enter image description here

Here is a circuit called LED chaser that I found the other day. I watched some videos on Youtube and saw that its function is to blink the LEDs in some kind of sequence:


But I don't fully understand what the circuit does or how it works. In the video the LEDs seem to blink in groups in somewhat random order (this could be cool for a disco light application). Is this the function of the circuit?

I also have a hard time getting my head around how this circuit works. There are the sections that consist of an LED with a series resistor, a transistor, and a capacitor. These sections are connected together with some kind of biasing resistors (?) going into the bases of the transistors. Then finally the collector of the transistor at the final section is connected back to the base of the first transistor as well as to the ground. But how do these sections work? I just see the transistors turning on when the circuit is powered, conducting current, lighting the LEDs.. Then the capacitors do something and I just get confused.

Help will be much appreciated! Googling hasn't produced any explanations.

  • \$\begingroup\$ re-arrange the leds into this order: 1 3 5 2 4 \$\endgroup\$ Aug 14, 2018 at 2:49
  • \$\begingroup\$ The reason they are in groups on the video, is because the video uses a modified version of this circuit, using just 3 transistors and 3 capacitors, and 3 leds per group. Otherwise it's the same principle. \$\endgroup\$
    – Passerby
    Aug 29, 2018 at 23:55
  • \$\begingroup\$ The Youtube design stresses the Vbe with about-7V then ramps up to turn on, which causes the collector to turn off the next stage for the same RC time constant. My design eliminates this by using emitter follower to drive the LED. tinyurl.com/y8xwtmgw I also included a switch and variable supply from 5 to 9V ( 30 to 65mA) \$\endgroup\$ Aug 30, 2018 at 5:05
  • \$\begingroup\$ Each stage is a 1-shot triggered by the previous stage and finally looping around after an odd number of stages, which then causes 2 or 3 LEDs to be on all the time as they cascade. But 2 adjacent LEDs can NEVER be OFF. \$\endgroup\$ Aug 30, 2018 at 5:12
  • \$\begingroup\$ Re the circuit in the tinyurl: Will it work with electrolytic caps? I tried the down log in both directions. Can't reproduce the result, the lights are all on, no pattern. \$\endgroup\$ Oct 17, 2021 at 2:14

3 Answers 3


This is a fun circuit! You will enjoy figuring this out. A couple of clues for you...

  1. The 470 ohm resistors are current limiting resistors.
  2. LED's operate on current flow, not voltage.
  3. Now, think about the RC time constant and what it is that controls the rate at which the capacitors are charged up enough such that the voltage drop across the caps rises sufficiently to turn on the transistors.
  4. Finally, note that this circuit can ONLY work with LEDs and not with light bulbs.


  • \$\begingroup\$ why not with light bulbs? \$\endgroup\$ Aug 14, 2018 at 2:50
  • \$\begingroup\$ Would it work with a light bulb and a diode in series? Points 1 and 2 I am aware of, but they don't seem to help me much.. I also know about RC time constants since I know how the BJT astable multivibrator works, but I'm still confused about this one. \$\endgroup\$
    – S. Rotos
    Aug 14, 2018 at 7:02
  • \$\begingroup\$ I haven't really been able to wrap my mind around this circuit, could you provide more assistance or should I offer a bounty? \$\endgroup\$
    – S. Rotos
    Aug 28, 2018 at 10:34
  • \$\begingroup\$ Have you tried simulating it? Looking at the voltages around the circuit could be a big help. \$\endgroup\$
    – Linkyyy
    Sep 2, 2018 at 19:35

At the beginning, we connect a resistor and an LED, and give them a voltage; the LED is lit. Next, put an NPN transistor in between the LED and ground, LED is OFF now. The transistor acts as a switch. Give a positive bias to the base of the transistor, it is switched on and LED is ON.

Simple LED circuit with current limiting resistor

Set up another pair of resistor and LED, with a transistor and a resistor to its base. Both LEDs are lit ON.

Two sets of LED circuit with transistor, connected by a capacitor

To disrupt the stable state, we introduce capacitor C2 to the base of transistor T2 with the other leg of C2 goes to the collector of transistor T1. When T1 is ON, voltage at its collector is very low close to ground. Voltage at base of T2 will be dragged low by C2, and thus T2 will be switched OFF. So now, T1 ON, T2 OFF. But after a short time, C2 is charged up through R2 and voltage at T2 base will recover to high enough level to turn T2 ON again.

2-way LED chaser or flip flop

When T2 is ON we use this opportunity to bring down the base voltage of T1 thus switching T1 OFF, by connecting base of T1 to collector of T2 with capacitor C1. Now, T2 ON, T1 OFF; only for a while, because C1 is charged up through R1 and T1 will be ON again, then the cycle repeats.

Now, if we set up another pair of resistor and LED, add a transistor T3 and resistor R3 to it, and then cascade this new pair to the previous two pairs via a capacitor C3, we will surely get a three way LED chaser with one LED OFF at a time. I tried it. When I expanded the circuit, with 4 transistors I only got two LEDs ON and two LEDs OFF in sync at a time. So there is no point adding 2 transistors when we can get same effect with only two transistors and put 2 LEDs in parallel to each transistor. When we use more than 4 transistors, I believe they will blink randomly, out of sync. I tried it with 5 transistors and 5 LEDs.

  • \$\begingroup\$ Very nice explanation. The "same schemativ with different look" makes it obvious that this is basically an astable multivibrator. Just some more stages are added. The randomness probably comes from the fact that at startup, there can be more than one LED lit and basically several "waves" chasing each other around... \$\endgroup\$
    – kruemi
    Dec 14, 2022 at 8:14

A few notes to ponder:

  1. KD Mann is basically correct in that once started it will run as long as it has power.

  2. There is the assumption that applying power will trigger at least one stage (LED) to come on, then time out which triggers the next stage.

  3. It is possible that if all resistors were equal in a very tight tolerance and all transistors had the same beta, or hFE, the circuit may not oscillate as expected. This circuit depends on slight imbalances to start up.

  4. If a 'no start' situation exist it can be avoided by making sure the LED and base bias resistors have differences of several percent. Likely the transistor with the lowest value base resistor will come on first, so unless you lower the value of one base resistor the startup will not be predictable. However it will always tend to start up with the same LED for the reasons I mentioned.

  5. The circuit is a loop that once started will not stop unless you remove power. Each stage triggers the next stage but the capacitor makes sure it times-out to trigger the next stage. Do not be surprised if you apply power and the starting LED seems to be random or fixed.


Your Answer

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

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