NEW: Learning electronics? Ask your questions on the new Electronics Questions & Answers site hosted by CircuitLab.
Basic Electronics » Powering Nerdkit using Super Capacitor
May 05, 2010 by rajabalu21 
Can any one share their experience of using a Super Capacitor? I have a 1F 5.5V Super Capacitor and would like to know the following.
Thanks Raja 

May 05, 2010 by bretm 
2: One farad equals one coulomb per volt. If charged to 5.5V, that's 5.5 coulombs of charge. One amp is one coulomb per second, so 30mA is 0.03 coulombs per second. 5.5 coulombs at 0.03 coulombs per second is 183 seconds. Actual time could be much less because of waste heat in the circuit. 
May 06, 2010 by rajabalu21 
bretm  Thanks for your answer. Now I understand the math behind the computations. Here is my thought, even though the maximum rating of the capacitor is 5.5 V, I would like to charge it to 5V as I can then power the nerdkit without any other regulatory circuit. Q = CV Q = 1 F x 5 V Q = 5 C = 5 As = 5000 /3600 = 1.388 mAh This is so low compared to a 9V battery in upwards of 400mAh. May be it is good enough to keep the controller in sleep mode for several hours. Can any one please explain how to charge the capacitor (what should be the max charging current)? I am not sure how to arrive at the charging current from the datasheet. Also comments on the expected life of the device compared to a rechargeable battery is appreciated. They say endurance 1000 hours. Does that mean that it will work for only 100 hours? I do not understand what ESR means and how the 30 ohms influence the rest of the design. Thanks Raja 
May 07, 2010 by mrobbins (NerdKits Staff) 
Hi rajabalu21, A few notes: You're overestimating the "equivalent mAh" of the capacitor because as you discharge it, the voltage drops, and at some point, the voltage has dropped enough to be unable to power your microcontroller and other circuit elements. For example, if you start at 5V, and maybe you design your circuit to work down to 3.5V, then you only get to use 30% of the capacitor capacity. In contrast, a battery's voltage won't change too much until basically fully discharged, after which point the voltage drops off precipitiously. That's because a battery is based on chemical reactions which have a particular energy associated with them, and as long as there are more reactants available, the voltage is roughly constant. ESR is Equivalent Series Resistance. You can pretend that the capacitor behaves like an "ideal" 1.0F capacitor in series with a 30 ohm resistor. This can be important when charging and discharging. On the charging side, imagine just connecting the (discharged) capacitor to a 5V source and letting it charge up on its own. The capacitor has its own RC time constant  in this case, (1 Farad)*(30 ohms) = 30 seconds! That's a long time constant, and generally we'd want to let something charge for several time constants to get within 99%+ of the final value. (Of course, there are faster ways of charging the capacitor  specifically by applying a slightly higher charging voltage, and stopping sooner. But this can damage the capacitor and/or shorten its lifetime.) On the discharge side, the 30 ohms can influence you in two ways. First, if you're drawing say 10mA from the capacitor, there's an extra 0.3 volt drop due to the ESR. Second, as with most digital circuits, even if the average current consumed is 10mA, it's actually not constant, and it's very "spiky"  most current is consumed on the clock edges, and almost zero in between clock edges. That ratio can be a factor of 10 or 100x, meaning that 100mA might be used by the circuit for 1/10th of the time, and 0mA the other 9/10ths of the time. With an ESR of 30 ohms, now you're dropping 3 volts during those spikes, which will certainly cause problems with the microcontroller and prevent its proper function. It basically means that you'll need to add other decoupling capacitors to handle the load. I'm not sure how they define endurance exactly in this case. It could refer to how long it holds a charge once it's been charged, or it could have some reference to the offstate lifetime of the device (although I doubt this one, since 1000 hours is less than 6 weeks). Mike 
May 07, 2010 by rajabalu21 
Mike, Thanks for the clarifications. I looked up the discharge curve of the capacitor and it looks like an exponential decay (may be linear with a steep slope) and now I understand that it cannot be used as a power source. Also many thanks for the clarifications on ESR and endurance. Yes. I was thinking why would someone buy a component that has a useful life of less than 2 months. Raja 
Please log in to post a reply.
Did you know that a microcontroller can measure an RC time constant? Learn more...
