Page 90 - Photodetection and Measurement - Maximizing Performance in Optical Systems

Interlude: Alternative Circuits and Detection Techniques

Interlude: Alternative Circuits and Detection Techniques 83

The paper by Liu et al. (1993) describes experiments made with such a system
of highly stable measurements of optical absorption for analytical chemistry.
Peak-to-peak noise levels as low as 3mAU were demonstrated (AU = Absorbance
Unit; 1mAU = 2.3·10 -6 intensity change). Hard-wired logic circuitry was used
to sequence the MOSFET switches. An alternative would be use a small micro-
processor such as a PIC, or even the more convenient BASIC Stamp from
Parallax Corporation. This has a built-in interpreter for a high-level (BASIC)
program and can very simply be arranged to drive the three switches using
the Stamp’s “high”, “low”, and “pause” commands. For low-speed applications
with minimum pulse times of several milliseconds and hence a data rate of a
few measurement per second the rapid in-circuit reprogrammability of the
Stamp is ideal. If higher speed is required a compiler is more suitable (such as
one of the several packages provided for programming PIC microprocessors).
Although more complex than a continuous-time transimpedance configura-
tion, the current-integration technique has a number of advantages. We have
mentioned the ability to easily vary sensitivity via the integration period. With
digital control and a crystal-controlled clock the integration period can be
chosen to be precisely 1/50s or 1/60s, giving good suppression of signals at the
50/60Hz line frequencies.
By using a capacitor instead of a transimpedance resistor, we can in princi-
ple avoid thermal noise. Purely reactive components have no thermal noise,
which is only generated by the real part of the component’s impedance, such
as lead resistance, leakage, and inductive eddy current losses. Capacitive tran-
simpedance amplifiers have also been widely investigated for reading out active
pixel CMOS image sensors. In this way very small photocurrents can be deter-
mined with good performance. See for example Fowler et al. (2001).

4.4 Forward-Biased Photodiode Detection
All the circuits described and analysed in this book so far operate the photo-
diode unbiased or reverse biased. However, this is not absolutely necessary.
Figure 4.4 shows the photodiode characteristic curves under various conditions
of illumination, originally shown in Fig. 1.9. While normal photodetection takes
place in the third quadrant, and the fourth quadrant illustrates solar cell oper-
ation, photodetection in the first quadrant is equally possible.
As previously demonstrated, every junction diode exhibits some photosensi-
tivity, including the common LEDs. Half-duplex optical fiber communication
systems have been built in which an LED was used sequentially as a source and
photodetector. Figure 4.5 shows one application of a self-detecting LED as a
sensor for an encoder or tachometer. The LED is operated under strong forward
bias from a load resistor R L. The emitted light illuminates a small region of a
retroreflective code plate with nonreflecting stripes. A small fraction of the
emitted light is reflected back into the LED chip where it creates carriers in the
usual way. Although the wavelengths of peak emission and peak photodetection
sensitivity for the LED do not occur at the same point, the overlap of the two

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.