Scint people at Gran Sasso during this period:
M. Goretti, E. Katsavounidis, N. Lu, D. Michael, D. Niccolo
R. Pazzi, P. Saggese
Reporter:
N. Lu
This is my very first shift at Gran Sasso, I've managed fulfilling
most of the shift duties without much problems. But according to
Doug, I've performed poorly, having not been able to understand
the commands that I was given to perform. I admitted that I've
learned much of the shift duties verbally from the previous shift
workers, and Doug has pointed out that it's exactly how the errors
got transmitted over monthes that could have cause the production
of totally useless data. I intend to remedy the defects in the
following shift that I'm taking right now.
The first week of my shift has been disastrous, the run crushed,
the HV system failed for SM1 and SM2, occasionally SM3, and WFD
system for SM4 gave acqusition system error messages. There were a
couple of mornings that I came in and found the run crushed two
or three in the morning and SM1 HV totally off. One morning
there was a wind storm outside the tunnel, and the power supply
was clearly affected. Every time the lights dimmed, SM HV would
go off, as if some ghost had come and switched it off. For that
morning, we had to leave uVAX 1 out of acquisition.
Investigation into the HV problem reveals that on uVAX 1,
the camac controller can only talk to one SM at a time. When HV
is being loaded, there's a discrepancy between the voltage read
and the voltage demanded. Paolo was certain that it was the
memory chip that has caused this problem. LeCroy has sent us
some memory chips and transistors, but we have not received the
MOS-FETs yet.
Paolo believes that another cause for the breakdown of
the HV system is the power line flickers and spikes. There
are four power supply lines coming from outside of the tunnel,
three phase lines and one ground line. The three phases lines
are passed through a stabilizer to have the flickers and spikes
smoothed out. But the ground line from outside is directly
connected with the one used inside the tunnel, so any flickers
and spikes could enter our electronic systyem via the ground
line. Paolo has requested to have a big inductor installed
in the ground line. Of course we have to wait for some
Italian electricians to come around to do it for us. Doug
suggested that we should also check if there's any discrepancy
between the HV ground line and fanout ground line.
Paolo has reminded us that the spare LeCroy mainframe
is not compatible with the 6 mainframes that are currently
in use. He thinks we'd better purchase a new compatible
mainframe, so we can replace any broken one in the future.
Doug likes this idea.
SM4 WFD systerm has given us problems before, and this
month was no exception. We kept on getting system error messages
on the acquisition console for uVAX2, message number 5C74. Erik
had to take SM4 WFD out of camac list. A couple of days later,
we got system error messages 5C74 on all uVAXes every few seconds.
The run seems to crush when there's too many system errors. We
recalled when the last time we had problems with SM4 WFD, it was
a blown capacitor on the backplane. So Erik, Doug, and I
proceeded to replace the backplane of SM4 WFD, and the problem
went away. It was once again the same capacitor that got blown.
Doug suspected the capacitor was installed backward and blew
after a few months.
I was responsible to bring WFD ADC led calibration back
to work. On Aug. 30th, Erik and I have performed a WFD ADC
calibration on uVAX1. On site I could see the HP pulsers and
switchboxes working as expected, but I'm not sure how the data
came out. And on Sep. 7th, Maurizio and I performed another
set of MON SIMU calibration on uVAX 2. Unfortunately we forgot
to put the correct camac list on. When we were performing
LED TED calibration, Maurizio found one cable to the fakebox
and another one to ADC GATE of scintllator trigger master
unplugged.
Erik has noticed that the acquisition system has
recognized the calibration runs as normal ones. He wondered
how that could happen if SPU and fakebox had detemined that
they're calibration runs. And the ON-LINE-LOGBOOK has recognized
the MON SIMU runs as SLOW MONOPLE SIMULATION. Erik doubted
the on line system has that much knowledge, he suspected that
someone has gone into the LOGBOOK and manually added the comment.
Maurizio has written e-mails to the Italian shift workers in
regard to this issue, since they are the only ones who were
seen editing LOGBOOK.
Erik has a program that collects error messages from
the acquisition console system, so the not-staring-at-the-screen
-all-the-time shift worker can have a statistical report of
the occurrence of error messages. This program also notices
any fake events, like same box firing twice in a single event.
It finds dead erp boxes and dead cspam boxes. It checks SPU to
see if fake latch has been generated. Doug likes Erik to make
this program user-friendly for more people to use.
On Sep. 6th, there was a power outage to the detector
for 12 hours. When the power came back to the electronics
system, Maurizio found that SM1 THOM crate controller has lost
its bit set for trigger, activity, and decay. On the following
day he and Paolo looked into this problem and found the cables
were plugged upside down. Erik warned that there are still
many calbes unlabelled or left out.
Doug has asked Maurizio to take charge of the
WRONGLIST from now on, Paolo to familiar himself with PAW to
make reference histograms for on line histrogram presenter.
Doug wants to clarify the shift worker's duties so everyone
on the detector can function efficiently.
That's all, folks!