Holds the second part of a circuit closed until pressure in the first reaches a set value, then opens. That makes the second actuator physically unable to start until the first has completed its work, with the pressure itself as the evidence of completion.

A timer assumes the first action completed; a sequence valve senses that it did. Doing it hydraulically needs no controller, works regardless of workpiece variation and cycle speed, and is harder to defeat - which on a simple machine makes it both cheaper and more trustworthy than software.

Clamp-then-cut. The workpiece must be clamped before machining begins, and the valve prevents the machining stroke from starting until clamp pressure has genuinely built. Any two-stage operation where the second stage must not begin early follows the same pattern. The valve is set once rather than programmed and maintained in software.

Against the pressure that genuinely indicates the first action is complete - high enough that it truly has finished, and comfortably below the relief setting so the system can still reach it. Setting against system pressure rather than against the completion pressure is the usual error.

Because the spring chamber must not see back pressure, which would add to the spring force and raise the effective setting unpredictably. Most sequence valves are externally drained for that reason, and an internally drained valve on a circuit with return line pressure sequences erratically.

A sequence valve is normally closed and opens at a set inlet pressure to pass full flow onward. A reducing valve is normally open and closes to maintain a lower outlet pressure. One controls the order of events; the other controls the pressure level in a branch.

A related valve type - the unloading valve - does that, opening at a set pressure to return pump flow to tank at low pressure once an accumulator is charged. It shares the pressure-sensing principle but the purpose is energy saving rather than sequencing.