Inefficiency of the Multiplicity detector

I have noticed recently, that the sizes of the silicon planes in the octagon and vertex detectors have changed. The planes became narrower in the x-y direction (perpendicular to the beam direction) without any change of the position of the center of the planes. I became curious if this does not result in gaps in the octagon for fixed phi emission angles.

To study this I have performed simple simulations of 20 central Hijet charged pions only without any physics but multiple scattering and without magnetic field. Thus the particles were never stopped and in principle all of them should hit one of the detectors. The simulations were done for two geometries: version 3.6 (with wider planes) and the latest version 3.7.

The results are the following:

There is some increase of the number of charged pions without any hits, but not very large. To check which particles were not registered the following plots were done:

• Fig. 1. Eta distribution of all pions (solid histogram), pions without hits for V3.6 geometry (dashed histogram) and pions without hits for V3.7 geometry (dotted histogram)
• Fig. 2. Phi distribution of all pions (solid histogram), pions without hits for V3.6 geometry (dashed histogram) and pions without hits for V3.7 geometry (dotted histogram)
• Fig. 3. Phi vs eta scatter plot of pions without hits for V3.6 geometry
• Fig. 4. Phi vs eta scatter plot of pions without hits for V3.7 geometry (vertex at z=0)
• Fig. 4a. Phi vs eta scatter plot of pions without hits in multiplicity and the first planes of spectrometer for V3.7 geometry (vertex at z=0); the position of the first plane of spectrometer is clearly visible
• Fig. 5. Phi vs eta scatter plot of pions without hits for V3.7 geometry (vertex at z=-10)
• Fig. 6. Phi vs eta scatter plot of pions without hits for V3.7 geometry (vertex at z=10)

There are four inefficient areas, which were already foreseen in the detector design:

• eta > 5.3 and eta < -5.3 - smallest angles, particles which exit the beam pipe after last multiplicity ring
• two windows in the octagon that allow the particles to go directly to the spectrometer arms (without hitting a multiplicity detector plane first) they are for vertex at z=0 in: -1.75 < eta < 1.75 and -0.4 < phi < 0.4 -1.75 < eta < 1.75 and phi > 2.74 or phi < -2.74

In addition to them there are many undetected pions for some fixed phi values (Fig. 2). The inefficiency for such phi bin exceeds sometimes 50 %. More details can be seen on the scatter plots (fig.3-6). The big black area in the middle is the spectrometer window. Dots on the left and right sides of the picture are due to particles emitted at very small angles. There are also lines or even bands of particles for fixed phi (horizontal) or eta (vertical) values. The horizontal bands are caused by the non overlapping of the octagon planes. The vertical bands prove, that the first vertex plane is separated from the neighbour octagon plane. In addition to the footprints of the windows or gaps in the detector there are several separate point in quite unexpected areas. These are due to particles (low momentum probably) that scattered at an large angle in the beam pipe and completely changed their direction.

In order to describe precisely the contributions of different inefficiency sources particles without hits were divided into groups depending on the area they fit:

The largest number of particles escapes undetected through both spectrometer windows, but a significant fraction of particles can also find the way between the octagon planes. After the last modifications this fraction increased from 3.4% to 5.0% of all charged pions.

Conclusions

It would be good to check once more if we can not make the octagon and vertex detector "hermetic" as many particles go between the planes. One may also consider to make the spectrometer window smaller, there is always a significant gap between the first spectrometer plane and the negative eta boundary of the window.

Figures location

The files containing postscript versions of the figures are in: Directory DISK$CSC:[WOZNIAK.MULT]

• INEF_ETA.PS;1 - Fig. 1
• INEF_ETAPHI36.PS;1 - Fig. 3
• INEF_ETAPHI37.PS;1 - Fig. 4
• INEF_ETAPHI37A.PS;1 - Fig. 4a
• INEF_ETAPHI37_VM10.PS;1 - Fig. 5
• INEF_ETAPHI37_VP10.PS;1 - Fig. 6
• INEF_PHI.PS;1 - Fig. 2