Analysis of run0072 (two hours accumulation).

Attachment 1:
Reconstructed beam image. The ring should correspond to the cell's frame.

Attachment 2:
Definition of cuts. The central region of the target was selected and then named "targ".

Attachment 3:
Particle identification. Deuterons are well separated.

Attachment 4:
Elab-theta plot. The target and deuteron cuts were applied. Kinematics curves of the (p,d) reactions are seen.

Attachment 5:
Elab-theta again. Calculated kinematics curves were superimposed.
An energy correction factor of 0.8 has to be introduced to reproduce the curves (probably due to the incomplete energy calibration).
In addition to the ground-state transition, excited states with spectroscopic factors larger than 0.4 are also shown. See PRC 48, 95 (1993) for the details.

Attachment 6:
Angular distribution of 40Ca(d,p)39Ca(1/2+) from the reference. A similar oscillating pattern was observed in the Elab-theta plot.

EDIT: the deviation from the kinematics was due to an incorrect beam energy.

Order-of-magnitude estimate of the target thickness.

We have about 5000 counts of 40Ca(p,d)39Ca(g.s.) in an angular range of 10-30 degrees.
The total beam flux is calculated as follows.

2 hours x 1.5 x 10^5 pps = 1 x 10^9 particles
About half is on target -> 5 x 10^8 particles

By assuming the 5-mb cross section (integrated with respect to the cm angle),
and ignoring the Jacobian and the detector coverage, we get

thickness = 5000 counts / (5 mb x 5 x 10^8 particles) = 2 x 10^21 cm^-2

c.f. 1-mm thick solid hydrogen 4 x 10^21 cm^-2

There was a sudden increase of the temperature at 4pm. Fortunately the cell is still filled with liquid.

Data analysis of run0128 (38Ca physics run). The multiplicity distribution of DSSD x-strips is shown in
Attachment 1. The second peak centered at around 128, which is the number of strips of one detector, corresponds
to events where beam-like heavy ions are hitting the detector.
This causes a huge energy deposit in the detector and eventually induces the mirror charge to be shared among
many strips. The number of these events amounts to about 6% of the total (note that only 100000 events were used
for the analysis). A typical charge distribution of such an event is shown in Attachment 2.
We may need to adjust negative thresholds to suppress them. 

EDIT : The same behaviour can be seen on Y strips (see attachment)