Wanted: Simple and Inexpensive Automated Pump Control System Solution

I am currently pumping water up to a tank that is about 700 meters from the pump shed. The control systems is manual. When I want to fill the tank, I go to the pump shed and flip on the power switch to the 220v, 5 HP pump. When the tank is full (usually overflowing by the time I notice), I go to the pump shed and flip off the power switch. There is no power at the tank at this time. Could someone suggest a simple and inexpensive means of establishing an automated pump control system that would not require power at the tank? If this is not feasible, how about a small solar photovoltaic powered unit?

I am a civil engineer, so don't expect me to understand too much electrical control jargon, but I should be able to handle a fair amount of technical detail. If someone knows of a standard kit sold by a particular manufacturer, that information would be welcome.

Bob Peterson

Does the tank fill from the bottom? If so, a standard pressure switch might do it. Think about how the pump on the well at your house works.

Curt Wuollet

You should think about a pressure switch if the tank full level is above the pump.

If it's not you can cheat if the tank is closed by putting a standpipe on the outlet.

That way when it gets full the head pressure increases rapidly as the
level rises in the standpipe. The pressure switch can be at the pump. The pressures can be quite low so diaphragm type pressure switches would be in order.

Sometimes you need a plumber not an automation engineer ;)

A simple ballcock (or float arm) valve at the tank and a pump with a built-in pressure sensor should do the trick. When the water level drops the valve opens and reduces the pressure in the line from the pump to the tank. The pump detects the pressure drop and starts automatically. I'd go to my local Farm Supplies store for this kind of stuff.
The tank is enclosed, but open to the atmosphere. The tank fills from the top with an air gap. The pump discharge pressure is constant during operation, even when the tank gets full. So a pressure switch at the pump discharge would not be effective in sensing how full the tank is, unless I put in a float valve as one of you suggested to artificially force the pump to shutoff head. That might work if I can find or contrive a float valve that can be configured to my tank and plumbing situation.

I'm not sure I understand the comment about installing a standpipe.

Thanks for these responses. I was hoping there would be an elegant, yet simple electronic solution, but these comments have stimulated some thinking along a different line.

Curt Wuollet

The standpipe won't work with an open tank. but with a closed tank, a 20 foot standpipe would add 20 feet of head pressure before overflow, enough to sense back at the pump even if they are on the same level.

> M. Jensen... how do you "notice" the tank is overflowing?


The tank has an overflow outlet pipe. I visually "notice" that water is flowing from that outlet. Then I drive down the road to shut off the pump by hand.
DesertRancher… how about additional detail::

1) What solutions have you thrown out, such as running a 2-wire circuit, radio-control, etc? And, what was the reason, for example, cost or other limiting factor?

2) Is pipe above or below ground?

3) Is pipe metallic or non-metallic material?

4) Can you see tank form your pump-shed?

5) Is operation 24/7, regardless of the weather, i.e., rain, sleet, or snow?

6) Where are you located?

I have two possible solutions, but need to know the answers to the above.

Regards, Phil
I should add that this system has been successfully supplying our house from a bore some 2km away for the last 20 years. Like I said, you need a plumber :p
>DesertRancher… how about additionaldetail::

> 1) What solutions have you thrown out, such as running a 2-wire circuit,
> radio-control, etc? And, what was the reason, for example, cost or other limiting factor?

Answer: I have not thrown out any potential solutions that might be practical and economical compared to the inconvenience of continuing current practice. Not being a control systems engineer, I just hoped for some advice leading to the best (most practical and economical) solution.

> 2) Is pipe above or below ground?

Answer: It is below ground.

> 3) Is pipe metallic or non-metallic material?

Answer: It is non-metallic (PVC)

> 4) Can you see tank form your pump-shed?

Answer: Yes, if I recall correctly, there is line of sight between the pump shed and the tank, but only from the top of the tank. I don't know if it might help to explain that the tank is made of steel and is cylindrical in shape, about 32 feet tall.

> 5) Is operation 24/7, regardless of the weather, i.e., rain, sleet, or snow?

Answer: Some disruption to the control system because of weather would be acceptable because I need to pump only about 10 to 20 hours per week during the peak water demand season and can cope with short term disruptions in pumping ability. Snow is very unusual and melts almost immediately, sleet is also uncommon, heavy rain does happen a few times a year, but high winds are common.

> 6) Where are you located?

Answer: Desert southwestern USA.
DesertRancher... following are two suggestions based on the details you furnished:

1) Using Level or Over-flow Detection at the Tank.
In an earlier life I manufactured SIS systems. One of the components was called Smart-A-Larm. It monitors the status of field-located contacts. In your case the minimum wire-size for the 700mt distance between the pump and tank is # 28 AWG (~0.1 mmq.) I will send you one at no cost.

