API standards establish Performance Verification Tests (PVT) requirements for subsea valves, but Petrobras'specification has considered additional requirements that contributes to evaluate the performance and improve there liability of subsea valves, applicable to Wet Christmas Trees (WCT), manifolds and Pipeline End Manifolds (PLEM).
Moreover, Petrobras' specification has included additional functional tests during qualification process. These functional test data combined with valve design data and acceptance criteria are used to compose acceptance criteria for functional tests that are performed during Factory Acceptance Tests (FAT).
This paper presents an overview of these requirements that compose Petrobras' new technical specification. The goal of this specification is to reduce subsea intervention due to valve failures by improving quality and reliability of this equipment. Test results and conclusion according to Petrobras' Valve Qualification Program are also presented.
Ultra-deepwater subsea production requires highly reliable equipment. The trade off between increased reliability and performance versus cost, establishes design manufacturing constraints that shall be verified. Prototype and product testing as verification method is broadly used by the oil industry. API standards play a major role on establishing Performance Verification Tests (PVT) requirements for subsea valves, which are considered critical components for subsea system reliability.
Although API 1, 2, 3 requirements for subsea valves are fully used by manufacturers for subsea valve qualification process, since 1989, Petrobras has included high cyclical endurance testing in the PVT for reliability assurance in each design. The number of testing cycles was calculated based on expected service life of a Wet Christmas Tree (WCT) and number of acceptable failures during PVT, using test plans of MIL-STD-7814, 5.
In the original technical specification, issued by the purchasing department of Petrobras (SERMAT) in 1989, the prototype was tested 3,700 cycles at hyperbaric condition with maximum differential pressure in the valve. The torque tests were measured in atmospheric and hyperbaric conditions.
According to test plan of MIL-STD 781 chosen at that time, up to two failures were accepted without resetting the qualification-testing counter, as long as these failures were not caused by design problems that could require design revision.
As testing experience proceeded, it was experimentally seen for those manufacturers tested, that the detected failures were not caused by cycling at hyperbaric pressure. Besides, the maximum torque measurements were observed during valve opening at atmospheric condition. This resulted in the first revision6 of valve PVT spec, which demanded the same 3,700 cycles, but 3,500 cycles were performed at atmospheric pressure. The remaining 200 cycles were performed at maximum water depth, as API-17D required.
Until year 2000, more than 60 valve prototypes were tested according to this specification, where Petrobras' R&D Center (CENPES) had performed the PVT in more than 30 valves and had witnessed other 15 ones. Problems or failures were detected in 25% of these tests, resulting in design reviews and improvements in an early stage.