Well Pump Expansion Tank

[ponytug]

I just had our pump and controller replaced two weeks ago. The last VFD controller and pump was replaced on 2009/10 (can’t remember). It was a one phase power unit; the new one is three phase, so I’m hoping at least to see lower power usage. But my 2hp pump is at 660’ so it works pretty hard. We did install 4 check valves in the pipe at equal distances along the pipe, so it stays charged between demand.

Our first pump went 35 years. Just curious, are you on three phase delta?

I saw the old pump and the new pump, both 2hp. And the flow of the new pump is very good. Maybe the check valves are the answer? It’s a 3 phase Franklin submersible and it’s putting out 9-10 gpm at the house. The well is 680’ total and the pump is at the bottom of 33 20’ pipe sticks.

Edit: I do irrigate all summer and run lawn sprinklers, but one circuit at a time.

Your numbers are spot on our numbers.

The check valves save energy as your pump doesn't have to refill the pipe, and help prevent a backwash that could stir up sediment in the well. They also help relieve the load on the check valve at the pump, keeping the pressure difference across any given check valve to a lower value. The only downside I can see would be an increase in moving parts, and therefore wear/maintenance over time.

@WinterDeere yes, I agree HP is the ability to do work (e.g. gpm x lift), but as a pump faces higher heads, the efficiency of any given centrifugal pump impeller drops off, so one needs to add impellers to the stack to generate the given head efficiently. It is easy to design in. For any given depth, there will be range of the optimal number of impellers for a given HP and desired flow. I would have a look at Franklin or Grundfos pages for specifics, if you are curious.

@WinterDeere I suspect that your prior issues with your well may have had other solutions besides pulling the check valve at the top of the well due to water hammer, besides fixing the leaking valves in the well. Water coming up several hundred feet of pipe is going to have momentum and if there is any is any air in the line, that's going to be a heck of a water hammer that's going to be hard on any parts of the plumbing. Ouch.

All the best,

Peter

 

[jyoutz]

Our first pump went 35 years. Just curious, are you on three phase delta?

Your numbers are spot on our numbers.

The check valves save energy as your pump doesn't have to refill the pipe, and help prevent a backwash that could stir up sediment in the well. They also help relieve the load on the check valve at the pump, keeping the pressure difference across any given check valve to a lower value. The only downside I can see would be an increase in moving parts, and therefore wear/maintenance over time.

@WinterDeere yes, I agree HP is the ability to do work (e.g. gpm x lift), but as a pump faces higher heads, the efficiency of any given centrifugal pump impeller drops off, so one needs to add impellers to the stack to generate the given head efficiently. It is easy to design in. For any given depth, there will be range of the optimal number of impellers for a given HP and desired flow. I would have a look at Franklin or Grundfos pages for specifics, if you are curious.

@WinterDeere I suspect that your prior issues with your well may have had other solutions besides pulling the check valve at the top of the well due to water hammer, besides fixing the leaking valves in the well. Water coming up several hundred feet of pipe is going to have momentum and if there is any is any air in the line, that's going to be a heck of a water hammer that's going to be hard on any parts of the plumbing. Ouch.

All the best,

Peter

I know that the new pump and controller are 3 phase power, and the one that was replaced was single phase. I don’t know what 3 phase delta is.

 

[jyoutz]

Maybe the Oakville table I referenced is not typical. It was just the first I found.


Check valves are probably a necessary evil at your depth, but they definitely don't help to compensate for pump capacity. You get a small pressure drop across each check valve, equivalent to the spring pressure divided by bore cross-section, probably just 0.5 PSI. Their function is to avoid system bleed-down, with larger vertical drops (higher pressure at pump).


Yep. But he has a 10 gpm pump. Remember, HP is just an ability to do work, that work being the product of both pressure and flow. For a given size motor, you can only achieve higher flow by transitioning to an impeller design that will cost you in pressure.

Choosing the right submersible pump for your water well.

