DIY Motorcycle ‘Doo-Doo-Doo’ Sound Circuit With Only Capacitors and Resistors and Transistors (Suitable for Teaching Purpose in Circuit Lab)

This circuit offers a valuable opportunity to explore key concepts in analog electronics, specifically through the use of a complementary self-oscillating multivibrator composed of PNP and NPN transistors. By integrating a frequency-selective capacitor, it generates a Motorcycle ‘Doo-Doo-Doo’ Sound.

A complementary self-oscillating multivibrator utilizes pairs of complementary components, such as NPN and PNP transistors, arranged in a cross-coupled configuration. This setup enables continuous oscillation without requiring an external trigger, producing an output rich in harmonic components. Such circuits are widely applied in signal generation, modulation, and timing systems.

This circuit has significant experimental value. By varying the parameters of its components, you can observe different results and explore the factors that influence the oscillation frequency of complementary self-oscillating circuits. This hands-on approach can deepen your understanding of analog electronics and the principles behind self-oscillating systems.

The necessary materials:

1 x 10kΩ Resistor

1 x 47kΩ Resistor

2 x 47uF Capacitors

2 x 1uF Capacitors

1 x S9013 NPN Transistor

1 x S9012 PNP Transistor

1 x Speaker (8Ω 0.5W)

1 x 3V DC Power Source

1 x 2-Bit Slide Type Switch

It is important to note that C1, C2, C3, and C4 are electrolytic capacitors. While they typically require correct polarity when connected to a circuit, they do not have polarity constraints when dealing with AC signals. However, in DC power supplies or DC signals, polarity must be observed: the longer lead must be connected to the positive terminal, and the shorter lead to the negative terminal. In this circuit, special attention must be given to C1 and C3 to ensure they are not connected in reverse polarity, as improper connection may damage these electrolytic capacitors during operation.

C1 serves as the charge-discharge capacitor in the bias circuit of Q1, while C3 functions as a decoupling capacitor connected in parallel with the power supply. The role of C3 is to quickly supply or absorb current when the circuit's load current fluctuates, preventing sudden voltage changes. By smoothing the power supply voltage waveform, the decoupling capacitor enhances the overall quality of the power source.

Once you have completed building the circuit according to the schematic and breadboard layout in Step 1 and connected a 2 to 3V DC power supply, turn the switch on. You will hear the speaker producing a sound similar to a motorcycle engine. The waveform of this signal is shown in Fig.speaker_wave.

In Fig.speaker_wave, you can observe that the point marked ① represents a negative spike voltage, with a magnitude close to the supply voltage. This occurs because, when capacitors C2 and C4 are in the charging state, their positive terminals effectively act as ground. Although, theoretically, the speaker behaves like a short circuit in a DC circuit, it actually has impedance. As a result, during the charging phase of C2 and C4, the current primarily flows to the capacitors' positive terminals rather than directly passing through the speaker to GND—since current naturally follows the path of least resistance.

When C2 and C4 finish charging at the falling edge of the waveform segment marked ③ in Fig.speaker_wave, transistor Q1 turns on. Since one end of C2 and C4 is connected to the base of Q1, when Q1 turns on, its base-emitter junction becomes forward biased. At this point, C2 and C4 effectively have their negative terminals grounded, causing their negative voltage to drop to 0V instantly. Due to the capacitor's inherent characteristic of resisting sudden voltage changes, a sharp voltage drop on one terminal will induce a negative voltage spike on the other terminal. Consequently, when the negative terminal of the capacitor suddenly drops to 0V, its positive terminal generates a negative voltage spike close to the supply voltage (in some cases, the peak negative voltage can be up to three times the supply voltage). This explains the waveform segment labeled ① in Fig.speaker_wave.

If you want a more detailed explanation of the capacitor's ability to prevent sudden voltage changes based on its internal structure, please refer to this capacitors blog

If the capacitance in the C2 and C4 regions increases, what happens?

As shown in the video in Step 2 , as the capacitance of this region increases, the interval between the speaker's sound occurrences becomes longer.

This is because the period of the RC frequency selection signal follows the formula T = 2πC. When the capacitance C increases, the period T also increases, meaning that the interval between each sound produced by the speaker becomes longer as the capacitance value rises.

With the circuit in Step 2 connected to the power supply and the switch turned off, use a multimeter to measure the voltage across the speaker terminals. You will find that the measured voltage is 0V because the speaker is constructed from a metallic coil, which behaves like a conductive wire under DC conditions.

Next, remove the speaker (as shown in Fig.speaker-removed) and insert two extension wires into the original positions of the speaker leads on the breadboard. Then, use a multimeter to measure the voltage between these two wires—you will find that the measured voltage is close to the supply voltage, instead of 0V.