2) Using Motor kW Output.
If the tank is filled from the bottom, the motor’s Hp output will increase with level. Thus, by installing a Wattmeter or transducer, the level at shutoff can be determined!

If you want additional info contact me off-forum: (cepsicon[at]aol[dot]com)

I doubt I could find a ready-made valve and ballcock/float device that would exactly fit my inlet pipe and tank configuration, but I probably could custom fabricate one with a little effort. As you say, that just a plumbing project.

However, there is the little control systems issue of instructing the pump to start again at the proper time. I don't know about your system, but if I understand your suggestion correctly, the pressure switch at the pump would shut the pump off when the closed valve causes the pump discharge pressure to rise. But, once the pump shuts off, the pressure would eventually drop off either because (1) leaks in the system would allow a gradual pressure drop or (2) the water level in the tank drops as water from the tank is used and allows the float to drop and unseat the valve. Wouldn't the pressure drop just cause the pump to start again? And wouldn't this just result in a damaging cycling of the pump on and off as the float valve seats and unseats itself within a very narrow range of water levels?
You could keep it pumping using an off-delay timer to avoid excessive stop/starts. This idea could also be used to prevent the pump starting again with an on-delay timer and set both to 1 minute as an example.

Tom Chubb
I assume the on and off pressure settings are different. In our case, the pump also supplies several water troughs and several water tanks. Leaks are a common issue ( and are usually only found when the water comes to the surface when it isn't raining ) So our main problem is that the pump runs constantly when the leaks get too bad. Obviously there's a difference between 700m of pipe to one tank and 2 km of pipe to several.

However, if you don't want to or cant run a cable from a level sensor on the tank to a relay on the pump then a ball cock is likely the most reliable way to ensure you get enough back pressure to switch the pump off. Otherwise you might find yourself struggling to keep the pressure switch on the pump adjusted. I guess you don't have standard 25mm or 3/4 inch alkathene pipe either.

And of course - advice on the internet is often worth as much as you paid for it - but I hope this helps.

If the water demands were relatively steady, a timer might be a workable idea. Unfortunately, the demands change quite a bit from week to week and season to season. But, a timer in combination with some of the other suggestions might be good. Thanks for the idea.

Yes, I think it would be best if my on and off tank levels were different, unless some sort of timer could be introduced to the control circuit and setting up the proper timing could be a continuous headache. After you explained how your system works, I understand now why your setup works well with a simple ballcock.

I initially started by investigating the idea of running a cable from a level sensor on the tank to a relay on the pump, but I was deterred by the challenge of identifying exactly what devices would be best from among all that were available, how to configure and wire the devices together, and how to power the remote sensor all while keeping the cost reasonable. That's when I turned to this forum. Thanks for your insights and suggestions.

Tony Gunderman

If your pump has a good separation between pumping current and dead-head current, you may be able to get by with a low cost current switch (with NO and NC contacts), a timer (with two sets of contacts), and the ball cock valve suggested previously. I am assuming your water usage is not such that you have to constantly monitor it since you are doing it manually now. So, if you install a delay on energize timer that will close its timing contact when the pump has been off for say 30 minutes, that can be used as a re-start request for the pump. The power to the timer will need to have a latch circuit that holds it on until the current switch makes to get through the initial motor start. Once the current switch makes, it will cause the timer to drop out but it will maintain the run request on the pump starter as long as the pump is pumping and the current switch is satisfied. When the valve closes, the pump dead-heads, the current drops off, and the current switch will stop the pump. Your 30 minute restart wait delay would then begin timing again. When the delay has elapsed, the seqeuence starts again and tops off the tank. AutomationDirect and others have fairly inexpensive current switches and timers.

Of course, you would probably want to add a manual/auto switch, too.

You could also do this with a current switch and two timers to get the additional benefit of protecting your pump from continuously running dry on loss of prime (though it would still try to restart once every 30 minutes).

Tony Gunderman
Tony... amperage-measurement represents a poor control method, if supply voltage fluctuates.

The reason is a running motor is a constant kVA machine, i.e, if supply voltage goes up, current goe down, and vice-versa. For this situation, then, power-measurement should be used.

Regards, Phil Corso

Tony Gunderman


I agree with your assessment 100%. That is why I stated it would require good separation between pumping current and dead head current. If he does not have that, then a true power measurement would definitely be the better way to go. It will probably be more expensive, though.

We typically use instruments from Load Controls (PMP-25 or PMP-1701) in these situations where we want true power measurements to monitor a pump.

A check with a clamp on ammeter would give a pretty good indication whether or not current alone would work.

Tony Gunderman