How to use a "super" capacitor?

I have a couple of these super capacitors that I bought back during the initial TI LaunchPad craze when I saw the Launchpad takes ultra low power to the extreme article.

However, I have no idea how to use them… How do I “charge” them? How do I run a device off of them? I’ve asked Google many times, but never found a satisfactory answer (or one that I could understand). And I hate joining forums just to ask one question. But, here I am, with lots of forum posts and super friendly, super smart folks who probably dreamt of this stuff in their sleep as toddlers!

Treat it somewhat like a battery with an RC discharging curve. You can connect it to the 3.3v or 5v and ground depending on the rating.

Initially when you turn on power, the cap behaves as though it is short-circuited and eventually gets charged to 5v.

Ah 2.5v… so put 2 in series and be mindful of the polarity! That’s a huge cap. Then to 5v. Never exceed 5v…or… *close eyes and cover ears*

I’d also make sure to limit current inrush as an uncharched cap appears like a shortcut to ground the first moment you apply voltage, which might overwhelm your power source (e.g. USB ports max. 500mA minus Core’s own consumption).

Also consider, when you put two caps in series, as @kennethlimcp suggested, you are reducing the over all capacitance by half.

Like this?

Some sort of resistor?

That’s about the only part I’m familiar with!

When it’s charged, does it begin to show a voltage on the other “side”? Can you disconnect it from a power source and power a device like a battery? Perhaps charge it with solar and let it run at night on the caps? I understand it won’t run nearly as long as a traditional battery, but I have them, so I might as well try to use them and learn about them!

other side? Yes so like i mentioned, treat it like a battery.

You can charge it with any source, take it out and place it on any other devices and run like a battery. Preferable a low power device or the discharging will be so fast and you get nothing out of it :smiley:

I wouldn’t recommend a resistor like what @ScruffR mentioned. What he said is valid for the inrush but i don’t want to charge my capacitor through a resistor and then to the spark core.

During normal operation with a power supply to the core and the capacitor attached, there will be power loss in the resistor. I don’t really like that :stuck_out_tongue:

The inrush will only be for a short period of time and it doesn’t cause much problem imo. Maybe i’m wrong :smile:

Yes and the capacitance becomes 5F due to them being in series. But you don’t have much of a choice since the rating is 2.5V and can’t do much unless you double it.

Datasheets are useful :smile:

I would probably limit your charging current to 1A.

If you don’t have a power supply with a constant current output, it’s easy enough to make one:

You’ll probably still need a variable power supply to accurately set the output voltage of the above circuit to 2.5V, or you could get lucky with the right combination of diode drops and a regulated SMPS wall adapter (maybe 6V?)

I have a drawer full of older technology super caps… I typically just set the voltage on my power supply to 5V (if the cap is rated for 5V, mine are 5.5V) … then turn the current limit all of the way down. Connect the cap, then turn up the current limit until you get to the desired rate.


Due to the ESR of the cap, it will naturally limit to some current… but this is bad to rely upon the cap’s internal resistance… you stress the cap, and it could fail. Best to design a proper inrush current limiting circuit like I posted above.


I used to do that as the power supply is already current limited so it’s like C.C mode so i don’t use a resistor. :smiley:

Do you charge each capacitor at 2.5V, or can you charge them in series with 5V? I don’t think I have any variable regulators, but I do have a handful of 3.3v and 5v each.

Err… scratch that. I do have a single LM317…somewhere.

I also have the SparkFun Breadboard Power Supply Stick, but I imagine that won’t help much if each cap has to be charged independently at 2.5V. What if I took a very sharp, very precise knife and cut a wire in half down the middle? Will that essentially cut a 5V source into two 2.5V sources? :wink: (I’m kidding)

I have a ton of various ways to source 3.3V and 5V. Not so many for 2.5V. Could I jerry-rig a voltage divider with resistors to get 2.5V? I guess I technically could, but would it actually work (or introduce too much interference or something)?

