I'm trying to designing a water level indicator using npn transistor(BC548). I tried simulating the circuit on PSpice, but I don't know what to add for the sensor. I want a graph. The circuit will look something like this- Water level indicator

I just want to know what can I use for the parts that is sub-merged in the bucket in PSpice/OrCad.

  • 2
    \$\begingroup\$ use four switches, each in series with a resistor \$\endgroup\$
    – jsotola
    Apr 21, 2022 at 23:35
  • \$\begingroup\$ Get a bucket of water, stick some wires in it, measure the resistance on your multimeter. Use a resistor with that same resistance. \$\endgroup\$ Apr 22, 2022 at 9:55

2 Answers 2


You will probably need more gain than just one BC548 can provide. Try using them as a Darlington pair for each level.

That said, normal tap water has a resistivity in the multiple k-ohm to multi-megaohm range depending on the probe size and water ion concentration. You could characterize using an ohmmeter and your proposed probes.

More here: https://www.labmanager.com/white-papers-and-application-notes/resistivity-conductivity-measurement-of-purified-water-19691

Once you know your immersed-probe resistance, you would then model it as a resistance with a switch in series for each sensor, assuming open circuit for a not-immersed probe.

That said, the probes could interact with each other. The common probe will have its own resistance, and if the individual probes are close to each other they could steal base current from one another.

My thought? You could also consider using an optical sensing approach instead. Here is an example (mods: not a product recommendation) of this kind of sensor: https://www.amazon.com/Taidacent-FS-IR02-Infrared-Stainless-Detector/dp/B08LGJ1L7X They are popular in the aquarium hobby.

These sensors have a prism and an IR emitter-detector pair. When in air, prism reflects the beam triggering the sensor; when in water, the liquid causes the beam to leave the prism, causing the sensor to turn off.

More about this here: https://www.abestmeter.com/electronic-water-level-switch/ (BONUS: there's a T1-3/4 'LED' type available too.)

And here: https://www.electroschematics.com/optical-liquid-level-sensor/

Why do it this way? Electrical isolation (so, safer), works with any liquid regardless of ion content, doesn't corrode over time. I propose that you could create the plastic pieces using off-the-shelf material along with suitable LED-sensor pairs. Or use 4 of these commercial units.


You may want to measure water depth with transistors , but unless you can guarantee the ionic content of water, you will not be able to calibrate or control it's resistance vs depth.

However you can reliably detect the dielectric constant of Dk=80 ore consistently. With two electrode separate with a fixed gap you can thus calibrate the capacitance between a pair of electrodes with a fixed capacitance in air and expect that impedance will reduce by a factor of 80 when immersed fully in water. For that to occur the part always above the top level must be smaller capacitance per unit length with a much greater length/width ratio which modulates capacitance.

Thus the task for you to choose are the quantum levels or linear scaling of capacitive impedance at some frequency. Astable clocks with a single CMOS Schmitt Trigger are one way to create a very low power C meter. Then attenuation levels rise with water height and may be detected with the low frequency carrier amplitude attenuation using a battery voltage with transistor comparators for a simple circuit.

I'l leave the details for you to recognize in existing circuits or try to create on your own.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.