Re: Load sharing of solar panels and batteries in solar boat.

Although I am a mechanical engineer by trade, not an electrical engineer, I 
do have contacts that can help you answer your question.  

First let me address your question from a mechanical standpoint, since I am 
also a card carrying member of the ASME, sponsors of the Solar Splash. I 
have also witnessed the vehicles that run in the solar car cross country 
race.

This Solar Splash contest is all about efficiency.  To win, you not only 
have to build a man carrying boat that runs on solar and battery power, you 
have to get the most speed for the energy available.  If all of the entries 
are equal in efficiency (they won't be, but the top few places will be 
close to equality), then the question is how to get the most from the 
incoming solar energy.  If it is a totally sunless day, the minimum speed 
would be that which would drain the battery totally during the 2 hour 
endurance race, but not a minute too soon.  (That's kind of like running at 
Indy and planning to run out of fuel on the victory lap.  To establish this 
minimum speed, you will have to test your boat for efficiency and measure 
the health of the battery during the race, and hopefully leave a little 
margin for error.

If it is a fully sunny day, then you would like to run your motor at a 
speed which will just consume the battery as well as utulize all of the 
current available from the solar panels.  If you get the 480 watts allowed, 
this would be roughly equivalent to the 0.5kw per hour you can get from the 
battery, or double the sunless energy condition.  Determining this speed 
experimentally can tell you what your total speed variation would be during 
the race.  Again, with the top few boats which are equal in efficiency, in 
these scenarios it won't matter how the motor is fed as long as you can 
drain the battery in both scenarios and not waste solar energy.

If the sky is intermittently sunny and cloudy, you boat speed might vary by 
the solar incidence if you hooked them in series. (Remember, mechanical 
engineer talking here.  There may be reasons why you wouldn't want to do 
that.)  When the sky was sunny, the boat would go faster, and slow down 
when the sun goes behind a cloud. This may be what you are trying to 
achieve with your DC-DC converter. 

Let's say you adopted your idea to vary the voltage to the motor to draw 
from the battery during the cloudy time and draw less battery energy during 
times of full sun.  But that assumes you know the weather patterns during 
the race.  You might actually find the local weather radar data on-line and 
look at the weather just prior to the race and what weather fronts might be 
coming by within the 2 hours.  If that sounds complicated, it is.  (The 
original Sunraycer car from GM used a similar strategy in it's original win 
in Australia.)

The only real advantage of a constant prop speed from such a strategy is 
that it might be more efficient than having the motor speed up and slow 
down.  I can imagine the prop churning when it gets a sudden boost of power 
from the sun, and the speed of the boat coming up to speed slowly.  
Assuming you have a working boat, you could vary the speed between the 
battery only mode and full sun mode and see how much efficiency loss you 
get versus a constant speed in between.  If there was truly a gain in 
efficiency by running at a constant boat speed between the lowest and 
highest operating speeds, then your scheme to control the voltage has some 
merit.  On the other hand, if the difference in efficiency is only a 
per-cent or two, then it might be better to spend time trying to reduce the 
drag of the boat or improve the prop efficiency by several percent.  Just 
determining the correct prop speed for a given size prop might take 
experimental time.

All of the above is really trying to answer the question of why you might 
want to control the voltage to the motor other than a simple parallel 
circuit.  I would say it is more important to get the boat up and running 
for trials and see how efficient it is before spending a lot of time trying 
to make an elaborate control circuit.  Your race strategy should be to 
exhaust the supply of the battery, but not too soon.  You might simply have 
a meter showing the solar panel output and battery health and let the 
driver provide the control, guided by people on the shore making 
calculations as to what is the correct setting for that portion of the 
race.

When designing anything, always take a step back and see how what you are 
working on applies compared to the big picture.  Sometimes you can't see 
the forest for the trees.

Now to attempt to answer your electrical question.  My electrical engineer 
associate responded to your question as follows:

"This  is an interesting problem and it appears that Nathan is fairly
astute.
I have never worked directly with Solar Cell/Battery  systems but that
doesn't slow me down in terms of making comments.  What comes to mind is
that there are DC-DC converters that are called Buck-Boost converters. 
(This refers to the circuit topology used) If the source voltage is higher 
than the required output the converter works in the Buck mode such that the
output is lower than the input.  If the input is lower than the desired
output then the converter works in the boost mode. Therefore it may be
possible to bring the voltage from the solar cells into the battery through
a diode, with the anode of the diode toward the solar cell and the cathode
toward the load. The converter would receive its input from the solar cell
and send its output directly to the load (battery). If the solar cell 
output voltage is less than the battery the input of the DC-DC converter 
would not see that because of the diode. If the solar cell voltage is more 
than the battery the converter would work in the buck mode.  Sounds like it 
might work although there are some details like filtering (which I can 
explain in more detail if necessary).  I believe that you could pass this 
along to Nathan to give him some hope.  Remembering that I am not familiar 
with these solar cell systems.  As he has correctly stated, there have been 
a lot of systems built so I'll bet someone has solved this issue."

Hope I have helped.  Feel free to ask more detailed questions if you like. 
 We may be able to correspond directly.  I assumed you have a team 
involving mechanical as well as electrical engineers.