Skip to main content

Closed-Loop Hydraulic Position and Force Control Integrated with a PLC

A closed-loop hydraulic axis regulates a cylinder's position, its force, or both, by measuring the result and correcting the valve command many times a second, and on a heavy machine that loop has to work alongside the PLC that sequences everything else. UTEC Industrial designs, engineers, machines, fabricates, and installs custom material handling systems for aerospace and heavy industry from its Spokane Valley, WA facility, integrating Allen-Bradley PLC and motion control with in-house CNC machining, heat treating, and stress relief. This article covers where the position and force loops close, how force is measured and handed over from position control, how a force-controlled axis is sized, and how a hydraulic motion controller exchanges data with an Allen-Bradley PLC. Valve selection is covered in the companion article on proportional versus servo valves; this one starts where that choice ends, at the controls and tuning links of the chain design → engineering → parts machining → fabrication → assembly → weld fatigue → stress relief → drives → controls → tuning → monitoring.

Where should the control loop close on a PLC-controlled hydraulic axis?​

There are three places the axis loop can run, and a heavy machine can use more than one of them. Bosch Rexroth's brochure for its electrohydraulic motion controllers describes them as scalable, from single-axis controls integrated in the drive and multi-axis controls installed in the control cabinet through to complete machine control with an integrated PLC. Its overview runs from valves with integrated axis control, through cabinet axis controllers (one family covers one to four axes), to modular motion control for 2 to 32 axes. Its architecture diagram for the one-to-four-axis controller shows a higher-level PLC for logic control above the drive controllers, connected over PROFIBUS, PROFINET, or EtherNet/IP.

  • In the valve. The valve's on-board electronics close the spool loop and, in the valve-integrated case, the axis loop as well.
  • In a dedicated hydraulic motion controller. A controller built for hydraulic axes closes the position and force loops at a fixed rate and takes setpoints and commands from the PLC. Delta Computer Systems' controller manual describes a deterministic controller that always runs at its loop time setting; on one model the selectable loop times run from 250 µs to 4,000 µs, and the 250 µs setting supports only one axis. Rexroth's brochure gives "variably adjustable scan times of up to 0.5 ms per axis".
  • In the PLC. Logix 5000 controller tasks can be configured as continuous, periodic, or event, and a periodic task performs a function at a specific time interval.

Moving a fast loop out of the PLC is engineering judgment, not a published limit, and the next three sentences set out that judgment. A servo position or force loop on a stiff hydraulic axis needs a sample period that is short and constant compared with the axis dynamics. A dedicated controller holds that period whatever else the machine is doing, while a PLC task shares the processor with the sequence, interlock, and communications logic. Slow loops, such as a clamp pressure or a power-unit pressure setpoint, can stay in the PLC. UTEC Industrial, a Rockwell Automation Recognized System Integrator, programs the PLC side of this split on Allen-Bradley ControlLogix and CompactLogix controllers (Bosch Rexroth R999000069 2011, pp. 4 to 7 and p. 11; Delta Computer Systems 2023, §3.2.2 pp. 54 to 55; Rockwell Automation 1756-RM094N-EN-P-2025, Ch. 5 pp. 39 and 41).

Which hydraulic loops can a PLC close on its own?​

A Logix PLC can close a loop itself when the process is slow compared with its task period. Rockwell Automation's manual says the easiest way to implement its enhanced PID instruction is a function block routine in a program in a periodic task, and that in periodic timing mode the instruction uses the periodic task's update rate as its delta-t. In engineering terms, the loop's tuning then assumes that the task runs on schedule. The manual's first example and its initialization notes describe three features that carry over to a hydraulic pressure loop:

  • Transducer failure. Optionally wiring the analog input's fault indicator to the instruction's process-variable fault input forces the loop into Manual when the input is faulted, and stops the output from winding up or down while the signal is not available, for example when a pressure transducer fails or its wiring breaks.
  • Startup bump. Initializing the control variable avoids a bump in the valve or pump command when the loop is first enabled.
  • Cascade. Primary and secondary loops can be cascaded, for example an outer loop that trims the setpoint of an inner pressure loop.