Some water wells can be very deep at over 800’ or more while others are shallow at 50’ or even less. Water demands vary from just a few gallons per minute to hundreds of gallons per minute. In this overview we are going to explore the factors affecting the selection of the correct submersible well

oakvillepump.com

I ran into this problem in my last house. What happens is that all check valves have a small leak-down rate. But if one farther down has a higher leak-down rate than one up higher in the system, the result is it pulls a relative vacuum on the line. The pressure switch up top doesn't see this, since it's isolated from the line by a check valve up top. But when the pump kicks on, it's slamming fresh water up into this vacuum, and "bang!"

The system where I observed this had only one check valve up top, just upstream of the check valve, as the whole system was only 150 - 200 feet deep. It made a hell of a racket, every time the pump kicked on. I had a well specialist come take a look at it, as my knowledge is just general plumbing (we didn't have wells where I grew up), and it took him about 8 seconds to look at the system and tell me to remove that check valve up top... not necessary. After that, everything behaved well, but I wouldn't know what to do with a 600 ft. deep well.

I’m certainly no well expert. I am just stating how the well company set up our system and it works well. I’ve never had an air hammer problem with the water. We do run sediment and carbon filter canisters where the service enters the house, but not where the line diverts to the barn and irrigation hydrants. The line from the well is 1.25”.

 

[ponytug]

I know that the new pump and controller are 3 phase power, and the one that was replaced was single phase. I don’t know what 3 phase delta is.

Three-phase electric power - Wikipedia

en.wikipedia.org

There are two fairly common kinds of three phase power in the US; "WYE", and "Delta", the latter coming in several variants. Wye is 208/120V, meaning phase to phase is 208V, while phase to neutral is 120V. Delta is 240V phase to phase and 120V phase to neutral for two of the phases. There are also higher voltage analogs. Three phase works much better at supplying power to motors.

VFDs pretty readily make 3 phase from standard 240V, and can be a much cheaper solution than wiring three phase in many areas. (We pay close to 1k/yr for the privilege of wired 3 phase. Yes, it is probably on borrowed time here, as the meter fees are incredible and getting worse.)

All the best,

Peter

 

[Pettrix]

Well drilled in 2016
Production = 25gpm
Depth = 508 feet
Pump Depth = 485 feet
Static Water Level = 205 feet

Motor: Goulds VFD 3 Phase 3HP - 230 Volts – Range: 13GPM – 22GPM (max)

Drop Pipe: 1 ¼” Schedule 120 PVC with Stainless Steel Couplers -
3 Stainless Steel Check Valves @ Pump Head, 160 feet and 320 feet

So far so good on my well setup. No issues, good water pressure and good production.

 

[steve tym]

What leads you to believe he has a 10 gpm pump? Is it a Franklin 3200 series, or 4400 series?
The check valve manufacturers, the motor manufacturers and the pump manufacturers recommend a check valve every 200', or every 100 psi (actually 231').

 

[jyoutz]

What leads you to believe he has a 10 gpm pump? Is it a Franklin 3200 series, or 4400 series?
The check valve manufacturers, the motor manufacturers and the pump manufacturers recommend a check valve every 200', or every 100 psi (actually 231').

So I have 4 check valves for 660’. A little overkill.

 

[steve tym]

Most pump manufactures build a check valve on top of the pump, but that's not a great place for it, it should be at least a foot above the pump, but that's not very practicle most of the time. One every 200'. If that's 4, then it's 4.

 

[Valveman]

I have learned in my many years of working with pumps, that recommendations made by the pump companies are for the benefit of the pump manufacture, which comes at the expense of the homeowner. What you can see on the pump curve is really the only facts you will get from the pump manufacturers. Product quality, run time, tank size, number of check valves, VFD's, and other things are recommended to maintain planned obsolescence.

There has always been a problem of a vacuum between multiple check. It is just a matter of time as to when a small problem with one check valve will cause the vacuum that makes the water hammer when the pump starts. Multiple check valves are just a Band-Aid for the problem caused at pump shut off. When a pump is producing full flow filling a tank or whatever when it is told to shut off, even 10 GPM flow will cause a 1" check valve to be in the wide-open position. The water in the drop pipe, be it 50', or 700' has to reverse flow for about 1" before the check valve closes. That much water slamming down against the check valve when it closes causes tremendous water hammer and damages the check valve. The best answer we had in the past was to install multiple check valves to split the water hammer load and backup each other when they fail. But using multiple check valves to prevent water hammer on pump stop, causes water hammer on pump start. Water hammer on pump start is worse than on pump stop as the check valves are open when the water hammer on pump start happens. The resulting water hammer is then free to travel all the way down and shatter the thrust bearing in the motor. It is also best to start a pump against high pressure as with a closed or almost closed valve. Multiple check valves cause a pump to start with very little load or even against a vacuum, which is not good for the pump.

