Most corrosion induced tube failures have a component that is directly related to water chemistry. The corrosion may have occurred as a result of impurity ingress, improper selection of control chemicals and/or control ranges for the material in service, or failure to control the chemistry in the designated range. Often, the intended water treatment program would ensure adequate component life and low rates of corrosion. It is the failure to adequately monitor and maintain that treatment program that results in the corrosion failures.
Because virtually all corrosion mechanisms take a period of time to develop, infrequent short term excursions from the desired water chemistry can be tolerated. However, this tends to lead to complacency among plant operators and management. The thinking is: “nothing failed the last time the plant was out of specification; therefore it must be ok to continue to run out of specification”. This type of thinking should be avoided. Corrosion damage tends to be cumulative and the longer the material is in a corrosive environment the higher the likelihood of eventual failure.
Along the same lines, there is a tendency for operators to “wait until things have settled out” before taking water chemistry samples. This is especially true for cycling units. Rapid and frequent changes in temperatures, pressures, heat flux, and flow rates make control of water treatment parameters more difficult. Operators of a unit that daily cycles, starting up in the morning and shutting down in the evening, have been known to wait until late afternoon before taking samples because they know that the samples will likely be out of specification if taken right after startup. As a result, the unit may spend most of the operating period in an out of specification condition, while management is being informed that everything is ok.
Because corrosion failures do not occur instantaneously, it is very important to monitor chemistry control parameters over long periods of time and look for trends. On line instrumentation tied into the plant DCS is particularly useful for this purpose as the sampling rate is very high. Monthly averaged values have limited usefulness since the unit may spend a considerable period of time in an out of specification condition and still have an acceptable average. Keep in mind that it is the cumulative time in a corrosive environment that matters most to the tube material and the further out of specification or the more corrosive the environment the shorter the time before failures will occur [1].
[1] Jackson, P. Moelling, D. Malloy, J. Taylor, M. Tube Failure Diagnostic Guide - Third Edition, 2013. ISBN 0-9719616-3-8
Stay up to date and refer to our large collection of technical white papers and articles
