Hello

In my country house, I have a drinking water system setup with the following:
- A water tank at a spring location (Lower water tank), water pump (lower water pump) and a float and contactor to prevent the pump from sucking in air
- The lower water pump pumps water from the Lower to the Upper water tank ~400 m away ( ~1300 ft)
- Upper water tank has two floats, one preventing the Upper pump from sucking in air, and a second float that when the water level drops in the Upper tank, it turns the lower pump on by sending power through 400 m of underground cable.

We have been occasionally getting some rough winters with temps dropping to -20 degrees Celcius / -4 Fahrenheit. Introducing a heating element in the currently insulated housing has solved the problem, but can only be done with a lot of manual fiddling, as the lower pump area only has power when the float is dropped in the upper tank ( water level drops in upper tank). So today we hike through the snow, Install the heating element/unplug the lower pump, and manually lift the Upper tank float to trick it into sending power to the lower pump.
I attached a simplified drawing to explain the situation.

I was hoping to get some ideas of possible solutions of running the heating element without laying a separate 400 m of electric cable to sustain a constant feed for electricity. The current location of the lower pump are not in line of sight, but in a deep valley compared to the house/upper tank, so RF/Wireless relay options might be a bit difficult?

I am happy to hear any ideas!

Sam

 

Welcome to AAC!

We have been occasionally getting some rough winters with temps dropping to -20 degrees Celcius / -4 Fahrenheit. Introducing a heating element in the currently insulated housing has solved the problem, but can only be done with a lot of manual fiddling, as the lower pump area only has power when the float is dropped in the upper tank ( water level drops in upper tank). So today we hike through the snow, Install the heating element/unplug the lower pump, and manually lift the Upper tank float to trick it into sending power to the lower pump.

1. What are the consequences of running the lower pump when the upper tank is full?
2. What are the consequences of having the heater on the lower tank while the lower pump is running?
3. How far from the house is the upper tank?
4. Are the contacts in the upper "full" float exposed?

 

Tricky little problem.

How much margin is built into the system? What is the max time you can wait between when the upper system calls for the lower system to pump, and when the lower system actually starts pumping? Depending on this, a possible solution is to modify the lower system controls so it runs the heater for (example) 15 minutes before pumping. The heater stays on while pumping, and the whole thing resets when the float trips or the upper system cuts power. Basically, just a 15-minute delay timer between when power appears and when the pump contactor is closed.

If you want to run the heater continuously, or if the minimum required heater time before pumping is longer than your max tolerable delay, then a carrier-current system such as X-10 could send a signal down to an appliance module that drives the lower pump contactor. In this way, AC power and the heater are on continually. With an insulated tank, the additional 24-hour power just to maintain 35 degrees might not be all that much. Or, put the heater on a thermostat. The next layer is another X-10 module for the heater. With that, you have independent control of both devices.

Note: Not sure if X-10 is reliable at this distance. If not, then maybe something like Zigbee. There are people around here with way more experience about this, but there are multiple approaches once AC is available continuously at the lower station.

I built a system for lighting in my basement where I flick the light switch briefly to cause multiple fluorescent lamp banks to go from one bulb to two, to four, but this type of control might be too much wear and tear on the contactor.

ak

 

You have to get "Un-Switched" Power to the Lower-Tank in any case.

My solution would be to not switch the Power for the Lower-Pump at the Upper-Tank-Float-Contactor,
this would mean that only Coil-Power for the Lower-Pump-Contactor would have to be
switched at the Upper-Tank-Float.
This would also mean that You could use 2 - ~16-Gauge Wires to inhibit the Lower-Pump.

And, if You control the Lower-Contactor with a small Solid-State-Relay,
much smaller Gauge Wires could be run from the House to control the SSR,
as a SSR only requires around ~20ma, at ~5-Volts to operate.
Telephone-Wire with a small Wall-Wart-Power-Supply of 12-Volts-DC would then be perfect.

This should be set up in a manner that disables the Thermostatically-Controlled Heater-Element
when the Lower-Pump is commanded to run.
( does your Lower-Pump-Contactor have provision for "Auxiliary-Contacts" ???? )
This will prevent overloading the long run of high-Current-Wiring to the Lower-Pump by
allowing only the Pump OR the Heater to run, at any given time.
You will not want both to be running at the same time, under any circumstances,
as this will create additional un-wanted Voltage-Drop in the very-long wiring run,
which could possibly burn-out the Lower-Pump-Motor.
.
.
.

 

Getting wireless control reliably to the lower water tank seems problematic.
There are power line transceivers such as this that might work, but it doesn't state their operational distance, and would likely require some electronics expertise to get it to work.

I think powering the lower tank continuously and running small control (telephone) wires to that tank to control solid-state relays for the pump and possibly the heater, as suggested in post #4, is likely the best way.
Does that sound feasible to you?

