September 2016
I am looking to experiment with “ultra capacitors” as a replacement for AA batteries. Is this possible to do and, if so, what kind of capacitors would be a good place to start?
Please log in to post an answer.
The short answer is NO. The main function of Ultra caps is for TEMPORARY backup of memory type devices (i.e., clocks in DVD players) during brief (i.e., less than a day) power outages. They are NOT suitable, nor designed, to replace batteries simply because, as they are electrolytic capacitors, they need recharging when their stored energy is depleted. Also, they're not really designed for current loads greater than a couple hundred microamps. This subject has come up a few times in the past couple of years and I believe N&V had an article comparing ultra caps to batteries (primary and rechargeable).
This is only practical when the current draw is very low. A coulomb is one amp-second. By definition a one-Farad capacitor charged to five volts can deliver five coulombs. A single AA battery can deliver at least 1350 coulombs. This is because the material of the battery is consumed when it is delivering current. The capacitor isn't consumed, it's just a tank for electricity. About the only practical use of a small ultracapacitor is to maintain a CMOS memory while the set is unplugged.
I have done this. I'm going to put the conclusion right up front for those not interested in reading the rest of this. Given the present prices for AA batteries and ultracapacitors, it makes no economic sense to do this at this time. But I rarely let such considerations stop me from having fun.
Ultracapacitors and AA batteries both store energy. How much? For a capacitor, the energy stored is 1/2 * C * V^2 where C is the capacitance in Farads, and V is the voltage across the cap in Volts. One farad-volt^2 is one Joule of energy. I take a Maxwell Technologies BCAP1500 P270 as an example. This is a 1,500 Farad, 2.7 Volt capacitor. It is cylindrical in shape and with a diameter of 2.4 inches and a height of 4.5 inches. Volume = 20.36 cubic inches. It is a little smaller than a soda can. When charged to 1.5 Volts it has 0.5*1500*1.5*1.5 = 1,687 Joules of energy. When charged to it's maximum 2.7 Volts is has 5,467 Joules. These cost $59.20 at Digi-Key in quantity of 1.
Batteries are not usually rated by the energy they store, but this can be figured out. I take a RadioShack AA NiMH rechargeable battery as an example. These are cylindrical in shape with a diameter of 0.55 inches and a height of 2 inches. Volume = 0.5 cubic inches. Batteries are rated in Amp-hours. These are 2.5 Amp-Hours or 2500 mA-Hours. To equate this to an amount of energy, I enlist a theoretical load that draws 125 mA regardless of the voltage applied. One can make an actual load like this using a current source, but it must be able to work at these low voltages.
Starting fully charged, I run this for 20 Hours. I note that the average output voltage of the battery during the run is 1.2 Volts. I do the following calculations: Average output power = 1.2 Volts x 0.125 Amps = 150 mWatt. Energy used = .15 x 20 = 3 Watt-Hours. There are 3,600 Joules in one Watt-Hour so Energy used = 10,800 Joules. These batteries cost $19.95 for a set of four, so they are $5.00 each. Also note that a battery will hold its output voltage fairly steady until it is close to fully discharged. When using a capacitor, the output voltage will decrease in a linear fashion as current is drawn. For a capacitor i = C * dv/dt so dv/dt = i/C where i is current, C is capacitance in farads and dv/dt is the rate of change of voltage.
Using the above numbers if I start at 1.4 Volts (1,470 Joules in the capacitor) and I have to stop at 1.0 Volts (750 Joules left in the capacitor) because my "load" (a portable radio for example) will not work below 1.0 Volts, then I have only gotten 720 Joules out of the capacitor. If my "load" can operate without problems at 2.7 Volts, then between 2.7 volts and 1.0 volts I can get 4,717 Joules (86% of a full charge) from the capacitor per cycle.
One of the great disadvantages of a battery is that any rechargeable battery has a cycle limit. Battery manufacturers don't like to point this out but one can only charge and discharge a battery so many times before it losses its ability to hold a charge and has to be replaced. It depends on conditions but I have seen numbers between 300 to 1,000 cycles. Whether this is important depends on what you are doing. This is where a ultracapacitor can really shine. According to the Maxwell data sheet, one can cycle these ultracapacitors one million times and it will not lose more than 20% of its initial capacity.
Battery Energy
Ultracapacitor Energy
Not looking good for the ultracapacitors.
Now let's switch to discussing an actual experiment. I was in a local discount store when I found some "Floating Ball Solar Light" gadgets. As an impulse buy, I bought three of these things. These consist of a clear plastic ball about 6 inches in diameter. Inside this clear plastic ball is what I will call a solar lantern. This is a translucent cylinder about 2.5 inches tall and 1.75 inches in diameter with a plate at the top and a plate at the bottom. On top of the top plate is a small square solar cell 1.75 inches square. Below the bottom plate is a small compartment. At the very bottom of the sphere there is a small rubber gasket that actually covers a push-on-push-off button which allows one to turn these things off when not in use.
My wife thought these things were the cat's meow as they floated around the pool glowing softly in the evenings. They each had different colors of LEDs in them. Unfortunately they only lasted a few days until the internal batteries went dead and they stopped working and went in the junk box.
Two other notes, locating the on-off switch at the bottom, below the waterline was a bad idea as the gasket tended to leak. The outer sphere was actually two hemi-spheres that were glued together at the equator. This also tended to leak allowing rainwater in.
A few weeks later I took one and cut it apart. Inside is a single LED that illuminates the lantern. In the base is a small PCB with a single "glop top" IC. The battery leads were soldered to the terminals of a AAA battery. I unsoldered the leads and disposed of the old battery. I made some measurements. Under conditions of darkness, the LED draws about 10 mA. Under conditions of full sunlight the solar cell can produce 38 mA of current. On the surface, it would seem this should work.
I wired a 1500 Farad ultracapacitor to the battery leads and set the unit out in the sun for a day. The ultracapitor had about 1.4 volts on it which is enough to keep the LED glowing all night long. I can't leave the unit outdoors for a long term test because it is not waterproof in its present state. I can't put it back together like it was because the ultracapacitor is 40 times bigger than the battery it is replacing. I'm sure that like all electronic components, ultracapacitors will get smaller and cheaper over time, but they have a way to go to match batteries. I hate to buy batteries because I know they will go dead and have to be replaced in a few years at most. There is a strong psychological appeal to something that you can buy once and have it last for your whole lifetime.
If ultracapacitors down get down to a size where they would fit where existing batteries fit, then they might start to displace existing batteries. At present ultracapacitors are used in other applications.
