Water Hammer on Pump Shut off

[John64]

Hello,

I have a residential well that experiences water hammer every time the pump stops (since I've lived there so not a new problem).  The well is 90' deep, the casing is 5" and the pump is a 3/4 hp 10 gpm Sta-Rite HS+ Signature 2000 4" submersible at 70' and run by a 40/60 pressure switch.  I replaced the pressure tank 5 years ago with a 52 gallon which has 14 gallon draw down at 40/60.  I timed the fill time and it takes about 1.2 minutes with nothing running.  I don't know if there are additional check valves to the one on the pump.  The supply line is 1" PVC that enters the basement, immediately travels up the wall, crosses about 12 feet under the floor joists and then down to the pressure tank tee in the furnace room.  The system feeds a 2000 sq ft house with 2 full baths and also occasional lawn watering.

I'm thinking about adding a CSV1W for the constant pressure and less pump cycling benefits and I believe this will also solve the water hammering but I'm wondering if this might just be a bandaid for the hammering issue.  I assume the water hammer occurs from the check valve slamming shut.  Is this indicative of a bad or worn out check valve?  I'm worried that if the valve is bad that the pump may get damaged at some point.  Could this just be a normal result of the water column slamming back against the check valve?  I don't lose any pressure so I'm pretty sure that the check valve doesn't leak.

Also I understand the concept and the benefits of the CSV but I do have a concern about the pump overheating at low flow rates.  I don't know if the pump has the recommended flow inducer and I'm not inclined to pull the pump and install one for obvious reasons.  I understand that the CSV pressure and the shut off pressure should be adjusted to keep the tank fill time under a couple minutes after water demand is stopped so I'm not concerned there.  I'm more concerned if a single faucet is left running for an extended period, during long showers say at 2 gpm or lawn watering at low rates.  Will this provide sufficient cooling if there is no flow inducer at the pump?  Do most pumps have a thermal protection shutoff if they start to overheat?

One additional question on the installation of the CSV1W.  Your information sheet talks about not having the pressure gauge and switch on the main supply line however most pressure tank tees that you buy have the 1/2" ports for these on the straight run from the supply intake to the downstream exit of the tee (not on the 12" pipe to the pressure tank) which is how my system is currently set up.  The install would be much easier If I could leave this as is and then I only need to plumb the CSV in at where the PVC feed line ends and the tank tee starts.

Thanks in advance for your help!

[Cary Austin]

Sorry for the delayed response.  Couldn't get off the phone yesterday.  Anyway, adding a CSV may stop the water hammer on pump shut off.  But if you have a check valve sticking, the CSV may not solve the problem.  When a check valve sticks in the open position, sometimes the water flow has to completely reverse to close that check.  This is what causes the loud "thump" you hear after the pump shuts off.

The CSV would reduce the flow to 1 GPM to fill the pressure tank.  This would normally mean the check valve is only open the thickness of a piece of paper when the pump shuts off while only pumping 1 GPM.  Then the check valve would not slam on pump shut down.  If the system would have had a CSV from the beginning, the check valve would still be good, as it would not have slammed itself to destruction in the first place.

But since the check valve is now sticking open, it may still stay stuck open when the pump is only producing 1 GPM, until the pump stops and the flow reverses.  It is about a 50/50 chance of the CSV solving this problem after the check valve has been damaged.  You may still have to replace the check valve.

The pump/motor cooling is just the opposite of what  you are thinking.  The pump/motor pulls less amperage and runs cooler at 1 GPM than at higher flow rates.  With a CSV you could run 1 GPM from a hose, and never shut it off for the rest of your life, and the pump would run cooler and last longer than if pumping higher flow rates.

I don't know if the CSV will stop the water hammer from your damaged check valve, but it would certainly help, and would keep that from happening again as well as making the pump, tank, and everything else last longer as well.

[John64]

Hello Cary,

Thanks for the feedback.  Actually replying the next day is very fast in my opinion since I'm sure you are busy.  So just a couple more quick questions based on your response.

