Wrapped up in a package

Outer packaging

I never like to make projects permanent, because then I don’t feel as able to make changes when I want to. But the solar charger system has just been working so well, and with the addition of my new “poor man’s li-ion” charging scheme, I decided to tack it down. I put it on a basic perf-board. It’s all contained inside an altoids tin that I had originally set up with my voltage panel meter. I had always planned on putting the entire circuit inside, but until today, it was just a glorified volt meter.

I left a lot of space on the board since I’ll be adding two other features to the charger once I acquire two key components. First, the over-voltage cutoff circuit. This is a simple circuit that will simply drain off most of the current once the voltage on the out line is greater than 5.1. To do this properly, I need to get another switch, since I want to be able to disable this feature under certain circumstances. For instance, under situations other than battery charging, it’s not very useful.

The second extra feature is a constant current out. For now, I’m accomplishing this with another LM317, but I want to be able to control the level of current (since various devices may call for more or less). To do this, I need a variable resistor rated for at least 1 watt, but preferably 5. That way I’ll have more than ample breathing room, and power could be scaled later.

I’ve also dropped off my capacitor buffer because I don’t have a lot of capacitors on hand, and to get the capacitance I wanted, I would’ve needed to use 4 of my biggest 150uFs. That also takes up a lot of space, so when I get the variable resistor, I’ll probably just get a few 1000uF capacitors.

I’ve also abandoned the heat sink I had been using on the first LM317. First, it took up too much space. Second, I had only really needed when the array was configured for 12 volts. Those were experiments only and I’ve found I’m using the array at 6 volts most of the time. I’m sure I could find a heat sink that would fit, but I’m not going to bother until I come across a situation that requires it.

Oh, and my poor man’s li-ion batter is basically an old cell phone that I pulled the battery from. This particular battery setup was fortunate enough to contain some raw power points on a small circuit board attached to the battery. So I soldered on two extra wires to the main battery connector points, re-assembled the battery into the cell phone, and added two holes for the new wires to come out. So now I can take the normal cell phone charger plug, and hook it right into my charger system and then charge the phone. To draw power out, I just use the two wires I soldered on. The phone never has to come apart. Li-ion charging made easy (for now). As an added bonus, the phone obviously still works, so I can just turn it on to see the state of the battery.

5.2 Watts

6.5V full parallel micro solar array driving high-draw incandescent flood light bulb

After a lot of soldering and changing stuff around, I made my solar array portable enough to move it off the windowsill and outside. Out from behind the IR-blocking glass, it really shows its true potential. The short circuit current in the full parallel configuration is 0.8A at 6.5 volts. The best load I could find to test the power output under load was an incandescent bulb from an old flood light. I always keep this bulb around as a test load because it takes some serious ‘oomf’ to drive it. As an idea of the power, 3V from a pair of AAs will make it glow slightly orange. To get it to light at full brightness like in this image, real power is required. So, as you can see, even under load (probably the heaviest kind of load I’d ever even feel the need to use) the power stays up there. I’m comfortable classifying my array as a 5-watt array. I’ve been able to do a lot more reasonable things with it too. It has more than ample power to fully drive my portable CD player. Oddly enough, the CD player is a bit of a pig especially when the motor is spinning up. Draw can be up to 0.5A for several fractions of a second. To help overcome this, I’ve added a modest capacitor bank (about 600uF at this point) to the output to help buffer loads like the CD player that burst power requirements. Even that small capacitance often means the difference between the CD player working and dieing before it can fully spin up. Ironically, this newfound power means I really can’t charge my batteries anymore. The charge current is simply too high to charge NiMH unattended and without charge circuitry. I’ve ordered a few different charge ICs to start experimenting with fast charging. I can dial down the charge current using the LM317, but the reduction is so great, the potentiometer is not precise enough over the narrow range.

I’ve actually designed a constant current driver using a second LM317. However, like the voltage control, I want the current driver to be variable as well. This presents a problem as the potentiometer used there needs to have a low low resistance (around a few ohms) but also a high power rating. A 1/4 watt would be out of the question. Even 1/2 watt would be pushing it. I have seen versions available that have the right range, and are rated to 5 watts which would provide more than ample breathing room.

Interestingly, I’m trying really hard to think of alternative uses for the power of solar arrays, especially uses that can be directly used (as opposed to the normal charging of batteries for later). One of the ones that’s more interesting to me is driving refrigeration or ice making. This is actually a form of very useful energy storage. As kind of a gag, once I realized the amperage I was getting with the array, I actually hooked up my thermo generator in reverse. I was shocked how cold it got, in mere seconds. Ambient was around 65F, and after less than 30 seconds, the burn chamber (now the cold side) was dropping below 50F. That’s a long way to go for getting ice (I think the current would need to be far higher) but it’s a very nifty re-use of a power generating device. I think I might be able to use it to cool a can of soda or something. The chamber is almost exactly the right size, and with a little insulation wrapped around the edges, it might work.