$r$-$\phi$ trigger

The CDC $r$-$\phi$ trigger is the main component of the charged track triggers. It has to identify tracks originating from the interaction point, discriminate against various background track sources, and make a fast determination of the track $p_t$, the track direction, and the number of tracks. The $r$-$\phi$ trigger is formed using discriminated axial-wire hit signals. Anode wires in each super-layer are grouped into track segment finder (TSF) cells and the hit pattern in each cell is examined by a memory look-up (MLU) table to test the presence of a candidate track segment. MLUs are latched periodically by a sample clock of 16 MHz. Fig. shows the TSF cells for the CDC super-layers. The numbers of wires in a TSF cell are 17 for the innermost super-layer and 11 for outer super layers, as shown in Figs.  (a) and (b), respectively. In the order of the increasing super-layer radius, the numbers of TSF cells are 64, 96, 144, 192, 240, and 288. The TSF cells in the innermost super-layer are very important for rejecting tracks that originate away from the interaction point and the MLU pattern for these layers must be determined with special care.

Figure: CDC track segment finder (TSF) cells for (a) innermost superlayer and (b) outer super-layers.

The TSF-cell outputs are logically ORed in a super-layer to form track finder (TF) wedges. Fig. shows one of 64 TF wedges. The number of TSF cells to be ORed depends on the $p_t$ threshold chosen. The hit patterns of the TF wedges are fed into the next MLU stage. The second-stage MLU provides several output bits for different track categories according to patterns:

• short track (track in the three innermost trigger layers, 200 MeV/c) or
  
• full track (track which goes through all CDC trigger layers, 300 MeV/c).

Figure: CDC track finder (TF) wedge.

The output signals from the 64 TF wedges are fed into the next stage to determine an event topology:

• counting of the number of short and full tracks,
  
• determination of the maximum opening angle between tracks, and
  
• recognition of back-to-back track topologies.

The conditions are under software control and have considerable flexibility. The topology signals are sent to GDL.

Figures  and show the efficiency of short and full tracks as a function of $p_t$ and $\theta$ obtained from beam data samples, respectively. The efficiency drops at $p_t$ = 0.2 and 0.3 GeV/c for the short and full tracks, respectively, as designed. The short track is efficient for the full angular acceptance ( ), while the full track covers the barrel region only ( ).

Figure: CDC $r$-$\phi$ trigger efficiency as a function of $p_t$ for the short (top) and full (bottom) tracks. For the full tracks, only tracks in are used.

Figure: CDC $r$-$\phi$ trigger efficiency as a function of $\theta$ for the short (top) and full (bottom) tracks. Only tracks above the designed $p_t$ threshold are used.