ASAE Conference Proceeding
This is not a peer-reviewed article.
Rotary, No-Leak Valve
,2005 Agricultural Equipment Technology Conference . ASAE Pub #AETC05006
Abstract. High flow rotary valves with low actuation torques and flow forces were evaluated. These valves maintained the ability to meter flow for a range of rotor positions. The one objective that was not completely met, however, was the leakage performance of the metal to metal sealed rotary valve. This aspect of the design was concentrated upon in the design revisions for the project. The result of these revisions was a rotary valve design that showed great promise in minimizing the leakage of the valve to satisfactory rates. Also, a special rubber seal design reduced leakage to zero during stationary valve position.Keywords. No-leak valve, rotary valve, no-leak seal
The rotary valve design that was produced from the original project was unlike any other hydraulic valve that has been produced or designed. The objectives of the Rotary Valve Project were to design and fabricate a hydraulic valve that could offer a combination of low internal leakage, no axial valve spool, not required pilot pressure to actuate or operate, and maintain high flow control capability. The low leakage requirement of the design was the most challenging. Low leakage was defined as less than 5 drops per minute of leakage flow through the valve while at full pressure drop. This design aspect would allow the valve to be used in applications where very precise metering and control of hydraulic flow would be required. The design was also conceived to be an alternative to the conventional spool used in many valves to control the direction and flow of the hydraulic fluid. Therefore, a new means of achieving this control had to be conceived. Additionally, the valve did not rely on pilot pressure for its actuation or operation. This meant that the flow forces and torques within the valve must remain reasonable in order to actuate the valve by mechanical means alone. Finally, the valve project goal was to maintain the ability to accurately meter the flow of hydraulic oil. The proposed design goals have not been achieved by any currently available hydraulic valves.
Many valve configurations were proposed, and five prototype designs were constructed. Of these prototypes, the rotary valve design concept that showed the best performance was the prototype which utilized an eccentric insert and eccentric rotor. This was seen as the simplest means of minimizing the clearance between the rotor and the housing, which greatly reduced the leakage. The results of the testing in this first phase of the project demonstrated the significance of clearance on the leakage in a valve. Utilizing the concept of eccentricity, the clearance between the rotor and the housing was minimized by simply closing the rotor. The images of Figure 1 and Figure 2 show the eccentric rotor and insert design that was developed for the rotary valve.
Figure 1: Eccentric Insert and Rotor Design Concept – Closed Position
Figure 2: Eccentric Insert and Rotor design concept – Open Position
The rotary valve design that is shown in these images was expected to produce excellent performance results. The actual prototype constructed is shown in Figure 3.
Figure 3: Actual Prototype Valve Constructed for Eccentric Insert Design
This image shows the discrepancy between the proposed design and the actual fabrication. The results of the testing of this design was greatly improved leakage performance, but also increased actuation torques. This was due to the line of action of the jet forces. The arrow indicates the direction of action of the jet forces. Since the jet force did not act directly toward the center as proposed, a torque was created. This increased torque was also seen when the rotor was held in the closed position. The following plots show the significant testing results of the prototypes.
Figure 4: Comparison of Leakage Performance for Prototype Valves
Figure 4 shows the improvement in leakage performance that was seen with Prototype III (PT3). This was the most important performance aspect of Prototype III, as the leakage of the valves made them unfeasible. The actuation torques of the valves, however, were unacceptable for the rotary valve that could be controlled by stepper motor, as Figure 5 shows.
Figure 5: Plot of Closing Torques for Prototype Valves
Figure 6 Plot of Holding Torques for Prototype Valves
Figure 5 and Figure 6 show the torques of the prototype valves. As can be seen these plots, the closing and holding torques of the Prototype III valve were much greater than the other prototypes. The magnitude of these torques could be greatly reduced or even negated with redirection of the line of action for the jet forces on the rotor. Figure 7 shows the flow metering ability of the valves at 1750psi. This was a concern, and can be seen from this plot, the flow metering ability of the valves was excellent. Prototype III showed the best flow metering ability, since it had the largest flow metering band of the first three valves.
Figure 7: Plot of Prototype Valves Flow Metering Performance at 1750psi
Another plan was to construct the valve design IV. This design utilized a lesser slope of fit, which was determined by the amount of eccentricity. This would allow for less actuation torques, as the rotor would now be shearing the flow direction. A tight fit would be maintained by applying a constant torque to the rotor to hold it in the closed position. In addition, the jet forces acting on the rotor would be redirected, as the images showed. The result would be a greatly reduced holding torque for the rotor and reduced effort from the closed-loop position control system, as the disturbance inputs to the control system would be minimized.
The final design V incorporated a face seal into the rotor. The face seal was constructed from a pliable material that allowed a slight deformation when pressed against the housing. The result would be that a very tight seal is created, which minimized the leakage by compensating for any machining surface errors or inconsistencies. The overall leakage performance of the valve was improved, as the seal would no longer be steel on steel, but instead sealing material on steel.
