Not Applicable
Not Applicable
Not Applicable
This present application claims priority from U.S. provisional application No. 61/520,630 having the same title as the present invention and filed on Jun. 13, 2011.
This invention relates to pumps, and more particularly, to hydraulic injection pumps used in injecting fluid into a vessel, the pressure inside which is controlled within a predetermined range.
Air-driven hydraulic pumps use compressed air to drive reciprocating actions in delivering pressurized liquid. With a stepped piston having its large diameter side contacting compressed air in an air cylinder and small diameter side driving liquid in a high pressure injection cylinder, an air-driven hydraulic pump is able to provide high driving pressure, which is multiple times of compressed air pressure, and the multiplication ratio is determined by the ratio of the large diameter to the small diameter. To complete a reciprocation cycle, which includes a suction stroke and a pressing stroke, it needs to control the air pressure inside the air cylinder by filling and releasing compressed air. Normally, the pressure control is realized by using relay valves that use sealed air and switches to fill and release the sealed air in controlling the relay valves (U.S. Pat. Nos. 3,963,383, 4,645,431, and 6,386,841). Therefore, the reciprocating rate is determined by air pressure, air filling and releasing rate, and switch position. As a result, fluctuations in compressed air supply pressure affect both reciprocating rate and driving pressure. Also, in the pressing stroke, compressed air expands and results in pressure drop. The pressure change in compressed air is then amplified by the pump and causes larger change in driving pressure.
To accurately control the driving pressure, a primary object of the present invention is to provide controls means to adjust the driving pressure, thereby with a closed-loop control, the driving pressure can be controlled within a predetermined range.
A further object is to replace the relay valve using the controls means set forth in the foregoing object to provide a simplified pump structure.
A further object is to provide controls means to avoid the effects of the suction stroke in controlling driving pressure.
In accordance with the present invention, a hydraulically actuated pumping apparatus with driving pressure controlled within a predetermined range is provided.
According to one embodiment of the invention, an air-driven hydraulic injection pump is provided that has a pressure multiplication ratio of 1.0, however, has no piston device inside. The stroke control and pressure control are accomplished by energizing and de-energizing two solenoid valves to control air pressure inside the pump according to pressure sensing values obtained from a pressure sensor.
According to another embodiment of the invention, an air-driven hydraulic injection pump is provided that has a pressure multiplication ratio higher than 1.0. This pump has a piston inside and strokes and pressure are controlled by energizing and de-energizing two solenoid valves to release and fill compressed air according to pressure sensing values obtained from a pressure sensor.
According to another embodiment of the invention, a hydraulic buffer is provided with an air-driven hydraulic injection pump. The hydraulic buffer decreases pressure drops associated with suction strokes in which compressed air is released from the pump.
According to another embodiment of the invention, a hydraulically driven pump system including two air-driven hydraulic injection pumps are provided. These two pumps work alternately to control driving pressure within a predetermined range.
a depicts an air-driven hydraulic pump system with two two-way solenoid valves and a pump without piston;
b depicts an air-driven hydraulic pump system with a two-way solenoid valve, a three-way solenoid valve, and a pump without piston;
Referring to
Stroke control and pressure control for the pump 100 are accomplished by using the combination of controls to the solenoid valves 122 and 126. The controls to the two valves have four modes shown in the following table:
In Mode 0, both of the solenoid valves 122 and 126 are not energized, and the pump is releasing air to ambient. In Mode 1, since solenoid valve 122 is energized, the pump is disconnected from ambient. At the same time, the solenoid valve 126 is not energized. Therefore, in this mode, the air is trapped in the pump. Mode 2 is a special mode needs to be avoided, since in this mode the compressed air is directly released into ambient. Mode 3 is an aspiration mode. In this mode, the solenoid valve 122 disconnects the pump from ambient, while the solenoid valve 126 fluidly connects the pump the compressed air supply.
The pumping control starts with a suction stroke. When fluid in the pump reaches a certain level, a pressing stroke is triggered and driving pressure is controlled in a range commanded by the user. The pump goes back to suction stroke when a refill event is triggered.
