Finding the lost megawatts

5 September 2018

Valve maintenance is a complex, yet lucrative opportunity for improved maintenance strategy. Frank Todd and Katelin Kohn offer an insight into the different methodologies that can be used to determine leakage rates and improve plant performance.

NUCLEAR POWER PLANTS OFTEN SUFFER from power loses due to leakage through valves that are faulty or do not seat correctly. Often these losses are significant – in the order of a few megawatts – but they can be difficult to determine and difficult to quantify.

Typically, there are two main causes of valve leakage, says Frank Todd, manager of thermal performance for True North Consulting (a subsidiary of GSE Systems since May 2018), who has more than 30 years’ experience in optimisation, efficiency and thermal performance.

The first cause is generic deterioration over time. The second factor is erosion of the valve seats and discs by steam and water. So-called ‘zero-leakage’ valves may leak in practice, but this is becoming less and less of an issue for nuclear plant. There have also been instances of leaks caused by foreign material getting caught in valve mechanisms, adds Katelin Kohn, a senior engineer at True North Consulting.

Understanding the impact of a leaky valve is critical to making effective maintenance decisions. Typically, there are about 150 high energy valves in a nuclear plant that are candidates for cycle isolation monitoring. These include level control valves on high energy drain tanks such as moisture separator reheaters (MSRs) or higher-pressure feedwater heaters; as well as drain valves on turbine isolation/control valves.

“Cycle isolation is the nemesis of many a power plant. Often valves are leaking and because you cannot see the steam that is flowing to a condenser, significant leaks go unnoticed. The solution is a combination of good techniques and a good engineer that can understand the impact of the plant configuration, and know-how to take the measurements at the right locations,” says Todd.

Checking for leaks

The current approach to uncovering a leaky valve typically requires operations to do a plant walkdown when the thermal performance engineer finds there are missing megawatts and cannot see why.

It is very difficult to calculate the flow through a leaking valve with any accuracy, as assumptions can significantly alter the results. There are a couple of basic methods to measure the flow. One is using an acoustic measurement system. The second and most common is to measure the temperature downstream of the valve and use that temperature to calculate flow rate.

Todd stresses the importance of identifying the correct measurement locations. In pressurised water reactors, where temperature is measured manually, operators can take multiple measurements and move towards the condenser if high temperatures are detected upstream or downstream of the valve.

In boiling water reactors (17% of the current fleet), many valves are installed in areas with no personnel access, so remote sensors are required. True North has worked at the Columbia nuclear plant, a 1207MW BWR operated by Energy Northwest, carrying out cycle isolation monitoring. Todd says that wireless remote sensors installed at the plant were used to help identify a leaking evaporator relief valve, which over six and a half weeks had led to $330,000 in lost electricity generation. The company has also completed cycle isolation valve testing following a turbine retrofit at DC Cook 2 in Michigan. Cook 2 required one valve to be partially open, so calculations were required to complete the turbine acceptance test.

Flow measurement methods

“Repairs to leaking valves can provide significant payback but it also has been the case where valves were identified as leaky but they were actually as tight as a drum,” says Todd. “It only takes one of these non-leaking events to significantly decrease the confidence of any leakage monitoring programme.”

Todd says that True North relies on a variety of techniques to make the best determination on what is happening with the valve and its impact on plant performance. True North utilises its TP-PlusTM CIM software, along with its experience in configuring a leakage programme, to quantify the effects of leakage, prioritise repairs and provide cost justification for the repairs. Data for each valve are entered into the software, which produces results from advanced cycle isolation loss calculations that can be graphed to monitor valve degradation over time.

The software uses up to five different methods to predict valve leakages (see box).

  • Grashofs
  • ASME Figure 14
  • Darcy-Weisbach equation
  • Sonic Flow Equation
  • Choke Flow Equation

The methods all use downstream temperature to determine the pressure inside the pipe and then calculate a leakage rate. Some of them also calculate a velocity in the pipe. 

When applied with engineering expertise these techniques can be used to estimate current valve leakage and also monitor systems for earlier detection of leaks. However, additional factors (flow resistance, moisture and distance to condenser) also need to be considered. Flow resistance (or head loss) in pipes, valves, and fittings is generally given in terms of the resistance coefficient. The resistance coefficient can be included in the Bernoulli equation to account for flow resistance between two points. Because it is proportional to velocity and flow it can be applied as a correction factor to the velocity and flow results of the Grashof, ASME, and Choke equations that do not account for flow resistance.

The five flow equations described only calculate dry steam flow and make no allowances for the presence of moisture. If the leaking steam downstream of the valve is saturated then the moisture flow must be accounted for. The relationship between total leakage flow and its steam and liquid constituents can be summarised with conservation of mass and energy equations.

Finally, ‘distance to condenser’ is the equivalent hydraulic distance, based on actual linear distance and the number of elbows, tees, valves, and other flow disturbances in the leakage path. By applying a correction to the calculated flow, a more accurate estimate for the flow can be calculated. This correction protects against over-promising the savings to be realised by repairing the leaking valve(s).  

Author information: Frank Todd, Manager of thermal performance at True North Consulting; Katelin Kohn, Senior engineer at True North Consulting 

Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.