Hypothetical Solar Cell Situation

[Astrokid248]

So I have a hypothetical situation I want you guys to tackle. A robot requires a solar panel array that can provide 300 watts of power at 28 volts. These panels have to fit in an area of 24"x48", and there can only be three panels. How would you build these cells? How would you wire them? Go as deep with this as you want.

[MrForbes]

Huh....tell us about how long the robot has to get power, and how long it can sit in the sun in standby mode recharging.

A solar panel that size will give you about 5 amps to charge a 12v battery, or about 2.5 amps to charge a 24v battery pack.

This is based on my experience with a solar powered underwater robot, which operates on 48v (four 12v 7ah batteries in series) at a max of 5 amps. We spend a couple hours charging it, then we can run it for about an hour. We have the battery box set up with switches to parallel the batteries for charging, and connect them in series to run the robot. Charging and running do not happen at the same time (since the robot was built for a night competition)

[Al Skierkiewicz]

Avianna,
Right now typical solar panels are in the range of 12 watts per square foot. (give or take a couple of watts.) Say you could find (and afford) a 15 watt/sq.ft. panel, you would need 20 sq. ft. of cells to generate 300 watts. The panels are available in several different output voltages which you can then wire in series/parallel to get the right output voltage. What you are looking for (at 28 volts) is a panel that can generate close to 11 amps total.
If you do not have a height specification, it is possible to mount two panels, one behind the other in the 24x48 space depending of course on what latitude you are designing this to be used at. The front one will shadow the rear at higher latitudes.

[sanddrag]

300 Watts in 8 square feet? All I can say is it must be high noon on a clear day at the equator, and even then, good luck. In this circumstance, if I'm not mistaken, the sun irradiates approximately 1000 watts per square meter on the surface of the Earth. You have about 0.74 square meters, so assuming 100% efficiency, that's a maximum of 740 Watts. A higher end PV panel might get you a 20% efficiency. so, you'd be looking at closer to 150 Watts best case scenario.

Move the panel out of the atmosphere (into space) and you'll get about 36% more power.

Anyhow, what Al said: what is the height available? I was assuming minimal height. If there's no height limit, you can do a lot more.

[jee7s]

I concur with the previous posters that you would need incredibly high efficiency (or a very large pocket book) to get that kind of power out of such a small area. For what I say here, let's assume that you either have a panel with such efficiency, or you've increased your area to reasonably achieve your power requirement with existing technology.

The electrical design depends largely on the type of solar panel you choose. Because solar panels are basically just diodes that are connected in series or parallel--series connections mean more voltage, and parallel connections mean more current -- the electrical characteristics vary quite a bit. You can get high current or high voltage panels (or somewhere in between), and that would determine if you principally wire them in series or parallel. .

Either way, you'd want to have maximum power point trackers (MPPT) on either each panel individually, or the array of panels as a whole. MPPT is like impedance matching, but a bit more sophisticated. An irradiated solar panel has relatively constant current production for a range of lower voltages. Without MPPT, you could end up with lots of current at a low voltage, which means low power. Having MPPT means extra hardware and higher voltages, but it optimizes the power transfer from the panel to the system that's being powered. Generally, this means having a power source or power storage on board (like a battery). Also, these higher voltages could introduce additional safety concerns.

(As an aside...you'd probably want a battery (or several) anyway so you can keep operating when you drive under a tree or a cloud passes in front of the sun.)

As one more in a litany of concerns, you'd have to prepare for heating effects. Basically, a solar panel is a big black sheet. That means it is very absorptive, so it heats up. As the panel heats up, performance changes. This means there would be some sort of temperature+irradiance+time de-rating relationship that would need compensation.

So, not an easy problem to solve...with many nuances and challenges.

[PAR_WIG1350]

sounds like a job for a scaled up solar engine, but that would require some considerably large capacitor banks, and I'm not even sure information on such a high powered version of the circuit exists since I've only seen them used in really small applications. Alternatively, one could try solar-thermal with a steam or hot air turbine or a Stirling engine, the panels are replaced by large curved mirrors or lenses that focus the light onto a point that one wishes to heat. Depending on how it is done, this might be a possible solution, but it could also be worse.

[Astrokid248]

Looks like I'm using bigger motors then, because it's meant for Mars, where the sun is definitely weaker. Thanks for the input everyone, I really appreciate it.

[techhelpbb]

Looks like I'm using bigger motors then, because it's meant for Mars, where the sun is definitely weaker. Thanks for the input everyone, I really appreciate it.

Can we ask what kind of motors you were planning on drawing that power with?

[DonRotolo]

A light concentrator could be useful here. Gathering the light from several dozen square meters to the allocated space can deliver 300 watts. But that's not practical for a robot...

Tandem solar cells have been measured with efficiencies around 37% (http://www.eecis.udel.edu/~honsberg/Approaches%20for%20Ultra-high%20efficiency%20solar%20cells.pdf) for a one-sun irradiation. So 8 square feet of panel could yield 300 Watts - but that's a very bleeding-edge proposition, likely to cost as much as a new house.

[Astrokid248]

I don't have specifics on the motors drawing power, but the panels themselves will be mounted on a folding structure. The motors I currently have are only powerful enough to lift the given dimensions. I'll need bigger motors to lift a larger panel. I ask because I don't know much about solar panels at all; I'm mechanical, so I designed the lift arms, folders, and lazy susan. Two others worked on the solar panels, but I'm not entirely certain we picked the right style or the right wattage.

[DonRotolo]

The motors I currently have are only powerful enough to lift the given dimensions. I'll need bigger motors to lift a larger panel.False. You can use a very tiny motor - say, one that would fit inside a "AAA" battery - to move a panel the size of a tractor-trailer.

It might take some time, though.

Do some research on mechanical advantage and understand the concept of Work (as in unit energy over time). Then revise your power requirements...

[Astrokid248]

Ack, you'd have to see the design. I know about mechanical advantage, and unfortunately, the space requirements mean mechanical advantage is actually nil. It's actually a really terrible design, but I'm sticking with it until I learn more about the panels themselves.
Also, what does gold spam mean?

[Tristan Lall]

Also, what does gold spam mean?
Some fool trying to get you to buy gold. Preferably from them if they're a multi-level marketer. From anybody at all, if they're a Ron Paul acolyte (http://www.youtube.com/watch?v=SvWBXKAYqyk&t=4m11s).

[DonRotolo]

Also, what does gold spam mean?
It's better than regular spam!

For your calculations, in one sun, a typical panel is around 10-13% efficient, and you get about 1 kW per square meter from the sun. Now, if you need 300 W in real time, that makes it almost impossible, but if the 300W is intermittent - meaning you sit and charge for a while, then expend the energy - then you have a potentially viable concern.

If I were designing a robot system that, say, autonomously chased deer from my back yard, I'd work to reduce my power budget before I tried to get 300 W from a solar panel.