In a suction stroke, Mode 0 is triggered. As mentioned above, in Mode 0, the pump releases air to ambient. After the air pressure inside the pump drops, under gravity or the pressure inside the fluid tank 130, the fluid 133 flows through the port 131, the passage 132, and the check valve 106 inside the port 105 into the pump. In the suction stroke, no fluid flows out of the pump.
After a suction stroke, the controller enters Mode 3, in which compressed air flows into the pump 100 through the solenoid valve 126, the passage 125, the Tee connector 127, and the gas port 101. Under the air pressure, fluid in the pump is able to flow out through the port 102.
Sensing values obtained from the pressure sensor 103 are used in controlling strokes and liquid driving pressure. One embodiment of a control algorithm is realized with a service routine running periodically in the controller 110 for a timer based interrupt. As depicted in
In the stroke and pressure control, to avoid momentarily going into Mode 2, in changing modes from Mode 3 to Mode 0, the controller should de-energize the solenoid valve 126 first, while in switching modes back to Mode 3 from Mode 0, the controller should energize the solenoid valve 122 first. To further avoid troubles caused by Mode 2, as shown in
According to Table 2, in the system depicted in
In the stroke control, two events, a refill event and a pump full event, can be used to switch strokes. A refill event is triggered when a low liquid level in the pump is detected or the calculated liquid volume is low. To detect liquid level in the pump, a level sensor can be further installed inside the pump (not shown in
Two methods can be used in calculating the liquid volume in the pump in a pressing stroke. One is calculating the amount of liquid being released from the pump. Under the driving pressure inside the pump, when liquid starts to flow out of the pump, the flow rate of liquid through the port 102 is a function of the driving pressure. If the driving pressure is controlled constant, the flow rate is a constant value. Therefore, when the driving pressure is controlled within a narrow range, the lost liquid volume in a pressing stroke is approximately proportional to the liquid releasing time, and the liquid volume thus can be calculated by using the following equation:
V=V
0
−K*t (1)
, where V is the current liquid volume inside the pump; V0 is the liquid volume when a pressing stroke starts; K is a constant, and t is the liquid releasing time. To more accurately calculate the current volume, liquid releasing rate, which is proportional to the square root of the driving pressure, can be used in the calculation:
V=V
0∫t
where C is a constant; P is the driving pressure at moment t, and t0 is the time moment when a pressuring stroke starts. When the flow through port 102 is further controlled by a solenoid valve (not shown in
The other method is using the ratio of pressure change in Mode 1 to the amount of liquid released during the pressure change. According to the idea gas law, in Mode 1, since the air is trapped in the pump, if the effect of liquid pressure in the pump is negligible and temperature is constant, then we have the following relation:
where dP is the change in the driving pressure; dV is the volume change when driving pressure changes dP, and Vt is the tank volume. The liquid volume then can be calculated using the following equation:
According to equation (4), unlike the first method, this method doesn't require the liquid volume at the beginning of a pressing stroke, V0, in calculation.
The pump full event can be triggered using the pressure sensing value. In a suction stroke, when pressure inside the pump is released, the gauge pressure obtained by the sensor 103 is a proportional to the liquid level in the pump. Accordingly, the liquid level can be calculated using the pressure sensing value through the following equation:
L=Kp*P (5)
where Kp is a constant. When the calculated liquid level is above a threshold, a pump full event is triggered.
In addition to the pressure sensing value, the refill time is also an indication to the liquid volume in the pump. The liquid refill flow-rate to the pump 100 is a function of the difference between the pressure at the port 131 and that at the port 105. Therefore, in a system of
r
f
=f(Lt) (6)
. In a control algorithm, the function in equation (6) can be realized using a lookup table, the values in which are populated with testing data. According to equation (6), the liquid volume in the pump then follows the equation below:
V=∫
0
t
f(Lt)dt (7)
, where tf is the refill time. A pump full event can be triggered when the calculated liquid volume is higher than a threshold.
