Charge Collection Time Research Articles

A xenon ionization detector for digital radiography

Xenon gas x-ray detectors have been used successfully in CT scanners; however, they have been found to be unsuitable for digital radiography. We have designed and built a new type of xenon x-ray detector array and tested its suitability for digital radiography. The detector consists of two parallel plates separated by a 0.5-mm gap, filled with xenon gas at a pressure of about 30 atm. One of the plates is the high-voltage electrode, while the other is a circuit board etched to form an array of metal collector strips focused on the x-ray source. Since there are no metal septa separating the individual detector elements, the dose efficiency of the detector is high, but image degradation will occur due to cross-talk between detector elements. Measurements of the cross-talk show that about an 18% reduction in contrast will occur, when a low contrast object, subtending one detector element, is imaged. We have also measured a detector MTF of 14% at 2 lp/mm, a signal of 10 pC for a 1-mR x-ray exposure at the detector entrance, a 6% nonlinearity in the detector signal over about 3 orders of magnitude in x-ray exposure, and a charge collection time (time response) of about 0.1 ms. From these results it is concluded that this new detector design is feasible for digital radiography.

Status at Harwell of Opto-Electronic and Time-Variant Signal Processing for High Performance Nuclear Spectrometry with Semiconductor Detectors

A review of the advantages and problems associated with these signal processing methods is followed by a summary of recent developments at Harwell. A new spectrometry system for use with solid state charged particle and gamma ray detectors which can handle input pulse rates in excess of 106 per second with a high throughput rate and almost complete insensitivity to detector charge collection time will be described. The paper will discuss the benefits obtained in various applications and the remaining limitations.

Msham - A Multi-Hit Sample and Hold Multiplexer

The MSHAM is a single-width CAMAC module intended to be used for dE/dx or Z-position measurements, with a density of 16 analog channels. It is designed to record up to four hits/event per channel but the design can be easily adapted to eight hits. The charge collection time interval allowed per hit is externally controlled in the range of 50-500 ns, according to the requirements of the experiment. Besides the electrical performance of the MSHAM, i.e., linearity, noise, crosstalk, etc., the goal was to design multi-function circuits and high density packaging in order to achieve low cost per channel.

A novel pulse processing system for HgI 2 detectors

A pulse processing system for HgI 2 detectors is described where, for each detector pulse, the charge collection time is measured. The gain of the shaping amplifier is changed according to the charged collection time measurement. In practice this method leads to a shaping amplifier having positive ballistic deficit. This produces a marked improvement in energy resolution.

High Counting Rate Spectrometer

A signal processing system is proposed, capable of accepting a counting rate which is limited only by the charge collection time of the detector and by the time of the sample-hold circuits employed. This system has not been assembled. The concept is presented in order to stimulate interest and development by others.

Point-spread functions and optimum filters for position sensitive detectors that use pulse-shape modulation and zero-crossing demodulation

To enable the later determination of nearly optimum, realizable position decoding filters for position sensitive detectors as well as to enable valid comparisons of various position sensitive detectors, we introduce in this work a productive, high-quality approximation to the point-spread function for position sensitive detectors that use pulse-shape modulation and crossover-time demodulation. This approximation is the result of a theoretical calculation and is determined as a general function of the input signal and rms noise at the input to each of the two crossover detectors. It is precisely applicable to position sensitive detectors that use any type of transmission line encoding. The effects of random variables, such as charge collection time, are included in the calculation, but the effects of correlated noise in the two channels are not included. In addition, for the broad class of position sensitive detectors that have electronic and thermal noise as the dominant causes of position uncertainty, general equations, based on our approximate point-spread function, are given for the optimum filter impulse responses, the optimum pulse shapes at the inputs to the crossover detectors, and the associated minimum positional full-width half-maximum (FWHM). General equations are also given for positional FWHM of electronic-noise-limited detectors that use nonoptimum filters. Finally, for electronic-noise-limited detectors that use an RC-transmission line encoder that is terminated in its characteristic impedance, a plot is given of the optimum pulse shape at the input to the crossover detectors.

Continuum reduction in the spectral response of Ge(Li) pair spectrometers

A system designed to improve the spectral response of Ge(Li)-Nal(Tl) triple-coincidence pair spectrometers is presented. The method employed is based upon suppression of escaping brems-strahlung and selection of charge-collection times. Results are discussed for the spectra from thermal-neutron capture in nickel and nitrogen. Typically, a factor of 6 improvement in signal to background was achieved in the region of 1.5 to 4 MeV. When the time-discrimination criterion was applied to a single-detector system an average improvement factor of 2.3 was achieved for γ-rays in the same energy range. A quantitative examination of rejected and residual components was conducted for purposes of both identification and characterization of the various spectral constituents.

