Skip to Content

Solar Cells + LEDs = Tricky!

As described in the last post, the game concept "The City" will light up LEDs (ie supply power to households  such that players must add capacity (ie build up their power plant) so they can light more LEDs.  In this way, players must manage their electricity production and distribution.  I hope that we can use solar cells as our power plants for a variety of reasons (number 1 reason: they are cool).  So this time we're going to dive into solar cells, specifically:

1) How they work generally

2) How they work with our LEDs (in parallel, in series, of different voltages, etc.)

3) How the system might integrate into the game concept

Most of what is described below is just a regurgitation of Dale Grover's (co-founder of Maker Works) explanations to me on how this stuff works, along with my own data collection, so thanks Dale so much for your ongoing help.  In addition, here's some additional background info to learn more:

- www.icpress.co.uk/etextbook/p276/p276_chap1.pdf - excellent overview of the sort of things discussed below relating to solar cells, many of the same or similar graphs.  If you've got 100 bucks lying around you can by more than just chapter 1!

Equipment and Setup

DC power supply for generating LED curves

6 different solar cells

Soldered small wires to the backs of the solar cells and plugged them into a breadboard!

Watch out for 500W light bulbs, they get hot! I cracked my solar cell when I stuck in a few inches in front of the lamp!

Six different types solar cells were purchased from Futurlec (http://futurlec.com/).  Data is on the 6 types are listed below:

6         SZGD4020         3.0V 25mA Solar Cell                                      $1.20
6         SZGD3131         2.0V 45mA Solar Cell                                       $1.25
6         SZGD6161         2.0V 92mA Solar Cell                                        $1.90
6         SZGD5433         3.0V 45mA Solar Cell                                       $1.60
6         SZGD4026         4.0V 20mA Solar Cell                                       $1.20
6         SZGD6030         0.5V 280mA Solar Cell                                    $1.50

The above voltages and current ratings are based on a light level equal to 1000 W/m2 (Watts per meter squared), which is "standard testing conditions" for rating solar cells.  The technical term for this is irradiance.  Since I don't have a spectrometer to measure the exact light level, I used a flood light and placed the cells close to the light.  Since the volt and current output in front of the flood light is fairly close to the rated amounts, I can reasonably assume that my standard setup with the flood light is also close to 1000 W/m2.

Solar Cells

Each solar cell and LED has it's own V-I curve.  In the case of solar cells these curves describe, at a given irradiance (light level), what the current and voltage will be in the circuit.  Using 4 of these solar cells in parallel (rated at 3V and 45mA under standard testing conditions of 1000 W/m2 irradiance), the circuit was tested for voltage and current under two light conditions: sitting ~6'' in front of a 500W flood light (as close to full solar conditions as I could get), and sitting on a table under ambient indoor light (florescent indoor lighting).  The data was obtained by placing a known resistor in series with the solar cells, and then using the multimeter to measure the current of the circuit.  Since we know that the current must be the same between the multimeter and the resistor (they are in series), and we know the value of  the resistor, we can calculate the voltage using V=IR.

The curves generated below ("V - I Curve (4 x 3.2V 45mA solar cells in parallel)") by these measurements are very different to the curves generated by a normal battery, which produces a constant voltage (CV).  An example 2V battery is overlayed in the graph below - notice that regardless of the amp draw (mA), it produces a constant voltage (NOTE: there is an internal resistance for all batteries which is defined by the battery chemistry, so there is a maximum current which a battery can produce without impacting the voltage, but it's not relevant at the current ranges discussed here).


LEDs

In the case of LEDs, the V-I curves describe at a given voltage what is the current draw of the LED will be.  Below are two curves calculated for a blue LED and a red LED (graph "V-I Curve for blue and red LEDs").  This data comes with most LEDs in the technical specifications, but it's always helpful to actually test and see the data points yourself.  The data was collected by placing an LED in a circuit with a DC power supply with variable voltage.  A multimeter, set to measure milliamps, completed the circuit.  The DC power supply provided the voltage measure, while the multimeter provided the current measure (see diagram to the left).

If too much power is pushed through either of these LEDs they will burn out immediately - for these particular LEDs this can happen above about about 100mA.  Operating above their suggested conditions (about 2V and 20mA for red, and 3.3 and 20mA for blue) will have decreased LED life.

The brightness of the LED varies with current.  So, for example, a red LED will light up even at 0.1mA, but will be very dim, 1mA is brighter, 10mA even brighter, and anything above 20mA is pretty much the same (full brightness).

As someone learning about electronics, it's mentally hard to accept that these diodes operate only along these curves.  For example, the LED can never operate at Point A on the graph because it's not along the LEDs curve.  So that means, in a system with constant voltage the LED will have a constant current.  For example, an LED hooked up to a power supply with constant voltage like a 1.8V battery, will draw a constant ~3mA.  If you decided to switch out the battery to a 2.1V, the resulting current draw would be ~25mA.  So using an LED with a battery is pretty straightforward - you need a constant voltage which produces a reasonable current draw in the LED.