Since air density is much lower than liquid density, if the liquid tank 130 is empty, then in a suction stroke, the pressure sensing value is low. Therefore, pump full event will not be triggered with the pressure sensing method. By detecting a failed pump full event, we then can detect an empty liquid tank. Steps 201 and 202 in
Also, if the liquid tank 130 is empty, then air will be released from the pump rather than liquid due to the interrupt of liquid supply. It is hard to establish the driving pressure due to high volumetric releasing rate caused by low air density. Accordingly, by detecting the time of Mode 1 in a pressing stroke, we can detect an empty liquid tank or a leak pump. Steps 203 and 204 in
The pump depicted in
Pl=(Pc*A303−ks*x−f0)/A304 (8)
where A303 is the area of large diameter surface 303, ks is the spring constant of spring 305, x is the distance from natural position of the piston 302 to the current position, f0 is the friction force plus the static spring force, and A304 is the area of small diameter surface 304. According to equation (8), if spring force and friction force is small compared to that applied by the compressed air on the surface 303, the ratio between the areas A303 and A304, A303/A304, determines the driving pressure.
In a pressing stroke, when compressed air establishes pressure in the space 340, the piston goes downward under the pressure, pressing the spring 305 and generating driving pressure in the space 330. In a suction stroke, when the compressed air is released, the piston goes upward under the force provided by the spring 305. Thereby liquid is pulled in the space 330. Compared to the pump 100 shown in
The controls for the pump of
Pl(Lm)=(Pc*A301−ks*Lm−f0)/A304 (9)
where Lm is the max. displacement of the spring 305 in the pump 300. Compared to the air-driven pumps in the previous arts (e.g. in U.S. Pat. Nos. 6,386,841, 4,645,431, and 3,963,383), in the present invention, in addition to driving pressure being controlled in closed loop, with the same compressed supply air pressure, the driving pressure in the pump is also adjustable within a broad range. Typically, the lower end of the range is limited by the check valve block pressure, while the upper end is determined by the compressed supply air pressure and the ratio of the large diameter and small diameter of the piston. The adjustable driving pressure separates driving pressure from compressed air supply pressure and thereby enables more flexible applications of the air-driven pump.
Air-driven pumps need to be refilled with a suction stroke, during which the driving pressure drops and fluid stops flowing out from the pump. For some applications with high liquid flow rate, the pump needs to be refilled from time to time. To decrease pressure drops and provide continuous fluid flow, a hydraulic buffer can be used with the pump. Referring to
When pressure is established in the space 330 of the pump 300, through the port 102, the check valve 104, the passage 407, and the port 408, liquid flows into the lower chamber 420 of the hydraulic buffer 400 and builds up pressure therein. The pressure inside the lower chamber 420 pushes the piston 403 moving upward and pressing the spring 402. In a pressing stroke, pressure inside the lower space 420 is controlled by the controller 110 (
The pressure drop in the hydraulic buffer 400 is affected by the suction stroke time. In applications where only low pressure fluctuation is allowed, the suction stroke time has to be short or a large hydraulic buffer is required. To further decrease pressure drop and continuously control driving pressure, two air-driven pumps can be used together to have the driving pressure controlled by at least one pump during working time. As depicted in
Liquid flows out of the hydraulic buffer 630 under a driving pressure controlled by the controller 640. Referring to
In the pressure control, the idling pump (e.g the pump 610) starts a pressing stroke (Mode 3) after the working pump (e.g. the pump 620) enables pressure feedback control. Therefore, there is a period of overlap time in which both of the pumps are in pressing stroke, e.g. between the moments of 711 and 712. During the overlap time, the working pump is in pressure feedback control, while pressure in the other one is not controlled though the pump is pressurized. Pressurizing the idling pump is to reduce the transition time after pump control is switched. The pressure in the idling pump should be close to but lower than the threshold (e.g. the upper limit of the predetermined pressure control range) above which the working pump goes into Mode 0 releasing air, since liquid in the idling pump cannot flow into the working pump, and high pressure in the idling pump could firstly cause a peak of driving pressure in the hydraulic buffer and then a valley due to feedback control in the working pump. The air filling time (Mode 3 duration time) of the idling pump is calculated using the net air filling time of the working pump, which is a function of the total air filling time (Mode 3 duration time) and the total air releasing time (Mode 0 duration time) of the working pump before its pressure control goes steady. The net air filling time of working pump is a reference. The calculated air filling time for the idling pump should be shorter than the net air filling time.
Although the apparatus and method of the invention are described herein in relation to the preferred embodiments shown in
Number | Date | Country | |
---|---|---|---|
61520630 | Jun 2011 | US |