Pump House on a Hill - Design

[Osage Hills Craig]

My 2-story house sits on top of a hill (roughly 100' elevation from main). I've lived there 18 years and the pump house that my builder installed has worked flawlessly with one short-coming. We get our water from a water main (no wells in our area, due to rock). He put a pump house about 60' (elevation) up the hill. The pump pressurizes the water to about 105psi, which delivers about 55psi at the top (house). The pump architecture is a typical bladder tank, with a pressure sensing solenoid controlling the pump.

Like I said, this system works flawlessly with 1 exception. There is no sensor/control if the pump is not receiving water from the main. In the case of a main break, or any source water interruption, we have no way of knowing that the pump is running continuously and not able to satisfy the pressure sensor.

The pump is housed in an insulated pumphouse about 200' from the house, 40' down the hill, in an area that isn't very easy to get to. I do have a breaker in the house, to kill all power to the pumphouse; but knowing to do it, is the issue.

Is there some type of control that I can install on the input side of the pump that can sense non-availability of water and shut the electricity off.

 

[Reach4]

Yes. You could have a pressure switch.

The Pumptrol 9013FRG series has contacts that operates the opposite of the common pressure switch. For example, the factory setting of 9013FRG2J23 would cut the power if the pressure fell to 20 PSI and turn it back up when the pressure rose to 40. You could change the setpoints.

You might need a small pressure tank, on the theory that whenever you have a pressure switch, you need a pressure tank. I don't know if that would be needed in this case.

 

[Sponsor]

 

[Storm rider]

I think you could also accomplish this with a pump protector, such as Coyote or Franklin.

 

[Valveman]

I don't think a low pressure cut-off switch will work, as the static level above the switch will not drop low enough to shut the system down. I am also not sure if the Coyote or Pumtec is sensitive enough to know the water has stopped, without actually losing prime. However, a Cycle Sensor would let you see the amp draw, and could be set to shut the pump off at the amps that happen when you shut a valve to the incoming water.

 

[Reach4]

I don't think a low pressure cut-off switch will work, as the static level above the switch will not drop low enough to shut the system down.

I think the pressure at the input to the pump will drop a lot while the pump is running and the city is not providing pressure. Then I would hope the check valve would prevent water from running backwards through the turned-off pump.

I guess there might be a need for a small pressure tank at the switch. I don't know.

 

[Valveman]

Pump producing 105 PSI to get 55 PSI to the house means 50 PSI static. An adjustable low pressure cut-off switch set to 60 PSI might work, but I wouldn't trust it. Low pressure coming from the city may mean the pump cannot reach the pressure switch shut off, but would be too high for a low pressure switch to be used as a safety.

 

[Reach4]

Low pressure coming from the city may mean the pump cannot reach the pressure switch shut off, but would be too high for a low pressure switch to be used as a safety.

This pressure switch would be before the input of the pump.

 

[Valveman]

Low pressure cut-off switches on the suction side of the pump don't work very well. Every time the pump starts the pressure drops low enough to shut the pump off. You have to use a time delay with a switch like that. You can either use a regular electronic time delay, or make a mechanical one using a small bladder tank with an orifice in the inlet. Still I think amperage is the best way to protect the pump without having nuisance trips.

 

[LLigetfa]

As reach4 said, the switch he proposes is not a regular pressure switch with low pressure cutoff, it is a reverse logic switch that would go on the intake side of the pump. My only concern is whether on pump start there could be some bounce to it.

The Pumptrol 9013FRG series has contacts that operates the opposite of the common pressure switch.

 

[Valveman]

Some people call those a "lose of prime" or a "reverse acting" pressure switch. I also use an Allen Bradley 836 pressure switch, which has a normally open contact and can be adjusted to any pressure needed. The problem is when the pump comes on it will pull the suction pressure down to zero or less for a second before the water from the city starts coming to replace the lose. There is basically no way to start a pump that will not cause a low pressure cut-off switch on a suction line to open up.

 

[LLigetfa]

There is basically no way to start a pump that will not cause a low pressure cut-off switch on a suction line to open up.

