Ответ: "Откуда колбочка знает?"
Andrey Frenkel сказал(а):
показывает вероятность запуска
Несмотря на кажущуюся бардачность, в физике есть своя терминологическая дисциплина. И я, например, не могу путать скалярные величины и векторные, не могу вольно мешать "вероятность" и "матожидание". Вероятность - это функция вида P(λ), а "количество прореагировавших фотонов" - уже скаляр (интеграл от spectral product P(λ)n(λ)dλ).
Приведу строгие определения по детекторам из радиометрии:
1.6 PHYSICAL DETECTORS OF RADIANT ENERGY
In many investigations in color science, the need arises to make physical measurements of the radiant power involved. Such measurements are made by radiometers or photometers of various designs, all of which employ a physical detector of radiant energy. There are several types of detector commercially available, but many separate performance characteristics must be considered when selecting one that is most appropriate for a particular application. The important performance characteristics of a physical detector relate to the following properties of usual concern:
- Spectral range of response to incident radiant power.
- Magnitude of response, distinguishable from noise, to a given level of incident radiant power.
- Capability of absorbing the incident radiant power without damage or severe aging effects.
- Functional relation between response and incident radiant power (linearity).
- Speed of response.
Other considerations of importance are those of cost and the kind of ancillary apparatus (optics, electronic circuitry, cooling) needed to operate a particular detector.
There are two main groups of physical detectors: thermal detectors and photon (or quantum) detectors. Within each group there are several types; those of some interest to photometrists, colorimetrists, and visual scientists are briefly described below. In describing the performance characteristics of these detectors, certain technical terms are commonly used (see, for example, Reed, 1971; Grum and Becherer, 1979).
(a) Spectral Responsivity. The spectral responsivity, denoted by R(λ), of a physical detector is the ratio of the detector output, the signal, measured in amperes or volts, to the incident spectral radiant power, measured in watts. If the incident radiant power is chopped (modulated), R(λ) is measured in terms of the rms amplitudes of the fundamental frequency components. It is not affected by the bandwidth, nor by the chopping frequency, if the latter is low.
The term "spectral sensitivity" has often been used for the same quantity, but is no longer recommended.
(b) Quantum Efficiency. The quantum efficiency, denoted by η, of a photon detector is the number of electrons released per incident photon. Quantum efficiencies for different photon detectors vary between 0.001 and 0.35 (approximately) and depend upon wavelength.
(с) Signal-to-Noise Ratio. The signal-to-noise ratio is defined as the ratio of the detector output, the rms signal, measured in amperes or volts, to the rms value of the noise, also measured in amperes or volts. The signal-to-noise ratio is a quantity without a dimension. Noise in a detector expresses itself as detector signals whose small values distribute randomly within a narrow range. Noise originates at one or more points in the chain of events that govern the detection process and can be caused, for example, by the incident radiant energy to be measured, the detector temperature, the detector bias voltage (if applicable), and the signal-modulation frequency (if applicable).
(d) Noise Equivalent Power (NEP). The noise equivalent power is the rms value of the sinusoidally modulated incident radiant power, measured in watts, required to produce a signal-to-noise ratio equal to unity; that is, SNR = 1. The quantity is a function of wavelength and its determination depends on the modulation frequency and the noise bandwidth. A greater NEP means a poorer detector performance.
(e) Detectivity (D, D*). The detectivity D of a physical detector is defined as the reciprocal of the noise equivalent power; that is, D = 1/NEP. A more commonly used quantity is the normalized detectivity D*, which is defined as the detectivity D multiplied by the square root of the active area of the detector and the square root of the noise bandwidth. This normalization factor was chosen because the NEP of many detectors proves to be proportional to the square roots of both the active area and the noise bandwidth. Both D and D* are functions of wavelength and their determinations depend on the conditions of measurement, such as the modulation frequency used.
(f) Response Time. The response time (or time constant), denoted by
τ, is the time, measured in seconds, required for the detector output to rise from a stated low value to a stated high value when a steady level of radiant energy is instantaneously incident on the active area of the detector. For detectors whose output rises exponentially with time, the response time is the time at which the output has reached a value equal to (1 — 1/e) = 0.632 of its final value. After a time of 5
τ, the detector output will be approximately 99% of its final value. At high chopping frequency f, the responsivity R
f of the detector decreases significantly in comparison with the dc value R
0 of the responsivity. The following relation holds...
Обратите внимание, здесь термин "спектральная чувствительность" настойчиво предлагают заменить на spectral responsivity. Впрочем, это не помогло, "spectral sensitivity" используется повсеместно.
Вместе с тем, степень усиления в палочке намного выше, поэтому колбочковый механизм (по логике) должен быть либо очень сходным, либо даже проще (меньше каскадов).