How to fix Water hammer in pump process system

[A Learning Engineer]

I just finished a project at my first internship/co-op that involves making a pump system automated. Basically, the operator will type in the pressure setpoint, and a variety of devices will work together to increase the pump to pressure. The pump I am using is a Haskal 4b-14 pump. I have a 1/2-inch air line running into a 1/4 in air supply pipe. On the 1/4-inch air supply pipe there is a 3/2 solenoid valve, and an Ip current to pressure transducer. Both have 1/4 inch in and out ports, and the transducer controls the pressure that is going into the pump. On the outlet side of the pump there is crossroad, where many different pipe sizes are, first there is a pressure transmitter that is 1/4 inch, the outlet side of the pump is 1/2 inch so a reducer is needed at the crossroad for it. Then there is a 1/2-inch mechanical relief valve that is inline and goes to the part we are testing. At the bottom of the crossroad there is a pressure gauge, and it is lead to a hydraulic solenoid valve and its ports are 3/8 inch in size. The hydraulic solenoid valve is used as a safety valve if the pressure transmitter detects an overpressure. There is also a fluid supply to the bottom of the pump which is 1/2 inch.

I gave you all of these details because I have only one problem with my system. It has water hammer. When the system pumps up to setpoint and starts holding the pressure, the pressure in the IP transducer stays constant. While either holding the pressure and starting testing time, or when pumping up to pressure, whenever the pump cycles it creates a water hammer effect. The pump cycles and the pressure on the gauge jumps to 100psi over the setpoint for 1 second before returning to its original point.

For example, I was testing the system at 200psi, I started the test and the IP transducer slowly increased pressure until setpoint was met. Once it reached that setpoint, the IP transducer held the pressure constant, and the PLC started its test time. After about 10 to 15 seconds the pump cycles and water hammer occur, the pressure on the gauge spikes to about 280psi sometimes 300psi and then comes back down to around 200psi. My pressure transmitter also reads this and sometimes thinks there is an overpressure, I have a setpoint for overpressure, say 100psi over setpoint then open the hydraulic solenoid valve. This water hammer is not only causing my overpressure device not to work, but also gets dangerous and unsafe as the pressure increases. As I increase the setpoint to let's say 800 psi, the water hammer will be even stronger and the pressure wave will be around 400 psi, which is not good for the system and the part we are testing.

My question is how do I fix this water hammer and why is it happening?
Why does the manual version of this system not have water hammer? Manual version uses the same pump but instead of electronic controlled valves and transducers everything is manual, so the operator needs to increase pressure on the regulator and hold it constant. When the pump cycles no hammer head happens, the opposite happens, the gauge on the outlet side of the pump drops to 0 psi for a second then comes back to the original pressure.
Let me know what you think the problem is.

 

[SwinnyGG]

You have a bunch of really small lines and a bunch of fast acting valves; that's the perfect recipe for water hammer.

The water hammer doesn't happen under manual control because you aren't snapping a valve shut. Water hammer is the result of moving water being closed off by a valve that closes very rapidly - like an electronic or hydraulic solenoid valve. When a valve closes very quickly, the kinetic energy in the moving water behind the valve has no place to dissipate, so you get crazy pressure spikes.

You can rectify this by adding a hammer arrestor - which basically adds an isolated pocket of compressible gas into the system to provide a place for that kinetic energy to go. You can also rectify this by changing your valves for slow acting valves - these are a thing, and they exist for this specific reason.

 

[A Learning Engineer]

@SwinnyGG I was thinking about doing this. I just don't understand why this is happening because my other pump system is a manual setup and it has no water hammering at all. Not sure why, but I want to think that I can fix this problem without a arrestor. I am going to add a check valve to the air inlet side, not sure if that will do anything. If not I guess I will look into a arrestor. I need to find one that can withstand between 1600-2000 psi. Also in my case the electronics are acting exactly as they would if done manually. The pressure to the pump stays constant and the outlet pressure stays constant (if there are no leaks) when the pump drops a bit of pressure it cycles a bit and gets back up to pressure, but with the electronic valves for some reason the pump starts water hammering even though its not closing any valves.

TLDR: The electronic system is doing the exact same thing as the manual one, it keeps air drive pressure constant to keep water outlet pressure constant. Only difference is when the pump drops a bit of psi at water output pressure and cycles again, the manual setup goes back to pressure, the electronic setup goes back to pressure and also has a pressure spike of 100psi before coming back down to pressure.

