HRSG User's Group: Improving steam-plant reliability, durability and profitability
Flow-Accelerated Corrosion and the use of Pulsed-Eddy Current for HRSG Inspections
Doug Hilleman, PE and Ryan Moore, Intertek Group, PLC
Flow-accelerated corrosion (FAC), and under-insulation corrosion (UIC) are two of the most troublesome damage mechanisms in HRSGs, because monitoring them typically requires a plant outage, and the laborious removing of insulation from pipes, or the shaving of fins from tubes. Fortunately, a new non-destructive evaluation (NDE) tool can monitor both mechanisms without any of that removing-or shaving-labor. You don’t even need to shut down the plant, thanks to this new NDE tool!
Pulsed-eddy current testing was demonstrated on actual samples of damaged HRSG pipes, at the combined-cycle industry’s first-ever conference to feature hands-on training: the April 2019 HRSG User’s Group conference.
Figure 1: FAC on the inner wall of tubes or pipes causes dangerous thinning.
Doug Hilleman—who has scuffed his knees in a whole lotta HRSG inspections—began this presentation with an explanation of FAC, and why it’s so important for HRSG users to monitor it. Hilleman explained that FAC is metal loss from the inner wall of pipes or tubes containing fast-moving water (Fig 1). The metal is lost, Hilleman continued, via dissolution of the oxide layer that normally protects this inner wall. If FAC is not closely monitored and responded to, he emphasized, it causes a dangerous thinning of the pipe-wall which has led to several fatalities at US power plants, most recently at Southern California Edison Co’s Mohave Generating Plant, and at Dominion Virginia Power Co’s Surry Generating Plant. “These fatalities,” Hilleman explained, “tell us that monitoring FAC literally can save lives!” Hilleman went on to explain that FAC can occur in any part of the HRSG containing water or a water/steam mixture (Fig 2), but that its most susceptible components are mild carbon steel components:
- Feedwater heater shells
- Extraction Inlet
- Feedwater piping
- Drain Inlet
- Discharge and drain piping
- Downstream of valves
- Attemperator sprays
- Other components based on unique contributing factors—specifically, the oxidation-reduction Potential (ORP), pH, temperature, geometry, velocity, and chemical-injection points.
Figure 2: FAC can attack any part of the HRSG containing water or a water/steam mixture, so your monitoring program must be thorough.
These most-susceptible components should be fabricated of a steel alloy containing at least 1.25 percent chromium, because the chromium steels are more resistant to FAC than plain-carbon steel. Unfortunately, plain-carbon steel is the material in many of today’s HRSGs, because they were designed for low capital cost, and before FAC was well-understood.
After giving this background on FAC, Hilleman then compared and contrasted pulsed-eddy current (PEC) testing with the traditional method of monitoring FAC, ultrasonic testing (UT). The traditional method is a quantitative examination, he explained, meaning that it provides a definitive measure—in thousandths of an inch—of remaining pipe-wall thickness, whereas pulsed-eddy current testing provides only a relative measure of pipe-wall loss—the percentage of remaining pipe-wall, relative to the pipe’s baseline thickness. Continuing with the pros and cons of each method, Hilleman stated that UT’s cons are that it requires (1) the removal of pipe-insulation (as mentioned above); (2) surface-preparation; and (3) the unit to be shutdown.
So, which method do Hilleman and Moore recommend? The answer is: BOTH!
Hilleman and Moore use pulsed-eddy current testing to quickly and inexpensively identify locations of concern—the locations with lower percentages—and here they cut insulation or shave fins to conduct the quantitative UT testing. This process of PEC for screening, then UT for quantifying, has been confirmed numerous times by cutting out and lab-testing the locations found to have FAC-damage (Fig 3).
Next, Moore stepped up to the actual HRSG pipe samples in the Conference Room, wielding his pulsed-eddy current probe, and began testing the samples, as the readings appeared on a big-screen for the crowd to see. Working from one end of the pipe-sample to the other end, Moore’s test-results clearly showed that most of this length-of-pipe was at or near the baseline—meaning it showed no evidence of thinning, while three locations showed suspiciously low percentages—in the 60% wall thickness and 70%. These locations appear to have FAC-damage, Moore explained, so here we would cut the insulation and perform a UT test, to quantify precisely how much wall-thickness remains. One user was so impressed he asked Moore if he could buy a pulsed-eddy current tester from him. Moore explained that, no, Intertek doesn’t sell the equipment, it offers the inspection service. He further explained that it was risky to use the instrument without a good understanding of its capabilities and limitations when making important decisions about the presence or extent of FAC damage. Then the two exchanged business cards, and agreed to stay in-touch.