Well runs out of water

[kondas]

I have a 3/4 HP Submersible pump that I installed in 1996. It has always worked very well. However, when we use a lot of water, say running it for about 15 minutes, water flow stops. The pump continues to run. For a long time I figured it was just pumping out water faster than the well was able to fill up. Recently, I discovered that when this problem occurs, if I shut off the power to the pump for about a minute & turn it back on, it starts to flow water again. What could be the problem?

[hj]

Probably what you originally suspected. The pump draws air and then, since it cannot pump air, it spins until you turn it off and the air can escape and water fills the pump.

[speedbump]

You would do good to either use less water or get some protection for the pump like the Subgard (http://www.pumpsandtanks.com/subgard.htm) or the Pumptec. (http://www.pumpsandtanks.com/motor_protection.htm)

Running dry is very damaging for a pump or motor.

bob...

[rshackleford]

A flow control would work well. You can get them in 1 gpm, 2 gpm, and so on. They use a rubber diaphragm to let just that amount of water through.

[Bob NH]

I don't like to put a flow control on a submersible because of the steep pressure/flow characteristic of submersibles. Some of them will put out a lot more head than you want in the up-pipe. You could put the flow control at the outlet of the tank but that would also reduce the pressure available to the user.

In most applications the user would just let the pump run longer and still run the well down to the pump. And they will be wasting electricity because of the head loss in the flow control and the fact that the pump is not operating at its most efficient operating point.

If the pump is running the well down in 15 minutes as the user said, then it will probably run it down in 30 minutes while the lawn is being watered.

I think the Pumptec is the best solution. You could check on that famous auction site to get some idea of the price to compare with local suppliers. Last I saw they were $128 including shipping.

Another possibility is that the water table has receded so much that it is below the head capability of the pump, though that is unlikely with a properly selected submersible that has been set at the proper depth.

If I owned the system I would want to know the model of the pump, the flow/head characteristic, the depth of the well and the depth to the aquifer, and the level that the pump is set at.

It might be possible to get more water if the pump is set at a lower level. You would probably have to ask the driller/installer if you don't already have that information. If the well can be pumped to a lower level then there is more head available for the aquifer to deliver water to the well.

The make/model of the pump should be on a sticker somewhere. In 1996 most submersibles were installed with control boxes which may contain the information. Can you post that information if you find it?

[rshackleford]

Water wells typically have a linear type drawdown. This is not an exact statement, but for all intents and purposes this is true. For example, take a well that has a static of 50’, a total depth of 150’, and the pump is set at 100’. Let’s also say that the driller performed a pumping test on the well and determined that the well would pump 10 gpm at a drawdown of 100’. Do the division and you would get a specific capacity of 1 gpm per 10’ of drawdown. This specific capacity is basically a linear function through the depth of the well. So, you could say I want 5 gpm, this would mean 50’ of drawdown.

Okay, on to my point. I stated above that the example pump is set at 100’. If I am trying to pump this well at 10 gpm, I can probably get that. I might get 10 gpm for 5 minutes, 15 minutes, or even 15 hours, but eventually the 10 gpm pumping rate is going to lower the water level in this well below the pump and I will run out of water at the surface. For this well, if I want to pump continuously and not run out of water I can only pump at 5 gpm or 50’ of drawdown. According to the drillers test pump results I can pump at this rate forever.

Alright, how does this apply to your well? You can continue to let your pump run out of water and only get 15 minutes worth of water or you can let it pump at 5 gpm and pump water forever. For example sake lets say it takes 15 minutes for the well to recover. In on hour you would pump 10 gpm for 30 minutes or 300 gallons. If you pumped 5 gpm for 60 minutes you would also get 300 gpm.

So what is the advantage of 5 gpm? The toughest thing on a motor is starting and stopping. This is the case for any electric motor. When a motor starts it builds up a lot of heat and works on destroying the insulation. On the 10 gpm pump you will start the motor 48 times in a day and you will only start the 5 gpm motor once in a day.

There is also a well development method called "rawhiding." This involves pumping water from the well until the well draws way down. Then the pump is stopped until the water returns. Then the pump is turned on again and this cycle continues. This is called “rawhiding” and it encourages sand to come out of the formation and into the well water. On a well screened well this will not cause sand in the water, but on a poorly screened well or a well with very find sand you will get sand in the water. Also, some wells will pump sand at higher flow rates and not at lower flow rates. This is another advantage to slowing down a well.

The best protection you could actually get would be to install a flow control and Pumptech or whatever Goulds calls it in the lower horsepowers.