Is the pipe outlet to the upper tank at the top or bottom of the tank?
I suspect it's at the top but if so, you could add an extension to the bottom of the tank (if it's not deeper than 30 feet).
That would add another possibility to provide continuous power to the lower tank, and use a pressure switch, such as this, to control the lower pump.
It could measure the pressure in the pipe to the upper tank, and turn on the lower pump when the pressure indicates the upper tank needs water.
What's the vertical distance between the two tanks and the depth of the top tank?

 

A Pressure-Switch is a genius idea, but it would have to have fairly high-resolution,
as the Elevation from the Bottom-Pump to the Top-Tank is substantially more than
the depth of the Top-Tank, ( ""Deep-Valley"", how ever many feet that may be ).

And, the pressure will "spike" wildly when the Pump turns-On, and turns-Off,
not to mention, "Water-Hammer-Effects" caused by the extremely long Pipe-Run.

Water-Hammer-Effects will virtually REQUIRE a large "Spike-Arrester" at the Lower-Pump-Outlet,
and possibly a second large Spike-Arrester around half way between the Lower-Pump and the House,
to prevent splitting-pipes, and blowing-apart various Pipe-Fittings / Joints.

This may be the perfect application for an automatic Ram-Pump, AKA, a Hydraulic-Pump.
.
.

 

Water-hammer only occurs if the flow is abruptly stopped at the output (top tank), due to the inertia of the water in the pipe..
Here the stoppage occurs at the pipe input (pump output), so there is no water-hammer effect.

Even if the flow was abruptly stopped at the pump end, the maximum negative pressure generated would be from forming a vacuum at the water vapor pressure of about 24mm of mercury at 25°C.

 

False .........

When the energy gets to the end of an open Pipe,
it will be reflected back and forth, up and down the Pipe,
until all of the Energy is absorbed by friction,
and it can be a tremendous amount of Energy.

Example .......
A ~2-inch Pipe, maybe ~4-feet long, contains somewhere around ~8-POUNDS of water.

A ~1300- foot long Pipe, ( as used in the above post ),
divided by ~4 equals roughly ~325-Pounds of Water.

Lets say that this ~325-Pounds is moving at a leisurely ~10-miles per hour,
and is abruptly stopped by your foot.
What do You think is going to happen to your foot ???

Water-Hammer is called "Hammer" for a very good reason,
it can easily break Iron-Pipes under the right circumstances,
and PVC-Pipes are even less of a challenge to it.

That's why all Plumbing-Codes require "Spike-Arresters".

The Pipe being open on one end is inconsequential,
there's still a Check-Valve at the bottom-end of the run.

When that Check-Valve slams shut,
the Pressure-Spike can easily be hundreds of PSI.
.
.
.

 

False not....
Water-hammer occurs only when the output is suddenly stopped, not the input, do you not understand the difference?
A check valve at the pump outlet will simply let the water keep flowing until all the water's kinetic energy has dissipated.
No hammer.

If water-hammer were a problem, than the TS would already have experienced that and remedied it, if necessary.
I see no reason to bring up that red-herring issue, which is not relative to the TS's problem here.

 

C'est la vie,

If You would do some research into the Ram-Pump referenced above,
You might change your mind.

I speak from experience.
.
.
.

 

I don't need a lecture on how water-hammer is caused or how a Ram-pump works, nor do I need to change my mind, since nothing I have said is incorrect, in contrast to your less than accurate pontificating about its cause.
I fully understand the causes and effects of water-hammer.

My Uncle had a ram operated water system on his farm and I determined how it worked as a kid.
It depended upon a large, spring-loaded flap-valve at the end of a long 5" pipe from a water spring, that would let water run through the pipe to the outside, and then suddenly close, which caused the water inertia (hammer) to drive the water through a check valve into a pressurized water tank.
It generated enough pressure to provide water to his house, which was a least 50' above the level of the tank.
In the house you could hear a faint "ping" from the pipes, every time the valve closed.

Experience means nothing if you don't draw the correct conclusions.
Yours hasn't, as shown by some of your comments about water-hammer.

 

If I understand the problem correcty, you have a 400 meter cable which carries the mains power from the upper tank to the lower tank which currently turns the pump on and off in the lower tank, with a lower tank float switch to stop the pump if the lower tank is empty. It may be the way you have drawn it, but I think the water pipe should go into the upper tank above the level of the water to avoid water hammer and water syphoning/draining from the top to the bottom.

So you need to communicate from the upper tank to the lower tank along this cable to select the various states required. I think you could do this by having the power to the lower tank normally on, with the heater controlled by a temperature sensor. This would turn on the lower pump as well. But if the upper tank is full, the power should turn off briefly and a low DC voltage should be connected to the line which would tell the pump to disconnect when the mains turns on again a few seconds later. When the upper tank level falls the mains supply to the lower tank would turn off briefly and a DC voltage with the opposite polarity would tell the pump to turn on when the mains is connected again.

A second "fail safe" float switch with alarm, should be installed in the upper tank, cutting the power, if for some reason the pump doesn't turn off.

I believe this low bandwidth communications protocol would manage the two states you need to control from the upper tank! I first thought of using DTMF (Dual Tone Multi Frequency for which there are ICs available, used for telephones) which would probably work over this distance but I don't think you need that level of complexity.