At this point, some beginners might wonder: since the switch is off, Q1 is in a cutoff state, and therefore Q2 should also be non-conducting. Theoretically, the voltage between the collector of Q2 and the negative terminal of the power supply should be 0V. So why does the measured voltage nearly match the supply voltage?

This discrepancy arises because real-world components are influenced by non-ideal factors such as leakage currents, parasitic effects, or design limitations within the devices. These factors can cause deviations between measured and theoretical values. Although Q1 and Q2 are not conducting in this circuit, the applied voltage across the emitter and collector of Q2 may still propagate to the collector due to these effects. As a result, when measuring the voltage between the collector of Q2 and GND with a multimeter, the reading approaches the supply voltage. However, the actual current is extremely small, limited to leakage current, which means that in theoretical analysis, Q2 in its cutoff state can still be treated as an open circuit.

Thus, when studying electronic circuit principles, hands-on experimentation is essential—it allows you to uncover additional insights beyond theoretical learning.

Now we know that even when the switch is turned off, there is still a voltage between Q2's collector and GND, with a value close to the supply voltage. However, when the speaker is connected, its behavior in a DC circuit is equivalent to that of a conductor, which results in Q2's collector voltage dropping to 0V.

At this point, some beginners might wonder: since there is a voltage close to the supply voltage between Q2's collector and GND when the switch is off, wouldn’t connecting the speaker—essentially acting as a wire—between Q2's collector and GND create a short circuit and damage the circuit?

The answer is no, it won't. As mentioned earlier, when Q2 is not conducting, the impedance between its collector and emitter is extremely high. According to Ohm’s law, the current passing through the speaker in this situation is very small, limited to leakage current. This negligible amount of current is not enough to damage the speaker.

We know that under DC conditions, the speaker does not produce sound. In the circuit from Step 2, the reason the speaker generates a motorcycle-like "doo-doo-doo" sound is due to the charging and discharging effects of capacitors C2 and C4, which drive the complementary self-oscillating circuit formed by Q1 and Q2. This process creates periodic switching signals, effectively transforming the DC circuit into one with AC characteristics, allowing the speaker to generate sound.

Now, what happens to the waveform across C2 and C4 if we use a wire to short-circuit the speaker? See Step 4 for details.

This step is an experimental extension to explore how the waveform at Q2's collector changes when the speaker is short-circuited compared to when the speaker is connected.

Short-circuit the speaker as shown in Fig.short-circuit_speaker, turn on the switch, and use an oscilloscope to measure the waveform between Q2's collector and GND. The measured waveform is shown in Fig.Q2_Collector_Wave.

From Fig.Q2_Collector_Wave, we can observe that when the speaker is replaced with a wire, the complementary self-oscillating multivibrator circuit formed by Q1 and Q2 no longer produces a waveform that contains sine wave components and capacitor charging characteristics, as seen in Fig.speaker_wave from Step 2. Instead, the waveform consists entirely of spike voltages.

Why does this happen? When the speaker is connected, it behaves like an inductor because a speaker consists of a coil wound around a magnetized base, forming an LC frequency selection network with C2 and C4. The coil itself possesses the ability to resist sudden voltage changes due to its inductive characteristics, which can be explained by Faraday’s law of electromagnetic induction. When the circuit’s current changes, the coil generates a self-induced electromotive force (back EMF) that opposes rapid current fluctuations. This is why, when the speaker is connected, the capacitor's charging and discharging waveform is clearly visible.

If you want to learn more about how inductors prevent voltage surges, please refer to this inductors blog

When the speaker is replaced with a short wire, capacitors C2 and C4 still go through the charging and discharging process, but due to the oscilloscope's Main Timebase setting of 2.5ms, the capacitor’s charge-discharge curve is not as easily visible. If the Main Timebase were set to a sufficiently smaller value, the capacitor’s waveform would be observable. However, since the timebase was 2.5ms in the experiment where the speaker was connected, it is necessary to maintain the same timebase of 2.5ms when measuring the waveform under the short-circuited speaker condition to ensure consistency in comparison.

The contrast between these two scenarios—with and without the speaker—forms a significant control group, providing a clear and intuitive demonstration of how changing a specific circuit component affects the circuit’s behavior. In this particular case, replacing the speaker with a short wire results in a drastic change in the output waveform.

This step helps demonstrate the dynamic nature of electrical systems and allows for a deeper understanding of the factors influencing variability in experimental outcomes

Result 1: As shown in Fig.reform-circuit, we made a slight modification to the circuit by adding a resistor R1 in series with capacitor C3.

Now, when the switch is closed, the speaker initially produces a sound but then quickly stops, as demonstrated in the video in step 5. Similarly, when the switch is opened again, the speaker momentarily produces a sound before falling silent once more.

