[en] On October 29, 1987, at Perry Unit 1, during performance of stroke time testing, three of eight MSIVs failed to fast close as designed. The stroke time testing was being performed in accordance with a startup test procedure. Two of the three affected valves were inboard and outboard MSIVs in the same main steam line, which would be a significant safety problem in the event of a failure of that main steam line. Subsequently, on November 3, 1987, at Perry Unit 1, during performance of stroke time testing, two out of eight MSIVs again failed to fast close as designed. The failure mechanism could not be positively identified, but the most likely cause was determined to be degradation of the Ethylene Propylene Diene Monomer (EPDM) elastomer seats due to exposure to a high temperature environment. As a result of the failure at Perry on November 3, 1987, the licensee began a detailed physical and chemical testing program in an attempt to pinpoint the failure mechanism. Results of the physical and chemical testing substantiated the previous conclusion of heat degradation as the root cause of the failures and eliminated hydrocarbon degradation of the EPDM as a possible cause. In addition, the chemical analyses revealed the presence of stearate compounds on the surface of the EPDM material