THE CHARGING CIRCUIT:
Let's go through this in steps. It is actually very simple but you have to follow along closely, especially as we go into the step on the following page. We start at TERMINAL BLOCK#1 and will continue clockwise around the circuit!
1) This is where you have options. We need a DC source of anywhere between 5VDC-20VDC for our charge. I use a 11VDC@1A power supply, but I occasionally use a set of mini solar panels that I have in my window. The choice is yours. Just make sure that when you plug in your DC source, you are making sure that you have the correct DC polarity for DC+ and ground (DC-).
2) We have a 0.1uf capacitor and a 100uf capacitor in parallel with the input DC line. We only really need these because this line is for the charging of the capacitor bank, but we will be using this input line to power our digital display and we want to make sure that this DC line is smooth and without extra noise. The 0.1uf capacitor takes care of high frequency noise, or rather, lessens it (Decoupling capacitor). The 100uf capacitor acts to smooth the input DC. These two capacitors are not really necessary but they are preferred.
3) The LM317 is a variable DC-DC power supply. Using a 240 Ohm resistor in parallel with the VOUT and the ADJ line, and a 5k ohm variable resistor from the ADJ line and ground, we can vary the charge voltage from the charge voltage itself, down to 1.25v. For instance, if we have 8v at the input, we can vary the output anywhere between 8v down to 1.25v. It is EXTREMELY important that your LM317 is properly heat sinked, as it will get HOT. The LM317 kit can be found here: http://cgi.ebay.com/DIY-LM317-Variable-DC-power-supply-kit-PCB-Parts-/180609634986?pt=LH_DefaultDomain_0&hash=item2a0d2c52aa
4) Varying the current to the super capacitor bank is the name of the game. This is where you have the opportunity to gamble. Since the super capacitors will literally suck up all the energy it is given until full (With >0.01 Ohm ESR), we have to limit the current from the supply, or else we're going to completely destroy our LM317 circuit. As you can see, we have two 2.2 Ohm, 5W power resistors, a jumper, and a SPST (Single Pull Single Throw) switch. If the switch is off (Recommended), and the jumper is not attached, then the charge limitation is 2.2 Ohms. Wait a minute! That is too small of a current limiter! You're still going to hurt your LM317!!!
Not the case! If properly heat sinked, the LM317 will get hot but it will withstand the stress if you have this 2 Ohm load. The output voltage will drop down but you will see it come back up as the capacitor starts to charge. We have three charge options here. If you have a charge of 4v or higher, make sure that you have the jumper off, and the switch off.A)
Charge limited by 2.2 Ohms when JUMPER=OFF/ SPST=OFFB)
Charge limited by roughly 1.1 Ohms when JUMPER=ON/SPST=OFF
When you add the jumper, you place the two 2.2 Ohm resistors in parallel with one another, bringing the parallel resistance down to half. Please note that these resistors get hot.C)
Charge limited by the line resistance and capacitor ESR only when JUMPER=ON or OFF/SPST=ON
If the SPST is switched on, it doesn't matter how the resistor jumper is configured. The only resistance between the output of the LM317 and the capacitor banks is the line (trace) resistance, and the ESR of the capacitors (Yet to be seen). This is where you have to have cohones! Again, your LM317 can handle this if properly heat sinked (Heat sink included in kit), as the output voltage will drop down to the cap voltage and start to charge. However, this should only be used for charges of 1.5v or less. If you are charging the bank from 0v to 5.4 v, it will charge relatively quickly using the 2.2 Ohm charge option. However, around 3v of charge, it will start to slow down. At this point, take the jumper off to limit the current to 1.1 Ohm. At around 4.5v, you will notice that the charge will slow down again. Flick the switch to charge the remaining 900mv, and you will have no problems. Truth be told, I've charged from 2v to 5.4v with the switch on, but it is NOT good practice, and I was risking my LM317.
5) We have two IN4001 diodes in series with the charge line. These are not used for any type of rectification, but rather to allow DC charge to enter the capacitor bank, but not allow for any DC to travel backwards through the circuit after the capacitor bank is charged. If we didn't have these diodes here, follow the circuit backwards. Regardless of whether the jumper is on or off, or whether the SPST is on or off, there is a path back to the LM317, and there is a 240 Ohm resistor in a series path with a 5k potentiometer and ground. If we stopped charging (without the diodes), the charge on the caps would leak back through the circuit to ground, making our batteries terribly inefficient. There are two diodes in parallel to share the current along the line. If you have 1N4007s, or any 1N400X diodes, they will work just as well if not better. There are factors such as thermal runaway that we could spend time worrying about with these diodes in parallel, but the charge time from start to finish for this circuit is literally 10 minutes or less , so we're not going to worry about that at all.
6) The jumper (JUMPER#2) like a lot of this circuit is a custom option. If you are not going to watch the digital display (Seen later) as your super capacitor bank charges, then you are going to want to follow this step. When you build this charge circuit, probe the output of the diodes (TEST POINT) with reference to ground using your multimeter. There will be a voltage drop along the diodes, so we need to make sure that we measure here, and not at the anode end of the diode. Since we have a 5.4v MAX capacitor bank, we DO NOT want to have a charge higher than 5.4v. Check the voltage here using the 5k potentiometer at the LM317. Turn the potentiometer until you see a voltage of 5.2v-5.4v, then consider using a bit of hot glue to set the pot to steady it. You may think, why use the pot, and not a fixed resistor? You can, by all means, but you may want to change the charge voltage down the road. Now, the jumper is here because on the other side of the jumper lies the capacitor bank. If you test the voltage here when you have the jumper on, you will read the voltage at the capacitor bank, not the voltage that it will be charging to. You only take the jumper off when you want to take a charged reading. Leave it on at all other times.