Given that the primary objective of filtration is to extend machine life by removing contaminants from the oil, it is a paradox for the filters in a hydraulic system to be located where they reduce the service life of the components they were installed to protect.
So when considering the possible locations for filters in a hydraulic system, the overarching principle must be: first, do no harm. In other words, “the cure must not be worse than the disease”.
With this in mind, let us consider the pros and cons of the various hydraulic filter locations:
Pressure filtration: Locating filtering media in the pressure line provides maximum protection for components located immediately downstream. Filtration rates of 2 microns or less are possible, due to the pressure available to force fluid through the media. But filter efficiency can be reduced by the presence of high flow velocities and pressure and flow transients, which can disturb trapped particles. The major disadvantage of pressure filtration is economic. Because the housings and elements (high-collapse type) must be designed to withstand peak system pressure, pressure filtration has the highest initial and ongoing cost.
Return filtration: The rationale for locating filtering media in the return line is this – if the reservoir and the fluid it contains start out clean, and all air entering the reservoir and returning fluid is adequately filtered, then fluid cleanliness will be maintained. The other advantage of the return line as a filter location is that sufficient pressure is available to force fluid through fine media (typically 10 microns), but pressure is not high enough to complicate filter or housing design. This, combined with relatively low flow velocity, means that a high degree of filtering efficiency can be achieved at an economical cost. For these reasons, return filtration is a feature of most hydraulic systems. The main disadvantage of return line filtration is that the back pressure created by the element can adversely affect the operation of and/or damage some components.
Off-line filtration: Off-line filtration enables continuous, multi-pass filtration at a controlled flow velocity and pressure drop, which results in high filtering efficiency. Filtration rates of 2 microns or less are possible, and polymeric (water-absorbent) filters and heat exchangers can be included in the circuit for total fluid conditioning. The main disadvantage of off-line filtration is its high initial cost, although this usually can be justified on a life-of-machine cost basis.
Suction filtration: From a filtration perspective, the pump intake is an ideal location for filtering media. Filter efficiency is increased by the absence of both high fluid velocity, which can disturb trapped particles, and high pressure drop across the element, which can force migration of particles through the media. These advantages are outweighed by the restriction the element creates in the intake line and the negative effect this can have on pump life.
Figure 1. The Effect of Tensile Forces Acting on Axial Piston Design
A restriction at the pump inlet can cause cavitation erosion and mechanical damage. And while cavitation erosion contaminates the hydraulic fluid and damages critical surfaces, the effect of vacuum-induced forces has a more detrimental impact on pump life.
The creation of a vacuum in the pumping chambers of an axial pump puts the piston ball and slipper-pad socket in tension. This joint is not designed to withstand excessive tensile force; and as a consequence, the slipper becomes detached from the piston (Figure 1). This can occur either instantaneously, if the vacuum-induced tensile force is significant enough, or over many hours of service as the ball joint is repetitively put in tension during inlet.
The piston retaining plate, the primary function of which is to keep the piston slippers in contact with the swash plate, must resist the forces that act to separate the piston from its slipper. This vacuum-induced load accelerates wear between the slipper and retaining plate and can cause the retaining plate to buckle. This allows the slipper to lose contact with the swash plate during inlet, and it is then hammered back onto the swash plate when pressurized fluid acts on the end of the piston during outlet. The impact damages the piston slippers and swash plate, leading rapidly to catastrophic failure.
In bent axis pump designs, the piston is better able to withstand vacuum-induced tensile forces. Piston construction is generally more rugged, and the piston ball usually is held in its shaft socket by a bolted retaining plate. However, tensile failure of the piston stem and/or buckling of the retaining plate still can occur under high vacuum conditions.
In vane pump designs, the vanes must extend from their retracted position in the rotor during inlet. As this happens, fluid from the pump inlet fills the void in the rotor created by the extending vane. If excessive vacuum exists at the pump inlet, it will act at the base of the vane. This causes the vanes to lose contact with the cam ring during inlet; they are then hammered back onto the cam ring as pressurized fluid acts on the base of the vane during outlet. The impact damages the vane tips and cam ring, leading rapidly to catastrophic failure.
Gear pumps are mechanically the least susceptible to vacuum-induced forces. Despite this fact, research has shown that a restricted intake can reduce the service life of an external gear pump by at least 50 percent1.
Pump inlet or suction filters usually take the form of a 150-micron (100-mesh) strainer, which is screwed onto the pump intake penetration inside the reservoir. In the 10 years I’ve actively campaigned against their use (for reasons outlined earlier in this column), I’m sure I’ve heard all of the counter-arguments. Most arguments for the use of suction strainers are premised on bad design, bad maintenance or a combination of both.
The argument that suction strainers are needed to protect the pump from debris which enters the reservoir as a result of poor maintenance practices is a popular one. Nuts, bolts, tools and similar debris pose minimal threat to the pump in a properly designed reservoir, where the pump intake is located a minimum of 4 inches off the bottom. Of course, the proper solution is to prevent contaminants from entering the reservoir in the first place.
A similar argument asserts that suction strainers are needed to prevent cross-contamination where two or more pumps share a common inlet manifold. Here again, if suction strainers are necessary in this situation, then it is only due to bad design; the manifold must be below the pumps’ intakes. If properly designed, there should be a head of oil above the inlet manifold, and the inlet manifold should be above the pumps’ intakes. For cross-contamination to occur in this arrangement, debris would have to travel uphill – against gravity and a positive head of oil. That would be highly unlikely.
But even in situations where a suction strainer is mandated, for whatever reason, the problem is: The cure can be worse than the disease.