Check Valve Causing Water Hammer 1

[jartgo]

Hope everyone is having a great new year. I have a small project I'm working on bringing on line and I'd like to get your opinions about how I might address an issue that is more aesthetic than physical I think, but nevertheless, I'd like to see if I can improve upon it.

I have a submersible well pump (grundfos SP) doing about 180 gpm into some 4" filter piping. The piping is SCH80 PVC. The general line of flow is as follows (imagine a standard well with submersible pump, pumping directly into a pressure sand filter then out to the system):
Pump - about 100 feet below ground
Air Release - 3 feet above ground level
Swing Check Valve - 3 feet above ground level
Blow Off Tee
Gate Valve
Piping into/out of filters
Discharge into distribution system through 6 - inch line.

The distribution system pressure is about 50 psi, and the pump discharge pressure when the pump is running is about 60 psi. The pumps are started with a soft starter and when stopped, I get a noticeable water hammer on the filter face piping. My thought was that I either need to slow down or speed up the check valve closing. It's currently a weighted check valve and my thought was to put a spring on it to get it close as soon as possible after the pump stops. Before I make any changes, I thought I'd see what you all might recommend. My thought is that if I can get the check valve to shut before water has much of a chance to reverse direction, then the hammer might be reduced. I'd rather not get into adding a surge valve. Kind of wishing I had used a VFD in lieu of a soft starter, but it's a fairly small (15HP) motor.

The surge doesn't appear to be threatening to the system at all but the pipes do move and you can hear the hammer, which is what I would like to prevent as much as possible. Looking at the gauges, when the pump stops, the gauges rise to about 70 psi during the hammer.

What are your thoughts? Thanks in advance.

 

[bimr]

You might consider using one of the check valves that are advertised as a silent check valve.

 

[QualityTime]

You need to slowly introduce the water into the piping system on startup and slowly stop the feed of water into the piping system on pump shutdown. In order to do this you need to install a motorized butterfly that has up to 6 internal limit switches and it should have a very long CLOSE to OPEN cycle time (2 minutes would be good...otherwise consider using a modulating rate control valve where you can control the rate of opening and closing).

When the pump needs to be energized, the pump discharge valve should move from 100% closed position to 95% closed position. When it reaches the 95% clcosed position, an internal limit switch in the motorized operator should be wired to the pump starter to allow it to energize. At this point the pump is basically starting up against a closed valve to prevent a sudden full blast of water from being pushed out into the system. The discharge valve should then slowly open to the 100% open position that you are looking for. This will prevent water hammer on startup

The sudden shut down is causing water hammer. When the pump needs to be de-energized, the motorized pump discharge valve should throttle back to say the 98% closed postion. Upon reaching that postion, the related valve operator limit switch should trigger the pump circuit to de-energize. The valve continues to go to the full 100% closed position. This will prevent water hammer on shutdown.

 

[greggy]

We have had very good success with a resilient flapper check valve on pump discharge settings such as the 2100 Series by AFC. Very quiet on pump shutdowns.

 

[BigInch]

greggy's solution will probably work quite well.
Here's another one that should do the trick too, a valve w/ surge relief.

http://www.cyclestopvalves.com/pdf/catalog.pdf

I hate Windowz 8!!!!

 

[hydrae]

While most of the above solutions above will suppress the pressure spike on pump shut down, a true solution is the installation of a properly sized pressure tank after the filter

What is the type (PVC, cast iron ductile iron steel) and length(s) of the distribution system piping and what are the controls to call and release the pump? I assume you are pumping to an elevated storage tank with level switches. With the above info, it can be calculated what size of pressure tank or tanks will contain the surge on both start up and shutdown.

Hydrae

 

[djs]

You can also try to adjust the ramp down speed on the soft starter. With a long enough of a ramp down, should eliminate the water hammer.

 

[QualityTime]

There are many ways of lowering the surge pressures in the system. Energizing and de-energizing a pump against a virutally closed valve as described above and use of a properly sized pressure tank (surge tank) is the best combination to use. The pressure tank is really for situations where there is sudden power failure becasue during this occurence, controlled closing of the pump discharge valve cannot be done.

Energizing a pump and de-energizing a pump against an open discharge valve is never a good idea. The use of PVC piping kind of makes the situation more risky over the long term. I am not sure what will happen over time with the constant surging.

 

[bimr]

Have to assume that you are leaving some details out because the scheme that you describe is unusual.

The silent check valve should resolve your immediate issue with the banging of the check valve. Silent check valves are designed to close immediately so that the reverse flow (and the banging) does not occur.

http://www.valmatic.com/silentcheck.html

The unusual aspects:

It is not typical to pump directly into a distribution system without some type of pressure tank or tower (standpipe) on the distibution system. With your arrangement, one would expect the well pump to cycle continuously.

The air release valve should be a well service valve.

http://www.valmatic.com/wellserviceair.html

Would also agree with QualityTime about the use of PVC piping.

 

[cvg]

agree with bimr, except possibly regarding pvc piping being a significant issue. without knowing how many starts / stops per hour/ per day it is impossible to predict any failure of either piping or valves in the near future. Fatigue failure which is a definite possibility with pvc pipe is related to the number of stress cycles and the severity of the stresses. And since you have used heavy pipe (sch 80) and have reasonably low pressures, I would not be terribly concerned unless your start / stops are every 5 minutes.

 

[QualityTime]

Usually one does a surge analysis on the system. Surge tanks are used when the positive pressure or negative pressure exceeds the rating of the pipe. I would not depend on a pressure gauge reading because the wave moves so quickly. The worst case condition is when all of the pumps are running and there is a power failure and all of the pumps shutdown. In this case it appears there is one pump.

I would still maintain that starting and stopping a pump against a closed valve is the best way to prevent the hammer. Also I am not sure why you would start full blast water flow into a pressure filter. The bump alone will push turbidity into the distribution system

Also with PVC piping you get what you pay for. Sounds like this is a municipal project. They usually like to see things last.

 

[semo]

Been down this road before when pumping into a sedimentation tank from a well. Some of the problem may come from the air release valve. Is there a possibility that it is actually and air/vacuum valve? That would allow air to reenter the line upon shutdown allowing the check valve to hammer as the water column went back down the pump column pipe. The correct valve is the one mentioned above by valmatic designed for well heads.

I always use a silent check valve on my wells to prevent the hammer. As discussed earlier, this closes automatically. It just creates a higher headloss in the piping.

There are also other special checkvalves designed for water hammer; but, the silent will work just as well in this application with a lot less cost.

 

[jartgo]

A little more info. The well is one well in a multi-well system, the distribution system maintains pressure via elevated storage tanks. The call for water is then dictated by the level in the elevated tank. When called, the wells feed into the distribution system which also directly feed the elevated storage tanks (and vice versa). There are five wells in the system so that each well is called no more than twice a week, normally.

The PVC piping is SCH80 which is tough stuff. The pressure ratings of this piping in the range of 4-inches and less is substantial. Its use is desirable when the flows and pressures are small due to its inherent resistance to corrosion. I have several small facilities in service for more than 10 years in this fashion and have exhibited no known issues with fatigue, etc.

To address the issue at hand, a simple programming adjustment to the soft starter was all that was needed in order to eliminate the hammer. Ramp down was set to 3 seconds as opposed to immediate off.

Thanks.