# How to increase current output of capacitor bank with buck booster?

I'm pretty novice when it comes to electronics given that my background is in mechanical engineering. But I am trying to design a circuit that can take a 5V (almost 0A current) to charge a capacitor bank, then have the capacitor discharge to power an Arduino. My Arduino circuit draws about 300mA at 5V for almost 30-40 seconds. In my circuit, I am using a TPS61030 breakout board (Adafruit Powerboost 1000), which is capable of boosting a voltage as low as 1.8V to 5V and supposedly capable of outputting 1A of current. This works great and I can get a constant 5V for plenty of time by charging a 0.1F capacitor to 5V and letting it drain into the TPS61030. The problem is, I'm getting very little current, not nearly enough to power the Arduino for more than about 1 second. Can anyone suggest anything I can use to get the current I need? Again, I know very little about electronics and have a lot of learning to do - any leads would be appreciated!! Thank you for your help.

• Do the math: how much energy does your Arduino need? How much energy can that capacitor store? Is this anywhere near enough for the Arduino? (hint: no) Nov 3, 2018 at 1:45
• Arduino is not very efficient. You should use a ultra low power 8 bit micro controller like a STM8L. ST has an \$11 eval board with Arduino Uno V3 connectivity st.com/content/ccc/resource/technical/document/user_manual/… Nov 17, 2018 at 19:13

The energy you want to deliver to your Arduino is

$$300\ {\rm mA}\times 5 \ {\rm V} \times 30\ {\rm s} = 45\ {\rm J}$$

The energy stored in your capacitor is

$$\frac{1}{2}\left(0.1\ {\rm F}\right)\left(5\ {\rm V}\right)^2 = 1.25\ {\rm J}$$

Of which you will lose some to inefficiency of the buck-boost converter, and some of which you will lose because you won't be able to drain the capacitor completely before the converter stops working.

You need more the 36 times more capacitance to have enough energy to power your load, or you need to charge your capacitor to a higher initial voltage (30 V or more), if it can take it.

Or you could choose a different microcontroller that can operate on much lower current. A microcontroller alone can be found operating on a fraction of a milliamp. The total current needed by your circuit will depend on what peripherals you need to attach to it.

• I strongly doubt there are any 100mF capacitors on the market rated for more than five or ten volts. And yes, looking on Digikey it seems the highest voltage rating they offer on a 100mF cap is 11V--but, you could also get enough energy out of 10V or so on a 1F cap, and 11V and 12V 1F caps are available, so this project isn't completely infeasible... I mean, I'd still go for a low-power μC, but. Nov 3, 2018 at 2:00
• (not to say that digikey represents the entire market, but they do represent most of the easily accessible market at least. Nov 3, 2018 at 2:25
• @Felthry, fair enough, and that's why I included the caveat, "if it can take it". At this point we don't totally know that the 0.1 F capacitor isn't 100 1 mF capacitors in parallel. Nov 3, 2018 at 2:27
• I have done the exercise with Aluminum electrolytics and Tantalums. Basically, within a technology, there's a pretty constant energy vs. weight and cost and volume. IOW, a 10uF 160V cap is about the same size and cost as a 1000uF 16V cap, etc. Nov 3, 2018 at 2:37
• I'm currently using a 0.1F 5V capacitor, but I have 1F 6V and 5.5F 5V caps on hand as well. I can't charge it at higher than 5V because of the charging circuit I'm using (an energy harvesting module) but I can certainly use the supercapacitors. Does the current at which I charge the capacitor matter for anything other than the speed at which it charges? It shouldn't affect the output current right? Nov 3, 2018 at 2:39