Hydraulic duties that suit a PLC loop include holding a power-unit pressure setpoint, holding a clamp pressure while a part is machined or welded, and trimming a pressure that changes over seconds or minutes rather than milliseconds. The manual also states that the enhanced PID instruction is not supported in safety applications, or on the ControlLogix 5590 family, where it directs users to a different PID instruction. A PLC loop that regulates pressure remains a control function, not a safety function (Rockwell Automation 1756-RM006P-EN-P, 2025, p. 3 and pp. 93 to 94; Rockwell Automation 1756-RM094N-EN-P-2025, Ch. 5 p. 39).

What does the position loop need from its feedback sensor?​

The position loop can be no better than its transducer. Delta Computer Systems' design note recommends a magnetostrictive linear displacement transducer mounted in the hydraulic cylinder. Temposonics' operation manual describes its rod-style sensor as suitable in all fluid power cylinders and explains the principle: a current pulse in a waveguide interacts with the field of a position magnet that moves with the piston, launching a torsional strain pulse back along the waveguide, and the time that pulse takes to reach the sensor head gives an absolute position.

Figures from one 2018 manual for an industrial magnetostrictive sensor with an EtherNet/IP interface show what to read on a data sheet. Most of them are limits, not typical values:

FigureStated value
ResolutionSelectable, 1 to 1,000 µm
Cycle time1.0 ms up to 2,000 mm stroke; 2.0, 3.0, or 4.0 ms for longer strokes
LinearityBelow ±0.01 % of full stroke, minimum ±50 µm
RepeatabilityBelow ±0.001 % of full stroke, minimum ±2.5 µm
HysteresisBelow 4 µm

The percentages scale with stroke. On a 1,500 mm stroke, the linearity limit is 0.01 % × 1,500 mm = 150 µm, while the repeatability limit is 0.001 % × 1,500 mm = 15 µm. An axis that returns to a taught position depends on the 15 µm figure; one that must hit an absolute dimension from a drawing depends on the 150 µm figure. The cycle time also rises on strokes over 2,000 mm. Delta Computer Systems' design note lists Start/Stop, PWM, SSI, and analog outputs for these transducers and does not recommend the analog output for best control, because of its susceptibility to noise; it also says not to run the amplifier's PWM valve wiring in the same raceway as the transducer wires (Temposonics 551283 Rev. D 2018, §4.1 p. 8, §4.3 p. 11, and §9 p. 35; Delta Computer Systems, DOs and DON'Ts of Hydraulic Position Control System Design, accessed 2026).

How is cylinder force measured: one pressure transducer, two, or a load cell?​

Delta Computer Systems' manual lists the pressure and force inputs its controllers accept: a pressure transducer; a load cell, either through a signal conditioner with a ±10 V or 4 to 20 mA output or connected directly as an mV/V load cell; or two pressure transducers, for example one on each end of the cylinder, for differential force. Bosch Rexroth's controller brochure also lists force control with differential-pressure evaluation.

The choice matters on a differential cylinder, because rod-end pressure pushes back on a smaller annulus. Net force is F = p_A × A_A − p_B × A_B. A worked example with assumed values:

  • Bore 6.0 in, so the cap-end area A_A = π × 6.0² ÷ 4 = 28.27 in²
  • Rod 3.0 in, so the annulus area A_B = 28.27 − 7.07 = 21.21 in²
  • Cap-end pressure p_A = 2,000 psi; rod-end pressure p_B = 500 psi

F = 2,000 × 28.27 − 500 × 21.21 = 56,540 − 10,605 ≈ 45,900 lbf. A single cap-end transducer would report 56,540 lbf, about 23 % high, and that error changes every time the rod-end back-pressure changes. The example assumes static conditions, no seal friction, and pressures measured at the cylinder ports rather than at the valve.

Seal friction is the other error in a pressure-derived force. Moog's sizing guidance takes seal friction as 10 % of stall force when it is not known, and a pressure measurement cannot tell friction from load. A load cell in the load path measures the force the part actually sees, which suits a press, a test fixture, or a clamp where the force on the part is the specified quantity, at the cost of a sensor that sits in the load path and has to survive it (Delta Computer Systems 2023, pp. 171 to 172 and p. 176; Bosch Rexroth R999000069 2011, p. 10; Moog CDL6566 Rev D 2002, pp. 4 to 5).

How does an axis hand over from position control to force control?​

Some heavy axes need both loops in one stroke: a press ram that approaches in position control and then presses to a force, a clamp that closes on a casting and then holds a clamping force, or a straightening or proof-load fixture that moves to contact and then applies a programmed load. Delta Computer Systems' manual calls this dual-loop position-pressure or position-force control, in which the axis runs in position control and then transitions to pressure or force control. Bosch Rexroth's brochure lists a comparable function, alternating control: positioning with automatic transition to closed-loop force control and back. Its programming example is an axis that moves toward a target position and, if a defined force threshold is exceeded while it is traveling (the brochure's example is a press), changes over to force control without a jerk.

