PIDSnap

Valve stiction symptoms

Stiction is static friction in the valve — the stem resists moving until the signal has changed enough to break it free, then slips further than intended. It is one of the most common causes of poor loop performance and one of the easiest to identify once you know the signature.

It matters because stiction is invisible to tuning. A loop cycling from stiction can be detuned until the cycle is slow enough to ignore, but the cycle is still there, the valve is still wearing, and the loop has lost its ability to reject real disturbances.

Valve stiction producing a limit cycle Trend showing valve stiction: the controller output ramps smoothly while the process variable moves in abrupt steps, producing a limit cycle that does not decay. high low PV SP Output PIDSnap time
Trend showing valve stiction: the controller output ramps smoothly while the process variable moves in abrupt steps, producing a limit cycle that does not decay. A sawtooth or square-cornered cycle on the process variable against a smooth, slowly ramping output. The cycle amplitude stays constant instead of decaying. Put the loop in manual and the cycle stops — that single test separates stiction from every tuning problem.

The signature

On a trend, stiction produces a sawtooth or square-cornered process variable against a smooth, slowly ramping controller output. The output ramps because integral action is winding up against a stem that will not move; the process variable jumps because when the stem finally breaks free it goes past the position that was asked for.

The defining property is that the cycle amplitude stays constant. It does not decay, because the stiction band is the same size on every cycle no matter how small the error has become. Every tuning-related oscillation decays. This one does not.

Three confirmations, in order of effort

First: put the loop in manual. Stiction cycling stops immediately, because nothing is winding up against the stem any more.

Second: make a series of small output changes in manual, all in the same direction, each a fraction of a percent. A healthy valve produces a small process response to each one. A sticking valve produces nothing, nothing, nothing, then a jump.

Third, if you have position feedback: plot commanded position against actual position over a period of normal operation. Stiction shows as a parallelogram rather than a straight line.

Why stroking the valve proves nothing

The most common false negative in this diagnosis is stroking the valve through its full travel and concluding it moves fine. It does — a 100% signal change overcomes stiction easily.

Stiction bites on the movements a controller actually makes, which around a steady operating point are fractions of a percent. Test at the scale the loop works at, not at the scale that is convenient.

What causes it, most likely first

Packing adjusted too tight

most common tuning cannot fix this

The most frequent origin, and it often appears shortly after maintenance. Packing tightened to stop a leak adds friction the actuator has to overcome on every movement.

How to confirm it
Check when the valve was last repacked against when the complaint started.
What to do
Have the packing adjusted, and check the actuator has enough thrust for the packing that is fitted.

No positioner on a throttling valve

common tuning cannot fix this

Without a positioner the actuator balances air pressure against spring force and friction, so the stem goes where friction allows rather than where the signal asked.

How to confirm it
Check whether a positioner is fitted at all.
What to do
Fit one. This is the durable fix for a throttling service.

Corrosion or deposits on the stem

common tuning cannot fix this

Product build-up or corrosion in the packing area, common on crystallising or polymerising services.

How to confirm it
Physical inspection at the next opportunity.
What to do
Clean and inspect. Consider a different packing arrangement.

Questions that come up

Can I tune around stiction?

Only in the sense that you can make the cycle slow enough to be less annoying, by detuning heavily. The cycle remains, the valve keeps wearing, and the loop loses its ability to handle real disturbances. It is a way of hiding the problem, not solving it.

How much stiction is acceptable?

It depends on what the loop has to do. A rule of thumb is that stiction much above a fraction of a percent of travel will produce a visible limit cycle on a fast loop. Slower loops tolerate more because the cycle period stretches out.

Related

Last reviewed 2026-08-01.

Run a free PIDSnap check on this loop

Two photos, thirty seconds, no signup. The pre-flight check finds the structural problems that no amount of tuning will fix — before you touch the controller.

Start a free check