POWER September 2011 - 58

WATER TREATMENT
or turbine. Each of the water chemistry
parameter limits set by these industry organizations
links directly back to specific
failure mechanisms in the steam and water
cycles. These published recommendations
give us guidance as to the parameters
monitored and their concentration levels.
This information also tells us the type of
equipment and the accuracy requirements
of equipment purchased for the plant water
laboratory. For example, the IAPWS
recently published recommendations for
the concentrations of key parameters used
to monitor the steam cycle (Table 1).
In the past, the concentration of some
of the critical water treatment contaminants
that can damage the turbine could
not be measured directly at the part per
billion (ppb) levels, so indirect measurements
were substituted. For example, the
familiar cation conductivity analyzer has
been a required parameter by turbine manufacturers
because it provides a simple
and robust means of indirectly detecting
ppb levels of chloride and sulfate. Another
indirect measure is sodium. Sodium ions
per se don't damage turbine components,
but sodium hydroxide does. We can't measure
the ppb levels of hydroxide, so we
measure sodium instead, and we assume
that all of the sodium in the steam is there
in the form of caustic (sodium hydroxide)
as a worst case.
But these indirect measurements can
be fooled by other contaminants that may
not be harmful. For example, dissolved
carbon dioxide in the form of bicarbonate
will raise cation conductivity, though it
has been shown that the dissolved carbon
dioxide in the steam is not harmful to the
turbine. During commissioning, this has
caused many a battle between the owner
and turbine manufacturer about the warranty
and how reasonable or unreasonable
the cation conductivity limits are.
Start at the Source
As the power industry continues to pursue
lower and lower detection limits, the critical
importance of good sampling and sample
conditioning is often overlooked. The
most accurate and reliable instrumentation
cannot provide good information on a bad
sample. Also, learning that there is significant
contamination in the boiler 30 minutes
after the boiler water pH has dropped
is unacceptable. Good chemistry control
starts with good sampling (see " Maintaining
Water Sample Panels Improves Plant
Availability, " March 2008 in the POWER
archives at www.powermag.com).
Progress has been made in sample conditioning
procedures and system design
September 2011 | POWER
to ensure that the sample that reaches the
analytical instrument accurately represents
what was really going on at the sample
point in the steam cycle, within seconds
after the sample leaves the steam line or
drum (Figure 1).
A properly configured sample line and
sample conditioning panel follows several
well-accepted design practices (see sidebar).
Most sampling problems begin at startup,
when particulates in the equipment can
restrict flow or even plug up the sample
lines and pressure-reducing valve. Provisions
to safely flush out the sample line(s)
during start-up are important to maintaining
a good sample. Even with these precautions,
small amounts of particulates
can accumulate in and behind the pressure
reduction value or variable pressure reducing
element (VREL) valve. These can be
manually flushed out by temporarily opening
the valve slightly to increase the pressure
(and flow) in the line.
The best design uses a VREL valve that
monitors total sample flow and pressure
and automatically adjusts the sample flow
for the upstream pressure. As particulates
build up on the valve or upstream in the
sample line, sample flow drops. The automated
valve senses the sample flow decrease
and opens the VREL valve, which
flushes out the particulates. It also shuts
off the sample completely when the pressure
on the line drops below a set limit
(when the unit is off-line) and opens when
the unit comes up to pressure. This is particularly
helpful on units that start and stop
frequently or that start when lab personnel
would normally not be present to manually
flush the sample lines.
This is not to say that all chemistry parameters
need to be monitored and controlled
on a continuous basis. There will
always be a subset of parameters for which
testing once a shift, or even once a week
will be sufficient. But the combination of
the availability of online instrumentation
and reductions in the number of plant staff,
particularly specialists such as trained lab
chemists, will continue the push toward
more online monitoring.
Important Monitor Trends
There are a number of large trends emerging
in the world of steam and water sampling
that cross instrument and electronics
manufacturer boundaries. A few of these
are mentioned below.
Integrated Sensors. Integration of the
measuring circuit, calibration and diagnostics,
memory, and analog-to-digital conversion
into a single, compact electrochemical
sensor is a rapidly growing trend.
www.powermag.com
57
For example, consider the familiar pH
probe. When measuring the pH of a solution,
the interaction between the probe
and the sample generates a millivolt signal.
This signal is sent to the meter, where
it is interpreted and the pH is displayed.
When the pH meter is on a lab bench, the
distance between the probe and the meter
is typically only a few inches. However,
in a power plant, a pH probe may
be located on a sample line coming into a
central sample panel that may be located
hundreds of feet away from the monitor,
which can cause signal degradation and
incorrect pH readings. Also, if the cable
is run adjacent to other instruments or
electrical sources, additional signal noise
and interference is possible. Usually, the
1. Smart sampling. A state-of-the-art
sample conditioning system provides a constant
sample flow regardless of supply pressure
at the recommended velocity. A fully
automated sample conditioning system, as
shown in the photo, will improve sample
accuracy and keep your chemistry within
EPRI-recommended guidelines. Courtesy:
Sentry Equipment Corp.
http://www.powermag.com http://www.powermag.com
POWER September 2011

Table of Contents for the Digital Edition of POWER September 2011
