Modern Designs of Large Bore Pump Stations
Pump Station designs have evolved over the years. This is due to new designs of valves becoming available and also a higher awareness of Energy efficiency – particularly in Large bore Pump Stations and their adjoining Pipelines.
This article investigates some of these design features and possible causes of failures of older systems. Whilst not specific to any particular installation, inspiration for this article comes from discussions with some Pump Station/Pipeline Operators, and the realization that better solutions are available today.
Pump Control Valves
Gate valves were probably the most popular choice for a Pump Control Valve in large bore applications for the following reasons:
- Robustness of design with assumption of long life.
- Low Pressure drop due to “Full Bore” features.
- Ability to handle Raw water with occurrence of particles in the water.
- Ability to handle the high velocities during slow opening of the valve inherent in the requirement of Pump Control Valves.
Low pressure drop assumptions of Wedge Gate valves are a fallacy as the bottom groove into which the gate moves, as well as the guides, causes large turbulence resulting in Pressure drops. Besides the Energy losses, turbulence releases air out of solution and which goes into the pipeline system.
The other problem is that the flow characteristics through a Gate valve is anything but Linear with the valve effectively fully open at 20% from closed. So if the Design Engineer calculates that a 10 minute opening and closing time is required to minimize water hammer, he might not realize that the gate valve effectively cycles in 2 minutes!
The evolution of RSV Gate valves into high quality and dependable products have reduced the turbulence effects to some degree, but there is still a limit in sizes and pressure ratings available, so Wedge Gate valves are still very popular.
Butterfly Valves. This valve has become a popular alternative as Pump Control Valve as quality of manufacture has improved and they are cheaper and easier to actuate than Gate valves. Most popular on large bore stations is the Double Eccentric Soft seated designs, although Triple Eccentric Metal Seated designs are better due to their more linear flow characteristics and also longer life of the metal seals.
Energy losses are a problem and turbulence results in the same release of air as in the Gate valve.
Ring Needle Valves. These valves are quite popular for the following reasons:
- Pressure balanced Piston makes for easy actuation and inherent low forces on components and leading to long life and good reputations of these valves.
- Excellent ability to open slowly and withstand the high velocities experienced.
- In applications where the Pump Control Valve has to fill an empty line on a regular basis, this valve can be supplied with a custom made trim whereby the initial opening stage provides high pressure drop and when the valve is fully open pressure drop is much lower.
Although better than Gate or Butterfly Valves, pressure drop is higher than some other designs available and discussed below.
Spherical Valves.These are considered superior as Pump Control Valves for the following reasons:
- Full Bore design, with pressure drop same as an equivalent length of pipe – ie negligible.
- Eccentric shaft resulting in the retraction of the Seal (on the ball) from the body seat, resulting in almost no rubbing between seat and seal and long life.
- The seal on the ball is inherently protected from damage from flow velocity.
The fact that there are only a few manufacturers of these valves in the world, and their expense to manufacture have held them back from popularity on a larger scale.
Standard Ball valves with Hard Chromed ball and PTFE Seals are not suitable for this application as any particles which may get in between ball and seals will damage the seals, causing leakage and fast destruction of the valve.
Hydraulic Considerations to reduce Valve Slam and resultant Water Hammer after Pump Trip.
The above comparison of Pump Control valve designs considers only the suitability of various designs from a Mechanical and Energy loss point of view. Both these factors also have a large bearing on inherent dependability and valve life, ie the easier a valve is to operate and the less pressure drop, the lower the inherent destructive forces on the valve components and both meaning longer life.
Considerations of Water Hammer reduction after a pump trip have to do with the speed of closure of valves on the Pump Delivery and to bring the Water Column to a stop during the short period between moving forward and reversing (microseconds in a lot of applications). Different strategies can be employed to achieve the same result of ideal valve closure time during this period of “no flow”. One strategy is to make the Pump Control Valve closure very fast and the other is to choose a check valve to achieve this fast closure. Let’s investigate both available options.
Pump Control Valves.
To achieve a high speed of closure for a Pump Control Valve, the correct actuator has to be selected.
With traditional Electric actuators, when there is a Power failure the actuator stays in the position where it was at time of failure. Electric Actuator companies have come up with designs which incorporate a Nitrogen gas tank powering a cylinder to drive the valve, and a solenoid and declutching mechanism to disconnect the Actuator from the valve. This is complicated and expensive and prone to failure at the time of emergency.