 

[IRstuff]

There are two things that should work to reduce water hammer

> Water hammer is the momentum transfer from the moving water to the piping when the valve is closed quickly, so reducing the flow rate of the water would reduce the amount of momentum of moving water, which should reduce the water hammer

> As discussed above, manual valve closures typically do create water hammer, IF the valve is like a faucet valve, which takes time to close, but it also means that the water flow is reduced relatively gradually. However, a diverter valve, such as the one I have on my shower, is a fairly rapidly responding valve, and I can consistently produce water hammer. Therefore, if the valve can be closed more slowly, the water hammer can be reduced. Like

https://wicvalve.com/1-2-Inch-Anti-Water-Hammering-Electric-Solenoid-Valve-NC-2SCD-1-2-D-V.htm

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[A Learning Engineer]

The problem is there is no valve that I am controlling that is causing this. The water hammer is all caused by the pump mechanically cycling. The Air valve is left open and the transducer is sending a constant psi to the pump, no valves are moving at this point, the pump is just trying to hold pressure. Then when the pump cycles again during holding pressure (not sure why it cycles if it's at constant pressure) the water hammer happens and my pressure gauges spike to 100 psi over what it was holding at (say I was doing a test at 200psi, it was holding it there by keeping constant air pressure of 9.5 psi, then the pump cycles again for some reason and the water hammer happens, my gauges spike to 330psi for a second then drop back down to around 220 psi, the air pressure stays constant during this time). Not sure why this is happening, I am literally doing nothing in the system, no valves open and the pressure transducer sends a constant pressure. Not sure why when the pump cycles again the hammer happens.

Like I said before I am not closing any valves when this happens, everything stays open and constant to keep the pressure constant. The pump is mechanically cycling again and creating the water hammer, not sure why. Using the exact same pump as the manual system.

 

[IRstuff]

OK, so the problem is the pump, so you need to figure out how to keep it from cycling when it's not supposed to be operating. Again, as suggested earlier, a hammer arrestor would be the simplest solution, if you don't, or can't, tackle the pump cycling problem

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[A Learning Engineer]

I guess I will need to find out how my pump works exactly, maybe a check valve is damaged because this only happens as everything stays constant (9.5 psi air inlet, 200 psi water outlet, pump cycles, 200 psi goes to 300 psi for half a second before dropping back down). How can the pump spike in pressure when the air drive pressure is constant the whole time?

 

[waross]

Look for pressure spikes in the air line feeding the pump.
The pressure transducer may be slow acting when feeding a small volume system.
You may need a device on the control air line.
You don't mention what you are using for controller.
Some controllers may develop transient pressure surges if the volume of the load is too little.
That is, your problem may be supply pressure spikes causing output pressure spikes.
I solve a similar issue with about 2 feet of capped 1.5 inch diameter pipe capped and teed into the output of the controller.
The added volume of the added 1.5" pipe smoothed everything out.
But first check for pressure spikes.
The first rule of problem solving is to determine what the problem actually is.
If you have a control air issue you will waste a lot of time and money chasing water hammer down the rabbit hole.
Hint. It works on manual but not with an electronic valve. Maybe the valve (transducer) is the problem.
The second rule of problem solving is to recheck the first rule.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[A Learning Engineer]

@waross, the controller is a Nitra NCP2-20-3120N IP transducer, this is being controlled by a click PLC. As I watch the water hammer occur 2 things happen to the output (water output) and input (air input) gauge to the pump. The pressure on the air side drops down to 0 when the pump cycles, while the water side pressure gauge has a spike in pressure that lasts for 1 sec, spiking 60-150 psi then dropping back down to its original pressure. Also, this water hammer happens only sometimes and occurs randomly. If I am doing a 150psi test hold the chance for a water hammer to occur is less likely than if I was doing a 200psi test hold. This only happens when holding the pressure constant, so while holding that 200psi outlet pressure constant the air controller is holding a constant pressure. Then the pump cycles and the water hammer occurs, sometimes it happens 4 times in a test and sometimes only once. Not sure how the outlet side is spiking when the air drive pressure stays constant, on the manual pump it cycles while holding pressure, but when it cycles nothing happens, it lets some air in and the pressure stays constant at both sides. I will need to figure out how the pump works, because I am not sure why it even cycles when the air pressure is constant.