Figure 8: Rotor Facing to be Incorporated with Sealing Material
Also, valve IV was modified with the insert to a pliable material, such as a Teflon or plastic. This allowed the insert material to experience a slight deformation as the rotor was closed against the insert. The result would be that the deformation of the material would compensate for the machining surface inconsistencies and provide an excellent seal to minimize the leakage at the inlet.
A variation of valve V incorporated a seal around the inlet orifice for the rotor to be pressed against. The seal that was proposed is a “quadra” seal. This concept used an eccentric design that pressed the rotor onto the housing, instead of the rotor that acted to shear the rotor against the insert to seal. This shearing action did upset the seal from the groove and lost the sealing effect. The seal around the orifice would provide the 360 o sealing action that is required to minimize the leakage. A representation of the seal around the inlet orifice is shown in Figure 9.
Figure 9: Seal Around Inlet Orifice
Conceptual Design of Prototype V, called DR-R
The V prototype that was designed and built was to concentrate on leakage at the inlet orifice of the valve. It was known that the full rotor design had too much torque and still some problems with leakage due to the required machining clearances needed to achieve acceptable value. This prototype valve then considered a prototype seal produced by Seal Master Corporation as shown in the dimensioned drawing in Figure 10. The seal was made with a hollow center so that pressure could be applied to the seal through this internal cavity causing the seal to expand. The objective was to have the seal reduce the clearance to practically zero so that no leakage would occur when it was pressurized. One problem with using seals is they wear, but by using an inflatable seal the wear rate is very low since it retracts back into the groove prior to movement of the rotor, which is a new concept. The seal was made out of a 65 durometer Nitrile compound. One major advantage of this seal was that it eliminated leakage due to small imperfections in the rotor face during actuation of the seal. This is an advantage because there will be imperfections in the surface finish of the rotor during machining. The seal was designed to seal a 0.002-0.005 inch gap that can be seen below in Figure 10 in the lower left corner of the drawing and is pointed to by the radius dimension of the drawing. The gap can also be seen in Figure 11 in more detail. This is the surface that actually moves out of the groove towards the bottom of the drawing and engages the rotor surface.
Figure 10: Dimensioned Drawing of Seal Master Valve Seal (inches)
Figure 11: 2-D View of gap that will be sealed by Seal Master Seal
The valve did require more control components than any of the previous valves due to the expansion-retraction the prototype seal sequences in the valve. The pressure intensifier was the main additional component added to the valve allowing for its proper expansion function. This was required by the seal so that when it was inflated it had a pressure of approximately 10 to 50psi greater than the pressure drop across the seal. Without this increase in pressure the seal could likely extrude causing the valve seal interface to fail. Also seal pinching and wear might happen along with risking injury to those operating the system in which the valve is being used. There is a slight possibility that the seal might be extruded, because it is not fully 100% supported and constrained by this design. The prototype rotor clearance was approximated by the machinist to be 0.0005 inches. The last drawback that was considered is the reaction time of the seal during pressure release to fully deflated and the retraction into the groove. This could cause problems if the rotor was prematurely actuated and appropriate control time constants weren’t determined.
One concern was the deflation time and rate of the inflatable seal. The deflation time and rate were modeled using a second order step input response technique based on vibrations (Doebelin, 1990), which used Equation 1 to calculate , which was required for Equation 2 to iterate until the equation equaled 0.001, which is the distance that the seal must travel to fully deflate. The variables of these two equations are as follows, is the damping factor, is the natural frequency of the seal and t is the time to deflate 0.001 inches. The deflection time was calculated to be 0.0022284 seconds, which equates to a seal return rate of near 0.04 inches per second. This time is faster than the discharge rate of the fluid exiting the seal via valve performance and bulk modulus affects.
This research revealed many interesting results for each of the prototype valves tested. The first conclusions that will be discussed are the leakage results. The main work was to obtain the goal of a leakage rate of five drop/min.
The first design was built twice by two different machine shops with the first version not showing any improvement over previous results that can be found in Brown, (2002). The second version showed great improvements over all previously tested valves with a maximum leakage of 10.22 drops/min at 300psi and a minimum of 5.61 drops/min at 700psi at the null valve position. This rotor was shifted at higher pressures creating maximum leakage rate at 300psi and the minimum at 800psi, which is not what was expected. The V valve built, DR-R, showed the lowest leakage rate that was possible as there was no measurable leakage produced when the valve was tested at a 1400psi drop.