The changeover trigger belongs in the specification: a position, a force threshold, or both. A badly handled changeover can show up as a force spike on the part at contact, or as a hunt between the two loops. In one study of force tracking, Liu and Alleyne, working on an electro-hydraulic actuator at the University of Illinois, developed a model-based nonlinear controller with parameter adaptation for force and pressure tracking, and discussed friction modeling and compensation for pressure tracking with experimental results. As engineering reasoning, because a force loop's response depends on the stiffness of what it pushes against, gains tuned on one part or one fixture may not hold on the next, so the acceptance test should cover the stiffest and softest cases the axis will meet (Delta Computer Systems 2023, §3.7.3 p. 176; Bosch Rexroth R999000069 2011, p. 10 and p. 15; Liu and Alleyne 2000, Abstract).

What is the difference between force control and a force limit during a move?​

A force limit and force control are separate functions, and confusing them produces the wrong behavior at the moment of contact. In force control, the controller regulates force to a setpoint, and position becomes whatever results. In a pressure or force limit, the axis stays in position control, and the controller holds back the valve command when force reaches the set limit, so an obstruction, a mis-seated part, or a jammed fixture does not see the full stall force of the cylinder. Delta Computer Systems' manual lists pressure or force limit as a separate feature from pressure or force control, describes it as a special type of pressure or force control that limits pressure or force during the position or velocity motion of an axis, and states that it cannot be used simultaneously with pressure or force control.

The distinction appears in a specification as two different sentences:

  • "Press to 40,000 lbf and hold for 5 s" describes force control.
  • "Move to 850 mm and do not exceed 5,000 lbf on the way" describes a position move with a force limit.

A clamp that must never crush a thin-walled aerospace part needs the second function even though it never regulates force on purpose. Neither function replaces the hydraulic circuit's own pressure protection, which is designed to ISO 4413:2010, the general rules and safety requirements for hydraulic fluid power systems and components used on machinery as defined by ISO 12100 (Delta Computer Systems 2023, p. 171 and p. 178; ISO 4413:2010).

How are the cylinder and valve sized for a force-controlled axis?​

Moog's guide to electrohydraulic valves sets out this sizing sequence:

  • Total force is the load force plus the acceleration force, the external force, and seal friction, with seal friction taken as 10 % of stall force unless it is known.
  • The actuator area is chosen so that stall force is 30 % above the required force.
  • For a servovalve, supply pressure is set so that the pressure drop across the valve is equal to one-third of the supply pressure.
  • The loaded flow at maximum velocity is converted to a no-load flow using the load pressure drop, Q_NL = Q_L × √(P_S ÷ (P_S − P_L)).
  • Rated valve flow is chosen at a 1,000 psi drop for a servovalve, or at a 150 psi drop for a proportional valve, plus 10 %.
  • For closed-loop control of systems using electrical feedback, the guide says optimum performance will be achieved if the servovalve's 90° phase point exceeds the load resonant frequency by a factor of three or more.

A worked example, with assumed inputs: a horizontal press axis must deliver 30,000 lbf of process force, accelerate a 20,000 lb moving mass at 20 in/s², and advance at 4 in/s from a 3,000 psi supply.

  • Acceleration force = (20,000 ÷ 386.1) lbf·s²/in × 20 in/s² ≈ 1,036 lbf
  • Load, acceleration, and external force = 30,000 + 1,036 + 0 = 31,036 lbf
  • Stall force: Moog sets the actuator area for a stall force 1.3 times the required force, and counts seal friction in the required force as 0.1 times stall force. Writing the two rules as one equation is UTEC's worked illustration, not an equation printed in the guide: F_s = 1.3 × (31,036 + 0.1 F_s), so F_s = 1.3 × 31,036 ÷ 0.87 ≈ 46,400 lbf
  • Required force F_R = 31,036 + 0.1 × 46,400 ≈ 35,700 lbf
  • Cap-end area = 46,400 lbf ÷ 3,000 psi ≈ 15.5 in², a 4.44 in bore; the next practical size up is 4.5 in (15.90 in²)
  • Loaded flow at 4 in/s = 15.90 in² × 4 in/s = 63.6 in³/s; load pressure drop P_L = F_R ÷ area ≈ 2,243 psi (unrounded, 35,674 lbf ÷ 15.904 in²), so P_S − P_L ≈ 757 psi
  • No-load flow Q_NL = 63.6 × √(3,000 ÷ 757) ≈ 126.6 in³/s
  • Rated flow at a 1,000 psi valve drop: the guide's rated-flow relation, Q_NL = Q_R × √(P_S ÷ ΔP) with ΔP = 1,000 psi, gives Q_R = 126.6 × √(1,000 ÷ 3,000) ≈ 73.1 in³/s, or 73.1 ÷ 3.85 ≈ 19.0 gpm, and with the guide's 10 % margin about 20.9 gpm for a servovalve rated at a 1,000 psi drop