Having only one check valve at the pump is the only way to stop water hammer on pump start and start the pump against full pressure. The problems caused by the check valve slamming shut on pump stop at the bottom of a 100' or 700' string can be eliminated with a soft stop of the pump. When using a Cycle Stop Valve the tank is only filling at 1 GPM when the pump shuts off. This means the check valve is only open the thickness of a piece of paper, and there is no water hammer when it closes, no matter if the pump is set 10' deep or a 1000' deep. I have installed single check valves on pumps that were set up to 2200' deep. You just need a check valve that is rated for the pressure and a soft stop to keep it from slamming. The pump will start much easier when there is only one check valve at the pump and no others in the system. A properly set VFD will also have a soft stop. However, VFD systems require special check valves with stems that keep them from spinning in response to the way flow comes out of a pump with reduced speed.

No house would have three phase power of any kind. The VFD just makes three phase out of standard single-phase current at a house. But it doesn't do a very good job of it. Making three phase by puling DC power to simulate AC power causes a 230V motor to operate on voltage spikes over 1000V. There are also problems with resonance vibrations at different speeds. Then there are the multitude of different types of VFD controllers, and they all handle low flow and on/off situations differently. VFD's are not easier on motors and do not save energy. Anyone who has gotten 10 years or so from a VFD is on the high end of the spectrum. VFD's are designed to make lots of money for the pump manufacturers by falling in line with planned obsolescence guidelines, making pumps and controls last and average of 5-7 years.

Many standard pumps, with extra-large tanks or other controls like the CSV, (not recommended by the factory) have lasted 30-50 years so far. I have one still working that I installed in 1982. You will never see a VFD system last anywhere near that long. I know, I know! "Look at all the fancy adds and marketing of VFD's by the manufacturers". Everybody that has a VFD wants to think they made the right decision. But eventually I hear most people say it was the worst decision they ever made.

 

[steve tym]

I have learned in my many years of working with pumps, that recommendations made by the pump companies are for the benefit of the pump manufacture, which comes at the expense of the homeowner. What you can see on the pump curve is really the only facts you will get from the pump manufacturers. Product quality, run time, tank size, number of check valves, VFD's, and other things are recommended to maintain planned obsolescence.

There has always been a problem of a vacuum between multiple check. It is just a matter of time as to when a small problem with one check valve will cause the vacuum that makes the water hammer when the pump starts. Multiple check valves are just a Band-Aid for the problem caused at pump shut off. When a pump is producing full flow filling a tank or whatever when it is told to shut off, even 10 GPM flow will cause a 1" check valve to be in the wide-open position. The water in the drop pipe, be it 50', or 700' has to reverse flow for about 1" before the check valve closes. That much water slamming down against the check valve when it closes causes tremendous water hammer and damages the check valve. The best answer we had in the past was to install multiple check valves to split the water hammer load and backup each other when they fail. But using multiple check valves to prevent water hammer on pump stop, causes water hammer on pump start. Water hammer on pump start is worse than on pump stop as the check valves are open when the water hammer on pump start happens. The resulting water hammer is then free to travel all the way down and shatter the thrust bearing in the motor. It is also best to start a pump against high pressure as with a closed or almost closed valve. Multiple check valves cause a pump to start with very little load or even against a vacuum, which is not good for the pump.

