Water level monitoring: simplest way

Question

I like to develop a water level monitoring for my cistern tank: empty level, low level, mid level, high level.

One of my booster pumps broke because it runs even when there is no water in the cistern tank. The cistern tank is filled via a float valve from the elevated raw water.

One of my option is some previous answer related to water level monitoring:

Is there a risk of 220 V AC flowing in the water, and how do I avoid that? I don't know microcontrollers or programming, so I choose hard wiring and an indicator light.

A second option is to use floatless relay like the ANLY A61F-G3.

But how to have an indicator light at the low limit (E4) and high limit (E1) separately? The indicator light option is available only for low (E3) and high (E2). How to have water level monitoring regardless if you have a motor or not the simplest way? Example: I want my pump to stop working anything below E4 and start pumping when it reaches E4. Is that possible with the floatless relay?

Note: it is not about the monitoring of the filling by the cistern pump, but preventing my booster pump from working if there is no water in the cistern tank.

Answer 1

Consider something simpler made from just switches.

From each switch lever, suspend a float on a fixed length thread/light chain. When the water is high enough, the float releases the switch. When water is too low, the weight of the float activates the switch. This kind of design has the advantage that all the electrical parts are above water; also it's easy to test: you push the switches manually and make sure all your signals work as desired. (I'm not entirely sure what exact functionality you want from your ladder drawing.)

If you use two pole switches, you can organise your indicators completely independently of your pump lockout. You can get switches which are okay for mains. Or if you want to keep them completely isolated, you could use a relay.

^{simulate this circuit – Schematic created using CircuitLab}

Alternatively, you could use magnets with reed relays, but I'd expect the microswitch with lever and float/weight is simplest.

Typical Switches and Floats

Frm ZF Switches

From Plastech

Answer 2

If the goal is to protect the pump, one of the submerged flow switches is enough.

If having mains voltage in a submerged wire is a problem, you can run the switch on 12V DC from a small power supply and use a 12V relay to control the pump.

If you want visual level indication... personally I'd go with analog.

On the well pump, I use a flow detection switch. Here's an explanation of how it works.

When water is needed, the pump turns on, and a timer starts. After about twenty seconds, if the flow switch does not detect flow, it automatically turns off the pump and it will remain off.

Unlike water level meters, this also saves the pump when the inlet filter is clogged...

These aliexpress paddle switches can also be used as a rugged version of the microswitch to build jonathanjo's solution, or as a limit switch for the above "analog" solution. You don't have to screw them into a tee on a pipe. You can also use it as a switch by replacing the paddle with a float or having something that goes up and down with water level bump into it. They're pretty robust, enclosed, have 15A contacts, and the pipe connection is obviously watertight.

Answer 3

If your main issue is protecting the pump, perhaps investing in one with its own dedicated line-voltage float switch is the way to go. Pump protection should be a proven, self-contained and low-tech approach, which is designed to be safe with immersed line voltage wiring.

This frees you up to choose a level monitoring approach that is safe, cheap, and reliable, without needing to deal with line voltage.

With that out of the way, here's some ideas for monitoring level:

• Multiple optical prism sensors, in a dip tube. Simple, inexpensive, low voltage, no moving parts, could give direct drive to LEDs or some simple display. Arguably the cheapest approach - no microcontroller needed, and runs on safe 5V. Gives the same functionality as you propose using float switches.

• Capacitive tank sensor, used in the RV and boating trades for fuel, water and wastewater monitoring. Again, for your cistern, in a dip tube. No moving parts, but more expensive than optical sensors due to the electronics. These can be self-calibrated to indicate fullness. Complex electronics so tend to be standalone systems.

More here about optical and capacitive sensing: Is it possible to do water detection with a single electrode?

• Laser time-of-flight (TOF) sensor, measuring distance to water surface. Problematic due to water's transparency; may be improved by using a tethered float with a cat's eye retroreflector to make a proper return signal. Maybe use a weighted bobber in a tube with the cat's eye on top? TOF sensors are not too expensive, and are conveniently available as modules. For a display application they need a microcontroller. Appears to be a popular hacking project.

• Hydrostatic (pressure) sensor, suspended from the top with the sensor at the bottom. Popular choice for cisterns; easy to install and service. Lots of off-the-shelf products from modest to expensive. Again, for a display, needs a microcontroller. Shows up as an Arduino project.

