"The roof tank overflowed again last night." Or: "There is no water on the top floors and the caretaker is not answering." In many apartment buildings, someone has to remember to switch the pump on and off. When they forget, water runs down the walls, or residents wake up to empty taps. Worse, if the underground reservoir is empty and the pump keeps running, it can run dry and burn out.

An automatic level controller takes that job out of human memory.

How automatic control works

Most buildings have an underground or ground-level reservoir, filled by the city supply or a deep tubewell, and a roof tank filled by a pump. The controller watches the water level in both:

  • When the roof tank drops to a low level, the pump starts, but only if the reservoir below has enough water.
  • When the roof tank is full, the pump stops.
  • If the reservoir runs low while the pump is running, the pump stops so it does not run dry.

Level sensing can be done with float switches, which are simple and mechanical; with non-contact sensors mounted above the water; or with sensors that measure pressure at the bottom of the tank. Each has trade-offs. Floats can stick, non-contact sensors can be confused by condensation or a narrow tank, and submerged sensors can collect sediment. The right choice depends on your tanks and your water.

Alerts worth having

A controller that only switches the pump is already useful. One that can also message the caretaker or the building committee is more useful. Typical alerts:

  • the reservoir has been empty for a long time, so the city supply may not have come
  • the pump has run much longer than usual, which can mean a leak, a stuck valve or a pump that is no longer lifting water
  • the roof tank is still low after the pump ran
  • the controller has lost power or its connection

Steps to set it up

  1. Map your system: tank sizes, pump type, where the starter or switch is, and who operates it today.
  2. Decide the levels at which the pump should start and stop, together with your plumber.
  3. Choose a sensor type for each tank.
  4. Have a qualified electrician install the controller alongside the pump's existing starter and protection devices. Keep a clearly marked manual switch so the pump can always be run by hand.
  5. Decide who gets alerts and how: app, SMS or both.
  6. Test it by draining and filling, including what happens when power goes and comes back.

Load-shedding, generators and weak signal

After a power cut, the controller should check the water levels before it restarts the pump, not simply switch it back on. If the building can run the pump on a generator, agree with the committee whether pumping should continue during outages. Mobile signal is often weak near basement reservoirs and pump rooms, so check coverage before relying on SMS from there, or place the communication part where the signal is better. A small battery backup lets the controller report that power has gone.

Mistakes and trade-offs

  • Sensors in turbulent water, such as right under the inlet pipe, give jumpy readings and make the pump start and stop too often.
  • Removing the manual switch "because it is automatic now". Always keep a way to run the pump by hand.
  • Forgetting the pump itself. Automation protects against dry running, but the pump, starter and wiring still need regular maintenance by a technician.

Where to start

Begin with one building and one roof tank. Run the controller for a few weeks while the caretaker still checks by eye, and let it take over once everyone trusts it. If you manage several buildings, a shared dashboard for all of them can come later. See IoT device development for how this kind of system is built, or get in touch to plan a first installation.

Related service IoT device development Connected devices from prototype to dashboard: sensors, firmware, connectivity and alerts that reach your phone.