As engineering reasoning from the guide's own definitions, its step (d) shortcut, Q_R = 1.1 × (Q_NL ÷ 3.8), agrees with that rated-flow relation only at a 1,000 psi supply; applied at 3,000 psi it would give about 36.7 gpm, a valve about 1.75 times the needed rating, and the guide's step (g) says it is preferable not to oversize the servovalve flow capacity, as this will needlessly reduce system accuracy. As a cross-check by the same reasoning, the loaded flow at the actual 757 psi valve drop, converted to a 1,000 psi drop by the same square-root law, is 63.6 × √(1,000 ÷ 757) ≈ 73.1 in³/s, the same rating. Assumptions: stall force equals supply pressure times cap-end area with the rod end vented to tank, the 30 % margin and 10 % friction allowance apply together, the full force acts at the 4 in/s advance speed, and line losses are ignored. A proportional valve rated at a 150 psi drop has to be converted to the actual valve drop by the square-root flow law worked through in the valve selection article, which also shows how moving mass and trapped oil volume set the load resonance.

The phase-lag rule then screens the valve class. The guide states the rule for a servovalve, so applying it to every valve class, as here, is engineering reasoning. If this axis has an assumed load resonance of 15 Hz, the rule puts the valve's 90° phase point at 45 Hz or more. Plummer's example values for a 4-way valve at 40 L/min and 70 bar put the 90° phase frequency at about 5 Hz for an open-loop proportional valve, 10 Hz for a position-controlled proportional valve, 50 Hz for a force-motor direct-drive valve, and 100 to 200 Hz for two-stage servovalves, so, on those example values, a 45 Hz requirement rules out the first two classes. ISO 10770-1:2009 describes methods for determining the performance characteristics of electrically modulated, hydraulic, four-port directional flow-control valves, so a specification for such a valve can ask for performance data measured to it (Moog CDL6566 Rev D 2002, pp. 4, 5, 8 and 9; Plummer 2016, Table 2; ISO 10770-1:2009).

Which installation details decide whether the loop can be tuned?​

Delta Computer Systems' design note lists the considerations for a high-performance closed-loop hydraulic position axis, and the mechanical and hydraulic ones are fixed before a controls engineer ever connects a laptop:

  • Valve. Use a servo-quality valve or a zero-overlap proportional valve with linear flow per input; an overlapped or closed-center spool gives poor control around zero and a "thud" at each change of direction.
  • Valve location. Mount the valve on the cylinder and keep the trapped oil volume small.
  • Hose. Do not put a flexible hose between the valve and the cylinder.
  • Accumulator. Put an accumulator near the valve.
  • Feedback. Mount a magnetostrictive position transducer in the cylinder; analog output is not recommended for best control because of its susceptibility to noise.

As engineering reasoning, a hose between valve and cylinder adds both volume and compliance, which lowers the hydraulic natural frequency and with it the gain the loop will accept. On the valve side, Plummer's review notes that electrical rather than mechanical spool-position feedback allows higher loop gains and correction of errors due to hysteresis or temperature, that industrial valves began to adopt it in the 1970s, and that a landmark 1973 valve had on-board electronics to close the loop. Where a hose is unavoidable, as on an axis that articulates, the same reasoning says the loop should be expected to run at a lower gain and the positioning tolerance should be set to match (Delta Computer Systems, DOs and DON'Ts of Hydraulic Position Control System Design, accessed 2026; Plummer 2016, §3).

