 |
| Fig. 5. Proportional electromechanical interface device (PEMID) causes an electrohydraulic valve (PEHID) to shift with two results: valve-control method (A), where valve output moves the load, and (B), where the valve changes the displacement of a pump with resulting pump output then moving the load, an example of the pump control method. |
 |
| Fig. 6. A small current must cause a small spool shift and a large current must cause a large spool shift in continuously variable electrohydraulic valves. To assure such proportional, stepless spool positioning, some valves use a spool-position transducer to measure actual spool position. The spool is made to stop in a position commensurate with the command voltage through feedback-loop closure. This closed feedback loop often is called the inner loop. |
Electromechanical actuators
It is possible to construct proportional electrohydraulic interface devices (IHs) only because of the invention of certain proportional electromechanical interface devices (IMs). The IMs commonly used in the fluid power industry include:
- torque motors,
- linear force motors, and
- proportional solenoids.
The IMs receive an electrical current input, convert it to mechanical force and motion, and then transform the energy into some sort of hydromechanical action. The direct mechanical action is always within a valve, although that valve may stroke a pump or directly power a load. A circuit designer would select one path or the other of the family tree, Figure 5, for a given application. Valve control is called the energy-loss control method in path A, because the valve, being a restrictive device, consumes excess power as a necessary part of its control function.
Path B, on the other hand, is called the volume-control or load-demand method that supplies only as much power as the load can use. The only losses encountered using this method are those caused by the modest inefficiencies of the pump and actuator. Altogether, these are nearly always less than those for the energy-loss method, all other things being equal. This leads to the following fundamental truths regarding proportional hydraulic systems:
- path B is always more efficient than path A, and
- path A always has lower initial cost because valves are less expensive than variable-displacement pumps, and one fixed-displacement pump can supply pressurized fluid to more than one valve and functioning circuit branch.
Electrohydraulic valves
Continuously variable electrohydraulic valves illustrate that a continuously varying control current always results in a continuously varying, controlled output variable. That output variable could be flow, pressure, or simply the position of a spool that affects final flow and/or pressure. Broad categories of these continuously variable valves are:
- direct-driven valves where the force of the proportional solenoid acts directly upon the main spool to provide the desired degree of hydraulic control, and
- pilot-operated valves in which the IM acts first on a primary hydromechanical device whose output acts on a main spool. These are also sometimes called multi-stage valves and are either 2-or 3-stage but never more.
Direct-driven valves
Further subdivisions include those valves that use some means of spool-position feedback and those that do not. The non-feedback types simply take the force of the proportional solenoid and put it against a restoring spring. Thus, the main spool would take a position commensurate with the force generated by the solenoid if those were the only two forces acting on the main spool. Unfortunately, there are two other significant forces that act on the spool: flow forces and stiction forces.
Flow forces are a natural phenomenon in all control valves that result from the momentum change that takes place as the result of the valve's throttling effect. This occurs when potential (pressure) energy is converted into kinetic (velocity) energy in the constricting region of the valve. In spool valves, the flow force always acts to close the valve regardless of the direction of flow. The consequence is that when using the valve, say at a low pressure drop (flow), the spool is in a position where the solenoid force is balanced by the restoring spring. As the valve's pressure drop increases, either because of a reduction in load restriction or an increase in supply pressure, the flow force increases so as to close the valve.
As a result, the spool takes a position not totally controlled by the control current (solenoid force). This does not mean the valve does not function; it does mean that the spool's exact position at any moment is harder to predict. It is true, however, that in some direct-acting valves, flow forces can be so high that they cause an automatic near-closure at high pressures and flows. Load-dependent spool position can be detected by mathematical analysis of valve pressure drop at a steady control current but varying flows. The curve, if there is no load-induced spool shift, will relate pressure drop to the square of the passing flow. If the data does not fit the square relationship, flow forces are probably causing a spool shift.
Stiction forces also act upon the spool and the solenoid's armature, so that spool position does not smoothly vary as control current continuously and smoothly changes. Instead, the flow (spool position) has a staircase effect. Additionally, the curve trace for increasing control current is not the same as the curve trace for decreasing control current, producing stiction-induced hysteresis. If the valve is to be used in manually operated control systems, this hysteresis is not a major problem because a human operator can compensate easily for such performance aberrations. But when automatic controls using feedback are contemplated, hysteresis can cause a continual hunting or oscillation rather than smooth and stable operation.
Incorporating electronic dithering - that is, causing the spool to be in a continuous but acceptably small state of agitation - can help significantly to reduce the detrimental effects of stiction-induced hysteresis. When properly implemented, a closed-loop feedback control system around the spool, called the inner loop in the hydraulics industry, Figure 6, can all but eliminate stiction and flow-force effects. This loop is closed by measuring actual spool position, usually with an LVDT position transducer, and comparing it to the commanded position. If the position is incorrect, the electric current to the valve's proportional solenoid is adjusted until spool position becomes correct. Thus, the spool is always in the exact spot commanded, dynamically induced lags notwithstanding.
While not exactly true, the foregoing statement is acceptable for all practical purposes. The spool position feedback transducer of choice is nearly always an LVDT. Because LVDTs must be operated with AC voltage, they must always be accompanied by a special electronic signal conditioner that has:
- an oscillator section that generates AC voltage to excite the transformer. This AC voltage is not derived from the 60 Hz power- company line; it is generated by a solid-state electronic oscillator usually outputting a few volts at a fixed frequency, generally between 3 kHz and 10 kHz, and
- a phase-sensitive demodulator section that converts a transduced AC signal into an equivalent DC signal with the full sense of the algebraic sign of the measured position.