The following is starting to border on the wacky, but a pressure tank could be used to buffer/dampen the bounce. Just locate the switch on the tank Tee and run a very small line to the pump inlet side. When the pump drops the inlet to zero or less, the pressure tank will slowly drain through the small line allowing time for the city flow to recover. This of course would only work if the city flow can sustain pressure on the pump inlet side.

If unsure, the Cycle Sensor may be the best remedy. It may however require some tweaking if the city pressure fluctuates a lot.

 

[Valveman]

You can either use a regular electronic time delay, or make a mechanical one using a small bladder tank with an orifice in the inlet.

Yeah that is the kind of mechanical time delay I was talking about. But you really need to have an orifice on the outlet only, so the city pressure can refill the tank as quickly as possible. Otherwise refilling the tank through the small line of orifice takes too long and the pump still goes off.

 

[LLigetfa]

Ja, I could steal an idea from a famous valve {wink} and cut a small notch in the seat of a check valve. That would let the tank refill quickly yet slow down the draining. The orifice would also be self clearing to reduce the chance of clogging.

 

[Reach4]

Yeah that is the kind of mechanical time delay I was talking about. But you really need to have an orifice on the outlet only, so the city pressure can refill the tank as quickly as possible. Otherwise refilling the tank through the small line of orifice takes too long and the pump still goes off.

I don't see that as a problem. If the system trips, turning off the pump, I don't see a problem with having a delay (20 seconds seems about right) when water pressure returns. You would not want the high setpoint on the pressure switch to be too close to then normal water pressure. But if we presume normal water is 60 PSI, low trip is at 20 PSI, and high trip is at 4o PSI, I think that should work fine. Setting the low/high to 10/30 might be even better if the switch can support that.

 

[Valveman]

Oh you can make the little tank work like a mechanical timer. I have done it several times. But if you don't get it right or the city pressure fluctuates, it can cause the pump to bounce on/off rapidly. But I do think it is the best way to prevent putting a vacuum on the city lines.

 

[Osage Hills Craig]

I've enjoyed all the dialog. Let me clarify one thing. My pressure switch is 80-100 psi; so there's lots of gap between the switch and anything being discussed. I lose about 40 psi between the pump house and the house, so I'm delivering ~55 psi to the house. My house is 40' higher than the pump, and 2 story.

I would also add that we don't see any bounce due to the pump cutting in. I don't know if this is due to the 40' of elevation and the line is 2". There's significant water in the line above the tank. I know, I've drained it.

 

[Reach4]

I would also add that we don't see any bounce due to the pump cutting in.

You would need a gauge on the incoming water to see the potential bounce we are discussing.

 

[Osage Hills Craig]

You would need a gauge on the incoming water to see the potential bounce we are discussing.

I plan on putting a gauge on the input side.

 

[Jadnashua]

Is this a hypothetical question, or does it happen to you? At least where I live, it's a very rare occasion when the utility fails to provide water. I've lived here 30 years, and it's never happened, nor has it at my mother's house where I grew up (and she's still there).

In theory, a water cooled pump could get ruined if it ran dry. If I were worried about it, I'd probably use one of the sensors that monitored the pump's run-time current, and shut it off if it sensed it was running dry.

 

[Osage Hills Craig]

Is this a hypothetical question, or does it happen to you? At least where I live, it's a very rare occasion when the utility fails to provide water. I've lived here 30 years, and it's never happened, nor has it at my mother's house where I grew up (and she's still there).

In theory, a water cooled pump could get ruined if it ran dry. If I were worried about it, I'd probably use one of the sensors that monitored the pump's run-time current, and shut it off if it sensed it was running dry.

It does happen, although not in the last couple of years. We get our water via a rural COOP, which buys it from a city municipality. A water main break, at the wrong place, can give us a disruption of service. It doesn't happen often, but the issue is I don't have a great way of sensing it. It boils down to noticing that you're not getting any water, and then going and turning the pump breaker off. I've never burnt a pump up, but I've went down to the pump house and it is extremely hot. On the flip-side, once I've thrown the breaker, there's no real good way to find out that service has been restored. We've been here 18 years, and it's probably happened a half dozen times. I'm getting ready to replace the pump, due to age, so I'm looking into how to close this short-coming.