Water hammer while well is pumping

[Mliu]

Property owner has an existing shallow, low-yield well feeding a 3,000 gallon atmospheric storage tank for landscape irrigation. He wanted more volume so he had a second shallow well drilled approx 150' lateral distance from the first well. The discharge from the first well flows through 1" PVC pipe and simply dumps into the top of the storage tank. The 1" PVC discharge pipe from the second well was spliced into the discharge pipe from the first well using a tee fitting: the discharge from the first well flows straight through the tee, while the discharge from the second well flows into the tee from the side port.

The first pump continues to operate as before with no problems. But when the second well is run with its output throttle valve opened to any reasonable amount of flow (>3 gpm), there is tremendous, continuous water hammer in the discharge pipe; this is while the pump is running, not when it stops. The water hammer occurs whether or not the first pump is operating. I suspect that the hammer is caused by the way the water enters the tee from the side and then has to turn 90 degrees abruptly to the tank fill pipe, while also creating standing pressure waves in the pipe leading back to the first well. Thoughts?

I'm planning to replace this tee connection with a wye so the water from the second well enters into the common tank fill pipe at a 45 degree angle. Thoughts?

 

[Valveman]

I have systems with a dozen pumps teed into a mainline. That is not what is causing the hammer. How fast of a hammer is it? Is it possible you pump is kicking on and off.

 

[Sponsor]

 

[LLigetfa]

My guess is the pump is cavitating.

 

[Valveman]

My guess is the pump is cavitating.

If the pump is not kicking on and off for some reason that would be my second guess. Since it starts as soon as the pump comes on it makes me think of cavitation or pumping the well dry as the second possibility.

 

[Mliu]

Some additional information:

1. The water hammer is rapid, violent, and loud. As soon as it occurs I stop it for fear of physical damage. So I don't have a lot of parametric information from during its occurrence.

2. If the throttling valve is partially closed to the point of about 3GPM or less (backpressure at the well head of ~75-80 psi), then when the pump comes on, NO noticeable water hammer occurs (although I've observed some rapid vibration on the pressure gauge needle of +/-2 psi magnitude).

3. In the case above, if I begin opening the throttling valve -- while the pump is running -- to the point that the back pressure drops to ~70 psi or less, the waterhammer immediately occurs. If I then start close the throttling valve back to about 75-80 psi back pressure, the water hammer stops.

4. There is a spigot between the well head and the throttling valve. If the spigot is open when the pump comes on, then no water hammer. I don't recall seeing any pressure gauge vibrations during this test.

5. The pump is not cycling.

6. The well is not pumping dry.​

So the water hammer only occurs when there is flow through the discharge pipe to the storage tank and when that flow exceeds a given threshold.

 

[Valveman]

Then my next guess would be you are using a gate valve as a throttle valve and the gate is bouncing.

 

[Craigpump]

I agree with Valveman on the gate valve either that or there is some type of obstruction before the T

 

[Mliu]

Thanks for the information. The throttling valve is a gate valve. In fact, both wells are set up exactly alike and there's never been any problems on the first well no matter how open or closed the setting of its gate valve for throttling. But I suppose the second well's valve could have a loose fitting gate.

I had intended to replace both throttling valves with Dole valves, but wanted to "balance" the output of the two wells first because they are both driven by a single master timer. We're trying to maximize the output of both wells without overdrawing them and damaging the aquafer. Right now, the first well is set so it delivers about 5 gpm while running, and it's currently set to run 5 minutes per hour because the well's past performance was roughly 600 gal per day. Since the master timer runs both well pumps simultaneously, any difference in recovery rate will have to be accounted for by the setting of the throttling valve. For example, the second well might get a 6 gpm valve if it produces more than the first well, or a 4 gpm valve if it produces less. Considering how close they are to each other and that they are both at about the same depth, I suspect they will both perform roughly equally, and I'm not even convinced the second well will net much additional water as I believe their zones of influence will overlap considerably. But I was not the one who dictated the placement of these wells.

Would you recommend a globe valve instead of the gate valve? I didn't chose ball valves originally because they are more difficult to fine-tune the flow rate through them. If I put in a globe valve (or some other recommended adjustable valve), should I bother replacing them with Dole valves later on?

 

[Mliu]

Yesterday, I replaced the gate valve with a globe valve. No difference -- the pipe to the tank is still hammering badly when the flow through the discharge line is increased over a certain level.

 

[Valveman]

Yesterday, I replaced the gate valve with a globe valve. No difference -- the pipe to the tank is still hammering badly when the flow through the discharge line is increased over a certain level.

I am afraid you may have what is known as a “hydraulic phenomenon”. In other words I have no idea what the heck could be causing that.

