Yeah I thought about that one, but I think your wrong. Size is not a problem, but the quality of the panel and how much percentage of the energy it can hold from the sun and pass it along. And since we only need energy for a toy car, we dont need much solar energy for a small canister of water. And the engine you are talking about, the fuel cell, can be buildt as small as possible as long as you have the room for an anode, cathode and an electrolyte, again depending on how much energy you need.
If the toy car is small and made by lightweight material, the energy needed to move it by the small motor, which I have built (ordinary electrical motor 1,5 V, similar to the ones in ordinary toy cars), is little. But these sizes and energy can be calculated and adjusted for the better with trial and error-
Now offcourse I can run the power straight from the solar panel to the electrical motor, but theres not alot of chemistry to be said in that experiment now would there, only physics (wavelengths, magnetism etc)
You must also remember that this is not a long circuit of electrical energy here, it goes form electrical to gauss, then from gauss to electricity then to the motor, so done correctly, the loss would be little.
No, the problem is not the size, but building the fuelcell which I have found a recipe on, and which is QUITE difficult (money, parts, equipment etc, simply an impossible recipe) and to build a lightweight solar panel. If anyone knows how to build a "simple" fuelcell which runs on hydrogen gaus and a way of creating an effective solar panel, I would be gratefull!