Filters for use in pulse-area sensing systems

Variation of the charge collection time degrades the resolution of both peak-amplitude and pulse-area measuring systems. However, the specific amount of degradation depends upon the particular measuring system used and upon the filter network chosen. The results discussed in an earlier paper are used in this study to demonstrate the existance of a general class of filter networks suitable for area sensing. The specific magnitude of resolution error for evaluating practical systems employing single RC differentiation and RC-CR filters is available from the included graphs. A comparison between peak-amplitude and pulse-area sensing techniques based upon equal resolution errors is included for the RC, RC-CR, and delay-line filters. This comparison demonstrates the superiority of particular filters for pulse shaping within various range of collection times and filter time constants.

Computer Algorithms and Detector Electronics for the Transmission X-Ray Tomography

This paper is addressed to the problem associated with transmission tomography and its implications in solving large numbers of simultaneous equations and associated parameters, such as multimillion weighting factors. Mathematical approaches to this problem in several directions are briefly reviewed and a technique (linear superposition technique with compensation) developed for this purpose, along with results obtained with this method is presented. Another important aspect of the multiple scanning tomography is counting statistics which requires the count rate be in the order of 5 MHz or more for the pulse counting to have statistical accuracy of 0.5% or better. Several alternative detectors besides conventional Sodium Iodine (NaI(Tl)) detector, namely, plastic scintillator (with enhanced photoelectric absorbtion by, for instance, tin loading) and various semiconductor detectors, such as CdTe and possible HgI2 are under investigation. Plastic scintillators will naturally be a suitable candidate owing to their fast light decay characteristics. A tin loaded plastic scintillator (HE 140 type) is under investigation and some of the preliminary results are reported. With the semiconductor detectors, the preliminary charge collection time observations and calculations indicate that these new exotic detectors (CdTe and HgI2) can be applicable to the high rate operations primarily by virtue of their short charge collection times and moderately good energy resolutions.

A technique for correction of pulse heights from thick silicon particle detectors

When detecting protons with thick silicon detectors, it is sometimes desirable to use short pulse-shaping time constants in the associated amplifier. As the charge collection time varies for electrons and holes along the proton track, an energy dependent signal loss is observed when pulse shaping and charge collection times are comparable. A method for calculating the resulting nonlinearities in the proton energy scale is presented and corrections applied to the energy calibration are compared with experiment.

A Risetime Discriminator for Low Level Tritium Counting

This paper describes the design of a risetime discriminator circuit (RTD), with associated anticoincidence/coincidence (AC/C) logic and its application to the counting of low-activity tritium samples in gas proportional counter systems. The circuitry, packaged in a double-width NIM, extracts signal risetime information which is used in conjunction with the output of an annular guard counter to electronically reject background counts due to high energy charged particles (?-mesons), and gamma-ray photons. The use of the guard and AC/C logic without RTD is effective in reducing the background due to charged particles because of the high probability of coincidental detection by the guard and proportional counters. However, the reduction of backgromd related to gamma-ray photons by simple AC/C techniques is ineffective because of the small probability of a coincidence event occurring in the two counters. The RTD circuit utilizes the property that gamma photons and high energy pmesons produce ionization tracks in the proportional counter that are typically longer and have lower specific charge density than those tracks produced by low-energy (tritium) beta particles. Thus, these events have longer charge collection times and result in slower signal risetimes at the output of a charge-sensitive (integrating) preamplifier. The RTD circuit differentiates the wide-band preamplifier signal which produces a pulse with amplitude proportional to the signal risetime. The derivative is then peak-detected, stretched, delayed and inverted. An amplitude comparison is then made between the stretched derivative and the linear input signal using a linear summation amplifier.

A Computer Study of Time Resolutions and Charge Collection Times of Ge Gamma-Ray Detectors with Currently Employed Time Pick-Off Techniques

An attempt to quantitatively calculate the time resolutions from different combinations of Ge gamma-ray detectors, filters, and time pick-off techniques, is described. The study is based on charge collection time calculations previously studied, optimizations of filters, and the choice of parameters in the different time pick-off techniques. Results of the time resolution simulations with a 3 cc planar detector and a 5.6 cc coaxial detector of the Ge (Li) type, are presented and are in good agreement with previously obtained experimental and theoretical results. A brief discussion of the most common trigger modes (and related time jitters and time walks) particularly in reference to the Ge gamma-ray detectors, is also given.

Trapezoidal filtering of signals from large germanium detectors at high rates

A filter with a trapezoidal weighting function is the optimum one when both noise and variations in the charge collection time are present. A trapezoidal filter was realized by a time-variant (gated filter) system. This paper presents the results obtained with large germanium detectors at high rates and higher energies, showing that trapezoidal filtering is essential for good resolution in this case.

