Some specialized literature regarding the maintenance of safety relief valves (SRVs) is available, but it is best to always consult the manufacturer’s detailed installation and maintenance manuals on the subject, as some type of valves might require special attention. It is highly recommended to always use genuine manufacturer’s spare parts. In some parts of the world, this is even required by law when it comes specifically to SRVs. This enables everyone to keep track of the valve, which is installed under legal conditions and will assure that the guarantee on the valve is not jeopardized. It also keeps its ‘ passport ’ up to date, and it can be verified if set pressure or backpressures eventually change.
Many users are concerned about the maintenance frequency of their SRVs installed base. Many major companies have their own internal procedures or are following codes and recommendations. However, there are no strict legal requirements that touch on maintenance frequency of SRVs. Once in some boiler installations in the past, it was a rule, or better a habit, for the firemen to pop the valves daily or weekly by means of the lifting lever, an unenviable job as it was extremely dangerous for the personnel and also not very good for the valve itself. It also required a minimum operating pressure of 75% of set pressure, usually of very hot steam, something you do not want to be very close to. This procedure also led to a lot of accidents and so is not used very frequently anymore. Other companies only tested and maintained their valves after an accident or major pressure upset, and all the rest was/is between both these extremes.
While preventive maintenance on this safety component is extremely important, it should also be noted that a valve cannot be endlessly tested, popped, overhauled, and so forth, and that maintenance (and testing) also cause wear on the valve. Therefore, a good compromise must be found. Regular inspections of SRVs are necessary to ensure safety. However, here we should define “regular”, taking into account that inspections are also costly and potentially destructive. A delicate balance between safety and cost must be obtained. This is a very complex problem involving multiple factors that include the individual valve’s application, pressure, temperature, medium, age, size, and type. The problem here is that it is difficult to generalize for the complete valve part.
Testing and maintenance is indeed a necessary evil but should not be done more than absolutely necessary. Here we will present a simulation model for determining the inspection policy for SRVs in a typical petrochemical plant, based on experience. In my opinion, it minimizes the total inspection and repair costs without jeopardizing the safety. The model is simply a result of 20 years of observation in the chemical, petrochemical, power and oil and gas industries. The maintenance frequency is, in my opinion, dependent on too many variables and combinations thereof (process conditions, environment, location, temperature variations, pollution, etc.) to completely generalize the recommendations. It is my experience that reliability of SRVs can only be determined on a historical basis for each individual installation, application, location and even type of SRV. The variety of applications and types of valve makes it impossible for us to correlate meaningful information with relation to such events as failure rates for all industries; however, we can track the individual valve and its ‘health’. I would compare the frequency for inspecting the valves with regular doctor check-ups. For instance, the older you get, the more frequent will be your doctor check-ups; if you work in a dangerous environment, the more frequently you should see a doctor.
Maintenance frequency
The best way of effectively determining a maintenance schedule is by keeping a detailed log of the history of each valve. Inspection frequency should be based on criteria which are explained hereafter. We will demonstrate how to build a schedule of inspection activities based on historical data for each individually installed valve. This procedure starts with installation: An SRV is tagged and is given a ‘ passport ’ containing all of its data, process data and revision dates with comments and the spare parts used and when. Actually, the basis of this passport already exists when it leaves the manufacturer (on the tag plate) and the basic records of the valve are also kept with the manufacturer for later reference.
Rationale
SRVs are generally inspected at the same time as the elements to which they are fitted. Each item of surface safety equipment should be allocated an inspection grading 1, 2, 3 or 4 which indicates the maximum intervals that may elapse between two inspections.
Each valve should:
- Initially, be given an inspection grade “1” and be given its first inspection after a short service period, typically maximum 1 year. This first inspection tells a lot about the condition of the valve and the application it is used on.
- Subsequently, based on built-up knowledge of service conditions and surface safety system parameters and conditions following the first thorough inspection, the inspection grading should be reviewed, allocating either grade 1 or grade 2.
- Subsequently, based on extended knowledge gained of service conditions and surface safety system parameters and conditions following previous extensive inspections, be graded to inspection grade 1, grade 2 or grade 3.
