Single-cell battery powered astable multivibrator

There is a problem with lighting an LED with a 1.5V battery because the forward voltage of the LED is higher than the battery voltage. The simplest improvement is to use a step-up DC/DC converter. This design example provides a simple and reliable alternative for low-cost applications. The circuit in the attached figure uses a classic astable oscillator composed of transistors Q1 and Q2. The square wave drive signal at the collector of Q2 turns the PNP switching transistor Q3 on or off. When Q3 is turned on, it charges the inductor L1, and when it is turned off, the inductor L1 releases the stored energy through the LED reverse and boosts the voltage. Therefore, any type of colored LED can be lit.

When the battery voltage is 1.5V, the oscillation frequency of the astable circuit is 1/T0. Where: T0=TL+TH, and TL≈0.76R2C2 and TH≈0.76R1C1, where T0 is the cycle time, TL is the on time, and TH is the off time. With the component values ​​in the attached figure, the frequency and duty cycle are approximately 28.5kHz and 50%, respectively. During the on period, transistor Q3 is turned on, and inductor L1 begins to charge at a constant voltage, so its current rises linearly to a peak value. This value is shown in the following equation: IL1PEAK=[(VBAT-VCESATQ3)/L1]*TL, where IL1PEAK is the peak current of L1, VBAT is the battery voltage, and VCESATQ3 is the collector-emitter saturation voltage of Q3. During the off period, Q3 is turned off, and the polarity of the inductor voltage is reversed, making the LED forward biased, and the inductor discharges through the LED with a higher peak voltage. The voltage is roughly equal to the forward voltage of the LED.

Since this cycle is repeated at high speed, the LED appears to be glowing steadily. The brightness of the LED depends on its average current, which is proportional to the peak current. The current of the LED is roughly a triangle wave. Since the cut-off time of Q3 is limited, its peak current is approximately equal to the current of the inductor. The average current can be simply estimated as: ILEDAVG≈0.5*IL1PEAK*(TDIS/T0), where TDIS is the discharge time of the inductor L1 through the LED. This value can be roughly estimated from the discharge slope of L1, which is VLED/L1, and VLED is the voltage of the LED.

To control the brightness of the LED, the value of the inductor can be changed to increase or decrease the peak current of the inductor. The inductor modification range is from 100μH to 330μH, which can achieve the best brightness for the LED model used. However, the charging slope of L1 is always less than the discharging slope, and since TL is equal to TH, L1 has enough time to discharge completely. When it charges in the next cycle, its current cycle always starts from zero. If this is not the case (for example, TH drops too much), the inductor current will increase in each cycle until Q3 is out of saturation. Since the final current value depends on the DC gain of Q3, it becomes unpredictable. The LED can be flashed by driving the base of the optional transistor Q4 with a low-frequency gate signal.

When selecting the circuit components, there are no special restrictions on the component parameters. For the transistors, any small signal transistors can be used. However, if possible, try to select a PNP transistor with high DC gain and low collector-emitter saturation voltage for Q3 to obtain the best efficiency. In addition, pay attention to the peak current not to saturate L1 and not to exceed the maximum rated peak current of Q3 and LED. The astable circuit can start working with a power supply as low as 0.6V, but the LED does not emit light. When the power supply voltage exceeds 0.9V, it emits weak light. When the power supply voltage exceeds 1V, the LED has sufficient brightness, but this depends on the forward voltage of the LED. If the LED in the attached circuit does not need to flash, Q4 tube can be omitted.