...Coming soon


555 Timer Based Boost Converter (Spoiler: Not Very Useful)

Boost circuits are an awesome way to power microcontrollers. Instead of 4AA batteries or a 9V (usually regulated down to 5V or 3.3V), you just boost one or two AA’s up to your desired voltage.  The only problems are that boost ICs (like these two) are a little expensive once you decide to make more than 2 or 3 of something, and the cheaper ones are hard to find in packages that can be hand soldered easily.

I stumbled upon one great solution here – use PWM from a microcontroller.  And Atmel recently introduced the ATtiny43U with a built-in boost converter – hopefully this will become a trend and all microcontrollers will come with built-in boost converters, the same way they (mostly) all support serial, I2C, and SPI.

But in the meantime I wondered if a simpler solution might be possible, a boost converter using a 555 timer.  The 555 timer is a very common $0.20 chip that can generate a PWM at a given frequency and duty cycle based on the charge time of some chosen resistors and a capacitor.  The circuit would need the same basic external components as a boost converter (inductor, diode, capacitor), but it would be a lot cheaper.   It wouldn’t be able to regulate the voltage, only multiply it using a calculated duty cycle, so the solution already starts off a bit limited since the voltage of a AA battery varies over it’s lifetime.  But despite this limitation, there are definitely applications where providing roughly 5V from a single AA or AAA is very useful (powering servos in a kite aerial photography rig, for example)

The circuit looks like this:

555 schematic

Wikipedia has a great explanation of how a boost converter works.  The basic idea is that the source voltage is run through an inductor and alternately connected to ground through the transistor and disconnected so that it will charge the capacitor.  The ratio of how long it’s connected to ground and how long it’s disconnected is the duty cycle, which determines the boosted voltage.

The math is pretty simple: D = 1 – Vi/(Vo+0.5) gives you the duty cycle you’ll need (the +0.5 part is to compensate for the forward voltage of the diode).  555 timer calculators are available all over the Internet for choosing resistors and capacitor based on the duty cycle. As long as the frequency is in the KHz it should work.


And when wired up it does work great… until you attach a load, then the voltage drops precipitously.  Higher frequencies seem to help a bit, but not enough.  A higher duty cycle is needed to compensate for the load, which is how a boost converter handles the problem, but our simple 555 timer with its constant frequency and duty cycle is not up to the task.

If the load never varies – powering a string of LEDs, for example -  then you can play with the duty cycle to get the voltage you need under those conditions.  Other than that, the only thing it’s useful for is charging a capacitor.  If you wanted to boost a 1.5V AAA battery up to 12V or 50V and release it all at once, this circuit can get it done.  Use it to shock your cat.

So a 555 timer isn’t very useful for a boost converter because there’s no easy way to adjust the duty cycle on-the-fly.  But inexpensive PWM controllers do exist… perhaps I’ll try that next.

3 comments to 555 Timer Based Boost Converter (Spoiler: Not Very Useful)

  • Adam G.

    Nice project. Have you thought about a charge pump?

    • Ben

      I have, but they seem to require a lot of parts – several capacitors, diodes, and transistors, plus a clock of some sort. Seems strange to me that they don’t exist as a $0.50 component like the 555 timer…

  • Mario

    Hey, I’ve tried to implement this circuit using PWM from an AVR MCU instead of a 555 but the voltage doesn’t seem to be boosted at all.
    There’s actually a voltage drop of around 0.5V which is probably from the diode.
    I’ve tried many different frequencies, from around 500Hz all the way up to 60KHz and a few duty cycles. I’m using a BD135 as the transistor, a 5V regulated power supply and a 47uF, 400V electrolytic capacitor.

    Any ideas of why it’s not working?