Introduction to Magnetite Exfoliation
Field TrialsA field trial was conducted at Wisconsin Public Service in Rothschild, WI in the spring of 2009. EPRI issued a report detailing the results. Shawn Gowatski gave a presentation at the EPRI BRIG meeting in June 2009. TesTex received a call from a client in Maryland asking if we could perform an inspection to detect magnetite in their 1.75” OD, 0.240” wall, SS-347 superheater tubes. The unit had its second tube failure in 3 days and they needed a method to detect the blockage in the tubes. TesTex had manufactured special LFET scanners to inspect these tube bends for the EPRI trial. These bend scanners were for a 1.5” tube. TesTex ran a quick study in the lab and determined these scanners would work on the 1.75” OD tubes. We only had two of these scanners in stock. TesTex sent one team to the site and the customer put some magnetite in a tube that was removed for us to prove the technique was valid. We also performed a rough calibration onsite by filling a 2” section of the tube full with magnetite and also half-filled with magnetite. The responses were linear. The inspection frequency was 1000Hz and low power level was used. The magnetite causes the LFET signal to decrease due to it magnetic properties. It shows up as additional metal.
The unit contained 171 Assemblies with 12 Loops per Assembly. The customer requested more teams but due to the fact we only had two scanners, TesTex decided to send another team to work nights.
A total of 2052 loops were inspected. Boilermakers were provided by the client to spread the assemblies when necessary. Thirty-eight bends were found to have magnetite inside the tube. The client cut the bends out that contained magnetite. The magnetite was removed from the tube and then the bend was welded back into place.
The LFET scanners used contain 4 LFET sensors spaced circumferentially alongside each other. The responses of the sensors are the 4 lines shown in the bottom two windows. When no magnetite is detected the sensors will provide a relatively flat line. When a magnetite deposit is encountered the signal will decrease due to the presence of additional metal. The signal decrease is directly proportional to the amount of magnetite detected. When wall loss is detected, the signal will increase. The simulated C-scan window changes to a darker blue where the magnetite is detected.
The plant has a sister unit that also recently had the superheater tubes replaced with the SS-347 material. TesTex inspected this unit when it became available and found over 300 bends containing magnetite. The unit had run approximately 7 months with the new tubes. TesTex has performed additional inspections on both units with less than 60 days run time and have found lesser amounts of magnetite. The follow-up inspections have shown about 30 bends that are 10-25% full of magnetite. A procedure has been developed to send 9 technicians to site to perform the inspection of 4104 bends in less than 12 hours using (4) two-man teams plus an analyst. A report detailing the findings is presented upon completion of the inspection.
TesTex has also performed some spot inspections for a customer in PA who has recently replaced their superheater tubes with SS-347 on two different boilers. The industry news of other utilities experiencing magnetite issues prompted them to perform the inspections. The inspection was focused on the pendants closest to the left and right sides of the boiler. No evidence of magnetite was detected in either boiler.