1 Like

Also, this thread is becoming a crash course in EE for me. Darn Arduinos have insulated me from a bunch of this stuff. When I got my TI LaunchPad “kits”, it was like someone ripped off the training wheels and seat. That was a very uncomfortable ride that I walked away from without much learned other than new heights of frustrations.

1 Like

If you are planning on running them in series for 5V 5F, then you would want to charge them in series with a 5V source. If you are starting to get serious about electronics, you definitely need a power supply. This looks like a nice unit if you just wanted to break down and buy something:

Typically you might go see if Adafruit and Sparkfun have anything they’ve vetted as “good”. Adafruit has nothing. Sparkfun has this beast which is way more expensive and a lot less capable. Not to mention it has no screw down binding posts (a must have for any power supply).

Sure there is a ton of cheap crap from china on ebay… I would get a recommendation from someone on any of it before buying though.

1 Like

Just for the records ;-), I didn’t suggest a resistor and since a part of the question was about the charging, I only refered to that (sorry could have bean clearer about that :blush:)
When dis-charging, the power flow would turn around and so the current could “take a different route” through the discharge circuitry bypassing any unwanted components.
This way you could perform “incircuit” charging much like an UPS.

For these reasons it would be great, @wgbartley, to know how you’d like to charge the caps. In-circuit or out-of-circuit?
@BDub’s suggestion with an adjustable power supply is great for the latter - and as an EE you are desperately in need of one anyway :wink:

Edit: As supplemental info about caps I found this video quite interesting (esp. timeline 0:12:00 onwards)

1 Like

So sad it is “only” 5F after you put two in series. I remember a Popular Science article back when I was a kid. Somebody had finally created a 1 farad capacitor. It was something you drove up to in your car. :wink:

Either, or both. I don’t have any specific project planned for them, but wanted to figure out how to use them before I attempted to use them. I guess I’d be charging them out-of-circuit to begin with. Maybe with an Arduino 5v output or the “power stick” I mentioned earlier. I’ll check the voltages coming out of those to make sure it’s not too high. If I only need 3.6v or so out of it, charging it with slightly less than 5v should be safe, right?

I cannot claim to be anything near an EE. Just a day-job programmer with an unquenchable appetite for tinkering. $150 for a power supply doesn’t seem too terribly expensive, but it will likely be a couple of months (at least) before my wife will let me spend a bunch of money on a new toy (I’m still having to replace crap parts on my cheap Chinese 3D printer). Too bad my birthday and Christmas are a long ways away. I’d beg for a power supply and a scope (and a desk!) so I can stop tinkering on the kitchen counter and/or dining table.

Also, thank you guys so much for helping me learn this stuff. I :heart: this :spark: community!


I’m finally getting around to trying to play with these things. I have hooked it up to the 5v line on an Arduino to charge. I have them in series. VCC to + and - to GND. If I put a multimeter from the first + in the series to the last - in the series, I get 0.17v. It’s been steadily going up. Does it take a long time for these things to charge, or am I doing something wrong?

You can analyze it as a simple RC circuit with a perfect voltage source and series source impedance. The RC time constant will determine the charge time and you need to get to 4*RC to get to ~98% of the supply voltage.

Just to make the math easy, assume the 5V supply has a 1 ohm source impedance, then 10F should take 4RC = 40 seconds, but the inrush current would 5A and I don’t think your little Arduino supply will give you 5A, and I don’t think your supply has 1 ohm.

So more realistically, lets say 5V with 5ohm source impedance and 1A delivered current. Now 4RC = 200 seconds.

Does that seem close based on your power supply?

Be careful–this is one time you should take the wedding ring off for safety! Tell the wife I said you had to!

It’s definitely taking longer than 200 seconds. The source is my laptop USB, so 500 mA max. I’m not sure what you mean by “source impedance”, and I’m not sure how or where to check for that!

And, yes, I have strong faith in the fuses in my laptop. I’ve accidentally done some very dangerous things to it, and all I have to do is reboot to enable the USB ports again.