Solar Camping Light

[Carbon2010]

Some ideas I have been trying to sort through.

I'm looking to construct a small solar light for when we go camping. As much spread as possible. I have selected a few components already - more so because it's what I have on hand.

-12v 2W solar panel (used to trickle charge a car battery- 2 available)
-2 3W cool white Luxdrive crees.
-(1)Source of power
-Lm317 chipset for current regulation
-Various resistors/case/wire/Ect.

(1) I have a few options here. I'm looking for something reasonably bright but with at least 6 hours run time. Options are:
-3 hybrid NiMH 9V batteries
-1 3.6V 4500MaH NiMA battery pack (gutted from an old box)
-Your personal recommendations


Now if you guys can give me an idea on how to go about this that would be great. The basic idea I have now is the panel is fed into a simple current regulation circuit using the LM317. Using C10 i'll select the correct resistor, although this doesn't take care of overcharging, just regulation (I think - please let me know otherwise)

The LEDS simply attach to the Power option I choose. Panels charge the battery - Leds use the power.

If anyone has a better idea PLEASE post it up. I would love the feedback.

Thank you!

 

[Carbon2010]

Well I looked into the 9v...Defently not suited for this project. With only 200MaH of juice, I can't expect full power for very long...

Battery options anyone?

 

[Carbon2010]

Well It looks like the 3.6V battery pack is going to be the winner. I'm looking to see if I can acquire a second pack and run them in series to the two leds. Current stays the same so the C10 for the two will be around 450Ma...which will put me right around a 3OHM resistor.

Although I'm still not 100% on an overcharge monitor. Will this circuit take care of that?

 

[Illum]

A couple questions

How much hours of sunshine is there per day? this is full unobstructed, cloudless days.
Does this project require both CREEs?

12V 2W is likely difficult. to compensate for loss, Voc of "12V" panels usually reside in the 16-17V range, with Vcc being around 13-14V, 2W at this range would make Icc give or take 150ma max, parallel panels would yield about 300ma max.

6 Hours? not likely, my portable solar setup uses 30W of panels but so far testing has not been able to produce 6 hours of runtime... granted I'm running 3 XPGs at 1A, but still... even in days I thought it was sunny enough my panels were having trouble pushing out the current I calculated it for.

 

[archer6817j]

This is a little out if my depth, but I just got back from a 10 day back country trip. I took 2 different 18650 lights, a charger, and a Brunton Solaris USB compatible solar panel. I'd throw a battery on the charger in the AM and when I'd come back from fishing in the afternoon...charged battery! It was awesome. The small panel only puts out 400ma (weighs 4 oz) but my USB charger maxes out @ 500ma anyway. I'm all about DIY but this is one commercial solution I'm very pleased with.

My buddy got the larger Solaris that is twice the size but can put out 1000ma. Both of them worked great and this is my first experience with any kind of portable solar panel.

 

[yazovyet]

You may wish to consider running fewer LEDs or running the 2 of them at lower current levels. Why? Because you have two 2 watt panels for a total of 4 watts. If you run two LEDs at 3 watts thats a total of 6 watts. So for every hour of run time you need 1.5 hours of charge time. And that assumed your charging and regulation all worked at 100% efficency and the panels get perfect sun, so 9 hours of perfect charging for 6 hours of run. If you don't plan on running it for 6 hours a day but just want the option then carry on, but if you want to run it 6 hours every day (or more likly night) then consider changing the LEDs.

I'v never looked at over charging and stopping conditions for battery charging so I cant help you there.

Lastly, I wouldn't normally complain about the use of capitals but I can't help but mention something about this (because you took the time to use capitals so I figured you would want to use the right ones and I just feel compelled to say so) :
"only 200MaH of juice"
'M' stands for mega or 1,000,000
'm' stands for mili or .001
'A' is the symbol for amps
'h' or 'hr' is generally used for hour
So you'd want to say '200 mAh' or '200 mAhr'.
Again, I'm not complaining or trying to be mean, I just figured you might want to know, I'd want to know if I was doing something like that... then again I would be too lazy to use capitals at all.

When you say "Current stays the same so the C10 for the two will be around" what does C10 mean? I've seen some thing else like that on these forums that i didn't understand.

 

[Illum]

The 12V 2W panels in essence are designed to trickle charge 12V automotive lead acids, which usually comprise of starting batteries rather that deep cycles. :thumbsdow The minute outputs suggest its role is simply to offset natural discharge displacement rather than contributing an actual charge replacement...I've not used them on depleted cells to see what happens, so I can't really say they don't work at all.