Having only one check valve at the pump is the only way to stop water hammer on pump start and start the pump against full pressure. The problems caused by the check valve slamming shut on pump stop at the bottom of a 100' or 700' string can be eliminated with a soft stop of the pump. When using a Cycle Stop Valve the tank is only filling at 1 GPM when the pump shuts off. This means the check valve is only open the thickness of a piece of paper, and there is no water hammer when it closes, no matter if the pump is set 10' deep or a 1000' deep. I have installed single check valves on pumps that were set up to 2200' deep. You just need a check valve that is rated for the pressure and a soft stop to keep it from slamming. The pump will start much easier when there is only one check valve at the pump and no others in the system. A properly set VFD will also have a soft stop. However, VFD systems require special check valves with stems that keep them from spinning in response to the way flow comes out of a pump with reduced speed.

No house would have three phase power of any kind. The VFD just makes three phase out of standard single-phase current at a house. But it doesn't do a very good job of it. Making three phase by puling DC power to simulate AC power causes a 230V motor to operate on voltage spikes over 1000V. There are also problems with resonance vibrations at different speeds. Then there are the multitude of different types of VFD controllers, and they all handle low flow and on/off situations differently. VFD's are not easier on motors and do not save energy. Anyone who has gotten 10 years or so from a VFD is on the high end of the spectrum. VFD's are designed to make lots of money for the pump manufacturers by falling in line with planned obsolescence guidelines, making pumps and controls last and average of 5-7 years.

Many standard pumps, with extra-large tanks or other controls like the CSV, (not recommended by the factory) have lasted 30-50 years so far. I have one still working that I installed in 1982. You will never see a VFD system last anywhere near that long. I know, I know! "Look at all the fancy adds and marketing of VFD's by the manufacturers". Everybody that has a VFD wants to think they made the right decision. But eventually I hear most people say it was the worst decision they ever made.

So, 946 pounds of pressure on a single check valve??
You and I don't work in the same world.
Most check valves are designed to be operated int he full open position, not partially open. VFD check valves being the exception, but who uses a VFD check valve with a Cycle stop?

 

[WinterDeere]

So, 946 pounds of pressure on a single check valve??

Is that math for the OP? I thought water ran about 0.43 per foot, so more like 300 psi for the 700 ft. rise, plus whatever pressure he's running in the house. Let's call it 340 - 360 psi.

 

[steve tym]

Is that math for the OP? I thought water ran about 0.43 per foot, so more like 300 psi for the 700 ft. rise, plus whatever pressure he's running in the house. Let's call it 340 - 360 psi.

I was referring to valveman at 2,200 foot setting. You are correct, .43 pounds per foot.

 

[Valveman]

So, 946 pounds of pressure on a single check valve??
You and I don't work in the same world.
Most check valves are designed to be operated int he full open position, not partially open. VFD check valves being the exception, but who uses a VFD check valve with a Cycle stop?

Yes, and 946 PSI is nothing for an oil field type check valve made for 3K PSI or so. Also, a pump set at 2200' needs to see 946 PSI on start up. Any extra check valves up the line would defeat the back pressure needed to keep the pump from upthrusting on start up.

You are correct a VFD check valve is not needed with a CSV. The varied flow rates from a CSV do not cause a spinning action at the pump like a VFD. Also, many VFD's like the sub/mono drive with the switch are hard on check valves as the check opens and closes every time the subdrive switch makes and breaks. This can cause a check valve to open and close 45 times a minute. For these reasons a better check valve is needed with a VFD system than a CSV system.

 

[Valveman]

Is that math for the OP? I thought water ran about 0.43 per foot, so more like 300 psi for the 700 ft. rise, plus whatever pressure he's running in the house. Let's call it 340 - 360 psi.

Yes, and if the pump was made for 700' it needs to see that 300 PSI on start up. Just need a single check valve on the pump that is rated for 300+ PSI.

 

[geteh]

That's a normal way to deliver them, but doesn't imply they shouldn't be adjusted. 40 - 60 PSI is a very common well pressure switch setting, and it's easier for a plumber to relieve a few PSI during installation, after checking with a gauge, than to haul a compressor or pump down to the basement to charge the thing. And if you're one still running 30 - 50 PSI, also easy enough to let 12 psi out thru the valve.