 

The phone company has been running DTMF for miles with reasonable success. (I am 1.8 miles from my CO.)

I agree that DTMF might be overkill for this application, but the idea of switching the existing wiring between power and signalling, rather than imposing the signalling on top of the power, is an idea that opens up many possibilities.

To the TS: Along a different track, two things -

1. What is the mains voltage and where are you located?

2. What is the heater power?

ak

 

The phone company has been running DTMF for miles with reasonable success.

DTMF does sound like a reasonably simple option, and I like that over trying to do some type of power on/off sequence signaling.
I see the on/off sequence as more complex to implement and probably need micros to easily do the logic.

DTMF is normally sent over a quiet, POTS current-mode connection but with sufficient signal amplitude and filtering at the receiving end, it should work over a noisy AC line, but coupling into the line would require some voltage protection circuits.
It only has to transmit a continuous state so that reduces the receiving problem.

 

If the mains were ac and I was an evil genius, I might propose a wild idea like:

Installing a 1:1 transformer at the pump and superposing a DC current into the line for the heater element.

But I'm not a genius...but sometimes I am evil.

 

That is an old balanced audio trick, and would work very well here. I think the technique has a name, but I can't recall it now. More questions for the TS:

3. What is the lower pump power / current?

4. Is the wiring down to the lower pump 2-wire (Line-Neutral) or 3-wire (Line-Neutral-Earth (domestic 120 V) or Line-Line-Earth (domestic 240 V))?

Depending on the power levels involved, a pair of center-tapped isolation transformers might be affordable.

ak

 

DTMF is normally sent over a quiet, POTS current-mode connection but with sufficient signal amplitude and filtering at the receiving end, it should work over a noisy AC line, but coupling into the line would require some voltage protection circuits.

One of the reasons various carrier-current systems work is that the carrier frequency is far removed from the current (mains) frequency. In round numbers, DTMF is 100x closer to the mains freq than something like a wireless intercom, X-10, etc. The power line impedance is way lower at 1 kHz than at 100 kHz. I think the lower DTMF frequencies would fare much better over the distance than X-10 or whatever, if you can pump in enough energy.

ak

 

Oh wow tons of discussions and ideas. Thanks for all the replies !

@dl324
-the consequences of running the lower pump when the upper is full is a very wet shed, this is prevented today with a float.
-the consequences of having the heater on the lower pump while the pump is running is that it does not run enough to generate enough heat, typically pump will run for 30 seconds or so.
- House is ~ 40 m (130 ft) away from upper tank.
- not sure what you mean by if the contacts of the upper float exposed, one float in the upper tank is wired the opposite of the other float so it detects when its empty instead of when its full.

@AnalogKid

There is not much margin built in the system, but a delay module sending power to the heater first to heat for ex: 30 mins before pumping can start can work but might also need manual adjusting to when things get hotter or colder.

@LowQCab

Running a small gauge wire would be a great solution combined with a solid-state relay, but the last time I tried running a second cable along part of the 400m underground cables, it just would work, a day of pushing and pulling cables wasted.

I am not sure if the Contactor has Aux contacts. Will have to check when I’m back at the Cabin.

@crutschow

The Power line transceiver idea seems genius! I thought about getting ethernet down to the pump via PLC but seems it is not reliable over 100 m distance. But the device in the link you shared seem to work with distances ~ 1 Km to 1.5 Km. It seems this could be an option. For this to work, it would need to have one device at the upper tank communicating with another device in the lower tank? Could the PLC be used to trigger a relay or more at the lower tank area to control multiple devices?

The pipe outlet is at the top, the drawing is just an oversimplification…

Currently the second float (when it drops) signals the lower pump to pump when the upper tank needs water.

@LowQCab

Upper and Lower tank have about 60 m / 200 ft of hydraulic head difference and are connected with a PE 1” pipe. There are 1 way check valves at the bottom and at the top of the system.

@Jerry-Hat-Trick

-Water pipe does go at the top of the water tank.
-When you mention a low DC voltage connected to tell the pump to turn off and on. Do you mean using the same power cables and communicating with a Relay at the lower pump that can control?

@AnalogKid

-Mains voltage is 230 V @ 50 Hz (Europe/ Bulgaria)
-Heater power is ~100 W

@ElectricSpidey

The mains are AC single phase, I’m not sure I follow you idea though…

@AnalogKid

The pump is rated ~ 12 Amps
The wiring down to the pump is by a Line/Neutral (2AWG) Aluminum stranded cables

 

What are you heating with the heater?

 

There is not much margin built in the system, but a delay module sending power to the heater first to heat for ex: 30 mins before pumping can start can work but might also need manual adjusting to when things get hotter or colder.

Control the heater with a simple thermostat. Now the heater might run for only 1 minute or all 30 minutes, but only as needed. Set the delay for the coldest conditions, and let the thermostat save power and prevent over-heating.

The nice thing about this is that there is no change to any of the upper station controls or the long-distance wiring. One thermostat for the lower heater, and one delay module for the lower contactor. That's about as low-tech as you can get.

ak