S1E3 Chatshaala with electronics gadgets | Exploring mobile charger - Diode in details

Today we discussed the diode component of a rectifier in little more details.


  • Diode as a Gate metaphor.

  • forward biased and Reverse biased

  • Conduction in both biasing

  • Diode with AC source

  • Diode with DC source

  • Diode with resistive load

  • Diode with capacitive load

  • effects on the waveform before and after through diode.

  • Forward/threshold voltage

We used the Falstad Circuit Simular to simulate the above circuits to understand the behavior and application of diode.

Some of the simulation image is shown below.




You can join us in live discussion on exploring Electronics gadgets during weekdays at 11:00 Am to 12:30 pm at CHATSHAALA.

You can also post your question and do discussion in this thread.

Happy Exploring!


Fullwave circuit. lushproject has a center tapped transformer. You can add a slider to each component by right clicking on a component and selecting slider. You can change the max and min values.
NOTE: Capacitor value for the slider is in Farads. Electronic capacitors are in the pico to micro range (except for the new supercaps used as coincell replacements). Change the value to microfarads.
Check the waveforms using scope in the simulator.Fullwave-circuit


Today’s session was really interactive and boosted the confidence to try out circuit changes without the fear of damaging any component, sounding dumb whilst asking doubts and suggesting changes. At the end of the session we left with an open question of how to get an output with both positive as well as negative cycles of signal visible, like in the case of full wave rectifier with 2 diodes…

I tried out this solution, where I connected 2 voltage supply sources, with opposite polarities, same resistance in series with the two but two diodes used here, so that one diode works in forward bias mode during positive cycle of first supply, and other works for positive cycle (as it goes in forward bias mode) of second supply coinciding with the negative cycle of first supply.

As we expect, the opposite supplies should cancel each other out, but the image shows that simulation resulted in our expected output. Now the question is, does this simulation hold true in practical implementation, or not!

Another interesting find, when tinkering the circuits: Check out the circuit we tried out and see if you (the reader) get the same square waveform! Additionally, let me know whether this will happen for practical circuit too without any damage to the components, or not?