I do agree that you should keep some records of your well and pump information.

[speedbump]

Good description of drawdown Bob,

I think Shack is talking about a Cycle stop or Smart Tee, but your right, with the wrong pump this could be disasterous.

bob...

[rshackleford]

Nah, I was just talking about a good old fashion Dole flow control. Pump the well and monitor drawdown. Find a pumping rate that keeps the water level above the pump and choose a flow control in that range. We use these quite a bit for low flow wells. We are in an area where people are happy if they get 3 gpm and 10 gpm is a smokin’ well. The flow control will be a simple, economic, and well friendly solution.

[speedbump]

Is that the coupling with the insert that has a little tiny hole on the middle? If so, the first one I saw was when I was about 15 years old ( a few years back) and couldn't figure what the hell it was. Especially since it was installed on a perfectly good well. I think someone put it in accidentally.

I think I would rather have something with a little more adjustment available if needed like an old fashioned backpressure control from a jet pump maybe.

bob...

[Bob NH]

We all agree that the flow rate should be reduced. I think it is better to do it at the pump than with a flow control.

The problem with a flow control on a submersible is that it drives the pump to very high pressure where you are wasting energy by operating the pump off its best efficiency point and you are wasting energy at the flow control. For 3/4 HP Goulds 4" models that a homeowner might use:

7GS07 Min/Nominal/Max flow rates are 1.5/7/10 GPM and the pressure at 1.5 GPM is 173 psi and 169 psi at 3 GPM.

10GS07 Min/Nominal/Max flow rates are 3/10/16 GPM and the pressure at 3GPM is 130 psi.

13GS07 Min/Nominal/Max flow rates are 4/13/20 GPM and the pressure at 4 GPM is 93 psi.

18GS07 Min/Nominal/Max flow rates are 6/18/28 GPM and pressure at 6 GPM is 78 psi.

Pressures given are AT THE PUMP, not at the tank.

If you can't find the model number of the pump, and maybe if you can, you can get a rebuilt residential size water meter ($31, stk #62276) and two unions ($5 ea, stk #48816) from USA BlueBook (800-548-1234).

Find out what your pump rate is and how much water you use, and you can put a smaller pump head on your existing motor (If it's a Franklin motor). Depending on what you now have for pump and well capacity, you might go to the 5GS05, 7GS05, or 10GS05.

Depending on your well capacity (which you can determine with the water meter), you may want the Pumptec that is discussed in previous posts. You can probably size the pump so you don't need it. If you match the pump to the well capacity you can save the cost of the Pumptec which would be a good bit of the price of a new pump head.

While you are replacing the pump head you may want to put on a longer down pipe if the well suits (For reason, see first paragraph of Post #6 by rshackleford). Ask your driller if he is still around.

If you lower the pump, you need to account for the required head from a greater drawdown. That is usually not a problem with a submersible because you will get about 30% more head at the minimum flow than you will at the nominal flow.

[rshackleford]

Bob,

Do you know of any 4” submersible curves that also show pump efficiency? It would be interesting to compare actually difference in HP for your example of flow versus pressure.

Also, do you know how much energy is actually used by a submersible pump? I have always considered it inconsequential, but it would be interesting to actually see some per month costs of running a submersible pump. I often hear people that live in town say that it must be nice not to have to pay a monthly water bill, but I disagree. I think if you added up all the costs of a well most people would realize their monthly water bill is okay.

[speedbump]

I'm not sure what Bob you mean Shack, but submersible pumps don't use much energy and for my money and my experience, I'll take the well any day over the water bill. I am sure it verys from one location to another, but for the most part the water bill will far out weigh the cost's of a normal water well system including maintenance and repair.

If you want to see the effeciency curves for submersibles, look at a distributors book. Instead of the graphs they show homeowners, the curves are far easier to use and more accurate. They also show an efficiency slot down the middle of the curve. If you like, I'll scan one and post it here.

bob...

[Bob NH]

The smallest submersible for which I have efficiency curves is the Goulds 33GS. I also have curves for up to the 80GS and for the Dempster Submaster Y and V series which are rated at 45 and 60 GPM.

For the 33GS30 the maximum efficiency is 68% at 32 GPM as near as I can pick it off the curve. It is producing 2.13 water horsepower and is using 3.13 shaft horsepower.

At 10 GPM, which is 31% of the flow at peak efficiency, the pump is operating at 40% efficiency, producing 0.97 water horsepower while using 2.43 shaft horsepower.