If my check valve is sticking then installing the CSV may result in the valve not completely closing because I will no longer have the high velocity of the water column reversing to slam it shut like i do now on pump shut off (soft stop with CSV I suppose)?  The telltale sign of this then would possibly be a slow loss of pressure in the pressure tank if the valve is indeed leaking?

Your comment on the pump drawing less amps at low flow is a little confusing to me.  I'm a mechanical engineer so maybe just not enough electrical experience but at low flow (CSV almost closed and keeping pressure at pre-set pressure) isn't the pump pushing less water at higher pressure and therefore working harder?  More work per a given period of time means more power and therefore more heat generated?  Does the pump actually slow down (less rpm's) with higher pressure and less flow as opposed to full flow at lower pressure?

Thanks again.

[Cary Austin]

If my check valve is sticking then installing the CSV may result in the valve not completely closing because I will no longer have the high velocity of the water column reversing to slam it shut like i do now on pump shut off (soft stop with CSV I suppose)?  The telltale sign of this then would possibly be a slow loss of pressure in the pressure tank if the valve is indeed leaking?

You are correct, it will not have the high velocity in the reverse direction that is now slamming the check valve.  However, that is the problem you are trying to solve.  But even though it won’t have the high velocity, the CSV cannot completely close, so you will have the max pressure going back to close the check valve, which is usually all you need.

[Cary Austin]

Your comment on the pump drawing less amps at low flow is a little confusing to me.  I'm a mechanical engineer so maybe just not enough electrical experience but at low flow (CSV almost closed and keeping pressure at pre-set pressure) isn't the pump pushing less water at higher pressure and therefore working harder?  More work per a given period of time means more power and therefore more heat generated?  Does the pump actually slow down (less rpm's) with higher pressure and less flow as opposed to full flow at lower pressure?

That would be true if this was a positive displacement pump like a piston type.  However, centrifugal pumps are counter-intuitive.  The horsepower is directly related to the flow rate.  So increasing the head pressure reduces the flow rate and the power consumption.  Many pumps will draw 50% less energy when restricted with a valve to low flow.  Some pumps do this better than others, as it is the impeller design that makes the difference.

Everybody thinks that way though.  I thought that way myself many years ago, until I started studying this natural characteristic of centrifugal impellers.  Restricting a centrifugal pump with a valve actually makes the pumps work easier, not harder.  This counter-intuitive aspect of centrifugal pumps makes CSV’s a hard sell.  Even many pump engineers do not know this, mostly because they just never tried restricting the pump flow to see what happens.

[John64]

Thanks for the explanation, I'm going to order one and give it a try.  May take a while before I can install it but I'll post back on the results once it's up and running. 

[Cary Austin]

Great.  We would love to hear back from you.

[John64]

So I finally got around to installing a CSV1W on my system.  I'm happy to say that my water hammer is almost entirely gone, but only if I set the valve so the pump runs for about 3 minutes to fill the tank after it turns on (40/60 pressure switch) instead of the 2 minutes recommended in the installation directions.  I'm not sure why this is but it seems like it takes that long for the water flow to really slow down.  It doesn't seem to be a problem though.

I do have one question however.  The instructions say to set the tank pressure to 5 - 10 psi below the cut-in pressure for an above ground installation.  I left mine at 2 psi below and everything seems to be working fine.  Why do you recommend setting the pressure so much lower and should I do this even if I have no problems at the current 38 psi?

Thanks again and I highly recommend your product!

[Cary Austin]

I still think you have a check valve sticking in the open position.  Otherwise the water hammer would be completely gone on pump stop.

2 PSI air pressure below cut-in is actually perfect.  We recommend 5 PSI less because most air pressure gauges and water pressure gauges don't read the same as the next gauge.  5 PSI lower just gives more room for pressure gauge error. 

Also an increase in temperature will increase the air pressure in the tank.  So if you air it up to 38 PSI in the morning, later on when the day warms up it could show 42 PSI.  We just use 5-10 PSI less air to be on the safe side.

Also the exact air pressure in the tank is not as important when you have a CSV, as we are only filling the tank at 1 GPM instead of full pump flow, so the draw down of the tank is not as critical.