Start of Shift Notes:  

Shot setup for store # 3271 in progress 
COT to be kept OFF, run JET_ST3_DECOUPLED[1,433,403]

AntiProton have been loaded. 
Ip = 9400E9 
Ipbar = 980E9 

set prescale bit 1 (L1_JET3_PS1) to 10.  
It significantly reduced dead time.

COT Gas Alarm. Cryo tech is taking care of it.

IFIX can't communicate with TOF. The TOF PC seems to be running fine. But the TOF alarm on IFIX
has been gray for more than 10 minutes.

Connection still not established with node.

Note off-diagonal entries in Trig-Mon L1 Cal. Paged ADMEM expert. Mark Mattson replied -- he thinks it is a monitoring issue.

I'm not sure what is up with these plots. If you look at this one (1.3.3) and the L1 Dirac Occupancy (1.3.2), it implies that there are absolutely *no* EM entries. The plot I usually look at is the 125 MeV DCAS Occupancy (1.3.4), which looks good (normal EM occupancy).

The Sci-Co asked if this might be related to what Bill B did earlier to recover ccal05. If so, I would expect problems in 3 phi on the west side, and not for the entire central calorimeter.

The next question was, should anyone else be consulted? As it is 1 AM, and there are no other "symptoms" to indicate a problem, I suggested it is enough to make a note in the e-log. Experts can look into this tomorrow.
-- Wed Mar 3 01:30:57 comment by...S.Miscetti --  

The only two persons working on the whole Central were Fotis and
B.Badgett. From the above plot looks that only the EM is having
problem with the trigger, while the Display shows also a multiplicity of small hits on CHA. Fotis
was working with the Laser and the only reason that could affect all the Central is LER downloading.
We will try to get in touch with both of them.

I talk to Fotis and I correc the above statement. Fotis did not
touch the LER setting; this is not required for Laser Cal!
LED are working properly has checked by HATD plot for CHA.

This problem has been noted a number of times already.
There is an option to control the hardware spike killer for CEM . The way this is controlled changed
recently

as for JDL request. The code to tell the trigger simulation 
if the spike killer is enable or not exists and it's finding 
its way to Trigmon.
If the errors appear in a run with a TEST trigger cable you 
they are not real.

circular pattern in calo - qcd instanton vortex bubble ?

 
L1_JET3_PS1[1]   / 10    12375 Hz 
L1_MB_CLCPS50K[1]/50000     25 Hz 
L1_MB_XING_PS1M[2]/1000003   1.6 Hz 
L1_JET3_PS1M[1]   /1000003   0.2 Hz 

Total L1 12240 Hz 
      L2   102 Hz 

ObjectMon has no data.  Plots are not filling with events.  I have killed the Mon and
restarted, but this did help. I am trying again.

TrigMon L1 Sum Et at Stefano's request

LOSTP spike

L1_JET3_PS1[1]   / 5       20900 Hz 
L1_MB_CLCPS50K[1]/50000       22 Hz 
L1_MB_XING_PS1M[2]/1000003   1.8 Hz 
L1_JET3_PS1M[1]   /1000003   0.1 Hz 

Total L1 21000 Hz 
      L2   100 Hz 

TOF was grey in the iFix Alarm window. 
TOF PC downstairs reported: ' SUMMARY LIST is overflowed' 
I restarted PC and went through all procedures to start SMACS 
but once I hit 'play' button. I get the same error: 
'SUMMARY LIST is overflowed' 
TOF is back green now, however heart bit remains grey.  
We take no action about this problem till the morning.

TOF heart bit is back green. Everrything is fine. We leave TOF
off for this run.

Please note:  Rainer and I manually switched SVXMon
to the cosmic setup for the tests tonight, because the data in not in stream A.  At the moment all
other Mons are running in physics mode.  SVXMon will need to be switched back to physics mode when
we begin taking physics data.  (Just source the script to put all consumers in physics
mode.)

CER_SVXMON_HALT_RECOVER_RUN_ERROR 

b0dap84.fnal.gov:ConsumerErrorRe:2:37:01 AM->Runtime Error 2, Event 301, R 
unNum 179578: SvxMon Halt Recover Run: Pipeline out of synch in 135 silicon read 
out chips . 

SCPU_BAD_CHANNEL_COUNTS Error 

Hardware EVB has detected a problem with data quality in 
 SCPU b0eb23. 
b0l3pcom1.fnal.gov:main:2:44:48 AM->Host b0eb23.fnal.gov, task tRec_0 
SCPU-P1-E-BadChannelCounts: In event 207985 the sum of channel counts is inconsi 
stent with the total count, VRB slot 20. 
b0l3pcom2.fnal.gov:main:2:44:49 AM->Error on L3 node b0l3098 (partition 1) 
 Wed Mar 3 02:44:48 2004 l3_node 
  in refoInt_reformatProc (l3_refevt.c:610) 
     ( L3_REFORMAT_ERROR: see Error Handler _reformatter.log file for full message) 

In both cases HRR worked.

adjusted prescale to ST3 trigger to be below 20 kHz. re-started SVXMon in cosmics mode since we are logging into stream I. several glinks seemed to be unsynch'ed in ISL, but HRR cured it. now seems to be taken data fine and safely.  - Rainer :: (run 179578)
-- Wed Mar 3 02:55:19 comment by...rainer --  see silicon elog

Please Note:  I also switched ObjectMon to run in
cosmic mode.  Kaori confirmed that both ObjectMon and SVXMon read only stream A in physics mode.  We
do not have stream A in the current configuration.

While speaking with Kaori, I created a new .tcl file for ObjectMon when running with the jet
trigger.  It has the physics reference plots but picks up all data streams.  Note that the jet
trigger data does not quite match the references, but it is within what we expect.


The .tcl file is "ObjectMon_jet.tcl" and is located with the other ObjectMon .tcl files.  This file
will need to be set manually when running in this mode again.

There were lots of persistent errors during last few hours: 

L1Mon: saw 210 L1 DMA transfers, expect 1 (buffer number 1) 

All of them were auto-HRR receoverd.

Shift started with shot setup in progress for run#3271 
Plan was to take data with JET_ST3_DECOUPLED[1,433,403]  
with Silicon and then test TEST_MET15_DECOUPLED[1,434,403].  
COT has to be OFF. 

Scraping done at 0:35 am.  
Ip/pbar = 8700E9 / 935E9, Lostp/Lostpbar = 7 / 0.5 Khz 
Initial Luminosity L = 5.4E31 

Gas alarm reset by Cryo. 

Test of JET_ST3_DECOUPLED[1,433,403] without COT crates,  
Silicon IN. 
Initial prescale L1_JET3_PS1[1]/10. RUN 179578.  
The prescales have been fixed to have a rate < 20 KHz  
(L1= 18KHz, L2=100Hz) 
Rainer fixed several glinks unsynch'ed in ISL. 

Still it is not yet clear if this is a valid test.  
Trigmon shows a not fully diagonal correlation  
between DCAS trigger tower ET and ADMEM Et. 
This is only in the central EM. Similarly(?) the  
display shows anomalous hits on the CEM, CHA.  
This has to be investigated more carefully today. 

Consumer processes were also having some problems  
since the data do not go on stream A:  
SVXMON and OBJECTMON run now in cosmic mode.  
Check Kaori comments on the e-log.