However, as we saw with our solar power curves, an LED in a circuit with a solar cell is a bit more complicated.  Overlaying the previously discussed curves (LED and solar cell) we can identify exactly what the current and voltage of our circuit will be by identifying the intersection points (see graph below).  Let's take a few examples:

1) Solar cells using flood light with 1 red LED - 2.1V and 63mA (intersection of red and yellow curves).

2) Solar cells using flood light with 1 blue LED - 3.1V and 8mA (intersection of blue and yellow curves).

3) Example 2V battery with 1 red LED - 2V and 10mA (intersection of 2V battery curve and red curve).

4) Example 2V battery and 1 blue LED - 2V battery curve and blue curve - these curves don't intersect!  So, the LED doesn't turn on, and no current flows.

5) Ambient indoor light and 1 blue LED - green curve and blue curve - these curves don't intersect!  So, the LED doesn't turn on, and no current flows.

Ok, I think we've got the idea.  Now, what happens when you have 2 red LEDs in parallel, or 1 red and 1 blue in series?   Oh gosh, at this point I sort of went crazy and I'm not going to explain in detail, because I'm at the point of recreating an intro level electronics textbook which probably isn't productive for the blog (though the mental exercises are helping me immensely).  But with the basic concepts outlined above, you can probably figure it out.

Application to board games

One of the key interesting elements here is that you can put a red LED or blue LED in the same circuit, without a resistor, and both will work and not blow out.  Because solar panels work as such low power, it is a challenge to blow out an LED.   You just can't do that with a constant current device like a battery - a 3V battery will never be able to run a red LED, it will always blow out (unless you put a resistor in series, of course).  Now think about how this relationship may apply to games (player 1 has a circuit with blue LEDs, player 2 adds a red LED and POOF all the blue LEDs turn off!  This is similar to a capture mechanic in go, for example).

On the negative side, the solar energy available inside (ambient indoor light) is really really low, and while a red LED will light up it's very dim (for example, a red LED and 68 ohm resistor dissipate .5 milliwatts of power under ambient light using 4 solar cells in parallel, while the same setup dissipates 143 milliwatts of power under the flood light!).  Also, it's hard to create a simple circuit with a clear binary LED on/off, because with sufficient voltage any number of LEDs in parallel will still be on, they'll just get dimmer and dimmer for each additional LED added in parallel.  For the purposes of the game, we want a cut off point - add one more LED and all LEDs go dark (a blackout).  To get that result, more complicated electronics is required (maybe that'll be a future post).

So what's next?  Here's my thoughts:

1) Having very long wires connecting the solar panels to the board, so that each player can position the solar panel right next to a light in the room (a lamp, overhead light, etc.).  Yes, this is kind of the "you need a lamp to play this game" solution, but it's required to get enough power for nice, bright LEDs.  Also, it adds an interesting competition for good "light space" in the room.

2) Screw solar panels and use a microprocessor to do the work.  I'm currently working to develop an Arduino based process so that each player plugs into the Arduino, and uses some simple switches to upgrade from a "coal plant" to a "nuke plant" to a " microwave plant" or whatever, with each upgrade increasing the maximum current available to their individual circuit.  Also, the current and voltage could be displayed on the unit, which is a really useful direct feedback mechanism so that players can develop intuition about the key variables in the game, and the Arduino can solve the binary LED on/off problem.  This requires more equipment and is less cool (in my mind) because I really like solar panels and think they are interesting little buggers, but it also adds some useful elements and everyone loves Arduino's which is a marketing plus.

3) Develop a game which may be more appropriate to the natural tendencies of these variables - we may be trying to put the proverbial square peg in a round hole combining electricity with The City.  A more abstract game may be more appropriate (like the blue and red LED example, I described earlier).  I'm certainly not ready to give up, but I'm sure there's many other applications of these concepts for other games.

As always, if any thoughts above piqued your interest or if you'd like to correct me if I erred, comment or email me at gbathree (at) gmail *dot* com.

Next Post

I'm hoping to have some preliminary artwork done, and to introduce the illustrator (who's doing great work).  Maybe make some progress on the Arduino front, and it's likely that I'll have some quality, easy to use, connectable telephone poles by then.  Yes, telephone poles are exciting.

2 Responses to “Solar Cells + LEDs = Tricky!” Leave a reply ›

  • Greg,

    I have an instructor who is working on a similar project with the NXT and green energy curriculum. Would you be interested in working with a group of people to develop a curriculum we could use during a class or better yet, an arduino summer camp week.

    I'll have him come this Friday and we can talk more but I would like to help you out with your project and will hopefully have some more time to do so in the near future. I am working on structuring my company so that I have a solid group of administrators who manage various education sectors. At any rate, I hope this will free up some time so that I can work on projects such as yours.

    Awesome!

    Katie

  • Profile

    That sounds great! Actually, in putting together this post I had some more thoughts on alternative, more educationally appropriate games which rely on the same kind of functionality. Let me know when you have time to get together and discuss -

    Greg

Leave a Reply

You must be logged in to post a comment

Contact

Please send questions, comments, and inquiries to greg (at) austiclabs.com

Office located at Maker Works near Ellsworth and State st in Ann Arbor at:

3765 Plaza Dr.
Ann Arbor MI 48108
tel: 919 545 1083

Featuring Recent Posts WordPress Widget development by YD