The reason the speaker only emits a brief sound upon switching on or off is as follows:

When the switch is closed, the power supply charges C1 through R2, gradually increasing the base voltage of the NPN transistor Q1. Once this voltage reaches approximately 0.7V, Q1 begins to conduct and oscillate. (In real-world conditions, because electronic components are non-ideal, manually toggling the switch introduces mechanical noise—a signal containing a mix of frequencies—which acts as an input to Q1's base, triggering oscillation.) As a result, the speaker starts to produce sound.

However, the speaker does not continue to emit sound indefinitely. This is because, as C1 accumulates charge, the voltage across it eventually reaches a threshold, causing Q1 to enter saturation. Once saturated, Q1 loses its amplification properties, and the circuit effectively behaves like a DC circuit, meaning that the speaker is no longer driven and stops producing sound.

Result 2: Remove the 103 ceramic capacitor at C3 and replace it with another capacitor of the same type and labeled 103. Then, close the switch.

As shown in the video in step 5, the speaker produces a brief sound and then stops, just like in Result 1. When the switch is opened again, the speaker once again briefly emits a sound before falling silent. However, compared to Result 1, the sound is sharper, and the frequency difference between the two results is noticeably perceptible.

Even though the circuit structure and component parameters remain identical, the experimental outcome varies. This is because electronic components, even of the same type and specification, inevitably have manufacturing tolerances. In a self-oscillating circuit, these tolerances can affect the frequency selection network formed by the RC components to some extent.

If you attempt to replicate this experiment, you might obtain different results, or you might get something similar or even identical to mine. The variation depends on the current state of your circuit and the degree of manufacturing deviation in your selected components compared to mine.

Result 2A: Adding a 103 Ceramic Capacitor (C4) in Parallel with C3

Now, building upon Fig.reform-circuit, we add another capacitor, C4, in parallel with C3, as shown in Fig.reform-circuit-1.

When the switch is closed, you will notice that as long as the switch remains closed, the speaker continues to emit sound. As demonstrated in the video in step 5, the complementary self-oscillating multivibrator generates an output signal at a frequency of over 500 Hz , which drives the speaker, producing a sound similar to the operation of a CNC lathe.

Why does the speaker continuously produce sound after adding a 103 ceramic capacitor in parallel? This happens because the RC frequency selection network's frequency has changed. The newly selected signal frequency can now propagate through the positive feedback network to Q1's input, allowing the circuit to maintain oscillation indefinitely without the need for an external input signal.

From the oscilloscope display in the video in step 5, we observe a spike voltage reaching 13V in the negative half of the waveform's coordinate axis, which is significantly higher than the 3V power supply voltage.

Why does the circuit generate a voltage four times higher than the supply voltage? This phenomenon arises due to two key factors:

Capacitor voltage stabilization – Capacitors inherently resist sudden voltage changes.

Inductor current stabilization – The speaker coil acts as an inductor, which resists rapid current fluctuations, indirectly manifesting these effects as voltage variations.

The combination of these two properties results in spike voltages far exceeding the supply voltage. If the speaker’s inductive effect were absent, the capacitor-generated negative voltage would be approximately equal to the positive peak voltage, as seen in Step 4. In this specific circuit, the observed negative peak voltage closely matches the power supply voltage.

Result 3: Removing the 103 Ceramic Capacitor (C4) to Restore the Circuit to Result 1 Configuration

Building upon Result 2A, we now remove capacitor C4, restoring the circuit to the exact configuration of Fig.reform-circuit from Result 1.

When the switch is closed, the speaker continuously produces a high-pitched sound, marking a clear difference from both Result 1 and Result 2. As shown in the video in step 5, the circuit generates a frequency of approximately 2.3 kHz, resulting in a sharper tone from the speaker.

Although Result 1, Result 2, and Result 3 use the same circuit structure, they yield three distinct outcomes. Since both Result 1 and Result 3 use identical components, the previously discussed issue of component variation due to manufacturing tolerances does not apply here—the capacitor used in Result 3 is the exact same one used in Result 1.

So, why do these identical circuits with identical components produce different experimental results?

The answer lies in environmental factors. After running Result 2A, the circuit experienced significant power dissipation, which led to a rise in temperature for components like the transistor. Since PN junctions inside transistors contain parasitic capacitance and parasitic resistance, changes in temperature alter their values, affecting circuit behavior. Due to this thermal shift, the circuit in Result 3, under the influence of its positive feedback network, remains in oscillation, causing the speaker to continuously emit sound.

However, when we turn off the switch and use a screwdriver to short the terminals of electrolytic capacitors C1 and C2, we discharge them completely. After letting the circuit sit idle for a period, then reconnecting the power supply and turning the switch back on, the experiment produces the same result as Result 1 once again.

The affordable DIY materials are available at https://www.mondaykids.com/monday-kids-educational-diy-complementary-astable-multivibrator-kit-for-stem-learning.html