What happens to a load-holding or vertical axis when the loop drops out?​

A closed-loop axis that holds a load against gravity, such as a lift table, a gate cylinder, or a tilting fixture, needs a defined state when the controller, the valve power, or the command signal is lost. Parker's operation manual for a closed-loop proportional valve with on-board electronics describes what the valve itself does when it is switched off. When there is no supply voltage or no enable signal, or the current command on the 4 to 20 mA option is below 3.8 mA, a zero-lap version of the valve takes its power-down position of approximately 10 % opening, according to the ordering code, which is not the same as its hydraulic neutral; an overlap-spool version powers down to its zero position. A valve that powers down 10 % open can let a load drift unless the circuit holds it.

That is valve behavior, not a safety function. Load holding is a question for the hydraulic circuit and the safety-related control system:

  • ISO 4413:2010 specifies general rules and safety requirements for hydraulic fluid power systems and components used on machinery as defined by ISO 12100.
  • ISO 13849-1:2023 covers the design and integration of safety-related parts of control systems that perform safety functions, in high demand and continuous modes of operation; it does not apply to low demand mode.
  • ISO 12100:2010 specifies principles of risk assessment and risk reduction for achieving safety in the design of machinery, including procedures for identifying hazards.

Delta Computer Systems' design note says not to use counterbalance valves, because they counteract the motion controller, and, where a safety valve is needed to keep a vertical load from falling, to use a blocking valve between the valve and the cylinder. The note's stated reason is control performance, and UTEC Industrial does not treat a controls supplier's design note as guidance on how to hold a suspended load safely; that is settled by the circuit design under the standards above. The valve selection article explains why a valve's own spool feedback must not be used to switch off a safety function (Parker Hannifin MSG11-5715-742/UK 2024, p. 13; ISO 4413:2010; ISO 13849-1:2023; ISO 12100:2010; Delta Computer Systems, DOs and DON'Ts of Hydraulic Position Control System Design, accessed 2026).

How does a hydraulic motion controller exchange data with an Allen-Bradley PLC?​

When the fast loops run in a dedicated controller, the PLC still owns the sequence, setpoints, permissives, and operator interface. Delta Computer Systems' manual describes ControlLogix and CompactLogix controllers exchanging data with its motion controllers over EtherNet/IP I/O at the requested packet interval (RPI), with one connection of up to 125 input and 124 output registers on two of its Ethernet models and up to three such connections on a third model. A magnetostrictive sensor with an EtherNet/IP interface joins the same Logix I/O configuration, added either from its EDS file or as a Generic Ethernet Module. EtherNet/IP carries CIP over standard Ethernet, and cyclic I/O of this kind is its implicit messaging.

The design points that follow are:

  • One register map. Each axis's command, setpoints, mode, and status words are mapped once, in a documented table, so the PLC never writes the same register from two routines.
  • RPI for the logic, not the loop. The RPI is chosen for how often the PLC logic needs fresh axis status; the fast loop does not depend on it, because it runs inside the motion controller at the controller's own loop time.
  • Loss of connection. A lost connection is treated as a fault that stops the axis in a defined way, not as stale data.

The rule of thumb for choosing an RPI for an EtherNet/IP device is worked through in the FANUC robot and Allen-Bradley PLC article (Delta Computer Systems 2023, §6.10.2 pp. 628 to 629; Temposonics 551283 Rev. D 2018, §7; ODVA PUB00138R8-2024).

How are the position and force loops tuned and proven?​

Tuning order matters on a dual-loop axis. Delta Computer Systems' manual instructs setting up and tuning the position control first, before the pressure or force transducer is set up and scaled and the pressure or force is tuned. The order makes sense because the force loop takes over from a position move that has to be stable before the changeover can be judged. Jelali and Kroll's text on hydraulic servo-systems covers physically based modelling, experimental modelling (identification), and control design in successive chapters. Its description says it details nonlinear identification and control developments such as feedback linearisation, and also reviews standard approaches such as linear state feedback, feedforward control, and compensation for static nonlinearities. Manring and Fales's textbook has chapters on valve-controlled and pump-controlled hydraulic systems.

On the machine, a tuning record should capture:

  • The step and ramp moves used to tune the position loop, and the following error at design speed.
  • The changeover trigger, and the force overshoot at changeover.
  • The force-loop response on the stiffest and the softest workpiece or fixture the axis will see.
  • The oil temperature during each test.