I would check the amps. If it is not electrical I would isolate the pipe at the well. I had a brake line act that way one time as it was weakened and would contract and expand quickly causing a hammer effect. Maybe you have some flexible pipe underground somewhere?

 

[Mliu]

I am afraid you may have what is known as a “hydraulic phenomenon”. In other words I have no idea what the heck could be causing that.

I took a video of the well head pressure gauge as I opened and closed the throttling valve and the spigot in order to illustrate the "hydraulic phenomenon." As soon as I get it uploaded, I'll post a link so you can see what's happening.

I would check the amps. If it is not electrical I would isolate the pipe at the well.

I don't think it's the pump. When the well was first drilled, the owner wanted to get an idea how much it would produce. So we lowered the pump into the well with a garden hose attached to the pump on one end (the well is only 30' deep), and with a PVC ball valve at the other end of the hose for throttling. The water flowed fine with no pump issues or other "hydraulic phenomenon." Even now, if I open the spigot at the well head (which is upstream of the throttling valve), then the water hammer goes away and the pump seems to run fine.

What do you mean by "isolate the pipe at the well"?

I had a brake line act that way one time as it was weakened and would contract and expand quickly causing a hammer effect. Maybe you have some flexible pipe underground somewhere?

I don't think there is anything wrong with the piping. The pipe coming off the well head is galvanized steel. After the throttling valve, there's a short length of Sch80 PVC until it transitions underground. Then it's Sch40 PVC all the way to the tank (~150'), bedded in sand and buried at a depth of 18".

 

[Reach4]

Is this controlled by a float switch only?
If there is a pressure switch, have you listened to the pressure switch while the oscillation was going on?

If there is a pressure switch, is there a pressure tank at the pressure switch?

 

[Mliu]

Have you listened to the pressure switch while the oscillation was going on?

Is there a pressure tank at the pressure switch?

There is no pressure tank and no pressure switch. The discharge pipe from the well empties into the top of a storage tank that is open to the atmosphere (an above-ground cistern). Pressurized water is provided to the irrigation system by an independent VFD pump inside the storage tank.

So the only back pressure in the discharge pipe, after the water passes the throttling valve, is the 7' of water head to the top of the storage tank.

 

[Reach4]

Going back to LLigetfa's cavitation idea, I guess the suspicion would be that the intake to the pump was partially blocked, creating an excessive vacuum when too much water was being drawn past the blockage. I have no experience, but it seems plausible.

 

[Mliu]

Going back to LLigetfa's cavitation idea, I guess the suspicion would be that the intake to the pump was partially blocked, creating an excessive vacuum when too much water was being drawn past the blockage. I have no experience, but it seems plausible.

Unlikely to be the problem. There is a spigot (garden hose faucet) located between the well head and the throttling valve. If I fully open that spigot (discharging the water on the ground by the well head), then there is no water hammer and the pump runs fine. The only time there is a problem is when the water is directed to the storage tank with too much flow rate.

 

[Mliu]

I still suspect that it has to do with the way the #2 discharge pipe tees into the #1 discharge pipe. The next step is to excavate that connection, cut the #2 pipe off that tee, and then try running the #2 pump.

 

[Valveman]

If the tee into the line has anything to do with the water hammer I will eat my hat. But at this point I don't know what else to tell you to try. You might move the throttle valve to the well head by the faucet, if it is at the tee connection now. It is going to be a simple and small problem when you find it because big problems are much easier to spot.

 

[Mliu]

You might move the throttle valve to the well head by the faucet, if it is at the tee connection now.

The throttling valve is at the well head; the tee is 150' away.

After the well seal, the water passes through a flow meter, past the spigot (hose bib), through the throttling valve, through a one-way valve, and then on to the tank. The tee is underground at the intersection of the discharge pipes from well #1 and #2; it is at the base of the storage tank.

 

[Boycedrilling]

When you say one-way valve I assume you mean a check valve. I would be suspicious that the poppet in the check valve is chattering and causing your problem. It is down stream from the hose bibb, and you don't have a problem at the hose bibb.

 

[Mliu]

When you say one-way valve I assume you mean a check valve. I would be suspicious that the poppet in the check valve is chattering and causing your problem. It is down stream from the hose bibb, and you don't have a problem at the hose bibb.

Yes, it's a check valve. Sorry, I wasn't thinking clearly this morning.

Is there a way to determine if the check valve has a chattering poppet? The check valve in question is a PVC check valve that has a spring-loaded poppet. This well head is set up exactly the same as the first well head (which also has the same type of PVC check valve), and the first well has been operating for several years without problem.

I can take out the check valve and replace it with a section of straight pipe for testing purposes. Perhaps I'll go there tomorrow to try that.