PIDSnap

Valve deadband and backlash

Backlash is lost motion: the signal changes, and nothing happens at the stem until the play in the linkage has been taken up. It is closely related to stiction and often confused with it, but the signatures differ in a way that is easy to see once you know what to look for.

Stiction resists movement in any direction and produces stick-slip. Backlash only bites on reversals — a valve moving steadily in one direction shows no backlash at all.

Backlash and deadband in a valve linkage Trend showing backlash: the process variable stays flat while the controller output reverses, then follows once the deadband has been taken up. high low PV SP Output PIDSnap time
Trend showing backlash: the process variable stays flat while the controller output reverses, then follows once the deadband has been taken up. Flat spots on the process variable exactly where the output changes direction. The width of the flat spot is the size of the deadband. This is lost motion in the linkage or the actuator coupling, not a tuning problem.

The signature: flat spots at reversals

On a trend, backlash appears as a flat section on the process variable occurring exactly where the controller output changes direction. The output reverses, the process variable holds still for a while, and then resumes following once the play has been taken up.

The width of that flat spot, expressed as a percentage of output travel, is the size of the deadband. That gives you a number to take to whoever will be doing the mechanical work.

How to measure it deliberately

In manual, move the output in one direction until the process variable is responding cleanly. Then reverse and make small steps in the opposite direction, noting the total output change before the process variable starts to move.

That total is the deadband. Repeat it in the other direction — the two are often different, which itself is diagnostic.

What causes it, most likely first

Worn or loose stem connector

most common tuning cannot fix this

Play at the coupling between actuator and stem.

How to confirm it
Physical inspection, or the deadband measurement above.
What to do
Tighten or replace the connector.

Positioner feedback linkage play

common tuning cannot fix this

Wear in the linkage that tells the positioner where the stem actually is.

How to confirm it
Deadband present with the positioner in service and different with it bypassed.
What to do
Inspect and adjust the feedback linkage.

Worn rotary valve shaft or coupling

common tuning cannot fix this

Ball and butterfly valves drive through a shaft and coupling that wear, producing rotational play that shows up as deadband.

How to confirm it
More common on rotary valves than on globe valves; measure the deadband as above.
What to do
Mechanical repair. This does not improve on its own.

Questions that come up

How is backlash different from stiction on a trend?

Stiction produces stick-slip in both directions and a sustained limit cycle even when the output is moving steadily one way. Backlash only appears at reversals — flat spots on the process variable exactly where the output changes direction, and nothing unusual while the output moves steadily.

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