The valves were analyzed using the flow gain coefficient called K q , which relates the change in flow to the change in rotor position and can be seen as
Table: Table of K q Values for IV and V
Comparison of Flow Data between IV and V
Tsukiji, T. 2003. Flow Analysis in Control Valve Using a Vortex Method, First International Conference on Computation Methods in Fluid Power Technology . Melborne, Fluid Power Net Pty Ltd, Australia
Brown, R.J., 2002. Design of a Two-Way Electrohydraulic Rotary Valve with Low Leakage and No Pilot. Unpublished Masters Thesis. West Lafayette, In.: Purdue University
Rusch, D., 2003. Design of a Two-Way Rotary Valve with No Leak and No Pilot. Unpublished Masters Thesis. West Lafayette, In.: Purdue University
Ahlgren, B.K., C.B.S., Fawaz, I. 2002. Fluid-directing multiport rotary valve. US, Calgon Carbon Corporation. U.S. Patent No.6431202
Babin, C.J. 2001. Servo motor operated rotary bypass valve. US, Eaton Corporation. U.S. Patent No. 6289913
Babin, C.J. 2003. Servo Operated Rotary Valve With Emergency Bypass And Method Of Making Same. US, Eaton Corporation. U.S. Patent No. 6588422
Baruschke, W.; O.K., Lochmahr, K. 1999. Rotary Valve. Germany, Behr GmbH 5967185
Baumann, H. D. 2002. Rotary Valve. US, Fisher Controls International, Inc. U.S. Patent No. 6024125
Brumm, R.S. 1980. Low Torque Control Valve. US, Cashco Incorporated. U.S. Patent No. 4193578
Calvin, D.G. 1998. Rotary Valve with Pressurized Energized Seal. US, Keystone International Holdings Corp. U.S. Patent No. 5765815
Godfrey, R. 1986. Ball Valve With Preloaded Seals And Method Of Manufacture. U.S. Patent No. 4603836
Groenefeld, H. 1981. Rotary Sperical Plug Valve. US, Honeywell Inc. U.S. Patent No. 5386761
Holtgraver, E.G. 1995. Rotary Valve Actuator. US, Saving by Design, Inc. U.S. Patent No. 5386761
Hotier, G. 2003. Rotary Valve. FR, Institut Francais du Petrole. U.S. Patent No. 6537451
Kalippke, H.; F.W., Renninger, E.; Meiwes, J.; Dick, D. (1994). Rotary Acuator. Germany, Robert Bosch GmbH 5275373
Kawano, S.; M.N., Katsuragawa, S.; Inoue, T. 2002. Rotary Valve unit in a pulse tube refridferator. JP 6640349
Kivipelto, P.J.; E.T. Y.-K. 1988. Rotary Valve with Pressure Urged Sealing Member. Finland, Neles OY. U.S. Patent No. 4747578
Krevald, W.R. 2003. Rotary Valve And Piston Pump Assembly And Tank Dispenser Therefor. US, Diamond Machine Works, Inc. U.S. Patent No. 6579079
Maeda, N. 1991. Flow-Control Valve. Japan, Hitachi, Ltd. U.S. Patent No. 4984766
Niessen, L.J. 2001. Valve locking mechanism and method. U.S. Patent No. 6568422
O’Reilly, P.B.; J.W.D., Latif, T.; Mowery R.W. 1999. Rotary Fluid Valve Systems. US, Barksdale, Inc. U.S. Patent No. 5934320
Piccirilli, D.F.; Liederman, K.E.; Vint, M.V.; Bejster, J.V.; Harmer, N.P.; Gree, M.; Haigh, M. E.; Jalilevard, A. 2003. Rotary valve for single-point collant diversion in engine cooling system. US, Visteon Global Technologies, Inc. U.S. Patent No. 6539899
Reed, C.L., J.A.G. 1984. One Piece Top Seal For A Valve. US, Xomox Corporation. U.S. Patent No. 4475713
Rihm, H.W.R. 1978. Rotary Valve. Germany, Honeywell GmbH 4073473
Runyan, G.L. 1981. Rotary Ball Valve Having Seating Rings. US, Celanese Corporation. U.S. Patent No. 4257575
Schmitz, C.M.M.R. 1984. Rotary Valve. US, Honeywell Inc. U.S. Patent No. 4431161
Stacy, P.C. 2003. Rotary Valve. GB 6571825
Sudo, H. 1981. Rotary Valve. Japan, Kutbota Ltd. U.S. Patent No. 4269218
Thiele, U.K.; D.G. 2003. Multipath rotary disc valve for distributing polymer plastics melts. DE, Gneuss Kunststofftechnik BmbH 6550497
Woodworth, R.D. 2002. Balanced Rotary Servovalve. U.S. Patent No. 6470913
Woodworth, R.D.; J.E.B. 2001. Rotary Servovalve and control system. US, Raymod Dexter Woodworth. U.S. Patent No. 6269838
Yagi, T.M.J. 1990. Rotary Valve Made of Ceramics. Japan, NGK Spark Plug Co., Ltd. US.: 4922949