Power plants increasingly “cycle” to meet market conditions. That’s especially true of many newer combined cycle plants that were designed for continuous operation. Cycling operation creates a host of issues from thermal fatigue failure to freeze protection. Preservation of boilers and heat recovery steam generators (HRSGs) during shutdown and cycling operation generates much discussion. Since most of these plants were designed for base-load and continuous operation, their designs don’t include the equipment one might include for a plant designed for cycling operation.

A typical market-driven shutdown may last overnight, a few days, or several months. The plant often doesn’t know. Rapid startup remains a priority, so the boilers or HRSGs usually remain filled with water. Condenser vacuum disappears and air enters the steam cycle as the plant cools. There’s often no steam available on the subsequent startup, so the drums are filled with oxygen-laden water. Copper and/or iron transport increase, corrosion can be severe, and failures can occur much more quickly than in a base-loaded plant that operates continuously.

Picture denotes through-wall tube failure caused in part by oxygen corrosion.

 Unless precautions are taken, corrosion may occur on external and internal surfaces of boilers that are out of service for any length of time. Proper downtime procedures are extremely important, since oxygen corrosion occurs more often during out-of-service and startup periods than during normal operation.

Picture denotes typical of the oxygen pitting corrosion that can occur during shutdown.

There are some general procedures than can be adapted and implemented for the proper short-term and long-term lay-up of HRSGs, boilers, deaerators, auxiliary boilers and associated steam cycle equipment in both traditional fossil fuel and combined cycle systems. The optimum lay-up procedure varies, depending on site-specific or lay-up specific variables including: anticipated duration of the lay-up; the equipment present at the plant; and economic conditions such as fuel costs and contract or market requirements. No general procedure can adequately address all of the different lay-up options. One can capture the most common and widely accepted procedures, but specific plant capabilities and situations may require site-specific approaches and, in some cases, the installation of additional equipment.

In general, cycling plants should maintain feedwater and drum pH at the upper end of their respective target ranges. Increase hotwell and/or deaerator levels at shutdown to maximize the inventory of deaerated water for the subsequent startup.

The choice of storage methods depends on the length of downtime expected. If the boiler is to be out of service for six to eight weeks or more, dry storage is preferable. Wet storage is usually suitable for shorter downtimes required for cycling units.

Water used for unit startup should always be deaerated (dissolved oxygen less than 30 ug/L). Many pure merchant designs cannot provide deoxygenated water unless the plant is running (these plants often do not have an aux boiler, nitrogen sparge system, or demineralized water storage tank deaeration capability, etc). Such plants often break vacuum during off-line periods requiring startup with oxygen-laden water. Plants suffering under these design constraints should be evaluated in detail and additional action taken to mitigate the damage caused by startup with oxygen-laden water.

The Importance of Monitoring Corrosion Product Transport
Monitoring of iron and copper transport is essential, continuous monitoring provides the greatest benefit. Traditional monitoring of metal transport and generation in the steam cycle relies primarily on periodic wet tests (either "patch" tests or photometric analysis). This approach leaves significant holes in the data stream. Every thermal, chemical, or hydraulic event liberates or generates metal oxides in the steam cycle (a "crud burst"). These events occur often and cannot be scheduled. They’re usually most severe during startup and shutdown, but also occur during normal operation.

Time-based testing (iron or copper sampling at a specific frequency) is important, but it cannot detect the majority of these events. Most plants are "flying blind" and cannot reliably monitor when and how much metal transport occurs. Lacking real-time data, the primary methods used to determine metal transport are periodic inspections and tube deposit weight densities. These methods are important and should continue, but they can only detect a metal transport problem after it has occurred. There's no opportunity to evaluate the impact on metals transport of mechanical, chemical or operational changes to startup, shutdown and operating procedures.

Online particle analysis can provide a window into metal liberation and transport as it occurs (not after the fact). This technology is available from several manufacturers. Recent improvements in instrument reliability and decreases in unit cost have made it a financially viable monitoring tool for most power plants.

This technology can be used to evaluate the impact of operation on iron liberation and transport by identifying those evolutions that result in the greatest metal transport. Once identified, plants can tune operation to minimize metal transport during these events. This tool allows real-time detection of any significant increase in metal transport. Plants can then tune startup and shutdown procedures to minimize transport with the equipment that’s installed and available. Equipment upgrades may still be required, but particle analysis allows plants to do the best they can with the equipment they have in the interim.

Choose the correct Lay-up Option
Merchant operation with variable dispatch creates special challenges for lay-up. Shutdown duration is unknown, yet quick start-up remains critical. In addition, many plants do not possess nitrogen blanketing or external water recirculation capability. Most shutdown units cannot be steamed to check chemistry without significant adverse economic impact. Wet lay-up procedures for these circumstances must be developed on a case-by-base basis to minimize damage at reasonable cost. If an auxiliary boiler is present, the drums can be sparged/blanketed with steam during idle periods to minimize oxygen intrusion and to increase circulation of idle boilers.

The optimum wet lay-up procedure utilizes steam to keep the unit warm and the drums pressurized, thereby eliminating the potential for oxygen ingress. This method allows for immediate start-up if dispatched unexpectedly. Other lay-up methods (cold wet lay-up or dry lay-up) require significantly longer startup times.

If there is no auxiliary boiler or other steam source, then the plant must select a dry or cold wet lay-up method that minimizes corrosion while still allowing startup within the required time. Even if a steam source is available, the selection of the optimum lay-up method may be influenced by the cost of auxiliary steam production. Cold wet lay-up is generally less costly (in the near term) than the cost of energy to keep the unit hot and under pressure, but frequent use of this method may lead to premature boiler failure or increased iron deposition. It’s important to complete a short- and long-term cost evaluation prior to selecting a lay-up method. Wet lay-up with steam provides the best corrosion protection and should be used unless economics dictate otherwise.

The expected out-of-service duration represents the most critical factor in determining the optimum lay-up method. Contractual issues also weigh heavily in method selection since some plants may be required to be able to return to full service in less time than it would take to recover from a specific lay-up condition. Therefore, plants must review their operational plans, implementation schedule, and contract requirements before selecting the most appropriate lay-up procedure.

Plants may elect to continue with business as usual and do nothing. That is an option, but only if plant management is willing to accept more forced outages and pays for additional chemical cleaning and repairs. A little capital and planning now can literally save millions in the years to come.