The HERMES silicon project: The radiation protection system
Nucl. Instr. Meth. 512 (2003) 44-51
The rest of this document describes the functional diagram of the module, the trigger circuit implemented in NIM modules, and the external connections and adjustments on this module.
The functional diagram of the beamloss monitor system. The blue box contains the circuit on the frontend, the red box is one channel of the beamloss monitor module. The two parts are connected via a 30 m long, twisted pair cable. Click here for a postscript file.
From the summer 2004 a BeamLoss monitor with an extra integrator was installed. The integrator (one per channel) is connected to the electronics just after the line receiver. It consists of an integrator (which changes the sign of the signal), an inverting amplifier (gain: 1x) to undo this sign change, a discriminator and an output stage (see figure).
Block diagram of the integrator. The integrator (timeconstant: 1s) is protected by a baseline restoring circuit, schematically depicted by the integrator in the feedback circuit. This circuit has a timeconstant of 100s. The testpoint (analog) has the inverted signal after integration. The output of the integrator is about 1V per Rad. Click here for a postscript file.
The current through the ionization chamber is proportional to the dosetempo on the chamber. This current is amplified in the frontend amplifier and converted to a voltage signal. The timeconstant of this amplifier is about 5 µ s, and the gain is 1 mV/nA. This signal is amplified by the line driver (gain = 8).
In what follows the numbers for both the new (summer 2004 and later) and old BLM (before summer 2004) are given. The values for the old BLM are mentioned in round brackets where they are different.
The cable to the electronic trailer attenuates the signal by a factor 0.78. The receiver in the beamloss monitor module has a gain of one. From the receiver the signal follows two different routes. In the first it is divided by two, buffered and fed into the ADC. The input resistor causes an additional attenuation with a factor 0.69.In the second route an amplification by a factor 1.5 (5.8) is followed by the addition of a 0 (204) mV offset. The resulting signal is compared with a threshold. The threshold cann be set in the range -1.25..1.25V (0..2.5V). The output of the comparator is subsequently tested for pulse duration. When the pulse is long enough an output pulse is generated. The levels for the integrating part can be set separately, but are not equipped with the time-over-threshold discriminator.
The trigger for the beamloss monitor. The outputs of the left hand (channels 5..7) and right hand ionization chambers (channels 1..3) are combined with a majority logic, requiring two out of three chambers with signal. A dead time circuit prevents triggers too close together, which might confuse the ADC card. Click here for a postscript file.
The outputs of the eight channels of the beamloss monitor are combined in the trigger circuit, implemented in NIM modules. A majority (two out of three) of the chambers on either side of the beam should have a good signal, while at the same time the dummy signal on this side has no valid signal. Left and right side are then combined with an OR-gate.
List of changes
- January, 2002. A deadtime was implemented by the first channel of the dual timer.
- June 2002. A second three-fold logic unit was installed. The last channel of this is used to implement the 12 GeV veto. The trigger of the BLM ADC was moved to the output of this.
- July, 2002. The pulse to the optical tranceivers was stretched with an extra timer, on request of the machine people who need a longer pulse. The trigger for the BLM ADC was moved to the first three fold logic (channel c in slot 5/6), to measure radiation during injection.
- July, 2002. The trigger of the BLM ADC was moved back to the 12 GeV vetoed gate. This facilitates a fast analysis of dump triggers. This modification will be temporary.
- March 3, 2003. The triggerdiagram in this manual was updated to reflect the changes above. No hardware was moved.
- November 1, 2004. The discriminator outputs are included in the trigger diagram. Some module positions have changed.
This picture shows the front of the beamloss monitor module. Point with your mouse at this picture to find out what is the purpose of a detail. The status (bottom) line of the new browser window will explain.
This picture shows the backpanel and its connections to the kicker. The two BNC connectors are shorted (4 relais are used to make a failsafe short) in normal operation. Open when the kicker should be triggered. The RS422 connection is connected to this circuit.
Click the list on the right of the little picture below to know where to perform the task mentioned there. These functions use JAVA, which should be enabled to make this work.
Points to measure the (FAST) analogue waveform (note: offset 0 mV)
Points to measure the (FAST) threshold level (no offset)
Points to measure the output of the (FAST) comparator (TTL)
Points to measure the (SLOW) analogue waveform (note: offset 0 mV)
Points to measure the (SLOW) threshold level (no offset)
Points to measure the output of the (SLOW) comparator (TTL)
Threshold adjustment FAST channels
Time-over-threshold discriminator adjustment
Threshold (integrating channels) adjustment
NIM output when (FAST) channel is over threshold
NIM output when (SLOW) channel is over threshold
ADC trigger input (TTL)
ADC clock input (TTL)
TTL inputs for testpulses to frontends
TTL inputs for dump kicker
Pilot lamps for ± 6V power
ADC inputs (connects to blm hostcomputer)
Connects to frontends channels 1..4
Connects to frontends channels 5..8
Supplies testpulses to frontends
Supplies frontends with power (5 V)
Stop blinking
top