BACKGROUND OF THE INVENTION
This application relates to a stability valve used in combination with a fluid pump.
Centrifugal pumps are used to move fluids through piping systems. Stability valves have been used to control centrifugal pump output. Prior art stability valves have been designed to be flow-dependent, such that the valves exhibit an increasing pressure drop as valve flow increases.
SUMMARY OF THE INVENTION
A stability valve assembly includes a centrifugal pump and a valve downstream of the centrifugal pump. The valve includes a valve member that is movable along a valve axis within a valve body to control a fluid flow from the centrifugal pump. The valve member has a orifice connecting a valve input to a valve chamber, and has at least one second orifice connecting a valve outlet to the valve chamber. A reference pressure upstream of the centrifugal pump applies pressure to a stepped portion of the valve member such that an amount of negative feedback provided by the valve assembly is proportional to a rotational speed of the centrifugal pump.
These and other features of the present invention can be best understood from the following specification and drawings, the following of which is a brief description.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 schematically illustrates a centrifugal pump fuel system.
FIG. 2 schematically illustrates a first, prior art stability valve.
FIG. 3 schematically illustrates a second stability valve.
FIG. 4 schematically illustrates forces being applied to a poppet in the valve of FIG. 4.
FIG. 5 schematically illustrates a third stability valve.
FIG. 5
a schematically illustrates pressure-dropping orifices of a poppet in the valve of FIG. 5.
FIG. 6 schematically illustrates forces being applied to the poppet of FIG. 5.
DETAILED DESCRIPTION
FIG. 1 schematically illustrates a centrifugal pump fuel system 10. In this case, the system 10 is operatively connected with a gas turbine engine 12, which includes combustors 14. The system 10 may deliver fuel to the gas turbine engine 12. A turbine shaft 16 of the gas turbine engine 12 drives a centrifugal fuel pump 18 of the system. The pump 18 takes suction from fuel tank 20, and delivers fuel through stability valve 22 (see FIG. 6) to the combustors 14.
Centrifugal 18 pump instability can occur when a change in pump discharge pressure with a change in pump flow (dp/dQ) is positive, setting up a dynamic system with a positive feedback, which is unstable. The stability valve 22 uses a pressure drop across the valve 22 to provide a negative feedback, cancelling the positive feedback from the pump 18 and stabilizing the fuel system 10. Prior art stability valves (see FIG. 2) have been designed to be flow-dependent, such that the valves exhibit an increasing pressure drop as valve flow increases. The valve 22, however, is speed-dependent such an amount of negative feedback provided by the valve 22 depends on an operating speed of the pump 18 and such that unnecessary amounts of negative feedback are not provided at low pump speeds. An amount of negative feedback (Feedbacknegative) provided by the valve 22 may be determined by equation #1 below.
where P1 is a valve inlet pressure;
- P2 is a valve outlet pressure; and
- Qvalve is an amount of fluid flow through the valve 30.
The operation of the centrifugal fuel pump 18 is governed by equations #2 and #3 below, which demonstrate that the centrifugal pump outlet pressure (Pact) at a given normalized flow is proportional to the square of the ratio of actual pump speed (Nact) to a reference pump speed (Nref).
where Pact is an actual centrifugal pump output pressure;
- Qref is an actual flow through the centrifugal pump;
- Pref is a reference centrifugal pump output pressure;
- Nact is an actual centrifugal pump speed; and
- Nref is a reference centrifugal pump speed. For example, Nref may be a maximum rated pump speed as defined by a manufacturer of the pump. Of course, other non-maximum pump speeds could be used.
FIG. 2 schematically illustrates a first, prior art stability valve 30 that includes a valve seat 32, a valve body 34, a poppet 36 movable along a valve axis 37, and a spring 38. As the valve opens to allow more flow, the spring 38 is compressed and a spring force (Kspring·Xvalve) is increased. A pressure P1 is applied to a first side 40 of the poppet 36. Pressure PX is applied to a second side 42 of the poppet 36. The valve body 34 includes a pressure-dropping orifice 46. P2 is a valve outlet pressure (i.e. pressure downstream of the stability valve 30). PX and P2 are related through the pressure-dropping orifice 46. During steady state operation, PX would be equal to P2. The valve 30 is “flow-dependent” in that its negative feedback depends on an amount of fluid flow through the valve 30, as determined by the stroke of the poppet 36. This results in unnecessary negative feedback when the pump 18 operates at low speeds.
FIG. 3 schematically illustrates a second stability valve 50. The valve 50 includes a valve seat 52, a valve body 54, a poppet 56 movable along a valve axis 57, and a bias member 58 (e.g. a spring). The poppet 56 includes a first pressure-dropping orifice 60 and at least one second pressure-dropping orifice 62. Equation #4, shown below, may be used to describe behavior of the valve 50.
0=(P1−PX)Avalve−Kspring·Xvalve equation#4
where P1 is a centrifugal pump output pressure;
- PX is a pressure inside the valve body 54;
- Avalve is an area of region 64 of the poppet 56;
- Kspring is a spring constant of the a valve spring 58; and
- Xvalve is a stroke 66 of the poppet 56.