So designers turned to Hydraulic actuators to overcome this problem with Electric actuators. In this method a Hydraulic Cylinder drives the valve, with the hydraulic power supplied by a standard Power Pack (some designers opt for one Power Pack driving all actuators in a Pump Station whereas others choose a Power Pack to drive each valve actuator). The valve also has a Counterweight attached to the shaft. In a Power failure scenario, a solenoid releases the hydraulic oil from one side of the actuator piston to the other side and together with the counterweight, the valve is closed in quite a fast time period.
Today this has become an accepted method of achieving fast valve closure in a Power failure scenario. This actuation method can be used on most Pump Control Valve designs. Although a very positive step in the right direction to provide a solution for the fast closure requirement, there is still a limit to closure time due to the inherent inertia of the Counterweight, and Water Hammer analysis might dictate a faster closing time than can be provided by this method. Some Pump Station designers and operators do not like the Counterweight as they are bulky and need to be installed behind covers from a safety point of view.
The Petroleum industry have for many years used “Gas over Oil” actuators for ESD (Emergency Shutdown) valves due to the very fast closure periods which can be achieved.
Check valve designs have evolved over the last 20 years (or so), and the main aim of designers have been to increase the speed of closure under its own power. So in cases where the Pump Control valve failure mode is not fast enough (or no failure mode at all in cases of Electrically actuated valves), a fast acting check valve can provide the solution. Different designs are used including Tilting Disc check valves and Swing check valves with counterweight (and hydraulic dampers), but in the case of very fast closure being required they simply are not fast enough.
The best check valve design which can close in a very fast period is the Nozzle check valve and which has gained in popularity over the last 10 years. The key to its design feature is the short stroke, inherent strong spring to drive the valve, and low pressure drop due to the “Nozzle” design which results in large pressure recovery. This valve also does not need counterweights and hydraulic actuators – fast closure speed is inherent in its design. Care should be taken in using this valve in Raw Water applications (where large particles could be encountered), due to its small flow passage.
Combination Pump Control/Check Valves.
With the use of Hydraulic actuation and Counterweights, most valve designs can be used as combination Pump Control and Check valves, and depending on reliability and faith in the designs, as well as cost of the chosen Pump Control valve, a Check valve can be eliminated altogether. This all depends on the design philosophy employed. In a lot of large bore and expensive Pump Stations, it may well be considered prudent to have both.
A.R.I (Air Valve manufacturers) have designed a Unique Actuator which uses the “Gas over Oil” principle and mounted on their Swing Check valve, provides a solution which achieves all the following advantages:
- Very fast (and adjustable) closure speed.
- Compact and self contained with built on Power pack.
- Fully retracted disc resulting in similar low pressure drop to Ball valves.
- Flow protected seats resulting in long life.
This valve has proven itself in providing solutions in Pump Stations where severe Water Hammer had previously been experienced. Almost infinite adjustability of speed of closure is possible, as the issue of high inertia of Counterweight designs has been eliminated.
Air Valves/Surge Tanks.
Whilst the problem of Valve “Slam” can be eliminated by employing correct design principles as discussed above, the problem of Column separation and resultant slam from the returning water column can only be reduced by the following methods:
- Bypass check valve to fill the low pressure zone.
- Air valves to suck in air to fill the low pressure zone and let the air out slowly, thereby dampening the returning water column.
- Surge Tanks to achieve the same results as Air valves.
In most cases bypass check valves operate too slowly to have the required effect and will be eliminated from this discussion.
Air valves and Surge tanks provide similar results and correct choices can only be made with modern Water Hammer software. Of course the correct moddeling of Air valves and surge tanks in the software is crucial and not always done accurately.
Choices between Air valve solutions and/or Surge tank solutions are also influenced by the confidence in either option by design engineers and clients.
The rest of this article focuses on Air valves, as it’s the author’s opinion that their uses are often misunderstood and under-utilized, and if the right designs types are used, can solve many problematic Pump Station and Pipeline issues, at relatively low cost.
Firstly, every Pump delivery should have a two stage closing Air valve to reduce the Column separation problem. Even if a Surge tank is employed which will inherently do the same, the Air valve still has the third function of releasing pressurized air bubbles released from the action of the pump, valves and fittings in a pump station.
Now for the controversial part.