This doesn't apply to the OP, if he's running a constant-pressure rig, but for everyone else, this is important:

Having your tank bladder over-charged can result in both short cycling AND massive pressure losses before cut-in. Round numbers for demonstration, let's say you have a 40 - 60 PSI well pump switch and a tank bladder pre-charged to 50 PSI. Turn on the system, and the well will charge everything up to 60 PSI cut-out, before turning off. Then you use some water, and get nice linear pressure down to 50 PSI, at which point pressure in your plumbing almost immediately plummets toward zero, as the bladder is completely expanded and there's no water left to push out of the expansion tank. If you're taking a shower, you'll notice great pressure, then a total loss of water, then it comes rushing back with a bang Ants for sale UK. Not good.

Having your tank bladder under-charged doesn't cause this problem, hence the recommendation to run 3 psi below cut-in. However, going too low also reduces useable tank volume, what they call the "draw-down" capacity, between pump re-starts. This is because there is less air space remaining in the tank, when the bladder is under-charged. So, even though there's water left in the tank, there's no bladder left to push that volume.

Setting the bladder right at cut-in would maximize the draw-down capacity, but run the risk of pressure loss right at cut-in. Hence the advice to pre-charge to 3 psi below cut-in.

For OP: I've never rigged a constant-pressure pump, they weren't a thing back when my family owned a plumbing business. But I would strongly suspect that one would want to set the bladder pressure (empty tank) right at your constant pressure setting, or just a few PSI below it, for the same reasons: maximizing draw-down while preventing pressure cycling.

I ended up replacing an expansion tank on my water heater and at the same time I did the same thing on my plumbing near my pressure switch. I had thought the tank was an expansion tank. (it was a 4.8 gallon tank) I had a plumber came over and he mentioned that he thought there was suppose to be a pressure tank here instead. Is there a big difference between the 2? My well system is also a constant pressure system. I've been having problems when my sprinkler comes on my pressure gauge goes to 0 and only goes up half back to the original pressure and I can hear the bladder in the expansion tank move and the ball in the tank slams hard and makes a big thug noise every time the zones change. Could this be caused by the wrong tank installed? i drained the system and checked the pressure inside the tank. it is what its suppose to be at?

 

[steve tym]

If it's not going to the preset pressure when the sprinklers are running, the pump can't keep up for one reason or another.

 

[WinterDeere]

@steve tym already nailed it. The fact that you're never returning to at least min switch pressure indicates pump is unable to deliver what's being asked of it. There could be three reasons for that:

1. Most likely there's a damaged line or soaker ring in your irrigation system. I've punctured full 1" lines with my plug aerator, and very few residential well pumps would ever keep up with a 1" leak. You'd think this would be obvious, but if the thing is running at 4am and you're never out in the lawn before 9am, then it may not be obvious.

2. Pump is failing or at end of life, and no longer delivering what was prescribed.

3. Pump is working properly, but just undersized for scale of irrigation system.

To debug, I'd do something like the following:

1. Ensure pressure tank is adjusted properly, by shutting off water, draining system down to zero pressure, and then checking pressure on air bladder. It should be 2-3 psi below pump cut-in pressure.
2. Charge up system, bleed all air out, then check draw-down volume of pressure tank. Since most are only 10-20 gallons, you can easily do this with a few 5-gallon buckets. First, run a sink until pump kicks on, then shut off sink. Once system is fully charged and pump turns back off, start filling 5-gallon buckets until pump kicks back on. Total water dispensed into 5 gallon buckets between pump cut-out and pump cut-in is your "draw down" volume.
3. Go on internet, look up spec's for your expansion tank, and make sure listed draw down is close to what you measured, for the given pressure range. This will tell you if your pressure tank is operating properly.
4. Run irrigation system, the offending zone, and time the draw-down. If you already know the gallons, and you measure the minutes, now you know the GPM draw of the irrigation system.
5. If draw is obscene (like > 10 GPM), then start looking for a damaged or leaking line in the system. If none is found, consider splitting offending zone into multiple zones.
6. If draw is reasonable (e.g. < 6 GPM), then try to compare to your pump spec's, if available. If not available, make a guess as to pump used, based on line size and system pressure, based on your favorite pump manufacturer's design tables.

Of course, the 6 and 10 GPM benchmarks I give are for a typical residential pump setup. Commercial or farm equipment may be a whole other class, so scale the numbers to your situation, the same theory / method applies.

