Vertex detector

Eng Up Kedr

vertex detector

The vertex detector (Fig.1) is a straw-type drift chamber made of mylar tubes with 20 micron thick walls. Such cells are mechanically isolated, the electric field in them is completely symmetric. Due to these features straw chambers are now rather popular. Usually the access to the vertex chamber is rather difficult so the chamber has to be reliable enough. Tubes suit that requirement very well. The radial dimensions of the vertex chamber are limited to 60 and 123 mm radius. The sensitive length is 670 mm. The vertex detector covers 95% of the total solid angle.

The chamber consists of two independent, self-supporting halves. The positions of the tubes are shown in Fig.2. There are 312 tubes in 6 cylindrical layers which form 3 double layers. Those in an odd layer are separated by as small intervals (0.5 mm) as possible. The angular intervals between the tube centers in the next layer are the same and tubes are shifted by half a period. Thus there are no holes in a double layer except at the edges. The inefficiency of a single odd layer results mainly from the inability to detect two particles in one tube. In this case only the particle closest to the wire will be detected. But if the particles come from the beam, the second one becomes the closest in the next layer and hence in is detected too. The inefficiency of the first layer is 13%, for the first double layer it is 3% (without the edge holes).

Main geometric parameters of VD given in the Table 1

Table 1

Tube's diameter10ÍÍ.
Length of sensitive area670ÍÍ.
Anode wireW, 20ÍËÍ.
First layer radius67.1ÍÍ.
Last layer radius116.84ÍÍ.
Number of the layers6
Full number of the tubes312

The geometry of the vertex chamber provides the accuracy of the impact parameter reconstruction at the beam region:

sigmad=2.2*sigma0,

where sigma0 is the r-fi resolution in a single tube. The contribution of multiple scattering is 40 mcm/P [GeV] and results mainly from the material of the vacuum chamber and the inner wall of the vertex chamber. The longitudinal coordinate is not measured by the vertex chamber.

The tubes are produced in the Laboratory by ultrasonic welding. The tubes produced by this method keep the cylindrical shape very well, especially after being glued to the end plugs. An unloaded tube with fixed plugs has a gravitational sag of about 0.1 mm.

The attractive feature of the module is that it can keep a pressure of up to 3 atm. Hence, in principle, the tubes may operate directly in air, so that there is no need for a leakproof gas vessel around the VC.

The slightly overpressurized tube becomes rigid and stable against twisting along its axis. This is essential, because the mylar wall shows very weak resistance to any sort of transverse tension, and special attention must be paid to keep the correct cylindrical form of the cell.

The resistance to radiation damage was studied. The collected charge was 0.5 C/cm corresponding to 10 years of operation of the collider at high energy. The decrease of the gas amplification is 25%. The aluminized mylar we use has been irradiated by 1.5 MeV electrons to a dose of 1 Mrad, which is 3 orders of magnitude higher than one could expect in an experiment. No visible damage or change of mechanical properties has been observed.

The electronics channel consists of a preamplifier, an amplifier-shaper and a digitizer. Preamps are situated near the wires on both flanges of the vertex chamber. They have a rise time of 5 ns and a response of 0.3 V/pC. The output pulse is transferred via 5 m long twisted-pair cables to the shaper which is installed near the detector. The effective threshold of the shaper together with the preamplifier is 0.01 pC. It is approximately two times lower than the average pulse produced by a single electron. The total slewing from twice the threshold up to very high values does not exceed 4 ns. It contributes to the spatial resolution mainly near the wire. The logic signals from shapers are transferred to digitizers through a 50 m line. The digitizers are direct counting TDCs with 2 ns resolution.

As a gas mixtures in the vertex chamber carbon dioxide with admixture of isobutan (up to 8%) is used. The usage of cool gas mixture minimizes the diffusion. Thsis choise has some other advantages. In a major fraction of a tube the electric field and the drift velocity are low. This results in a low contribution of the timing accuracy to the spatial resolution. For a cool gas the time of drift is almost the same with and without magnetic field.

On the other hand, the long drift time makes it difficult to use the vertex chamber in the trigger system. The main problem of operation with cool gases, of course, is the strong dependence of the drift velocity on temperature and pressure, and one has to control them very carefully.

To study different gases, electronics and reconstruction procedures, we have built a prototype "Etap", consisting of 14 tubes which were positioned in 6 layers, just as the VD geometry. The mylar thickness and length were different, 50 and 200 mm respectively. The chamber was tested in a 1 GeV positron beam at the storage ring VEPP-3. The temperature and the pressure were measured during the prototype tests, and their total variations did not exceed 0.3%.

In Fig.3 the resolutions versus radii for different gases are shown.

The vertex detector based on thin mylar tubes is placed in the KEDR detector now. It was in operation during the J/Psi scanning in 1998. An example of multihadron event detected in the vertex detector is shown in Fig.4.

Last modified on 26 September 2000