UTEC Industrial proves these results on the assembled machine during factory acceptance testing (FAT) and repeats the key tests at on-site commissioning, when the production oil, the building temperature, and the real parts are present (Delta Computer Systems 2023, pp. 177 to 178; Jelali and Kroll 2003, Ch. 4 to 6; Manring and Fales 2019).

What sensing, controls, and diagnostics does a closed-loop hydraulic axis need?​

The intelligence layer of a PLC-integrated hydraulic axis has more parts than the loop itself:

  • Position feedback. An absolute linear sensor, such as a magnetostrictive transducer, on each closed-loop cylinder.
  • Force feedback. Two pressure transducers for differential force, or a load cell where the force on the part is the controlled quantity.
  • Valve electronics and diagnostics. Parker's closed-loop proportional valve with on-board electronics accepts a ±10 V, ±20 mA, or 4–12–20 mA command, returns a spool-stroke diagnostic signal, lists sensitivity below 0.03 % and hysteresis below 0.05 % (below 0.1 % on one variant), and can be parameterized over IO-Link. Its solenoid-current monitoring switches the valve's actuator off, to prevent overheating, when the actuator current time interval exceeds 10 s; the manual says this state will not be reached under normal operating conditions but may occur with a contaminated, sluggish valve.
  • Fluid condition. The same manual requires a fluid cleanliness code of 18/16/13, and ISO 4406:2021 specifies the code used to define the quantity of solid particles in the fluid of a hydraulic fluid power system.
  • Controllers and HMI. The motion controller for the fast loops, the PLC for sequence, interlocks, and slow loops, and an operator interface that shows mode, setpoint, actual value, and faults for each axis.
  • Safety-related parts. Emergency stop, guarded-zone entry, and load-holding functions on safety-related control parts designed to ISO 13849-1:2023, kept separate from the motion loop.

Plummer describes integrating self-tuning functions, condition monitoring, and increased communication capability as a continuing trend in industrial valves. Trending the valve diagnostics alongside following error and standstill command, as the valve selection article describes, can catch contamination before it stops the axis. UTEC Industrial builds this layer on Allen-Bradley ControlLogix and CompactLogix controllers with PanelView and FactoryTalk operator interfaces over EtherNet/IP, in control panels built to UL 508A (Parker Hannifin MSG11-5715-742/UK 2024, pp. 4 to 6, p. 13 and p. 18; ISO 4406:2021; ISO 13849-1:2023; Plummer 2016, §5).

Where does closed-loop control sit in the chain from frame design to monitoring?​

Tuning and monitoring are the last links in the chain, but a force loop inherits every earlier one. The structure between the cylinder and the workpiece is a spring in series with the oil column and the load cell. Moog's guide ties hydraulic stiffness to the bulk modulus, the piston area, and the trapped volume (its equations are for an equal-area, double-ended cylinder), and combines it with the structural stiffness to give the overall stiffness that sets load resonance. The rest of this paragraph is engineering reasoning. The frame is part of that structural stiffness. A press frame or clamp arm that deflects under the 46,000 lbf of the sizing example stores energy the force loop has to push through, and how quickly force rises per unit of cylinder travel is what the force-loop gain acts on. A weldment that relaxes as residual stresses redistribute can also move the position the transducer was calibrated to.

  • Design and engineering fix the force budget, cylinder area, valve location, and load-cell position.
  • Parts machining produces the cylinder mounts, clevises, and any bore for a rod-style position sensor.
  • Fabrication, weld fatigue, and stress relief produce a frame that stays stiff and stable under repeated force cycles.
  • Drives, controls, tuning, and monitoring close the loops within the stiffness the structure allows, and watch for drift.

UTEC Industrial stress-relieves weldments, including by automated vibratory stress relief, and machines to tolerances as tight as ±0.001 in on the mounts and bores that locate a cylinder and its sensor. From the controls side, Delta Computer Systems' design note says to mount valves on the cylinders and not to use flexible hose between the valve and the cylinder (Moog CDL6566 Rev D 2002, p. 5; Delta Computer Systems, DOs and DON'Ts of Hydraulic Position Control System Design, accessed 2026).

