Page 83 - Photodetection and Measurement - Maximizing Performance in Optical Systems

Fundamental Noise Basics and Calculations

76 Chapter Three

2x
2N3904
BPW34
Variable Red AlGaAs
supply 4.7k 330 LED V 1M To scope,
0–12V sp. analyzer
2x -
100μF + LMC7101
Use micro-manipulator 2x
to balance LED beam on 10 Ohm
photodiodes for Interleaved
zero DC output
array

Figure 3.27 Two-photodiode shot-noise generator. By balancing the two pho-
tocurrents the opamp output is at zero volts but shows the added noise power
of the two photocurrents. An illuminated interleaved isolated diode array might
be a useful noise source.

the opamp. The noise power is two times that of a single diode ( 2 times the
noise voltage). By varying the LED brightness it was possible to vary the noise
almost from the noise level of the 1MW resistor (-124.9dBm in 1Hz) up to the
shot noise current of the two photodiodes passing 200mA (-85.9dBm), coupled
though the 1MW load. With more or brighter LEDs, or use of an optical con-
figuration with improved coupling to the photodiodes, even higher photo-
currents and noise powers should be achievable. The actual photocurrent
required for calibration can be obtained from a sensitive voltage measurement
across small load resistors connected in series with the free photodiode leads
(Fig. 3.27). Perhaps an array of interdigitated photodiodes (inset of Fig. 3.27)
would make a good, practical, single-component noise generator. It might
operate with adequate balance even without the use of manual balancing. Of
course if wide-band white noise operation is needed, some effort should be spent
to provide the receiver with adequate bandwidth, using all the techniques dis-
cussed in Chap. 2.

3.13 Summary
With its fundamental and its avoidable contributions, the difficulties of obtain-
ing meaningful measurements at low levels, disagreement about what kinds of
noise should appear under what conditions, and the need to calculate in com-
plicated circuits with marked frequency dependence, noise is a complex subject.
It is even possible that new measurements can show new effects not seen before.
Nevertheless, noise is so central to the performance of an instrument or exper-
iment that it is essential that it be estimated at the start, calculated in detail
as the design progresses, and measured in practice as soon as hardware is avail-
able. When just 3dB more power from an LED, laser, or plasma source often

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.