1KCube

With a little searching on Ti and Maxims sites, there appear to be many possible candidates for power components. I found a large number of regulator chips and modules that are within the range that I need. The battery charger options are significantly lower though, not many chargers have built in MPP for solar arrays, although Ti does have a few (3) purpose built chips for that. I'll need to dig through the datasheets a bit more to ensure I can get enough output from one of them. Most of these chips are designed for smartphones, and while that is all well and good, I anticipate needing a bit more power than that, especially during transmit cycles.

These logs are going to be 'stream of conscious' type ramblings for now as I get my thoughts in order on this project. In reality I probably don't have time to attempt this, but we'll do the best we can.

So, sourcing of the power components. 

Batteries: Most likely I'll go after hobby grade cells. Why hobby grade? Well, they are cheap, readily available, and most importantly, come in many different physical sizes and capacities. Flexibility in battery choice will make it easier to hit the weight target and give me some options in chossing the best capacity for the needed up-time. HobbyKing, BPHobbies, and similar discount RC retailers will probably be my main sources for now. 

Charge controllers: Hmm, most likely this will be a Maxim/Dallas part, but there is a chance that that Ti might have something usable in their product lineup. Ti has starting releasing a lot more power components since their acquisition of National Semiconductor a few years ago. I've used Maxim charge controllers in the past, but Ti's products are usually solid, so their could be some good competition here. 

Regulators: We've got even more options here. Ti, Maxim, and Linear (along with several others) make parts that could be good fits here. There are even some companies that make switching regulators that are drop in replacements for TO-220 package linear regulators. While that probably won't be the best option for the final flight config, it might be useful for early prototyping. 

SolarCells: This is a tough one. Spectolab (Boeing subsidiary IIRC) makes some darn fine cells (>25% eff), but even their B stock stuff is pricey. Stepping down to hobbyist grade cells will probably be a necessity, but I'm very worried about consistency of stock. It will be very hard to design a system if I can't get enough of the right types of cells to put together a full system with extras for spares. Lots of questions up in the air about this at the moment. 

So, how do we put together a high efficiency power system? Well, there are 4 main parts of a CubeSat power system. Solar Cells, Batteries, Chargers, and Regulators. All of these components need to be chosen and designed to maintain the highest possible efficiency. 

How do we do that? For starters, we need to be smart in out component selection. Lithium batteries are an obvious choice for this due to their high power density. Unfortunately, they have a nominal output voltage of 3.7V. While that isn't really a problem for 3.3V electronics, it's unlikely we will be able to design a full satellite that only utilizes a 3.3V bus. 

Unlikely, but possible. This is something that should be investigated. If a 5V bus is not needed, it could simplify the design of the power system tremendously. 

Overall, the fewer regulators used, the better. And oh boy, these will be some regulators. You won't find linear regulators here, no sir, not at all. This will need the highest efficiency switching regulators that can be found at a reasonable price point. Will this create other design challenges (noise, increased component count, debugging, etc), probably. Despite the increased effort needed, this project isn't really freasible without switching regulators. 

This is a rather ambitious project, especially for me considering I don't really have the full skill set necessary for it. But hey, that's part of the fun, right?

In my mind, the biggest hurdle in this project is going to be power. Satellites live and die by the amount of power they are able to absorb, and if you look at the available surface area on a 1U cube, it really isn't that much. Especially when you consider that you will loose some large chunks of the surface area to cutouts for connectors, structural elements, and the fact that your solar cells will not be able to cover the surface perfectly. When it is all said and done, I will be surprised if I end up with something close to 70% surface coverage. In reality, it will probably be lower.

What does this mean? It means that the Power System has to form the foundation of this project. Without a robust high efficiency power system, this doesn't stand a chance at succeeding...