NiMH 9V won't work. This is because internal construction uses AAAA cells, which have a nominal capacity of less than 300mah and will sag like crazy if you pull anything more than maybe 1/2C out of them. :thinking:

The easiest solution I can give is as follows:
Parallel up the two 2W panels and make sure your output is 250ma short circuit
Invest in one of these: http://ledsupply.com/03023-d-e-350p.php

• The potentiometer is supplied on a wire harness, bridge both ends of the potentiometer to a switch. The device will shutdown when the switch closes the circuit. The switch will pass virtually no current so you don't have to worry about insulating it. :hahaha::tinhat:

Invest in one of these: http://www.batteryspace.com/sealedleadacidbattery12v33ahs.aspx

• Why Lead Acid? It's cheap, easy to use, :twothumbs very tolerant of noobish charge/discharging practices, and relatively safer than li-ion. Why 3.2AH? Because its cheaper than 2.3AH and it is impossible to predict if you may need it on "high" sometime in the future. Yeah it weighs 3 pounds but hey, it does what you want when you want it to.

How did I come to this conclusion? Read below.

Finding dedicated LED drivers that will yield less than 350ma is not particularly easy if you are not inclined to design the circuit yourself.:green: A cheap and effective way would be using a buckpuck as these drivers are encapsulated and dimming function integrated. :naughty: Lets just say you've decided to drive your LEDs in series. Assuming your CREEs are XREs, :thinking: based on its datasheet you will need to maintain 3.3V at 250ma per LED. But since you are driving them with less current than that its safe to assume 6.6V in your maximum. With a buckpuck, The minimum battery voltage to drive a 6.6V chain of LEDs is 6.6+2 = 8.6V, maximum is something like 30v for these devices so we don't need to worry about overhead. :laughing:

Suppose we have determined that using 12V lead acids and assuming the battery is fully charged with a closed circuit voltage of at least 12.8V. The buckpuck has an efficiency of at least 90% at 5-15V input. Driving two CREE XREs at 350ma [lets consider that as "high"], it will draw around 200ma per hour it is running. In 6 hours you will have drained 6x0.2=1.2AH out of your battery on high. :candle: Since it is not recommend to discharge more than 40% of a lead acid's rated capacity to promote longer life, your minimum battery size will have to be 1.2/0.4 = 3AH. This is on "high" of course, but suppose you are content with running the LEDs at only 175ma instead of 350ma, based on the CREE XRE datasheet they have a forward voltage of ~2.9V at 175ma [extrapolation] [2.9x2]+2=7.8V. With the buckpuck dimmed halfway, it will only draw about 120ma from a battery with its voltage around 12.8V. 0.12x6=0.72 or 720mah of its capacity consumed. Minimum battery size is rerated to 1.8AH.

Now lets look at your environment, suppose you are lucky enough to have at least 4 hours of uninterrupted sunshine above you. :thinking: You would only need to gain 720mah/4 = 180mah of juice per hour to break even. Lets just say to compensate for losses the value is derated to 200mah per hour. Since solar panels rate their output current and ideal wattage to 16V instead of 12V, :sigh: you will need at least 3.2W of solar panels to break even. If you are running the LEDs on "high" [or 350ma] you will need to leave 5W of solar panels outside for 4 hours to break even. Does this make sense?:shrug:


[FONT=&quot]If you don't operate the device up to 6 hours and using a 5W panel on a small battery, a charge regulator may have to be used to prevent overcharging. If you run over 6 hours then your break-even point might land on the 5th hour on the next day.:fail: If you still only have 4 hours of sunshine tomorrow, derate the runtime for that night. [/FONT]

 

[Carbon2010]

Wow Illum - As always your replies are great.

I thought about going down the Buckpuck route (especially as I have a spare 1A unit that i'm able to dim down quite low.)

Doing some tests I have anywhere between 8-9 hours of sun...on average.

I'm going to be using for tests this setup.

ONE Luxdrive 3W star
2= 3.6V - 4500mah battery packs (wired in parallel)
1= lm317
1= 1.3 5watt resistor
2= 12v solar panels (wired in parallel)

So far I ran the light for 6 hours during the night. In the morning I turned the light off and hook the batteries up the the panel charge system on a breadboard. Left the setup by the window for aproxx 7 hours then hooked up the system. I was able to get another 6 hours out of the setup without dimming. I will do the same again tonight and see my results.

 

[Illum]

One reason why I had suggested a buckpuck was that it is a DC-DC converter, which can convert watts to watts essentially recomposing the input/output current/voltages to fit the need. :devil:

LM317 is an LDR, :duh2: it is linear, like your resistor. Yes the wiring is simple but you'll get alot of loss as power is converted into heat, which I am not confident your meager solar setup can afford. :shrug: After your initial 6 hours, what was the 3.6V 4500mah battery voltage reading? What was the voltage after the 7 hours of charge? I ask, partially out of curiosity, but also the fear that the battery pack was capable of a 12hour runtime since the start and that even after 7 hours the solar cells were not able to recharge it. You mentioned dimming towards the end of the second 6-hour run. If the solar cells weren't able to recharge it fully, your cells are going to be more depleted everytime a runtime is done. Whether this is so is contestable, especially since no voltage reading [or better yet, voltage reading under load] is provided.:thinking: