There are four sets of ECAL gains. The first were made with data from the runs 1200 and is
a complete 49 crystal calibration. The runs 1700 resulted in a calibration for the inner 5x5
crystals and the runs 1900 were used to calibrate the inner 3x3 crystals. The gains made for
the runs 2300 were made after the wedge change. Gains for epoch 2600 look similar with the
exception of crystal 25 (see explanation below).
The calibration procedure assumed that all energy is contained in a block of nine crystals
centered on the seed crystal. The gains of each crystal in the seven by seven block were
determined using an iteration method.
A plot of energy distirbution of 100GeV electrons going into the ECAL shows that energy
resolution for 100GeV e- maximum 2.6GeV. Note that noise for 3x3 sum is 0.6GeV (in spill
pedestal events).
This histogramshows the 3x3 sum of energy for electrons with seed crystal within central 3x3
matrix. Nine center crystals where selected from each epoch (1200,1700,1900 and 2300)
Here is a histogram of the crystal with the lowest sigma using only the f_1>0.5 cut:
crystal 33 run 2673
Gains vs. Time in days plot of the inner 3x3 crystals
The gains are normalized to 1 starting at day 0.
A new plot was amde after an error in the pedestals for epoch 1700 was found and correccted.
Thiss led to a smaller drop in gain. Now the variation for the outer crystals can most likely
be explained with temperature variation. This is to be examined soon. There is still a drop in
gain for crytstal 25 and between epoch 1700 and 1900 there is a significant drop for crystals 18,
33 and 25. This may be due to the 300 GeV pion runs that were taken between these two calibration
sets.
The realtionship between pedestal and gains is being investigated. Especially interestig is
the gain recovery which can be be observed. The last set of gains where determined after a
long period in which ECAL was not in the beam. This shows the gains tending towards recovery.
(Why crystal 24 shows a rise in gain followed by a fall back to 'normal' must still be
investigated, there is the suspision that there was a malfunction of the QDC.
see more)
There is a sudden rise for the gain of crystal 25 after 20 days. This is not a sudden increase
in crystal responce. The base of the photomultiplyer died for this crystal and had to be
excahnged. This will casue a change in respose, however it is surprizing that the change is
over 30%. This equiptment failure is especially bothersome because crystal 25 was hit with
multiple spills each containing over 4 million 225 GeV pions. Now it is no more possible to see
a correlation (if one exists).
It is possible that the remaining variations in the gains is dominated by temperature fluctuations
and not by the apperance of color centers due to radiation effects. When the temperature data has
been properly analyzed it will be possilbe to check the correlation.
Pedestal vs. Run plot for crystal 25
Pedestal vs. Run plot for crystal 24
These are plots of the pedestals and the rms of the pedestal for every run.
The later runs are quite stable with the exception of isolated runs. The correlation
between a large rms and fluctuating pedestal is obvious and therefor the rms can serve
as a pedestal quality control.
The pedestals of crystal 24 are quite stable but the RMS is exceptionally high (on
average about 3 times that of the RMS of crystal 25). This is something that should
be further investigated.
Temperature vs. Time plot for Thermocouplers 0 and 1
This is an initial plot of the thermocouplers to determine if and how well they work.
The probelms with castor are solved so this is a more complete set of data. Unfortunately
ECAL was unpluged for a long time so there is a large gap in the data. Also a filter was
applied that removed temperature readings that where taken within 30 minutes. An additional
cut was made on thermocoupler number 0 because there was a sudden drop in temperature that
could not have been real. For the TC_0 plot the time scale was zoomed in on where the data
is complete, while the second plot TC_1 shows the entire range of temperature data (still to
be expanded). The maximum temperature change is almost 4 degrees celcius, this will correspond
to a 8-10% change in crystal responce. This shows that the temperature must be accounted for
when monotoring the gains.
Run 1222
Run 1724
Run 1925
Run 2367
Run 2665
These are histograms for crystal 25 using the cut f_1>0.5. This means that at least 50% of
the energy for each event was in the seed crystal. Sigma ranges from as low as 1.8 up to 2.6.
A high sigma is due to insufficient calibration or poorly centered beam. If the beam is not
well centered the energy should still be summed up correctly. However if there is energy loss
down a crack this is not possible and a low end tail can be observed. The tail can be removed
by applying a cut on HCAL. This is yet to be done.
Here is a summary of the calibration efforts for the epoch 2600. All runs that where used for
calibration are plotted in nine histograms (one for each crystal).summary of epoch 2600