What should a specification for a PLC-integrated closed-loop hydraulic axis define?​

A specification that says only "closed-loop hydraulic control" leaves the architecture, the sensors, and the acceptance criteria to the supplier. For each closed-loop axis it should state:

  • The controlled variable in each phase of the cycle (position, force, or pressure), the changeover trigger between them, and any force limit during position moves.
  • The positioning tolerance as absolute accuracy or as repeatability, since on a long stroke the example magnetostrictive sensor's linearity and repeatability limits differ by a factor of ten.
  • The force range and tolerance, and whether force is measured by differential pressure or by a load cell.
  • Where each loop closes (valve, dedicated motion controller, or PLC) and the loop time or task period each needs.
  • The PLC data interface: the register map, the RPI, and the behavior on loss of connection.
  • The valve class, sized to the force budget and the three-times phase-lag rule, with performance data measured to ISO 10770-1:2009.
  • The power-down state of each valve, and the circuit and safety-related control provisions for load holding under ISO 4413:2010 and ISO 13849-1:2023.
  • The tuning and acceptance tests at FAT and at site, with the oil temperature recorded.

Each item is then a measurable acceptance criterion rather than a description (Temposonics 551283 Rev. D 2018, §9 p. 35; Delta Computer Systems 2023, §3.7.3 p. 176 and p. 178; Moog CDL6566 Rev D 2002, pp. 4 to 5; ISO 10770-1:2009; ISO 4413:2010; ISO 13849-1:2023).

Related Articles

References​

  • Delta Computer Systems RMC70/150/200 and RMCTools User Manual, Version 4.23.2: RMC70/150/200 Motion Controllers and RMCTools Software User Manual. Delta Computer Systems, 2023.
  • Delta Computer Systems: DOs and DON'Ts of Hydraulic Position Control System Design. Delta Computer Systems (undated web documentation, accessed September 2026).
  • Moog CDL6566 Rev D: Electrohydraulic Valves... A Technical Look. Moog Inc., Industrial Controls Division, 2002.
  • Manring, N. D., Fales, R. C. Hydraulic Control Systems, 2nd ed. John Wiley & Sons, 2019.
  • Liu, R., Alleyne, A. (2000). "Nonlinear Force/Pressure Tracking of an Electro-Hydraulic Actuator." Journal of Dynamic Systems, Measurement, and Control, 122(1), 232-236. DOI 10.1115/1.482466
  • Rockwell Automation 1756-RM006P-EN-P: Logix 5000 Advanced Process Control and Drives Instructions. Rockwell Automation, 2025.
  • Rockwell Automation 1756-RM094N-EN-P-2025: Logix 5000 Controllers Design Considerations. Rockwell Automation, 2025.
  • Temposonics 551283 Rev. D: Temposonics® R-Series EtherNet/IP Operation Manual. Temposonics, 2018.
  • ISO 10770-1:2009: Hydraulic fluid power — Electrically modulated hydraulic control valves — Part 1: Test methods for four-port directional flow-control valves. ISO, 2009.
  • Plummer, A. (2016). "Electrohydraulic servovalves – past, present, and future." Proceedings of the 10th International Fluid Power Conference (IFK2016), Dresden, Germany.
  • Parker Hannifin MSG11-5715-742/UK: Proportional DC Valves Operation Manual, Series DFplus, Design > 50. Parker Hannifin, 2024.
  • Bosch Rexroth R999000069: HNC100-3X Programmable motion control for electrohydraulic drives. Bosch Rexroth AG, 2011.
  • Jelali, M., Kroll, A. Hydraulic Servo-systems: Modelling, Identification and Control, 1st ed. Springer, 2003.
  • ODVA PUB00138R8-2024: EtherNet/IP — CIP on Ethernet Technology. ODVA, 2024.
  • ISO 4406:2021: Hydraulic fluid power — Fluids — Method for coding the level of contamination by solid particles. International Organization for Standardization, 2021.
  • ISO 4413:2010: Hydraulic fluid power — General rules and safety requirements for systems and their components. International Organization for Standardization, 2010.
  • ISO 13849-1:2023: Safety of machinery — Safety-related parts of control systems — Part 1: General principles for design. International Organization for Standardization, 2023.
  • ISO 12100:2010: Safety of machinery — General principles for design — Risk assessment and risk reduction. ISO, 2010.

Ready to Discuss a Material Handling System?​

UTEC Industrial designs, engineers, machines, fabricates, and installs custom material handling systems for heavy industry, from the stress-relieved structure and drives to the Allen-Bradley PLC controls, tuning, and monitoring that run them, at its Spokane Valley, WA facility. Send UTEC the application, loads, and duty cycle to start a system review.

Request a Quote →

Questions? Call (509) 922-1832 or email sales@utec.co

Last updated on