRT, see CheckForDeadUSBs. The expert might ask you to do this.

How to Get Help

• Contact the CrtExperts by phone or on Slack according to the severity of the problem.
• If there is a CRT-related problem when running, please mention the CRT in the end-of-run comments window. This will help CrtExperts track down the problem later.
• Write CRT messages on the e-log. There might be useful information about recent CRT maintenance there too.
• During beam data-taking, the CRT is not essential to our trigger conditions. So, if the CRT has problems and no other systems are affected, please keep running and send a Slack message to the CrtExperts. If the CRT is affecting other components while beam is available, please ask Rob about starting a new run without the CRT out of the run and contact the CrtExperts.