Why do turbine valves need their own maintenance program?

Because a turbine valve is a safety device first and a flow device second. Your bearings and blades degrade gradually and give you condition signals — vibration, temperatures, oil debris. Valves sit still for months, then must move once, fast, under the worst possible conditions. That duty cycle creates failure modes that standard rotating-equipment PM never touches: stem binding from oxide growth, seat leakage from wire drawing, sluggish response from galled bushings, and servomotor issues that only appear under a full-stroke demand.

The four valve classes carry different stakes:

• Main stop valves (MSV): the final overspeed trip barrier. A failure to close on demand is a catastrophic-event scenario, which is why trip testing is a compliance item, not a preference.
• Control (governor) valves: modulate constantly, so they wear — stems, bushings, seats — and their friction directly affects load response and valve positioning accuracy.
• Intercept and reheat stop valves: protect against overspeed from energy stored in the reheat circuit after the HP valves close. Often neglected because they cycle less.
• Bypass valves: erode and leak from throttling duty; a passing bypass valve costs heat rate every day and can mask itself as a condenser or attemperation problem.

Treating all four as "turbine valves — inspect at outage" is how plants end up with a sticky intercept valve discovered during a NERC or insurer-mandated overspeed test. Each class needs its own task list, interval and pass criteria.

How do you keep control valve stems and bushings from becoming your next forced outage?

Control valves are the wear items of the group because they throttle continuously. The two failure modes that matter most are rising stem friction and seat leakage. Rising friction shows up first as hunting or deadband in valve positioning — the governor calls for a move, the valve overshoots or lags, and load control gets sloppy. Left alone, friction becomes binding, and binding on a control valve during a trip event is an overspeed contributor.

A practical control valve PM program looks like this:

• Trend valve position vs. demand signal through your DCS historian. A growing gap between commanded and actual position is friction developing — catch it months before the outage.
• Exercise valves through full travel during planned load swings where the unit design allows it, and note any sticking points in the stroke.
• At outage, measure stem-to-bushing clearance and inspect for galling and oxide scale. Steam oxidation on stems operating at high temperature is progressive — if clearances are at the tight end of OEM tolerance, dress or replace now rather than next cycle.
• Inspect seats and plugs for throttling erosion. A control valve that passes steam when closed costs heat rate continuously and erodes faster the longer it leaks — a self-accelerating problem.
• Check servomotor and positioner calibration after any valve work; a mechanically perfect valve with a sloppy positioner still hunts.

Here's the objection worth naming: "We don't have condition monitoring on valves — we just overhaul them at major outages." That's fair for plants without instrumentation, but you almost certainly already have the data. Valve demand vs. actual position, hydraulic pressure during stroke tests, and closure times are all condition indicators — they just need somewhere to live and someone reviewing the trend. That's a CMMS workflow, not a sensor purchase.

What are you looking for in a bypass valve inspection?

Erosion and leakage, in that order. Bypass valves live a hard life: when they operate, they're throttling high-energy steam across a large pressure drop, which is exactly the duty that wire-draws seats and erodes plugs. When they're closed, any seat damage from the last operation becomes a continuous leak — steam you're paying to make, dumping straight to the condenser.

Inspection priorities at outage:

• Seat and plug surfaces: look for wire drawing (the thin, threaded erosion grooves that indicate a leak path under pressure) and cavitation-style damage on the plug. Minor damage can often be lapped out; deep cutting means replacement or weld repair.
• Leak-by evidence while running: a downstream pipe temperature that stays elevated with the valve commanded closed is the classic field indicator. Log it as a work order, not a mental note.
• Stem packing and actuator: bypass valves cycle under thermal transients; check packing condition and actuator response, especially on valves used for startup duty.
• Attemperation interface: spray water control problems downstream of the bypass valve often get blamed on the valve itself. Verify both before condemning either.

Decision rule: if a bypass valve shows measurable leak-by, repair it at the next available window rather than waiting for the planned major. The heat-rate penalty of a passing valve accumulates every operating day, and the erosion accelerates as the leak path grows — the repair only gets bigger.

What can you do this week to tighten your valve program?

Three concrete actions, none of which require a purchase:

• Pull the last 12 months of valve test records for your main stop and intercept valves. Can you produce closure times and pass criteria for every test within ten minutes? If not, that's your gap.
• Check where your test intervals came from — the OEM manual, an insurer requirement, or habit. If it's habit, compare against OEM recommendations and adjust with engineering sign-off.
• Walk the DCS trends for control valve demand vs. actual position over the last quarter. Any growing deadband or hunting is a work order today, not an outage surprise next year.

Then put the program where it can actually run: valve-by-valve PM schedules, test intervals, pass criteria and compliance records for every steam turbine valve type, managed as trigger-based, auditable work in OxMaint — replacing the scattered spreadsheets that make every audit a scramble.