- Subsequently, based on the further extended knowledge of service conditions and surface system parameters gained from the previous extensive inspections, be graded to inspection grade 1, grade 2, grade 3 or grade 4.
- Have an inspection review of the kept records of the valve carried out during each inspection in order to determine the inspection grade to be allocated during that inspection period.
The purpose of this review is also to identify any changes in the system or service conditions that may affect the inspection grading of the item and to build up an inspection history of corresponding surface safety systems. It is recommended to return to grade 1 if the service and process conditions have changed significantly. The recommendation for process change before changing grade would be approximately 5% to 7% from original process conditions.
Factors affecting selection of an inspection grading
The following factors can affect the selection of an inspection grading:
- Design constraints
- Operating constraints and conditions
- Legislative constraints
- Certifying authority requirements
- Modes of possible failure and consequences
- History of a particular item
- History of similar items in similar service
- Current inspection grade
- Period elapsed since the previous inspection.
The above factors should be taken into account by the inspection engineer in the process of assessing which inspection grade each item should be awarded.
Inspection grade awards guidelines
Inspection grade 1: All surface safety systems should be awarded inspection grade 1 until a system history can be built up by a series of inspections or, at least until the first major inspection has been affected.
Inspection grade 2: This inspection grade may be awarded where the following conditions are met:
- The valve was under grade 1 and opens within a tolerance band of 5% of the cold differential set pressure (CDTP). The leakage rate is acceptable according to API 527 or to the company specifications. The internal conditions of the dismantled valve show no or minor defects.
- The valve was under grade 3 and fails to open or opens outside a tolerance band of 5% of the CDTP. The leakage is outside the tolerance of API 527, the company specifications or the internal inspection of the valve shows defects which require further investigation, replacement or repair.
- The valve was under grade 4 and fails to open, the leakage rate is excessive and the internal inspection of the valve reveals serious defects such as galling and possible seizure which require further investigation, replacement or repair.
Inspection grade 3 : This inspection grade may be awarded where the following conditions are met:
- The valve was under grade 2 and opens within a tolerance band of 5% of the CDTP. The leakage rate is acceptable according to API 527 or to the company specifications. The internal inspection shows no or minor defects.
- The valve was under grade 4 and fails to open or opens outside a tolerance band of 5% of the CDTP. The leakage rate is outside the tolerance of API 527, the company specifications. The internal inspection reveals defects, which require further investigation, replacement or repair.
Inspection grade 4: This inspection grade should only be awarded where the following condition is met:
- The valve was under grade 3 and opens within a tolerance band of 5% of the CDTP. The leakage is acceptable according to API 827 or to the company specifications and the internal inspection of the valve shows no or minor defects. Grading transfers should only be considered after extensive inspections. The first extensive inspection can form the basis on which a valve may be transferred to grade 2 only if the parameters for this grading are met.
Subsequent extensive inspections of the valve can form the basis on which a valve may progress through the inspection grading system, taking no more than one upwards stepper inspection, only if the parameters for the grading step can be met. In NO case should the interval between inspections of safety devices exceed the interval between inspections of the pressure vessels involved. The converse of this also applies, with the downgrading of valves if inspection results indicate the current grading parameters are not being met. Sample inspections on the surface safety systems are not recommended. Each system must be inspected at the specified interval.
Inspection requirements and reporting
It is recommended that every complete replacement valve, either withdrawn from the stores or returned from a valve specialist, be inspected. New valves coming from an ASME or PED-approved SRV supplier should already be tagged and leaded and will most probably meet all code requirements without requiring further inspection. The only question is how transport could have affected the setting and operation of the valve. Careful verification of how the valve was packed and shipped is important. Therefore many users test all valves, regardless, before putting them on the system.
Actually, it is good practice that any valve received from an off-site location should be bench tested to verify set pressure and leakage rate. However, please be aware that if the tag on a new valve is broken, the supplier forfeits on the warranty of the valve unless it is done by or in the presence of a notified body who is able to tag or lead the valve again.
The set pressure may then be adjusted accordingly and this inspection may be used for inspection grading purposes as described above.The valves don’t need to be stripped for this commissioning inspection unless problems are encountered requiring further overhaul.
