Integrating valve maintenance with LDAR programmes

1. What LDAR tells you about valves

Leak detection and repair programmes, usually called LDAR, monitor fugitive emissions from equipment in service. Valves matter here because they form a large share of monitored components.

One study, reported in Chemical Engineering magazine, traced 60 per cent of refinery fugitive emissions to leaking valves. That figure belongs to one refinery population, and distributions vary by plant.

LDAR produces evidence, not repairs. Detector readings and screening methods indicate that a valve leaks. Confirmation follows your site leak definition and monitoring method.

The survey output is a tagged list of leaking components with dates and readings. That list is the input to valve integrity management, and the start of maintenance planning. For how leaks develop and how evidence is read, see valve fugitive emissions and maintenance response.

2. From leak tag to work order

Each leak tag needs a decision, and the leak path drives it. Common paths include the stem, the body joints and the seat.

Stem leakage to atmosphere is a common LDAR finding, and it often responds to sealant injection. Seat leakage, or passing, is a process loss rather than a fugitive emission, and it needs its own test. Body leakage is serious and usually needs prompt assessment.

Online treatment is not universally possible. The gates come first:

1. The valve must have injection fittings in good condition.
2. The valve design and the service must suit injection.
3. Site rules and the permit-to-work must allow the job.

Where OEM documents exist, consult them for limits and warranty terms. Treatment then means controlled amounts of cleaner, lubricant or sealant, within equipment and fitting ratings.

You watch the pressure response, and you stop if pressure climbs without progress. Our clean, lubricate and seal service works this way, on valves in service.

3. One record per valve

Integration depends on records. The LDAR component tag and the valve tag must match. Every reading and every action then lands in one history.

That history should carry the survey reading and date. It should also record the action taken, the product and quantity, and the recheck result.

With that record, the next survey becomes an audit of your maintenance. A valve that leaks again after treatment tells you something. Either the product did not suit the service, or the leak path needs repair rather than injection.

Records also pay for the programme. Survey findings, treatments and recheck results are real evidence for justifying a valve maintenance budget to management.

Standards help at the edges. ISO 15848 type tests valve designs for fugitive emissions, and says nothing about your installed population. Your own LDAR data does. See ISO 15848 for valves for what the standard covers.

4. Survey, treatment and recheck in order

A workable cycle keeps both programmes honest. The steps are simple, and the order matters.

1. Survey the population and tag every leaking valve.
2. Triage each tag by leak path and by the injection gates.
3. Treat in service where the gates are met, using controlled amounts.
4. Recheck within the repair window your LDAR programme defines.
5. Record the recheck result against the same tag.

The recheck separates screening from confirmation. A detector reading after treatment suggests the leak has stopped. It does not prove it. Confirmation follows your site leak definition.

For a seat leak, the confirming test is a seat leak test under agreed conditions. Examples include pressure decay across the closed valve, or double block and bleed monitoring. A witnessed leak test also serves.

API 598 and ISO 5208 are factory acceptance references, not field test results. A valve condition survey can start the cycle where LDAR data is thin.

5. The limits of integration

Integration has honest limits, and naming them saves money. LDAR is periodic, so leaks that start between surveys stay unseen until the next round. A monitoring programme does not replace maintenance between surveys.

Injection cannot fix everything. Damaged stem threads, eroded seats and body leaks usually need repair or replacement. If a treated valve fails its recheck, stop. The right next step is a proper assessment, not another injection attempt.

Product choice carries its own risk. Industry knowledge reports that organic sealants can harden in dry gas service. A hardened sealant can seize a valve, so match the product to the service before injecting.

Some findings belong to engineering, not to maintenance. The record should say which.