Next: SVD readout system
Up: Readout
Previous: Readout
  Contents
We adopted a charge-to-time (Q-to-T) technique to read out signals
from most of the detectors. Instead of using ADC to digitize the
amplitude of a signal, the charge is once stored in a capacitor and
discharged at a constant rate. Two pulses, the separation of which is
proportional to the signal amplitude, are generated at the start and
stop times of the discharge. By digitizing the time interval of the
two timing pulses with respect to a common stop timing, we can
determine both the timing and the amplitude of the input signal. The
hold timing is generated either by a self-gate method or the trigger
signal from a timing distributer module, TDM. The mechanism is
illustrated in Fig. .
Figure:
Concept of the Q-to-T and TDC based digitization.
|
For time digitization, we use a multi-hit FASTBUS TDC module, LeCroy
LRS1877S. Up to 16 timing pulses are recorded for 96 channels in a
single width module with a sparsification capability. The least
significant bit is 500 ps. A programmable time window has a 16-bit
range, which corresponds to a full scale of 32 s.
Most of the detectors, CDC, ACC, TOF, ECL and EFC, are read out by
using the Q-to-T and TDC technique. The use of the Q-to-T technique
reduces the number of cables for CDC wires into a half. In the case
of TOF, the time resolution of 100 ps is achieved by using a time
stretcher which expands the pulse width by a factor of 20. In the case
of ECL, a 16-bit dynamic range is achieved by using three ranges. A
signal is divided and fed into three preamplifiers of different gains,
converted to the time pulse and merged with an exclusive- logic
circuit. For a small signal, all the three ranges give short pulses
and the final output has 4 time pulses. By contrast, for a large
signal, high and middle gain pulses overflow and the final output has
only 2 time pulses from the low gain range. After digitization, the
range is identified by the number of time pulses for the pulse.
The KLM strip information is also read out by using the same type of
TDC. Strip signals are multiplexed into serial lines and recorded by
TDC as the time pulses. These pulses are decoded to reconstruct hit
strips. Similarly, a large number of trigger signals including those
for the intermediate stages are recorded using TDC. A full set of
trigger signals gives us complete information for trigger study.
We developed a unified FASTBUS TDC readout subsystem that is
applicable to all the detectors except SVD. A FASTBUS processor
interface, FPI, developed by us controls these TDC modules, and FPI is
controlled by a readout system controller in a master VME crate.
Readout software runs on the VxWorks real time operating system on a
Motorola 68040 CPU module, MVME162. Data are passed to an event
builder transmitter in the same VME crate. A schematic view of the
system is shown in Fig. . The overall transfer
rate of the subsystem is measured to be about 3.5 MB/sec.
Figure:
Schematic view of the FASTBUS TDC readout system.
|
Next: SVD readout system
Up: Readout
Previous: Readout
  Contents
Samo Stanic
2001-06-02