GORMAN-RUPP VG3B3-B, Инструкция по установке, эксплуатации и обслуживанию

Страница: 787 / 1180

laser turbidity measurement will begin to increase in response to a
particle event. In addition, the RSD parameter has been shown to be
more sensitive to a turbidity event in addition to serving as a precursor
event to a turbidity spike. The Hach Technical Information bulletin
"Introduction to Laser Nephelometry: An Alternative to Conventional
Particulate Analysis Methods", Literature piece 7044, Appendix B
provides more information on the application and use of the RSD
parameter.
The sensitivity of the RSD parameter is directly related to the instrument
design of the FT660 sc nephelometer. This is due to the optical creation
of a very small analysis volume within the turbidity sensor. (The analysis
or view volume is that volume of sample within the turbidimeter body that
is in view by the detector window.) This volume is small but well defined
by the optical design. This volume also contains a high energy density
from the incident light beam, which can easily be scattered by a single or
low number of particles. When a particle passes through the view
volume, there is a rapid increase in the scattered light signal while that
particle is in the view volume. When the particle passes out of the view
volume, the signal decreases rapidly. The change in signal is infrequent
and this instability of signal is quantified by the RSD parameter.
The RSD value is calculated as the standard deviation divided by the
mean for a given set of measurements. The result is multiplied by
100 and is expressed as a percent. Equation 1 provides the calculation
used to determine the RSD value:
RSD = (standard deviation
n
÷ mean
n
) × 100
Where n = number of measurements used
The RSD calculation is accomplished through a process of evaluating
the most recent seven displayed turbidity measurements. From the
seven measurements, the standard deviation and the average are
calculated. The value is then displayed on the secondary measurement
line of the SC controller. When a new turbidity measurement is
displayed, the value replaces the oldest of the seven measurements and
the RSD is recalculated and displayed. This is referred to as the
continuous RSD measurement and it is updated once every second.
The RSD parameter is treated as a separate and independent
monitoring parameter relative to the laser turbidity measurement. The
parameter is updated every second, which is the same rate the laser
turbidity value is updated. The parameter is best used as an early
warning parameter to an impending turbidity event and as a
complementary parameter to the turbidity parameter. (A turbidity spike
will also be complemented by a spike in the RSD parameter.) The
parameter has been designed to be very responsive to particles in the
1–10 μm range at very low concentrations. While the laser turbidity
parameter is currently approved for regulatory monitoring, the RSD
parameter is not a regulatory approved monitoring parameter.
Correlation study between kaolin standards and
measured turbidity
A study was done on the FilterTrak 660 sc (FT660 sc) laser
nephelometer to identify the specific correlation between kaolin and the
NTU value recorded using the FT660 sc. Two sets of kaolin standards
were prepared for turbidity measurement. Each set was prepared from
an independent lot of kaolin stock standard that was supplied by the
manufacturer. Each test standard was prepared using ultra-filtered water
immediately before measurement.
After preparation, each standard was put in the FT660sc, starting with
the lowest value. Measurements on the standard were done at 30-
second intervals, and 15 to 25 measurements were recorded for each
standard. From the measurement on a specific test standard, the
respective average and standard deviation was calculated.
Table 5 shows the:
• Averaged turbidity measurements from each of the two lots of kaolin
standard
• Pooled standard deviation from these two lots of kaolin standards
• 95% confidence interval (predicted) and the percent error expected in
the preparation for each standard value
The results show a strong correlation between mg/L kaolin and NTU.
The correlation is better than one percent over the measurement range
of the FT660 sc. The summarized correlation values for the entire study
are shown in
Table 6 .
Even though there was a strong correlation between the mg/L and the
NTU values when using the FT660 sc, the reproducibility of these
measurements was not good at the lowest two kaolin standards. This is
shown in
Table 5 , where the pooled standard deviations and predicted
English 23