Recently Thielsch Engineering evaluated a valve stem that was reported to have experienced rapid erosion in service. Another valve stem had also failed in service in a similar manner at the same facility. The steam turbine stop control valve was manufactured by General Electric (GE). The valve was operated in a GE two on one combined cycle unit. GE has attributed these valve stem failures to hard particle erosion. A service bulletin issued by GE states that the stem should be protected by a sleeve at the locations where the failures have occurred.
The owning energy supplier had retained an engineering firm to perform an analysis on the previously failed stem. Their report did not support hard particle erosion as the cause, instead citing a possible bromide corrosion issue with the nickel alloy from which the stem is manufactured. The energy supplier is currently monitoring the performance of another similar valve stem that has been in service for approximately four months. After only two starts and limited throttling, this stem is also exhibiting surface erosion.
The chemical composition of the valve stem Thielsch evaluated was typical of a nickel-iron-chromium super alloy such as Alloy 901, and was in compliance with the requirements for UNS N09901. The microstructure and hardness values were also typical for Alloy 901 in the solution treated and precipitation hardened heat treatment condition. There was no evidence of a material discrepancy that would have caused or contributed to the erosion that rendered this valve stem unfit for continued service under the intended operating conditions.
EDS analysis performed on the damaged and adjacent surfaces of the valve stem revealed elements typical for a nickel-iron-chromium super alloy such as Alloy 901. The elements sodium, chlorine, and sulfur were also identified in the scale deposits at and near the damaged area. These elements are present in substances corrosive in nature to many elevated temperature alloys. However, they do not appear to have been the primary cause of the surface degradation experienced by the valve stem in service. Bromine, suspected to be a corrosive agent present in the steam, was not identified by the EDS analysis performed by Thielsch Engineering.
Strain lines were observed in the microstructure adjacent to the damaged areas of the stem surface. These likely resulted from hard particle impingement that fostered the erosion and material wastage noted.
The metallurgical examination revealed that the surface damage incurred by the valve stem was the result of hard particle erosion. The hard particles likely originated from exfoliated oxide scale generated along the internal surfaces of the high-temperature steam tubes that supply steam to the combined cycle steam turbine. In particular, those high-temperature steam tubes fabricated from T91 (modified 9 Cr – 1 Mo) alloy steels have shown a propensity to generate and exfoliate heavy oxide scale during the early stages of operation. Over time in service, these hard particles eroded the affected surface of the valve stem through particle impingement.
A possible solution to this chronic valve stem erosion issue may be to increase the hardness and strength of the material by switching to an Inconel 718 material.
For additional information about oxide scale exfoliation and solid particle erosion please contact Peter Kennefick at firstname.lastname@example.org.