My drone has a flight time of 15-20 minutes, but nevertheless goes through batteries at a tremendous rate. I can have extras standing by, but after a while it gets to be a bit ridiculous with the amount of batteries (lose track of which ones are dead/alive, have to carry them all around, etc).
Lithium Ion/Poly battery chargers are fairly expensive, and for the most part very very slow. Also consumer level models designed with lots of generalities, not taking advantage of special aspects of a given battery role, so they tend to charge conservatively. And even then they can only charge one "battery" at a time, even though the battery may consist of one or more cells.
I wanted to bypass all of this and use my electrical knowledge and design a super fast charger. My first challenge I had to solve was I needed a high power DC source. You can buy expensive lab versions of these, but a computer PSU works just fine! So I acquired an old 1.2 kW server power supply for the task at minimum expense off of ebay. The power supply indicates that it supplies a rail voltage of +12. My batteries are 3 cell lithium ion batteries have a terminal voltage fully charged of around that. Great! I might not even need to convert DC-DC.
If any of you have ever worked with computer/server power supplies before, you know that you have to get the power supply to turn "on" before you can start pulling power from it. So I had to find the signal pins on the PSU, and make a switch to enable it to turn on at will. After some short searching online, I found the method with little issue. After measuring the terminal voltage of the supply, I came out with 12.4 V. This is almost ideal! The 3 cell batteries have a fully charged voltage of 12.6. So what this means if you are looking at the charge curve of a lithium battery cell, I will be able to charge each cell to about 90%.
Another big assumption I am making that I had not mentioned yet: the batteries I am charging are very high discharge RC batteries. Fortunately this means that they also recharge quickly as well and can take a lot of juice safely. For instance, the batteries all have a discharge rate of 30C. So 30C every second is 30 amps (1 Coulomb/second). Pretty crazy... Also to be as safe as you could be, the terminal voltage of the battery should never be much lower than the voltage of the charger. Otherwise the difference in voltage multiplied by the current has to be burned off somewhere, usually in the battery. If the battery overheats, you could have problems (ie 10 foot tall flames gushing out). I will have to test to see if heat is a problem in the final implementation. I experimentally verified that the batteries can take at least 30 amps of juice from my lab power supply. But I could not test over that because my lab power supply caps out at that wattage.
Next step was to build the charger. I built on wood for now. Wood does not conduct well, and that is a bonus. As long as heat doesn't get too bad and there are no sparks, I should be somewhat fire resistant.
I got some terminal blocks for power distribution. I used 14 gauge wire for interconnecting wire. This should be able to handle quite a bit of DC safely, and they will be really short. Ideally I should add more capacity at some point. Maybe two sets of 14 gauge, I don't know I will have to run the numbers at some point.
I also added a voltage indicator, because I need to know the charge status of the batteries while the process is ongoing. These can be acquired cheaply on ebay, I highly suggest ordering some if you like projects like this. They will come in handy.
So from here, I can hook up four different 3 cell lithium batteries and charge them in parallel! Turns out this actually saturates the power supply which is insane. The supply can give roughly 100 amps at 12V DC (1200W). I determined earlier both theoretically and then experimentally that the batteries when empty will pull upwards of 30 amps.
This actually works out in my favor! Let me explain. The power supply, when being close to saturation, drops its voltage to keep with the maximum power its capable of. This in turn closes the voltage gap of the supply (charger) and the batteries. I verified the batteries heat up no more than about 3 C throughout the process with my infrared thermometer.
I can watch the voltage meter go up until it hits 12.4 (terminal voltage) and then know that the charger has done all it can do. I can then top them off with a real (hehe) charger or I can fly as is. I usually fly as is. Sacrificing 10% of battery space for ease of use is OK with me. I can charge all 4 to that point in about 15 minutes. Pretty freaking legit.
I occasionally run a balance charge with my professional charger for the health fo the batteries. Especially before long term storage. This is necessary because while a + b + c = d, a probably wont be the same as b and c. Everything has been good for I would estimate about 40 or so charge cycles between 8 batteries. If there is any loss in capacity so far I do not see it.
WARNING: if you build one of these you do so at your own risk. You are playing with fire, literally. I leave them charging outside always to reduce fire hazard. That or I am in the same room as them throughout the charge process. My theory and understanding of what is going on is pretty sound i feel, but there is no logic in this charger. It is a brute force solution. So if something goes wrong be sure you are ready to pull the plug!
- Log in to post comments