The efficiency curve is pretty flat near the design operating point of the pump so the pump operating without restriction tends to be self-adjusting for variations in pressure at the tank.

But the efficiency numbers by themselves are deceiving. If the lower flow is achieved with a flow control valve, much of the energy is wasted. The flow-controlled pump will take 3.2 times as long to pump the same water while using 2.43/3.13=0.776=77.6% as much shaft power. Therefore, the total energy to deliver the same amount of water to the tank is 0.776x3.2=2.48 times as much energy.

So throttling the pump requires 2.5 times as much energy to get the same result. The energy ratio goes to 3.8 if the flow of the same pump is throttled to 6 GPM, or about 20% of the design flow. That corresponds to operating a 10 GPM pump at 2 GPM.

Besides using more power, the pump is being operated 3.2 times as long at 1.5 times the head. And for every gallon delivered, the impellers are forced to work against another 2 gallons of internal bypassing that can't be discharged because of the flow restriction. That has to diminish the life of the pump.

[speedbump]

Then there is the debate on whether it is better to backpressure a sub and keep it running or cycle it and take out the motor.

The best thing is to have all the well stats to start with, pick out the best pump for the job, install some safety equipment just in case (most homeowners won't spend the money) then go away feeling pretty good about your install. Too bad this isn't a perfect world. God knows I've tried to sell the protection many times and the home owner just knows it couldn't happen to them. Just like the teenager in his fast new car.

And I think we have lost Kondas.

bob...

[rshackleford]

I would definitely be interested to seem some efficiency on the smaller 4” pumps if they exist. I have sold a lot of 5, 7, and 10 G’s and GS’s but I don’t really know how efficient they are.

I don’t disagree about finding the right pump. A designer should absolutely pick a pump that matches the well discharge.

Sometimes, though, a well might be produce less than 5 gpm and that is when a choke might be used given certain head conditions. The other case where a choke is good is on shallow stock water well. The well is only going to be used a couple months out of the year the 10 gallon series pump is the cheapest available.

My calculations tell me that operating a ½ hp pump at $0.07 per kW would only cost $0.026 per hour. This isn’t much.

[speedbump]

I attached a curve sheet for the Myers Rustler. You will see the green efficiency slot down the middle of the curve.

Tell me how you arrived at the cost per hour for operating a pump. I have never known exactly how to do that.

[rshackleford]

Thanks for a look at the curves.

That cost per hour is very crude.

0.75 kW = 1 hp

0.75 kW/hp x 0.5 hp x 1 hr x $0.07/(kW * hr) = $0.02625

Go through all of the units. Horsepower goes away (divide out). Kilowatts go away. Hours go away. All you are left with is dollars. This is a little crude I known. I have also been told that most of the pump manufactures push the limits of the sub motors, so a ½ hp might actually be pulling more than its name plate value. Also, I guess on the power cost based on some three phase work I have been doing where I am at.

[speedbump]

I guess what I didn't understand is the kWh or kw hours. Does this mean the electric company charges the rate for one hour at the rate they charge. Or in better (I hope words) If the electric company's charge per kilowatt hour is Z. Do you have to run a pump motor for one hour at X watts per hour times the value of Z to obtain cost.

Sorry the images weren't better Shack, but the image size restrictions got me and I had to shrink it a bunch. It looked real good before I started shrinking it after the scan.

bob...

[Bob NH]

A pump operating at 60% efficiency (a pretty good submersible) with a motor operating at 75% efficiency, has a "wire-to-water" efficiency of 0.6x0.75=0.45, or 45%. That is a reasonable estimate for a pump operating near peak efficiency with no throttling.

If you pump 1000 gallons of water to 200 feet of head, the water energy is 1000 Gal x 8.34#/gal x 200 ft = 1,668,000 ft-#.

Now 1 kWHr = 2,650,000 ft-#.

So to pump 1000 gallons of water to 200 ft of head with a wire-to-water efficiency of 45% requires:

(1,668,000 ft-#)/(0.45 efficiency x 2,650,000 ft-#/kWHr) = 1.4 kWHr per 1000 gallons pumped to 200 ft of head.

As I showed in a previous post, throttling the pump can require 2.5 to 3.8 times as much power, so you could easily require 4 kWHr/1000 gallons if you throttle the pump. That is probably a dime a day to a home owner who might use 250 gallons per day, but it's a lot more if you are irrigating alfalfa.