Of all these, the hydrostatic approach seems to offer the most ready-to-go options for display, including wireless monitoring to your phone if you wish. All from a single sensor from no moving parts. How cool is that?

If you want to do your own controller, vertical tank math using a pressure sensor is straightforward: pressure is proportional to tank fill height (we'll see how below); volume is cross-section area x height.

How to compute tank fill height from pressure? We start with the physical properties of a water column:

• Pressure = height x density x g

so,

• Height = pressure / (density x g)

Where g is gravity acceleration (9.81m/s^2).

(see here)

Conveniently, we see that that 1m water height is 9.81 kPa in pressure. This gives us the linear relationship of 9.81 kPa/m. Thus, we compute metric tank volume, in m^3, as:

• volume (m^3) = area (m^2) * (pressure in kPa) / (9.81 kPa/m)

Simple, right?

This pressure method gives precise, real-time volume data. Do some post-processing and you can calculate dynamic usage, and even detect unusual activity like leaks.

You could also do this with the capacitive and laser sensor approaches. But at the risk of overselling it, as I said the pressure approach appears to have many more ready-made solutions for cisterns than those other methods.

Answer 4

Simple is to place a plastic tube (vertically) inside the tank and a long bobber inside tube. Right above tube place a contact/button so once the water level reach certain point the bobber touches the contact. Bellow this level the contact is open. All electric parts are outside the water so water can’t touch with electricity.

Answer 5

If pump location allows, a potential alternative is to sense the presence of water in the tube leading to the pump, on the pump input side.

You'd need to splice in a section of clear tube, or alternatively a short bypass to the main pipe. FEP polymer tube material ideal for this, but I don't think it's too picky.

The PLC logic may have to cope with occasional air bubbles or air pockets, from dissolved air vs temperature swings. It would be more to stop false alarms - your pump would be exposed to these regardless of sensing technology.

Example sensor, for the universe of PLC applications: https://www.digikey.com/en/products/detail/EX-F1-C5/1110-2624-ND/3895295

In-tube, non-contact, liquid presence sensors, of total-internal-reflection principle (images from Panasonic):

The tank-level sensing scheme pictured is even better, and can use multiple sensors on the same external-loop of clear tube. But would presumably not work be a good fit for an underground tank that's already buried.

The underlying technology is an LED, a phototransistor, and a light guide. So I'd expect an Alibaba version of it, but I've never looked.

Answer 6

The lower limit float switch (E4) should be active low and high limit float switch (E1) should be active high.

What does it mean? It means that when E4 switches off (assuming it switches off when it's above water level), it should turn on the low level indicator light and/or give a signal to stop the pump.

On the contrary, E1 should to turn on the high level indicator light and/or give a signal to start the pump when it switches on.

The main advantage of electronic is that we can build smart systems without sending 220V into the whole circuit through bulky and expensive components and therefore we avoid most of the hazardous situations. It's perfectly possible, and even easier, to use low voltage and limited current in a pump control system. But for some reasons, with pumps, 75 years old technologies are still widely used as if time had stopped in the 50's.

Instead of sending 220V to the float switches, you can build a circuit based on 12VDC using the same float switches (albeit with some precautions) or, better, with devices designed for 12V (or similarly low voltage). Then all the risks are definitely away. 220V should only be at the relay actuating the pump and at the pump itself. (Note that 12VDC and 24VDC pumps also exist, but their output is very slow.) Everything else could and should be on safe, very low voltage.

Turning on or off one of the two light when E4 is open or when E1 is closed, if it's just to be aware of the level of the water, is relatively easy. However, these signals are not enough to turn on and off the pump because the signal from E1 is not permanent. So the E1 signal should trigger the permanent powering of the pump and the E4 signal should trigger the permanent turning off of the pump.

Here is a simplified schematic with only two indicator LED's telling you when the water ir high enough to turn on the pump and when it's low enough to turn it off.

When E1 is open, LED1 is off because it doesn't receive current. LED1 turns on when E1 is closed. It's active high.

When E4 is closed, LED4 is off because all the current supposed to go to the LED is going to the negative pole of the power supply (GND or V-). This current is very small because it's limited by the resistor. LED4 turns on when E4 is open and the current goes to the LED instead of GND. It's active low.

This is good for small indicator LED's. For powerful lights, and for using signal outputs, a more complex circuit with transistor MOSFETs are necessary. But it gives you a first view of the concept.

^{simulate this circuit – Schematic created using CircuitLab}