Commissioning and running of sPHENIX

[Stefan Bathe <stefan.bathe AT baruch.cuny.edu>]

General (your name)

• 
  

Work Control Coordinators ()

• 
  

Plan of the Day (Stefan)

• 1st prio:  trigger  (day:  trigger meeting)

• 2nd prio:  calo data taking (night)

• Trigger on central events (this evening)

Kin

• 2 h access tomorrow

Evening (Sasha)

• Collisions most of the shift. Two stores, three and a half hours each, with 12 bunches.

• IR humidity alarm firing every few minutes at the beginning of the shift. Caused by IR humidity fluctuating just around the alarm threshold. Increased threshold from 50% to 60% (see corresponding entry in the logbook about how to do it).

• IR air handler is out of service, waiting for parts

Night (Brett)

• Inherited evening’s shift beam, went for couple more hours

• MCR called to dump, but never managed to start up again

Day (Ben)

• Beam right away, problems with sextupole PS, figured out tune to keep it in store, until 1:00pm, new fill 30 min later, running since

• Otherwise quiet except for EMCal chiller alarms, called Rob, had John top up chiller

• PS supply persistent (related to access)

• Kin:  STAR does not see a phase shift in timing (J.H.)

MBD (Mickey)

• Seeing unstable phase in clock from store to store, reason currently unknown

• In addition to the remote scope for monitoring, I’ve written a GUI /home/phnxrc/operations/mbd.  Will set up for the shift crew so they can adjust the timing at the start of each fill if needed.

ZDC (Peter)

• (from logbook) Ejiro took 100k with both ZDCs in the USB scope (Ch1 and analog sum from each side), to see if we see any obvious backgrounds when triggering on each side.

Instead, things look surprisingly familiar - photonuclear and Glauber contributions

Still, a large asymmetry seen in the trigger rates, potentially from PMT gains

Planning to increase gains on ZDC N (for us, not RHIC) but requires either swapping module or drilling a hole in the front panel of the current one

Background Counters (John)

• Background counters south tomorrow access?

Trigger (Jaebeom)

•       Tried yesterday to setup similar condition as yesterday with new conditions (firmware 35)

• See some MBD time difference north vs south in both collisions & laser
  

• Mickey:  it’s a physical thing:  seen also with scope

• Couldn’t get as good a fraction of good collisions relative to total events as before

• Plan:  (i) debug; (ii) even if performance is not as good, we can keep taking data and add HCal; (iii) gather information for tomorrow’s trigger meeting

• Jamie:  would be good to decouple two issues:  (i) want full MBD trigger working, but have hit limit; (ii) time in calorimeter

• Run MBD trigger with no vertex cut:  just cut on 40 tubes on each side -> clean AuAu central trigger -> overnight EMCal/HCal timing in, data, noise

John:  serial:  (i) trigger primitives; (ii) time

Mickey:  need information on trigger conditions

• Peter:  working on it

DAQ (Martin)

We switched to the GL1/GTM firmware version 35 yesterday.

That version promises

• The re-activation of the coarse delay feature by 56MHz (~17.7 ns) whole clock ticks that had fallen by the wayside in an earlier version. (We call this l1_delay). The fine delay that can then shift by sub-clock values has always been working

• (partially reported already yesterday) The fiducial tick, also referred to as revtick (as in revolution) works. John and I tried this - the unused vGTM 14 was connected to a clock master which in turn was connected to the scope with a CHISB dongle so we can see if this is started or not. Stopped the vGTM. We pulled the revtick input off, set reg 2 to 1, and started vGTM 14. It remained stopped. John then reconnected the revtick input, and then the gtm started. All good.

• We can read back the RHIC clock lock status in reg. 14

• We can now set the beam clock to a start value (see https://sphenix-intra.sdcc.bnl.gov/WWW/elog/DAQ/32) after a reboot or have a power cut

• We can read back the state of the GTM busies. In normal running, that gives whatever status you happen to catch the GL1 in. It is very useful to recognize if a GTM is persistently busy (then you don’t get triggers). 

      Virginia,Olivier, and improved the HCal DAQ setup with bringing seb16 and 17 rcdaq under   systemd control as the others, and made other improvements for the evolving standard operator DAQ procedures. Seb17’s readout (never 16 - West) still crashes now and then. Will investigate at the next occasion.

      

HCal (Oliver, Virginia)

• Bias currents are stable

• Issues with the MBD trigger overnight, we could not time in the HCal yet

• Currently we are trying to look at correlations between HCal and MBD

EMCal (John)

• Turned on bias this morning to whole detector under John’s control

• Logs in EMCal elog

• Bias currents increased from 25 to 200 uA

• Gunther:  when will EMCal crew show up?

• John:  Tim and Anthony have been done a lot of work already

• Next step:  timing once trigger is clean

TPOT (Hugo, Tristan)

Been operating the TPOT HV the whole day, monitoring dark currents, and trips. 

A good fraction (11 out of 16) of the detectors are very stable. For the other 5 detectors we have seen trips, some of which we could recover, for some others we had to lower the voltage.

We are still learning 

• 
  

TPC (Jin)

• During the GTM-GL1 tests last night, TPC FEE went into a stuck state after an accidental high rate trigger sent to the TPC at 9:32 PM. Not a big problem. Reset today via power cycling all FEEs between fills, and recovered GTM links on all sectors. 

• (Same as Fri-Sat owl shift) not much FEE drop out last night during beam

• Daily pedestal check run taken

INTT ()

• 
  

MVTX (Jo)

• Nothing directly related to shift

• Yasser implemented analysis on threshold tuning and produced plots (telescope data)

• Cooling seems to slowly decrease in flow; not critical yet, FEE cooling started 8.7 l/min, now down to 7.5 l/min, 

• Temperatures in the FEEs haven’t increased

sEPD(Rosi Reed)

• Waiting on interlock cables to be installed (Chris P.)

Gas/Cooling ()

• 
  

Magnet (Kin Yip)

• Cryo group continues to cool down the Magnet — at the moment, most reliably by reading the Cryo SUPPLY (TE1784H) and RETURN (TE1785H) temperature sensors.

- 

---------------------------------------------------------------------------------
Stefan Bathe

Professor of Physics
Baruch College, CUNY
and RIKEN Visiting Scientist

Baruch:                                     BNL:
17 Lexington Ave                      Bldg. 510
office 940                                  office 2-229
phone 646-660-6272                phone 631-344-8490
----------------------------------------------------------------------------------