Results of this inspection are also to be recorded and added to the “Valve History Record” or passport. From this point, a credible valve history must be established for each unit. This will create a reliable overall safety system for pinpointing trouble spots on surface safety systems and will facilitate future material selection or system modifications. Any system set up to record valve history and reporting should be simple to operate and clear to any user. Each SRV in service will have its own unique history of problems, lifts, and repairs. It is important that records of each stage of the valve’s history are documented as these will form the valve history.
This valve history or passport will at a minimum contain:
- Original valve specification data
- Valve commissioning report
- All subsequent rectification/overhaul/recalibration reports
- Any material changes/spring changes or changes of specification
- Eventual process changes exceeding 5% to 7% from the original
From this information, control sequences and planned maintenance routines may be designed and altered to suit the particular process area with which that specific valve is involved. A database such as this is also important where valve interchangeability is required in order to determine a minimum stockholding for maintenance purposes. Valves of similar body and trim materials can possibly be utilized in many different locations and services, thus removing the need for one-to-one valve stocking. This system will also mean that valves can easily be sourced from non-essential systems in order to maintain essential system viability.
TESTING TERMINOLOGY
Bench or test stand testing:
Testing of a pressure relief device on a test stand using an external pressure source with or without an auxiliary lift device, to determine some or all of its operating characteristics, without necessarily flowing the rated capacity. This is required on a regular basis when the valve is taken into the maintenance cycle at least to see that there is no shift on the set pressure and that the valve would open correctly during a pressure upset.
Flow capacity testing:
The usually special testing of a pressure relief device to determine its operating characteristics, including measured relieving capacity. This tests whether the valve flows the capacity as stated in the literature or as per given flow coefficients, or to simply determine the flow coefficient of the valve as such. This is done on a spot-check basis by independent notified bodies in limited locations worldwide especially approved for that purpose.
Hydrostatic testing:
Before assembly, each valve body is hydrostatically tested at the manufacturer at standard 1.5 times it’s maximum rating, typically during a period of 1 – 3 minutes. This is also called the shell test and reveals eventual
deficiencies in the castings. Several manufacturers have different procedures, which can usually be obtained for review. In -place testing: Testing of a pressure relief device installed on but not protecting a system, using an external pressure source, with or without an auxiliary lift device to determine or check some or all of its operating characteristics; mainly opening. Also, set pressure can sometimes be obtained by calculation.
In -service testing:
Testing of a pressure relief device installed on and protecting a system using system pressure or an external pressure source, with or without an auxiliary lift device to determine or check some or all of its operating characteristics; mainly opening and set pressure. Usually, this requires about 75% of set pressure present under the valve while testing (also known under commercial denominations such as Trevitest, Sesitest, etc.).
Leak test pressure:
The specified inlet static pressure at which a quantitative seat leakage test is performed in accordance with a standard procedure (e.g. API 527, see Section 4.2).
Pre -start up testing:
It is highly recommended that all valves be visually inspected before installation for dirt and particles, and the same goes for the system the valve will be installed upon. Especially new installations are prone to contain welding beads, pipe scale and other foreign objects, which are inadvertently trapped during construction. These foreign materials are devastating for the valve, and it is recommended that the system be purged carefully before installing the safety relief valves (SRVs) as these are very destructive when the valve opens. Also, caution should be taken that all protective materials, such as flange protectors, are removed before installing the valve. It is also recommended that the valve be isolated or gagged during pressure testing of the system, but make sure the gag is removed after testing. Some companies or local customs require the valves to be tested just before start-up, but normally the valves have already been set and sealed correctly at the manufacturer. This is not a recommended practice, but if needed on spring-operated valves, crack pressure can be checked by applying a suitable pressure source at the inlet of the valve. However, in the usual case on site, the volumetric capacity upstream is insufficient, and therefore a false reseat pressure (usually lower than actual) will be obtained. On non-flowing-type pilot valves with a field test connection, the set pressure can be easily checked. It is recommended that the manufacturer’s instructions, which should accompany the valve, be carefully followed.
Shell test:
See hydrostatic testing.