Fluid level indicator

Thread Starter


Joined Jun 29, 2006
Recently I observed a stainless steel rod with a magnet encased in a stainless steel ball which moved up down the rod as the fluid level changed. The changes in the level was recorded via galvanometre.
I would like to construct such an item but not sure of the circuitry. Can someone point me in the right direction . Thanks.


Joined Dec 13, 2003
If it used a magnet in a float then it most likely used reed switches. I have worked on several level indicators like this.

Reed switches are only rated for low currents although some may be rated for 240V AC - of course with water level use, 12 volts or lower would be safest. They can handle operating an LED directly. A reed switch is a glass cased device a few centimetres long & a few millimetres in diameter. The most common type has a wire connection exiting each end. Inside is a fixed connection & a very thin wire "reed" that will move sideways when a magnet is applied. This moving reed is the switching operation & normally open or normally closed types are available so that they can be open circuit or short circuit when operated. Reed switches are available as just bare components or by obtaining reed switches used for burglar alarms.

Although rare, you can also get reed switches with a pair of contacts that the reed moved between - a change over or SPDT operation, and a kind that applying the magnet in one polarity causes it to latch closed & requiring the opposite polarity magnetic field to release it.

The float may be stainless steel, or plastic (like fishing float) or even made of glass. A lot of water level switches in "sterilising" units for medical / vet / barber use had a glass float with the magnet in.

The float may be on the outside of a tube containing the reed switches or the float inside a tube with another tube beside it containing the reed switches. Some are only a single switch with a pair of wires, but they can have several switches with one common side & an individual wire for each switch. Another form has a series string of reed switches with a resistor soldered across each of them. The magnet causes a reed switch to operate, short circuiting the resistor across that switch allowing a level indication - or vice versa operation depending on the switch type - can't remember which was used.

The reed switch is often mounted just by fully encapsulating it in a neutral cure silastic. Neutral cure is important. If you open the cap of the silicone sealant & it smells like vinegar, then it is acetic acid cure & WILL EAT away wiring, circuit board traces & even corrode other harder to damage materials.

There are high / low water level switches available consisting of a floating collar with the magnet in it sliding over a central stem containing a single reed switch. The switch has a very limited range of movement & may be mounted pointing upward in the bottom of a container or mounted pointing down in the top of a container. With these, the collar can often be turned over adjusting the magnet positioning to set the operation to be the reverse of what is occurring ie reed turns on with water low, turn collar over & reed turns off with water low.

Some other side mounted switches with a swinging actuator also use a reed switch. I even worked on an industrial water heater where a float on a swing arm with a magnet inside the boiler operated a reed switch clipped on the outside - a neat way of avoiding wires through a pressure housing.



Joined Jun 23, 2004
I had worked on similar unit, the operator called it the mud meter being that it was in operation for one of the local mining groups an it was in a 2metre deep tank of mud, this makes sense. The sensor was as you describe it Brian, this one was just under 2metres long and had the external S.S. float donut/toroid shaped. inside was 3 or 4 pieces of PCB, with reed switches mounted about every 40mm, the sensing arangement had the reeds switching in resistance with a common connection on one side of the reeds and the other side of the resistors on another rail, I think there was a mimimum fixed resistance at the top as well. though that arrangement might seem to have jumps in resistance from minimum to the current level position this did not seem to happen at the metering, must have had enough capacitance to hold the current level voltage long enough between jumps. also the magnet was positioned such that it usually caused a make before break from one position to the next.
the fault was that there was not enough sealant in the device to stop the tracks on the PCB from corroding away.