[rshackleford]

When you buy electricity from the power company you are buying energy not power. Power (hp or kW) have units like length*mass/time. When you are consuming power you are doing it for a certain amount of time. Therefore the meter reads energy or length*mass.

Hp measures the same thing as kilowatts. Power.

Joules, Btu, calories, and kilowatt*hours are all measure the same thing. Energy.

So any way to get the amount of power usage you would:

(Motor power) x (length of time of usage) x (energy cost from rec) = cost of operating motor

This does not include things like minimum meter charges and demand charges, but it makes the general point.

BobNH...keep in mind that a motor can only use as much power as its nameplate indicates. I agree with you completely that there is no replacement for proper pump design. Proper pump design is sometimes outweighed by practical and economic issues, though.

[speedbump]

I'm glad Bob NH finally put the cost of a dime part at the end. Now I understand all that math better. I know there is not a lot of cost involved in operating a 3/4hp sub for a normal residence. And also that irrigating alfalfa could be quite costly. Here the fish farmers are the one's that cry about the cost of operating circulating pumps.

Can anyone put together an average cost of lets say a 3/4hp motor giving the average homeowner's household needs minus sprinkling the grass in normal america with normal water levels and normal electric power costs per day, week, month? I think this would be a great selling feature for homeowners looking for the difference in operating a well over buying water from the government. I realize repair costs would have to be estimated as well.

bob...

[Bob NH]

"Can anyone put together an average cost of lets say a 3/4hp motor giving the average homeowner's household needs minus sprinkling the grass in normal america with normal water levels and normal electric power costs per day, week, month? I think this would be a great selling feature for homeowners looking for the difference in operating a well over buying water from the government. "

That is why I calculated the kWHr per 1000 gallons at 200 ft of head. That is a reasonable head for a well where the water rises to within 60 ft of the surface and you are trying to get 50 psi in the tank, with the pump operating near its nominal flow rate (not throttled).

If the pump you need is operating at 300 ft of head at the nominal flow rate, multiply the 1.4 kWHr/1000 gallons by 300/200=1.5, so you get 2.1 kWHr per 1000 gallons.

It makes absolutely no difference what the horsepower of the motor is if you have the correct motor and the correct pump. You are working with total water pumped and the bigger pump uses more power and produces more water.

A typical homeowner may use about 100,000 gallons per year, and where I live the cost of municipal water is about $3.00 per 1000 gallons and we are paying about $0.12/kWHr for electricity. Therefore the electrical cost per 1000 gallons would be 1.4x0.12=$0.168, so the municipal water costs about 18 times as much as your pumping power. The one costs $300 per year and the other costs $16.80.

It takes 6233 gallons to put an inch of water on 10,000 square feet, or 27,150 per acre-inch. For the rates given above it will cost about $1.05 of pumping power to put an inch of water on your 10,000 sq ft lawn, or about $4.50 for an inch on an acre of your alfalfa. Everyone can scale electicity rate and usage numbers to fit their situation.

The above doesn't apply if you are operating a jet pump or are throttling your submersible.

Horsepower doesn't matter if you do the math based on how much water is being pumped.

[rshackleford]

If you strictly have irrigation pump, horsepower is less of a consideration. You start the system up and it runs and runs and runs at a consistent flow rate. In this case it is easy to calculate power or energy, its just physics.

However, in a household there will always be throttling. You take a shower and as you deplete the storage reservoir the pump will turn on. In my house the pump might continue to run while I shower. In your house it might cycle on and off several times during the shower. In both cases the pumps are moving all over the pump curve and therefore the efficiency is also all over the place. Also to play the devil’s advocate the water level in the well is changing. Again my well might drawdown 300’ and yours only 2’; this would also affect the efficiency and energy use. It is for this reason that I took the pump horsepower and length of time operated approach.

BobNH, I agree that the method you purposed would provide a good estimate of energy use and I guess all we are looking for is a good estimate. Perhaps the only good way to get a better estimate on power usage for household water systems would be to perform a study on a group of homes. The sounds like an excellent academic study and it would be useful to the consumer.

[speedbump]

I'm sorry Bob Nh, I don't follow your math, but I do like the part of the 18 times more for the city water than the electric to run the pump.That's what I was after and the fact that it would cost on average $300.00 per year for normal city water usage and $16.80 for the well pump.

I have customers here that have city water and are spending more than $500.00 per month just to use in the home and water the lawn. (These folks may be a little fanatical about their grass but still, that's a little ridiculous.)

Thanks,

bob...

[speedbump]

Ok Shack,

Does this mean your volunteering? I like your idea.

bob...