Timing with high purity germanium coaxial detector

The timing properties of a high purity Ge coaxial detector are studied and compared with those of a Li-drifted detector of similar size. The 1/ r field distribution in a Li-drifted device is modified by fixed space charge in the high purity Ge coaxial detector and this results in a higher and more uniform field distribution in a large portion of the detector volume. The effect is felt particularly in an improved hole collection due to their velocity not being saturated at the 100–500 V/mm fields. It is found that hole mobility obtainable in such fields is reduced by a factor of approximately 8 from the value normally specified at low fields. An immediate advantage of high purity Ge coaxial detectors is found, therefore, in a smaller spread in their charge collection times from which an improved time resolution and smaller energy dependent time walk results, particularly at low energies.

Trapezoidal Filtering of Signals from Large Germanium Detectors at High Rates

A filter with a trapezoidal weighting function is the optimum one when both noise and variations in the charge collection time are present, A trapezoidal filter was realized by a time-variant (gated filter) system. This paper presents the results obtained with large germanium detectors at high rates and higher energies, showing that trapezoidal filtering is essential for good resolution in this case.

Timing System for Ge(Li)-Detectors

A timing system for coincidence and nuclear lifetime measurements with Ge(Li)-detectors is described. The timing discriminator, based on the ARC-principle, is designed using integrated circuits. Uncompensated pulses from low energy events near the threshold of the timing discriminator are eliminated by a special internal coincidence circuit. Output pulses caused by noise, are excluded by duration discrimination. The detector signals are amplified by a preamplifier and a fast amplifier with 2.5 and 3.5 nsec risetimes,respectively. The time resolution has been obtained 15.6 , 2.7 and 1.1 nsec FWHM for energies 50, 511 and 1330 keV, respectively, by using of a planar detector (1.9 cm3) with a charge collection time of 40 nsec. The differential peak shift in a range from 10 to 1330 keV was 1.2 nsec.

A New Germanium Photodiode with Extended Long-Wavelength Response

A new germanium photodiode having an extremely uniform quantum efficiency from the visible to 1.65 micro. has been developed. The device consists of a lithium-drifted junction in which the light enters in a, direction parallel to the junction, thereby allowing absorption path lengths of several centimeters, if desired, with essentially zero dead layer. Sensitive areas up to about 10 mm in one dimension and several centimeters in the other dimension are possible. Typical characteristics for a device 2 cm long having a sensitive area 7 mm x 7 mm are: diode capacitance, 3 pF; charge collection time, 75% of charge collected in 25 nsec at 500 V; measured NEP, 3.7 x 10(-15) WHz(-(1/2)) at 1.625 microm and 77 K for frequencies up to 30 kHz, and 5.5 x 10(-13) WHz(-(1/2)) at 1.625 micron and 195 K. for frequencies up to 40 kHz. At low frequencies, the detectivity can be background limited by 300-K blackbody radiation. The device must be used and stored at reduced temperatures. A convenient cryostat, capable of maintaining the device at or near 77 K for 150 h without refilling, is described.

Low Noise Preamplifier for Simultaneous High Resolution Energy & Time Measurement with Semiconductor Detectors

A new charge sensitive preamplifier permits state of the art timing with semiconductor detectors without compromising energy resolution. The compromises that are involved in previous methods of obtaining amplitude and time information are avoided by using a single input FET followed by fast, low noise circuitry. With the same input transistor the noise in the "energy" channel is the same to within a few per cent as for the best room temperature preamplifiers. There are two timing outputs. The faster of these is voltage sensitive and has a rise time of 2 nsec. The slower output, which is obtained from the output of the charge sensitive loop, has a rise time of less than 5 nsec for external input capacities up to 50 pF with 0.5 pF feedback capacitor. The leading edge response can be adjusted to approach an exponential function with a time constant as low as 6 nsec. This property is desirable when timing with thick detectors that have a fluctuating charge collection time, where it is necessary to trigger near the base line.

A method to correct for charged-particle signal loss in lithium-drifted detectors

In semiconductor detectors, the charge collection time will vary with the physical point of deposition of energy within the intrinsic region of the detector. This leads to a geometrical dependence of the signal risetime so that if the signal risetime is not negligible relative to the amplifier differentiation time constant, varying signal loss will occur. For charged-particle spectroscopy, this loss will vary with incident particle kinetic energy. An energy-scale correction method is described for charged-particle spectroscopy in detectors of the p-i-n type. This method reduces energy scale distortions by about a factor of ten.

Time measurements with Ge(Li) detectors

A method of obtaining improved time resolution from coincidence measurements utilizing large volume Ge(Li) detectors is presented. The analysis indicates that a time resolution of the order of 1 ns is obtainable if a preamplifier with a short rise time is used. An experimental time resolution of 3.8 ns is obtained using a preamplifier with a rise time considerably longer than the charge collection time constant of a Ge(Li) detector. These results all apply to detection of gamma rays having a large range of energies detected with a large volume coaxial detector.