73
PART
II
Understanding the Technology
In this part of the book we turn our attention to a more in-depth look at particle
counters, from the basic construction of the instrument to the features of the complete
particle counting system. Particle counters are presented generically in this part,
while in Part III the actual manufacturer’s product offerings are presented. The
chapters for Parts II and III are designed to run in parallel, to allow for easy cross-
reference.
L1306/frame/pt02 Page 73 Friday, June 23, 2000 1:54 PM
© 2001 by CRC Press LLC
75
CHAPTER
6
Specifications
All instruments are designed for a specific task. Specifications are the measure
of how well the instrument performs the various parts of that task. Some specifica-
tions are common to all types of instruments (dimensions, weight, power require-
ments, etc.) and are easily understood. Specialized instruments have specifications
that are unique to the task for which they are designed. In many cases, common
specification terms such as resolution and sensitivity must be understood in relation
to the unique properties of the application.
In this section, the primary specifications employed for evaluating particle
counter performance are defined and discussed. Understanding of these few speci-
fication parameters is essential to making informed decisions about particle counters.
If these are not clear at first reading, readers should refer back to this section as they
continue through the book. It will become evident that these specification parameters
are interrelated, and will only be understood completely when the whole picture of
particle counting is clear.
A. SENSITIVITY
Sensitivity is the smallest measurable amount that an instrument can detect. In the
case of the particle counter, this refers to the smallest particle size that can be reliably
detected. The general rule in the industry is that a particle can be “reliably” detected
and measured if it produces a signal with a minimum 2 to 1 signal-to-noise ratio.
B. SIGNAL-TO-NOISE RATIO
All electronic instruments are subject to various forms of “noise,” or random,
meaningless electrical signals. The term
noise
comes from the analogy to sound
levels in a given space. Noise is meaningless sound, and every space where sound
can exist will have some level of noise present. Our ability to discern specific sounds
L1306/frame/pt02 Page 75 Friday, June 23, 2000 1:54 PM
© 2001 by CRC Press LLC
76 A PRACTICAL GUIDE TO PARTICLE COUNTING
depends upon those sounds being at a higher level than the noise present. The amount
by which the specific sound exceeds the noise level is called the signal-to-noise ratio.
The sound found between stations on a radio is a good example of electronic
noise. Even the highest-fidelity audio system produces some of this type of noise,
which becomes audible as the volume is increased. Every type of electronic device
is affected by this phenomenon. In the case of the particle counter, noise is produced
by the electronic detector circuit, the laser light source, the power supply, and is
induced by other devices. The imperfections in the structure of the flow cell
windows cause minute scattering and blurring of the laser light source. The com-
bination of all these factors results in a noise level that can be observed and
measured with an oscilloscope. (An oscilloscope displays electronic signals in a
manner similar to a television.) See Figure 6.1 for a representation of noise as
displayed on an oscilloscope.
This noise level is measured with no particles in the water flowing through the
particle counter sensor. (This is not exactly true. There are millions of particles well
below the detection limit of the particle counter sensor. Some of these contribute to
the noise level in a small way.) Once the noise level has been determined, particles
of a known size can be sent through the particle counter sensor, and measured. The
resulting amplitude produced by the particles is compared with the noise amplitude
to determine the signal-to-noise ratio. See Figure 6.2.
Figure 6.1
Noise on oscilloscope.
L1306/frame/pt02 Page 76 Friday, June 23, 2000 1:54 PM
© 2001 by CRC Press LLC
SPECIFICATIONS 77
As mentioned is Section A, the smallest particle that can be reliably measured
must produce a signal amplitude at least twice that of the noise, or a 2 to 1 signal-
to-noise ratio.
C. RESOLUTION
This term refers to the degree to which an instrument can distinguish between
differences in the object of measurement. In particle counting, resolution refers to
the accuracy with which the sensor can distinguish and measure differences in
particle size. This is stated in terms of percentage. A (+ or –) 10% resolution error
would mean that particles ranging in size from 4.5 to 5.5 µm could be measured as
5 µm in size.
Resolution measurement is a somewhat sticky subject in the particle counter
industry. It will be discussed in further detail in relation to calibration and other issues.
D. COINCIDENCE
Coincidence was discussed in Chapter 5 in relation to sample dilution. Basically,
it is the “coincidence” of two or more particles in the sensor view volume at the
same time, which is a function of the concentration of particles in the sample. There
are other factors that affect this, but they are beyond the scope of this book. Coin-
Figure 6.2
Pulse signal-to-noise measurement.
L1306/frame/pt02 Page 77 Friday, June 23, 2000 1:54 PM
© 2001 by CRC Press LLC
78 A PRACTICAL GUIDE TO PARTICLE COUNTING
cidence is usually presented as the number of particles for which the measured
counts fall within 10% of the actual number of particles present. This number is the
total number of particles per milliliter that are measurable by the sensor.
E. SIZING RANGE
This measure may appear under various names, but it is basically the range of
measurable particles that can be sized and counted by the particle counter. It extends
from the minimum size particle (usually 2 µm) up to 400 µm or so. The upper limit
is generally not too important, as most particles of interest will fall well below 50 µm.
F. SAMPLE FLOW RANGE
All particle counter data must be collected for a known volume of sample, which
usually means the flow rate must remain constant or be measured continuously. The
specified sample flow range refers to the flow rates at which the particle counter
sensor can measure particle size accurately. If the particles are flowing through the
sensor at too high or too low a velocity, they will not be sized properly by the
detector electronics. The reasons for this are covered in Chapter 7.
Figure 6.3
Resolution error as a function of cell width.
Cell Illumination
L1306/frame/pt02 Page 78 Friday, June 23, 2000 1:54 PM
© 2001 by CRC Press LLC
SPECIFICATIONS 79
G. FLOW CELL DIMENSIONS
The flow cell is the pathway through which the sample must flow. A larger flow
cell will clog less often and be easier to maintain. On the other hand, a smaller flow
cell will permit a higher concentration sample to be measured with less coincidence
error. A trade-off is necessary.
Resolution is also a function of flow cell size. The laser light source will not
maintain an even intensity across the entire flow cell, and this error grows with the
width of the flow cell. Particles passing through the edges of the beam will not
produce the same amplitude pulse as particles of the same size passing through the
center. See Figure 6.3.
H. VOLUMETRIC
The term
volumetric
refers to the coverage area of the laser light source. If the
beam covers the entire width of the flow cell, the sensor is volumetric. If only a
fraction of the width is covered, then it is non-volumetric, which is sometimes called
in situ. An in situ arrangement allows for a much higher concentration of particles
to be measured because it reduces the effective volume of the flow cell. The drawback
is that resolution is diminished because particles may partially pass through the
beam and thus be undersized. Figure 6.4 provides an illustration of this.
Figure 6.4
Volumetric and in situ sensors.
Volumetric
In- Situ
Cell Illumination
L1306/frame/pt02 Page 79 Friday, June 23, 2000 1:54 PM
© 2001 by CRC Press LLC