Let's Make Robots!

Need help with a solar-powered boat robot


Two friends and I are starting to make a solar-powered boat for a competition (here's the link: http://techfest.org/competitions/torque/solarsplash/), and we need a little help. We're beginners with no previous experience.

A few important rules:
  1. The objective is to construct an autonomous solar boat that completes a water track in the least possible time
  2. Length of the track is 1400 mm
  3. Source of light is three halogen bulbs of 500 W
  4. There's a tunnel on the track, so there will be relatively less light under it
  5. Size constraint for the robot is 250 mm x 200 mm x 200 mm
  6. Area constraint for the solar panel is 150 cm^2
  7. Needs to have a mechanism to discharge any stored energy before the race

Will (4) mean that we'll need to have a battery to store some energy that can be used when under the tunnel? Or can we get around it by placing the solar panels inclined at the ends of the boat?

As for (6), is 150 cm^2 enough to run the motors? We plan on using 2 or 4 pager motors (each operating between 1.5 - 3 v). If yes, what specifications should we be looking for while buying the solar panels? We think that buying 4 small panels instead of 1 large one will be better as it will allow us to place them at appropriate positions and inclinations.
But if 150 cm^2 isn't enough, we'll have to use a solar engine. We tried looking for the 1381 voltage detectors in local shops but to no success. What's a good alternative? If we do decide to order a 1381 online, what rank (i.e the trip voltage) would suit us best? And what will be the specs for the solar panel in this case?
What do you think? 

Coming to the propellers, we are undecided between fan-like propellers (like the ones in SMPS) to be used under the boat, and paddle wheels to be used on the sides of the boat. We could test both out and see which one's better, but it would be helpful to get some advice beforehand. And any advice on how to waterproof the motors/panels if we decide to go for the under water propellers/paddle wheel propellers respectively.

Implementing phototrophism (automatically turning the panels towards maximum light) is an option, but we thought it would complicate the bot. I guess this will introduce the need for a microcontroller? Also a part of the solar energy would go towards turning the panels.

How do we put it together? We were thinking of placing two inclined panels at the back (to get light when under the tunnel), one slightly inclined in the front, and one flat in the centre. Should we go for 2 solar engines (one connected to the panels at the back, and another to the panels at the centre and front), or just go for 1 and supply current to all 4 motors with this one? The former will help in keeping the boat running most of the time as the rate of incoming solar energy will be different for the two?
Sorry for so many questions! And thanks in advance :)

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Thank you both for your replies.

I don't know much about servos, but from what I've seen I suspect they need a microcontroller to run? We planned not to use one since it introduces extra complexity. This is our first bot, so we want to keep it simple.


Big capacitors look like a good idea. But how exactly do I use them? How should I set it up to 'fire' at the right moment? Will I have to use something like a current detector (or light detector?) that will close the capacitor circuit when the current / light goes below a threshold value? Looks sort of similar to a solar engine.

Since you want to run the motors all the time, there's no need to trigger the caps, they'll work automatically.
When the solar panels are lit up, they'll charge the capacitors, but when the solar panels get dark the caps will discharge back into the circuit. As long as the solar panels are powering the circuit, the caps will just sit there and hold their charge. The lower the solar panel output voltage is compared to the cap voltage, the faster they'll discharge, conveniently 'topping up' whatever the solar panels can't provide (until they run out anyway).

Make sure to put a high-current diode (1A or higher) on the solar panel leads going to your circuit. When the panels are lit they're like a battery, but when they go dark they'll actually consume power stored in your capacitors instead of charging them. The diode makes sure that current can flow out of the solar panels, but can't flow back the other way.

You don't need microcontroller for servo. You have to modify servo anyway so you can take away controlling electronics and use just the motor and gearbox.

3 x 500W Halogen bulbs should make for quite good lighting for the solar panels, since they're pretty close. If possible you should get a similar bulb and see what the voltage and current output are for your solar panels under similar conditions. The solar panels you linked to seem like a good choice, ~11.4mW/cm² is pretty high for common encapsulated panels, and with 4 panels you'll be close to but not over the 150cm² limit.
A solar engine will actually cost you some of your energy, so don't worry about using one unless you really need to. Hacked servos are also nice, since the built-in motor drivers are usually quite efficient, although you will probably find that you'll need over ~4.8V to drive them well.
Pager motors are cool because they're small, but they usually aren't very efficient compared to other small motors. I'd look into hacked servos and high-efficiency lightweight gearmotors like the ones Solarbotics.com stocks.

The tunnel is short and quite far down the track, so you can probably just coast underneath it using momentum, provided your 'boat' is fairly streamlined. The tunnel is also very low, so you won't get much light inside at all. If you angle your panels to catch light inside the tunnel they still won't generate much power, but on the other hand they'll generate less power during the open parts of the course - most likely a bad tradeoff.

If you want to store a little energy just to keep the boat going inside the tunnel, grab a few big capacitors. They'll charge and discharge quickly enough to be worthwhile, unlike rechargeable batteries which are too slow. Capacitors are also lightweight and efficient, so they won't slow you down.

I personally would go with the paddle wheels. I am implementing them into my own project. You could use a servo to power it and get some pretty decent speed out of it. The servo would obviously need to be modified for continuous rotation and you can pretty small and cheap servos with low power draw. There are many ways to waterproof a servo. Just google it and find what works best. I use a slightly varied method of this. http://www.societyofrobots.com/actuators_waterproof_servo.shtml