Accumulators used for fast response and over-pressure control of pressure-compensated pumps
Because most pressure-compensated pump circuits have closed-center or two-position directional valves (such as the one shown in Figure 1-16), they stay at full-pressure, no-flow until a valve shifts. After any directional valve shifts to start an actuator’s movement, pressure in the circuit starts to drop. When the pump sees a pressure drop, its internal mechanism starts shifting as fast as possible to start fluid flowing. Pump shifting times vary, but no matter how fast they shift, the actuator’s initial response will be slowed down.
With an accumulator installed, as shown in Figure 1-17, the pump is still at no-flow when the circuit is at rest. However, there is a ready supply of oil at pressure available. As a cylinder starts to cycle, as seen in Figure 1-18, fluid flows directly to the actuator from the accumulator and pressure starts to drop. This pressure drop causes the pump to go on stroke, but now pressure drop is minimal. The cylinder takes off quickly and smoothly, and the pump has time to respond to the flow need.
On the other end of the cycle, if the pump is at full flow and all valves center or all the actuators hit the end of stroke, the flow requirement suddenly drops to zero. The pressure-compensated pump is still flowing at the maximum rate and pressure starts to climb. The pump will continue at full flow until pressure reaches 80-98% of the compensator setting. There has been zero flow needed for some time, but the pump does not know this until pressure is near maximum. When pressure reaches compensator setting, the pump starts to shift to no flow. All pump flow during shifting time has no place to go, so this excess flow generates a pressure spike of five to ten times the compensator setting. This pressure spike can cause premature failure of the pump, plumbing, and actuators.
A common fix for this pressure spike is to add a relief valve near the pump outlet, set 150 to 200 psi higher than the pump compensator (as shown in Figure 1-16). This relief valve should reduce the pressure spike, but it does not lower it as much as it appears. A relief valve remains closed until pressure reaches 90 to 98% of its setting. Once the relief reaches maximum pressure, it starts to open, but by the time it actually relieves, the pressure may be 11/2 to 3 times its set pressure. This reduced spike is better, but still is not as good as what an accumulator could provide.
Other problems can occur with relief valves. For example, if the relief valve setting is at or near the pump-compensator setting, the pump can start oscillating on-off flow. As the pump nears its pressure-compensator setting and starts to compensate, the relief valve starts to relieve. A flow path is created when the relief valve begins to open, so downstream pressure drops, causing the pump to go back on stroke. The drop in pressure allows the relief valve to close, so downstream pressure builds up again. This oscillation cycle repeats rapidly, causing damage to the pump and possible line failure due to shock. In another example, if the relief valve setting is lower than the pump compensator, all pump flow goes to tank at relief pressure, generating excess heat. To avoid these problems, use the correct procedure when setting pressures on a relief valve used to reduce pressure spikes.
An accumulator absorbs excess pump flow with minimal pressure override or shock. While fluid from the pump compensates from full flow to no flow, as seen in Figure 1-19, it has a direct path to the accumulator. Because the accumulator has a compressible gas in it, it takes in the small amount of excess flow produced while the compensator is reacting. Pressure increase from this additional fluid is imperceptible.
To size an accumulator for fast response of the circuit, plan to have somewhere between 1 and 5 sec of actuator flow before pressure drops below the minimum it takes to move it. A rule of thumb is to have 1 gal of accumulator for every 10 gpm of pump flow.