Modifications have been proposed to effect further improvement of the device described in “Improved Piezo- electrically Actuated Microvalve” (NPO-30158), NASA Tech Briefs, Vol. 26, No. 1 (January 2002), page 29. To recapitulate: What is being developed is a prototype of valves for microfluidic systems and other microelectromechanical systems (MEMS). The version of the valve reported in the cited previous article included a base (which contained a seat, an inlet, and an outlet), a diaphragm, and a linear actuator. With the exception of the actuator, the parts were micromachined from silicon. The linear actuator consisted of a stack of piezoelectric disks in a rigid housing. To make the diaphragm apply a large sealing force on the inlet and outlet, the piezoelectric stack was compressed into a slightly contracted condition during assembly of the valve. Application of a voltage across the stack caused the stack to contract into an even more compressed condition, lifting the diaphragm away from the seat, thereby creating a narrow channel between the inlet and outlet. The positions of the inlet and outlet, relative to the diaphragm and seat, were such that the inlet flow and pressure contributed to sealing and thus to a desired normallyclosed mode of operation.

The Valve Incorporating the Proposed Improvements is depicted here in a simplified and partly schematic cross section, and not to scale.
The basic principles of design and operation of the proposed improved valve would be the same as those of the prior valve. However, there would be important differences in design details, leading to improvements, as summarized below:

  • The piezoelectric stack would be highly miniaturized (only 0.9 by 0.9 by 10 mm) and manufactured with high precision. The interior volume of the valve would be only 0.1 cm3.
  • Whereas the prior valve consumed a power of 2 W when actuated at a frequency of 100 Hz, the proposed improved version would consume only 0.1 W at 100 Hz. The combination of miniaturization and decreased power demand would be made possible by, among other things, utilization of a mode of piezoelectric actuation known in the art as d31. (The term “d31” signifies one of three independent moduli of piezoelectricity as well as the mode of actuation to which this modulus applies. In the d31 mode, the application of an electric field along one axis produces a longitudinal contraction along a perpendicular axis.)
  • Unlike in the prior valve, the piezoelectric stack would be isolated from the fluid to be controlled. Hence, it would not be necessary to take special measures to protect the stack against the fluid and, even more specifically, it would not be necessary to coat the stack with a dielectric material for protection against an electrically conductive liquid.
  • The design would include several features that would increase the ability of the valve to control a fluid at high pressure.

Like the prior valve, the proposed improved valve (see figure) would include a base that would contain a seat, an inlet, and an outlet. The piezoelectric stack would be connected to a valve boss at one end and to a rigid valve cap at its other end. In the absence of an applied potential, the valve boss would be pressed against the valve seat, so that flow would be blocked. The application of a potential of 60 V across the stack would cause the stack to shrink, pulling the valve boss away from the seat and thereby opening a flow channel between the inlet and the outlet.

In order to increase the spring bias of the valve toward the closed position and thereby help to minimize leakage in the absence of an applied potential, the boss plate would be slightly stretched. The force generated by the piezoelectric actuator would be about 100 N — enough to overcome both the tension in the boss plate and the pressure-aided valve-closing force at an upstream-to-downstream differential pressure as large as 300 psi (≈2 MPa).