 

[Valveman]

I ended up replacing an expansion tank on my water heater and at the same time I did the same thing on my plumbing near my pressure switch. I had thought the tank was an expansion tank. (it was a 4.8 gallon tank) I had a plumber came over and he mentioned that he thought there was suppose to be a pressure tank here instead. Is there a big difference between the 2? My well system is also a constant pressure system. I've been having problems when my sprinkler comes on my pressure gauge goes to 0 and only goes up half back to the original pressure and I can hear the bladder in the expansion tank move and the ball in the tank slams hard and makes a big thug noise every time the zones change. Could this be caused by the wrong tank installed? i drained the system and checked the pressure inside the tank. it is what its suppose to be at?

Years ago I would see expansion tanks with a 3/4 outlet and a tiny hole inside for the water path. Any expansion tanks I have seen lately have a 3/4 outlet and a normal size opening for the water flow into the tank. If there is no tiny hole for the water flow, there is no difference between an expansion tank and a pressure tank. Just the small ones are labeled as expansion tanks as they are too small to be used as a regular pressure tank. However, with the new constant pressure systems those small tanks are as large as needed, and many pressure tanks are just labeled as expansion tanks.

Even if you got one with the tiny inlet hole it would not cause the problem you describe. As was said, using more than the pump or well can supply will cause the pressure to go low and stay low when irrigating. A break in the line is usually only on one zone.

More than likely it is a problem with the constant pressure pump, as they are not designed to be reliable, long lasting, or inexpensive. Very likely a problem with the variable speed controller or VFD is causing the pump to not get up to full speed. Amp and flow test as mentioned would determine if the problem is pump, well, or VFD controller.

BTW, the thump noise you hear happens when the system pressure drops below the air charge pressure in the tank and the diaphragm hits the bottom of the tank. Pressure should never get low enough for the diaphragm to hit the bottom.

 

[geteh]

Years ago I would see expansion tanks with a 3/4 outlet and a tiny hole inside for the water path. Any expansion tanks I have seen lately have a 3/4 outlet and a normal size opening for the water flow into the tank. If there is no tiny hole for the water flow, there is no difference between an expansion tank and a pressure tank. Just the small ones are labeled as expansion tanks as they are too small to be used as a regular pressure tank. However, with the new constant pressure systems those small tanks are as large as needed, and many pressure tanks are just labeled as expansion tanks.

Even if you got one with the tiny inlet hole it would not cause the problem you describe. As was said, using more than the pump or well can supply will cause the pressure to go low and stay low when irrigating. A break in the line is usually only on one zone.

More than likely it is a problem with the constant pressure pump, as they are not designed to be reliable, long lasting, or inexpensive. Very likely a problem with the variable speed controller or VFD is causing the pump to not get up to full speed. Amp and flow test as mentioned would determine if the problem is pump, well, or VFD controller.

BTW, the thump noise you hear happens when the system pressure drops below the air charge pressure in the tank and the diaphragm hits the bottom of the tank and **** unblocked. Pressure should never get low enough for the diaphragm to hit the bottom.

thank you so much for your suggestion

 

[Johnmash2]

Yes, and 946 PSI is nothing for an oil field type check valve made for 3K PSI or so. Also, a pump set at 2200' needs to see 946 PSI on start up. Any extra check valves up the line would defeat the back pressure needed to keep the pump from upthrusting on start up.

You are correct a VFD check valve is not needed with a CSV. The varied flow rates from a CSV do not cause a spinning action at the pump like a VFD. Also, many VFD's like the sub/mono drive with the switch are hard on check valves as the check opens and closes every time the subdrive switch makes and breaks. This can cause a check valve to open and close 45 times a minute. For these reasons a better check valve is needed with a VFD system than a CSV system.
If you’re into word games while waiting for the pump to cycle, check out:
Word List | Wordle Answer

You’re spot on — a well pump with the right check valve setup can easily handle that kind of pressure, and avoiding extra valves helps maintain the necessary back pressure to protect the pump from upthrust. With a CSV, you don’t get the same rapid open/close cycling that can wear out check valves in VFD systems.