As shown in FIG. 3, pressure P1 is applied to a first side 68 of the poppet 56. Pressure PX is applied to a second side 70 of the poppet 56. P2 is a pressure downstream of the stability valve 50. In the example of the valve 50, PX depends on both P1 and P2, as shown in equation #5 below.
where Fn1-x is a flow through orifice 60 at 1 pounds per square inch differential (“psid”) differential pressure; and
- Fn2-x is a flow through orifice 62 at 1 psid differential pressure.
FIG. 4 is a free body diagram that schematically illustrates forces being applied to the poppet 56. On the first side 68 of the poppet 56, a force of P1·Avalve is applied, and on the second side 70 of the valve 40, a pressure PX force (Pk·Avalve) and a spring force (Kspring·Xvalve) are applied. An amount of negative feedback provided by the valve 50 (see equation #1) would provide less inefficiency than the prior art valve 30 of FIG. 2. However there is still room for improvement, as demonstrated by the embodiment of FIG. 6.
FIG. 5 schematically illustrates a third stability valve 80 that may be used as the valve 22. The valve 80 includes a valve seat 82, a valve body 84, a poppet 86 movable along a valve axis 87, and a bias member 88 (e.g. a spring). The poppet 86 includes a first, non-stepped portion 78 and a second, stepped portion 79 (see FIGS. 5a-b). The poppet 86 includes a first pressure-dropping orifice 90 and at least one second pressure-dropping orifice 92. For example, the valve 80 may include a plurality of pressure-dropping orifices 92 spread along a circumference of the poppet 86 (see FIG. 5a). Equation #6, shown below, may be used to describe behavior of the valve 50.
0=(P1−PX)Avalve−Kspring·Xvalve equation #6
where P1 is a centrifugal pump output pressure;
- PX is a pressure inside the valve body 84;
- Avalve is an area of region 94 of the poppet 86;
- Kspring is a spring constant of the a valve spring 88; and
- Xvalve is a stroke 96 of the poppet 86.
FIG. 6 is a free body diagram that schematically illustrates forces being applied to the poppet 86. As shown in FIG. 6, pressure P1 is applied to a first side 98 of the non-stepped portion 78 of the poppet 86. A reference pressure Pd, which is a pressure upstream of the centrifugal pump 18, is applied to a first side 99 of the stepped portion 79 via line 24 (see FIG. 1). In one example, the reference pressure Pd enters the valve 80 through a passage 81 that passes through an O-ring 83. Pressure PX is applied to a second side 100 of the non-stepped portion 78 and is applied to a second side 101 of the stepped portion 79 of the poppet 86. P2 is a pressure downstream of the stability valve 50. In the example of the valve 80, PX depends on both P1 and P2, as shown in equation #7 below, and also on the reference pressure Pd.
0=P1Aup+PdAmod−PXAup−PXAmod−KspringXvalve equation #7
0=(P1−PX)Aup−(Px−Pd)Amod−KspringXvalve equation #8
where Aup is an area of region 94 (i.e. the area of first side 98 of the non-stepped portion 78); and
Amod is an area of region 102 of the poppet 86 minus the area of region 94 of the poppet 86 (i.e. the area of first side 99 of stepped portion 79).
PX may be determined using equation #9 below.
where Fn1-x is a flow through orifice 90 at 1 psid differential pressure; and
- Fn2-x is a flow through orifice 92 at 1 psid differential pressure.
By substituting the value of P1 shown in equation #10, one reaches the formula shown in equation #11.
where Nact is an actual centrifugal pump speed; and
- Nref is a reference centrifugal pump speed.
As shown in equations #10 and #11, the value of P1 is explicitly dependent on an operating speed of the centrifugal pump 18. Referring to equation #1. P1 is a term used to determine negative feedback (Feedbacknegative). Thus, unlike the prior art, where an amount of negative feedback has been “flow dependent,” the amount of negative feedback provided by the valve 80 is instead “speed dependent” in that it depends on a speed of the centrifugal pump 18.
Negative feedback can be considered to be wasted energy in the system 10, so it is desirable to have as little as negative feedback possible. The valve 80 achieves this by only providing negative feedback when it is needed. At low speeds of the pump 18, the valve 80 provides almost no negative feedback, and by providing increased negative feedback at greater rotational speeds of the pump 18. Thus, the valve 80 optimizes the operating conditions at which large pressure drops are realized by the stability valve 80 around those operating conditions at which the pump 18 requires the pressure drops for stability, but minimizes the pressure drop when it is not required for stability in the system 10.
Although Fn1-x and Fn2-x have been described as corresponding to 1 psid differential pressure, it is understood that these variables apply at other differential pressures as long as they corresponded to the same differential pressure. For example, Fn1-x and Fn2n-x could correspond to 2 psid or 3 psid instead of 1 psid.
Although embodiments have been disclosed, a worker of ordinary skill in this art would recognize that certain modifications would come within the scope of this invention. For that reason, the following claims should be studied to determine the true scope and content of this invention.