Gas bottle installations
Remote gas storage offers flexibility in large and small systems, Figure 6. The gas bottle concept is generally described with this simple formula: accumulator size minus required fluid output equals gas bottle size. For example, an application that calls for a 30-gal accumulator may only require 8 to 10 gal of fluid output. This application, therefore, could be satisfied with a 10-gal accumulator and a 20-gal gas bottle.
An accumulator used with remote gas storage generally has the same size port at the gas end as at the hydraulic end to allow unimpeded flow of gas to and from the gas bottle. The gas bottle has an equivalent port in one end and a gas charging valve at the other. These two-piece accumulators can be configured or bent at any angle to fit available space.
The gas bottle concept is suitable for either bladder or piston accumulators. Note that bladder accumulators require a special device called a transfer barrier at the gas end to prevent extrusion of the bladder into the gas bottle piping.
Again, a piston accumulator should be sized to prevent piston bottoming at either end of the cycle. Bladder designs should be sized to prevent filling to more than 85% or discharging to more than 85% empty. The flow rate between the bladder transfer barrier and its gas bottle will be restricted by the neck of the transfer barrier tube. Because of these drawbacks, bottle/ bladder accumulators should be reserved for special applications.
Flow rates and response times
Table 2 suggests maximum flow rates for representative accumulator sizes and types. The larger standard bladder designs are limited to 220 gpm, although the rate can be boosted to 600 gpm using an extra-cost, high-flow port. The poppet controls flow rate; excessive flow causes the poppet to close prematurely. Multiple accumulators mounted on a common manifold are needed to achieve flows that are greater than 600 gpm.
|Table 2 - Maximum recommended accumulator flow rates|
|Gpm at 3,000 psi|
|1 qt |
|larger than 2½ gal||1,200 |
Allowable flow rates for piston accumulators generally exceed those for bladder designs. Flow is limited by piston velocity, which should not exceed 10 ft/sec to avoid piston seal damage. In high-speed applications, high seal contact temperatures and rapid decompression of nitrogen that has permeated into the seal material can cause blisters, cracks, and pits in the rubber.
Bladder accumulators respond more quickly to system pressure variations than do piston types for two reasons:
1. Rubber bladders do not have to overcome the static friction which a piston seal must, and 2. The piston mass does not need to be accelerated and decelerated.
In practice, though, the difference in response may not be as great as commonly believed, and is probably insignificant in most applications.
Tests at the University of Wisconsin, Madison, indicate that shock control does not necessarily demand a bladder accumulator. With system flow at a nominal 30 gpm in the test circuit, Figure 7, an internally piloted directional control valve, 118 ft away from the pump, closes to generate a shock. As the shock wave travels from the valve back through the hydraulic lines and around corners and various restrictions, some portion of its energy is consumed while accelerating the mass of fluid in the lines.
With 1¼-in. tubing, a 2,750-psi relief valve setting, and no accumulator in the circuit, oscilloscope trace A, Figure 8, shows a pressure spike of 385 psi over the relief valve setting. Adding a 1-gal piston accumulator at the valve reduces the transient to 100 psi over relief valve setting, trace B. Substituting a 1-gal bladder accumulator cuts the transient to 78 psi over relief valve setting, trace C, only 22 psi better than the piston-type protection.
A second, similar test with 5/8-in. tubing and a relief valve setting of 2650 psi results in a pressure spike of 2011 psi over relief valve setting without an accumulator, trace A, Figure 9. A piston accumulator damps the transient to 107 psi over relief valve setting, trace B, while a bladder accumulator damps the transient to 87 psi over relief valve setting, trace C. The difference between accumulator types in shock suppression again was negligible.
Another common misconception says that all servo applications require a bladder accumulator. Experience shows that only a small percentage of servos require response times of 25 ms or less, the region where the difference in response between piston and bladder accumulators becomes material. Bladder accumulators should be used for applications requiring less than a 25-ms response, and either type when response of 25 ms or greater is adequate.