Detectors deadtime

Nelms SM, Quetel CR, ProhaskaT, Vogel J and Taylor PDP (2001) Evaluation of detector deadtime calculation models for ICP-MS. J Anal At Spec-trom 16 333-338. 

The low MW power levels conuuonly employed in TREPR spectroscopy do not require any precautions to avoid detector overload and, therefore, the fiill time development of the transient magnetization is obtained undiminished by any MW detection deadtime. (3) Standard CW EPR equipment can be used for TREPR requiring only moderate efforts to adapt the MW detection part of the spectrometer for the observation of the transient response to a pulsed light excitation with high time resolution. (4) TREPR spectroscopy proved to be a suitable teclmique for observing a variety of spin coherence phenomena, such as transient nutations [16], quantum beats [17] and nuclear modulations [18], that have been usefi.il to interpret EPR data on light-mduced spm-correlated radical pairs. 

The large area, 2m x 2m, of AGATE has lead to the selection of drift chambers for the tracking detector, rather than the spark chambers used in EGRET. Drift chambers have fewer wires and much less deadtime per event. The power per wire is low enough to use many layers in order to reduce the multiple scattering of the electron and positron before the gamma-ray direction can be measured. 

Since deadtimes in this type of spectrometer are quite long ( 60 fis), the system must normally operate with deadtime losses in the 10 to 60% range. Consequently, most multichannel analyzers are equipped with an electronic means of deadtime correction, such that the observed spectrum represents the true number of photons arriving at the detector during the period of data accumulation. In addition to the ability to display the spectrum on a cathode-ray tube or television monitor, the analyzer can usually drive an X-Y plotter to produce a permanent copy. Alternatively, the contents of the analyzer memory can be printed as the number of counts in each channel, listed by channel number. Most quantitative fluorescence spectrometers include a personal computer with approximately 2-6 megabytes of memory plus some form of mass storage. In such a system the computer may control specimen presentation, the excitation conditions, and data accumulation in the multichannel analyzer. At the end of data acquisition for each specimen the computer analyzes the spectrum in the multichannel analyzer, computes the raw element intensities, corrects for interelement effects, and computes the concentration of each element. 

It should be noted that n/t is the calculated true counting rate at the detector (before deadtime losses), which is defined as It. Consequently,... 

Since the k different series of photons were assumed to be independent random variables, Eqs. (4.78) to (4.80) show that the p can be considered as the population mean rates for k different energy components of a spectrum, all being counted simultaneously by the same detector. Each component has its own population mean rate pi and variance which can be estimated in spite of the presence of the other components. In fact, with zero deadtime, or with an ideal livetime clock, the population parameters Pi and for each component are independent of whether or not the other components are present. The population parameters of each component can be estimated from the measurement of the n counts recorded in the time t. The estimates of the population mean count and variance for the i-th component are... 

A number of methods are available for calculating the expected mean rate r at which the deadtime intervals are generated and counted, given that the arrival of photons at the detector is described by Eq. (4.44) with an expected mean rate p [14,43]. Only the results will be quoted here. 

The most easily applied technique for measuring the effective deadtime for a wavelength-dispersive x-ray fluorescence spectrometer assumes that the expected mean counting rate of x-ray photons at the detector is proportional to the x-ray tube current i. A stable specimen that will generate the characteristic lines and the pulse height spectrum of interest is placed in the fluorescence spectrometer. All... 

The major difficulty encountered in using the x-ray tube current with the graphical method of measuring the deadtime is the failure to achieve proportionality between the x-ray tube current and the true counting rate at the detector. This problem may arise because the instrument does not permit accurate reading or setting of the tube current. More fundamentally, one cannot expect the proportionality to hold... 

With a real-time clock the average number of counts recorded in a real time t is defined by Eqs. (4.97) and (4.100), depending on whether the deadtime is nonpar-alyzable or paralyzable. The expected number of measured counts nm is related to the expected number of photons at the detector nt by... 

Note that the variance in the measured counts differs from nm by a factor depending on the product of the expected mean input rate at the detector, p, and the deadtime td. This product is referred to as the normalized input rate. The relative variance <7n /nm from Eqs. (4.133) and (4.134) is plotted as a function of the normalized input rate ptd in Fig. 4.53. In general, it can be seen that an < For deadtime losses less than 20% it is reasonable to use the approximation orn v > since this estimate of the standard deviation will be in error by an amount less than 22%. For normalized input rates above ptd = 0.2 it is important to use the more rigorous expression for cTn . 

The effect of deadtime on the random error leads to the question of whether there is a counting rate that provides the best precision in estimating the true counting rate at the detector. The relative precision in the calculated true counting rate arising from the random error is given by Eqs. (4.137) and (4.138) and is plotted in Fig. (4.55). 

The probability of observing a single livetime interval of length Xi between two deadtime intervals can be computed as follows. As a result of condition (1) in Sec. 4.5.1 and Eq. (4.44), the probability that no photons will arrive at the detector in the time interval of length t, starting with t< = 0 at the end of a previous deadtime interval, is... 

The use of a remote-echo detector allows r values shorter flian the spectrometer deadtime to be employed [55]. This is important in two-pulse ESEEM experiments where the deadtime prevents the signal for times r < from being recorded. Also in the deadtime-free four-pulse experiments described in 3.3, a small T value is often needed to avoid blind spots. Bhnd spots are a particular concern for flie measurement of proton spectra at X-band, where flie signals typically extend from 5 to 25 MHz, and with a r = 100 ns blind spots occur at nh = 0, 10, 20,... MHz. 

A number of experimental parameters have to be optimized in order to obtain the best SPECT image. These include attenuation, scatter, linearity of detector response, spatial resolution of the collimator and camera, system sensitivity, minimization of mechanical movements, image slice thickness, reconstruction matrix size and filter methods, sampling intervals and system deadtime. In a hospital, calibrating and monitoring these functions are usually performed by a Certified Nuclear Medicine Technician or a medical physicist. 

Critical parameters for selection of an appropriate detector are efficiency, energy resolution and deadtime, i.e. the pulse processing time of the detector. A measure for the energy resolution is the full width of half maximum (FWHM). 

Fig. 7. Schematic diagram of the events during a two-pulse Hahn echo sequence. Re and Im refer to the so-called real and imaginary NMR signal components, that is, the two channels of the quadrature phase-sensitive detector. Ideally, 0, = 90° and 02 1 0°, with phase cycling of and 02- dashed regions of the NMR signals following the pulses represent the deadtime of the receiver. From Ranee and Byrd (1983).

---