What does valve hysteresis mean in hydraulic proportional valves?

If you've ever tweaked a hydraulic proportional valve and observed that the valve doesn't return to the same position after you return the signal to where it began, it is because you've experienced valve hysteresis. It is among the most significant yet less well-known performance features in proportional hydraulics. Knowing about it is crucial to anyone speculating on commissioning, repairing, or troubleshooting high-precision hydraulic systems.

The definition of hysteresis in simple words

Hysteresis in relation to proportional valves is the term used to describe the difference in output of the valve regardless of whether it is pressure, flow, or spool position between the increasing command signal and an inversely decreasing command signal for the same input. Also, the valve is not following the same route in the direction of ascending as it does when going down.

Imagine directing a proportional directional control valve by the voltage of 0-10 V. You increase the signal from 0 V to 10 V while observing the flow output, then you ramp it back down from 10 V back to 0 V. If your control valve was equipped with zero hysteresis, then the flow curve rising would be perfectly in line with the flow curve that is coming down. However, in the real world, they are not. The downward curve is separated from the ascending curve, and the size of the gap expressed in percentages of the input signal's full range is the hysteresis number. A valve with a rating of one percent hysteresis with an input voltage of 0-10 V means that the output of the descending channel can differ by as much as 100 mV in command variance from the output that is ascending at any point.

Manufacturers usually express hysteresis in a percentage of the input signal rated range or the rated output range The values for conventional proportional valves generally vary between 0.5 percent and 3 percent. High-performance servo-proportional valves that incorporate electronics and feedback on position can attain hysteresis as low as 0.1 percent.

The reason why hysteresis happens is due to physical reasons.

Many physical mechanisms come together to create hysteresis within proportional valves. Recognizing these causes helps engineers to address the root of the problem.

The most common cause in proportional valves is the magnetic hyperhysteresis within the solenoid that is proportional. The ferromagnetic core in the solenoid is able to retain its magnetic force even after the coil's current shifts direction or diminishes. This flux remnant means that the solenoid force doesn't respond immediately and in a linear manner to the current of the coil, and it is a little delayed and is not symmetrically distributed in the direction of the coil. As a result, for the same current command, the solenoid will produce somewhat different force levels, based on whether the current is decreasing or increasing.

Mechanical friction is a second primary cause. The spool of the proportional valve moves within an exact-machined bore. Despite the tight tolerances, dynamic and static friction, commonly referred to as "stiction," creates an area of dead space around the command-specific position. When the command signal changes, the spool has to overcome stiction before it can move, and, therefore, it remains in its current location even though the command changed. This delay and error in position are key components of hysteresis measured.

The hydraulic forces that act on the spool can also have a part to play. Forces induced by flow, often referred to as Bernoulli forces or hydraulic reaction forces, are able to exert pressure on areas of metering on the edges and hinder its movement. Because these forces alter direction and strength based on whether the spool is closing or opening in unidirectional resistance that expands the band of hysteresis.

Additionally, in valves with no integral feedback of position on the spool, thermal effects, as well as wear and tear over time, could gradually raise hysteresis levels by changing the characteristics of the solenoid or increasing the bore clearance.

How does hysteresis impact the performance of your system? 

The actual consequences of hysteresis rely heavily on the method of application. In open-loop proportional systems, it is a perpetual uncertainty in the position. If the controller transmits the exact command signal two times at the same time—one after an increase from a lower value and then again after reducing from a higher value, your valve's spool is in two distinct locations as well as the actuator, which would generate two outputs. If you have applications that require repeatable positioning, uniform flow metering, or steady pressure control, this ambiguity will directly affect performance.

In closed-loop systems, hysteresis communicates to the controller through more complicated ways. A PID controller will often make up for moderate hysteresis by constantly correcting the errors between the actual and commanded location. But a high degree of hysteresis can increase the effort needed to integrate as well as slow the response time. It also may cause limited cycling, which is a situation in which the actuator flies back and forth between the setpoint since the controller is constantly overcorrecting to compensate for the shift caused by the hysteresis bands. For applications that require precise servos, like injection molding, die casting, or aerospace actuation, even a hysteresis of less than 1% can be inconvenient without proper compensatory strategies.

