5- 2.1 Introduction

A typical start-up channel, shown in block form in Fig 5 2, consists of the following major components (1) sensor, (2) pulse amplifier, (3) high-voltage power sup­ply, (4) amplifier—discriminator unit, (5) count-rate meter with control functions, and (6) readout equipment.

4- 2.2 Sensor

Because the neutron flux in a shutdown reactor is low, the output of the neutron detector (see Chap 2) is a series of pulses proportional to the neutron flux resulting from a neutron source in the reactor The detector must have a high neutron sensitivity with a very low sensitivity to gammas

The fission chamber is widely used because of its inherent ability to discriminate against gamma-generated signals A fission chamber 2 in. in diameter and 12 in long typically will have a neutron sensitivity of 0 7 (count/sec)/ (neutron cm 2 sec 1) with a gamma sensitivity of 4 X 1014 amp/(R/hr) The neutron signal generated m the fission — chamber circuit is of the order of 100 qV with a pulse width of approximately 0 5 qsec The gamma pulses produced in the chamber are smaller in amplitude A typical ratio of gamma to fission pulse height is of the order of 10 2 to 1 This ratio makes successful discrimination possible

Gamma pulses generated m the chamber can become a problem when the detector is located m a gamma field of 5 X 104 R/hr or more A pulse pileup in the discriminator circuit causes gamma pulses to be counted as neutrons Special techniques, such as pulse clipping or reshaping of the pulses, are used to minimize and protect against this Typical pulse shapes are shown in Fig 5 3

5- 2.3 Pulse Preamplifier

(a) Introduction. The pulse generated in the sensor must be amplified for transmission and for driving standard counting electronics The ideal pulse preamplifier must have high gain, wide bandwidth, stability, and low noise charac­teristics

Until recently fission sensor preamplifiers used vacuum tubes and were located as close to the sensor as possible In the last few years, good-quality solid-state preamplifiers have replaced the vacuum-tube units There are advantages and disadvantages with both types

A vacuum-tube preamplifier can be placed near the fission detector, і e, in the same radiation field as the sensor However, radiation damage to the preamplifier components decreases their life and increases spurious noise so that maintenance is required Vacuum-tube preamplifiers produce a large output pulse, ideal for counting equipment, but the vacuum tubes deteriorate with steady use and must be replaced regularly or suffer loss of gain. Generally, substantial maintenance is required per hour of successful operation

Solid-state preamplifiers, usually charge-sensitive, can be located at considerable distance (up to 90 ft) from the sensor but, in any event, must be out of the radiation field. This makes maintenance more convenient and increases channel availability However, this arrangement is suscepti­ble to noise pickup in the cable between the sensor and the preamplifier The output pulse from solid-state preampli­fiers, typically 10V, is generally smaller than that from a

vacuum-tube preamplifier Because of the lower signal output, the counting equipment must be capable of accepting a low-level signal and processing it for use in the readout and control equipment that follows it in the channel

(b) Vacuum-Tube Preamplifier. A vacuum-tube pre­amplifier is shown in Fig. 5.4. The amplification is provided by the four RCA 7586 (nuvistor) vacuum tubes The preamplifier has a gain of 30, a rise time of 5 X 10 8 sec, and a fall time of 2 X 10 7 sec Because vacuum tubes are used in this preamplifier, it can operate m a radiation field. The tube-filament heaters are d-c powered to minimize a-c noise pickup The capacitors are ceramic, both for im­proved temperature stability and for reduced susceptibility to radiation damage Capacitor Cl is used to isolate the detector high voltage from the preamplifier and to couple the sensor pulse to the preamplifier input circuit. The life expectancy of this preamplifier is approximately 600 hr in a 106 R/hr gamma flux (0 5- to 1 5-MeV gammas)

(c) Solid-State Preamplifier (Current Input). The

solid-state preamplifier shown in Fig 5 5 is a voltage amplifier, as opposed to the charge-sensitive amplifier to be discussed in the next subsection The active elements are transistors The input stage is a grounded-base transistor, which has low input impedance, high output impedance, wide bandwidth (high frequency), high bias stability, and good voltage gain.

The interconnecting coaxial cable between the sensor and the preamplifier is terminated into its characteristic impedance. A resistor in series with the emitter is selected to do this matching. This circuit eliminates cable ringing or reflections and provides a low impedance path for the current pulse generated in the sensor The pulse is capacitance-coupled to the input stage, and the capacitor also blocks the sensor d-c voltage The input current pulse from the detector is converted in transistor Q1 to a voltage pulse The remainder of the preamplifier is a high-gain standard operational amplifier with feedback. The pre­amplifier can amplify pulses at a repetition rate of 106 Hz or 106 neutrons/sec Typical preamplifier characteristics are

0 5 volt//iA 50 to 120 ohms ±3 volts into 50 ohms <5% for 200-nsec pulse width ±15 volts at 40 mA

As noted above preamplifiers mounted at a distance (20 to 40 ft) have noise problems associated with cable pickup Every effort must be made to shield against stray noise in the form of electrostatic or electromagnetic voltages be­tween the sensor and the preamplifier

In summary, the principal advantages of locating the preamplifier at a distance from the sensor are (1) sensor cooling is not so critical, (2) maintenance is simplified, and

(3) system availability is increased.

(d) Solid-State Preamplifier (Charge-Sensitive). A third type of preamplifier is a charge-sensitive unit (Fig. 5.6). This preamplifier is a fast-rise-time charge-sensitive pre­amplifier with dual-polarity output The input signal is coupled to the pulse-shaping and amplifier input module A1 by capacitor C4, which blocks the high voltage. A Shockley diode connected to ground at the input to amplifier A1 protects the input from momentary break­down of the detector or cable shorts. The output of A1 feeds a cable driver with dual-polarity output

Amplifier module A1 is a special fast-pulse amplifier connected in a charge-sensitive configuration. This is accomplished by connecting a stable small-value capacitor from the amplifier output back to its inverting input. This negative feedback will attempt to keep the amplifier input very near zero volts, thus making it a virtual ground.

Incoming charge is collected on the input plate side of the feedback capacitor, Cf This will cause some voltage shift at the input of A1 and, as a result of its inverting gain, a much larger inverted shift at the output Negative feedback action through Cf will restore the input to its normal zero-volt level. The magnitude of the output voltage shift is directly proportional to the amount of charge received, V = Q/Cf

It is very important that the amplifier have a very short rise time with respect to the incoming pulses, otherwise the virtual ground cannot be maintained, and charge may be diverted to ground through shunt protective devices or other circuit elements.

When only the feedback capacitor, Cf, is used, the circuit has become a charge-to-voltage converter. However, this configuration is limited by the ultimate saturation of its output as more and more charge is accumulated. For this reason the negative feedback resistor, Rf, has been added to discharge Cf between incoming input pulses The parallel combination of these two components, Cf and Rf, forms the clipping-time constant of the preamplifier, T = Rf X Cf This clipping time also serves to provide the best pulse shape required by subsequent circuits

The output of A1 is fed through C5 to Ql, which provides two outputs of opposite polarity by means of the output driver amplifiers. These driver amplifiers consist of parallel-connected Q2—Q3 and Q4—Q5 transistors that function as emitter followers to provide low output impedance suitable for driving any reasonable length of terminal coaxial cable