Air valve designers have concentrated on their large orifices over the years for the following reasons:
- It was the main item they could influence to achieve commercial advantages ie size of orifice and body size. The smaller they could make the orifice the less material on the valve body and the cheaper to manufacture.
- Modern designs incorporating two-stage closure concentrated on “switching” points between different designs and advertise that their design is better than others. The most popular SA design which everyone has specified over the years switches at between 40 and 70 kPa and which means their switching point is almost never achieved in properly operated pipelines!! This should be a shocking revelation for users of these Air valves, who have spent large sums of money thinking their Pipelines are protected.
Some new designs now advertise adjustable switching points. Which Engineer or user is going to adjust his switching point and based on what knowledge ? It is the author’s opinion that adjustable switching points are used purely as marketing “hype” and serves little purpose. Rather use a design where the 2nd stage can either be modified by the manufacturer to suit a specific special requirement, or leave it out altogether where they serve no purpose.
- Attempts to achieve non-leaking large orifices have largely been unsuccessful due to the high hydraulic forces causing leaking seals after a short period of time. Often the end result of a leaking Air valve is to close the isolating valve underneath it, to get rid of the problem !
So whilst all the marketing and design activities revolve around the large orifice, the small orifice is largely ignored, reason being that designers cannot overcome the physical problem that orifices which release pressurized air bubbles have to inherently be very small – so small that they easily clog up with dirt and become totally ineffective. And this is a big problem when you consider that 90% of the time in the life of an air valve the small orifice does all the work – release air from an operational pressurized pipeline !
And the Design Engineer and Pipeline operator are never aware of whether the small orifice is actually working or totally blocked and not working at all. Another factor to consider is that the relationship of pipe pressure, small orifice size and weight of the float is so critical that sophisticated users who are aware of this and who want to make sure that the float does release and let out air, specify “drop tests” to verify that the small orifice operates effectively. Even though the valve might pass the drop test, what happens after being in operation for a period of time ? Any minimal wear on the small orifice causes the size to be bigger than the original new orifice and which causes the float to not release from the orifice – hence another potential failure of the small orifice.
The author has performed some random tests on valves in the field as follows:
If the valve has a small drain valve on the bottom (as required by a lot of users), open it. In a high percentage of valves it is found that air is blown out for a short period and then water. This is the sure-sign test that the small orifice is not functional and the air valve is full of air !
So, does the small orifice really serve a significant purpose ?
The answer is an emphatic yes !!
Why do all Design Engineers specify an Air valve at least every 500 or 700m even though the pipe gradient might be relatively flat and the sizing (for pipe filling and draining) requires air valves at much larger intervals ? Because the small orifice performs the critical function of releasing pressurized air during normal pipe operation. If the small orifice does not function properly, chances are good that the pipeline has air trapment of large bubbles of air. Besides reducing the capacity of the pipeline and all the other negative factors of having air pockets, localized high velocity areas occur at these air pockets. During any Water Hammer causing event such as a Pump trip, the pipeline is bound to fail at the high velocity spots as the change in energy from Kinetic to Static (Pressure) is greatest at these spots. Most Design Engineers are very aware of this and that’s why they install air valves in these areas.
A.R.I have since many years ago designed a totally different concept for the small orifice. It works on the “rolling seal” principle and results in an orifice which is between 10 to 16 times larger than most air valves. It never suffers from “drop test” malfunction as the seal is broken initially in a fraction of a millimeter – so the orifice area/pressure/float weight relationship is almost irrelevant with this design. This concept works so well that many manufacturers have tried to copy it, but the patent has been so strong that they have been prevented by legal action.
Many users who have first started using the ARI air valves have noted that this air valve makes a continuous intermittent noise which they have not experienced with other air valves. This is because the small orifice is actually doing its job of releasing pressurized air all the time !
Water Hammer events can be overcome by proper design of Pump Stations. In a lot of older designs, water hammer is tolerated when the pressure spikes are within the design rating of the pipeline. What is often neglected is that the continuous high pressure spikes eventually weaken pipes and components, particularly when inferior material (such as GRP) is used. It is often worthwhile to re-examine potential Water Hammer relieving mechanisms in these older Pump Stations and their adjoining pipelines, as new methodologies and equipment can achieve much better results. With a review of the Pump Control Valves, Check Valves and the type of Air Valve installed and the new designs available, most problematic Water Hammer issues can be solved successfully.