In mobile hydraulics as well as less sensitive to precision applications, such as the extension of the boom on cranes or blade positioning in a motor grader, moderate hysteresis can be tolerated and can be ignored by the user. However, it can affect energy efficiency because it prevents it from returning to a location, which could cause an unintentional flow of bypass and even heat production.

Hysteresis vs. Similar performance parameters

Engineers often confuse hysteresis with other performance parameters for valves It is therefore important to distinguish them from each other.

Deadband is the amount of change in input signal at the null point that does not result in any output movement. It differs from hysteresis even though both are a factor in uncertainty of position. A valve may have a low hysteresis, but a high deadband or the reverse.

The term "repeatability" refers to the valve's capability of returning to a similar output for identical input signals in the same conditions over multiple cycles. Hysteresis is an element of the error in repeatability; however, repeatability also incorporates effects from variations in temperature, pressure fluctuations, and wear over time.

Linearity refers to how closely the relationship between input and output is an unbroken line through the entire signal spectrum. A high hysteresis doesn't necessarily indicate poor linearity—the ascending and descending curves may be very linear; however, they are different from each other.

"Threshold" or "sensitivity" refers to the minimum change in input signal that is required to cause an output change. For solenoid valves, this can be connected to magnetic hysteresis.

Reducing hyperactivity: Design and strategies for controlling

Valve manufacturers deal with hysteresis using different engineering strategies. The most efficient is to integrate the linear variable differential transformer (LVDT) or a similar position sensor directly into the spool and incorporate this signal as an inner closed-loop position control unit inside the amplifier chip of the valve. This method—which is common in servo-proportional valves of the highest quality—actively corrects for the spool's location error, regardless of friction or solenoid-induced lag, dramatically reducing the effectiveness of the hysteresis rate to a minimum of 0.5 percent.

Dither, an extremely small high-frequency oscillating sound that is superimposed over the command current, is a tried-and-true method of reducing the impact of magnetic hysteresis and stiction. Dither ensures that the spool is moving in a continuous micro-motion, which prevents it from staying in a single position and also reduces its threshold and the hysteresis value simultaneously. The amplitude and frequency of the dither must be carefully controlled to avoid overstimulation. Too high and it can cause undesirable actuator vibrations; too low, and it will have very little impact.

In the level of the system, adjusting the valve amplifier using separate characteristic curves that ascend and descend and thereby incorporating a hysteresis compensation table within the control loop -- may minimize the apparent hysteresis observed by the outside control loop. Modern electro-hydraulic controllers that have digital amplifier cards typically include this option as a standard option for commissioning.

Testing and specifying for hypertrophy

In the event of specifying proportional valves for high-precision applications, hysteresis needs to be explicitly mentioned in the document of purchase along with other important parameters like the rated flow and pressure drop, as well as response time and null leakage. ISO 10770 and related standards offer guidelines on testing methods for measuring constant-state valve characteristics, including the hysteresis.

During factory acceptance tests or commissioning on-site, hysteresis is determined by gradually increasing an input from a minimum to maximum and then back again at an established speed—usually slow enough to be regarded as quasi-static—and recording its output. The largest distance between ascending and descending curves multiplied by the input range determines the percentage of hysteresis. Modern valve test equipment automates the process and creates full characteristic curves, which serve as a baseline for condition monitoring throughout the life of the valve.

The phenomenon of valve hysteresis goes beyond an indication number in a sheet of data; it is a crucial representation of the physical flaws in solenoid actuation as well as mechanical sliding contacts. Engineers working on precision proportional systems, understanding the causes as well as determining its extent and deciding on the best compensation method, can make a difference between an effective system and one that hunts, is prone to drift, and frustrates users. It could be through feedback on spool positions, dither optimization, or controller-level compensation; regulating the hysteresis process is essential to achieving the maximum value from mechanical proportional technology.