Won't work very well for grow lights.
Since we leave the lights on for 16hrs the batteries fischarge quickly for a decent amount for lights.

Check candlepowerforums.com
Flash lights are primary use of LEDs with batteries (unless you talking big banks charged from solar)

We get a LOT longer daylight than most parts of the country so it would be more to extend the daylight by 2-3 more hours like summer time here. Mostly for a variety of seeds I want to grow.

As far are growing goes I'm pretty sure a decent light spectrum can be had with emitter selection. There's a whole lot of LED's out there with an individually narrow spectrum enough to be combined with others for plants.

Anyway at 2 hours a day, I'm looking at possibly 2-3 days before replacing batteries.

I want to use the say a half dozen (or more as needed) 4.2 volt 18650's in parallel to avoid huge losses in regulation that you'd get with say a car battery.

Thx for the reply.

I think, unless you want to grow to maturity, florescent shop lights are more practical to raise seedlings. You can aim and adjust them exactly to distances you want evenly. Since I use 3 of them, I can divide my seedlings into 3 groups and adjust the height accordingly.

So as they say, to each his own and YMMV.

Seysonn

Well, first thing I'd do if I were you would be to find the amp-hours on those batteries. Did some googling, and it looks like they are generally rated at about 5000 milliamp hours, or 5 amp hours. You say they're 4.2v each, so you're looking at lets round it off to 20 watt hours per battery.

The smallest grow LED I have is a 7w, and I barely consider it sufficient for a single seedling. Assuming it would light on 4.2 volts (which it most likely won't), it would burn for about 2.5-3 hours before it drained the battery.

And you said you wanted to run them 2-3 hours a day. So, on average, you'd be looking at depleting 1 battery per small plant per day. Thats not efficient at all.

If you really wanted to do a battery setup, you'd be better off looking into a deep cycle marine battery, and maybe some solar panels to recharge it during the day.

This post was edited by Edymnion on Tue, Jan 6, 15 at 17:45

Thanks for the replies, but... so many critics with so little information as to my purpose? I know some may just be trying to be helpful and I appreciate it but a little less heat and a bit more "light" would be more helpful. Remember I asked for schematics not technology options?

I do not plan on an AC supply nearby so shop lights and anything requiring AC power is out. Anyone with a couple of hundred bucks can buy an LED grow light on EBAY that runs 200 watts on up if you have AC power. Ditto a fluorescent grow fixture. My location is next to my house, and I plan on the LEDs being close to the seedbed as you can with cool emitters. A fluorescent bulb would keep me awake and probably my neighbors.

I do not plan on having dozens of plants growing assuming I am or need to is incorrect. Newton did not need ALL the apples of the tree to fall simultaneously did he?

The YMMV comment is unnecessary, I have a reason for batteries... what your reason is for lighting with shop lights is yours not mine.

Anyway, I've considered deep cycle batteries but they have several disadvantages some being cost, charge time, losses in voltage regulation, portability and size. Later on it may be profitable to go that route but not yet.

Solar panels are an option for recharging, but charge time is faster (albeit a bit less convenient) and more reliable if I charge them for now with AC. At some point I just may deploy light sensor (for turn on time) and Solar charging systems, but as before not yet.

Like I said, I plan on using several cells (I have a supply of them) and sufficient means to charge them. Amp hours delivered are multiplied by the number of cells so extending the time is simply a matter of adding cells. NP there either.
Multiple batteries should not be a problem with supply voltage, it's a simple matter of run time, current supply and regulation.

I have sources for 660nm and 450nm emitters all run at ~2.2 -2.6 volts. I may need to experiment with IR and FR emitters too.

Linear IC's (linear.com) has integrated circuits that I can source but, I had hopes there would be some hobbyists here that could have smoothed the path a bit.

you need to choose an led emitter, look at its forward voltage. you don't want to use the max of your forward voltage, more like 80% or something.

example:
these 3w leds on ebay we gonna run one deep red, one red, one blue thats 7.8-8.9v..
so put 2 batteries in series to get 8.4v thats good.

now the batteries say have 4200maH
you will run at 750ish so 5 hours.

like dis (whoops fixed)
{{gwi:2130088}}

This post was edited by OKgrowin on Tue, Jan 6, 15 at 22:19

I got my 35ah deep cycle at Battery Mart. Use it with solar panels and am well pleased with it. They are very reasonable too.
I am thinking about LED lights too because of their frugal power consumption.

Well the problem with giving you build information is that we don't know what LEDs you're going to be building with. There's a lot of math to figuring out what resistors you need based on the LEDs in each path of the circuit and the forward voltage of the power source.

You would probably get better help from a dedicated circuit design forum, like this one:

http://forum.allaboutcircuits.com/

you don't need resistors if you match the voltage.

I was bored while waiting for thermal adhesive to dry on my led strip so i made you an actual test circuit to show you how easy it is!

1. get leds..
I got these white leds 3.2-3.8v 700ma

2. get batteries.. got some AA's they are 1.5v Tested them and they actually read 1.6 yay.
{{gwi:2130089}}

3. match them. so i can hook up 2 batteries for 3.2v. which is one led.
{{gwi:2130090}}
{{gwi:2130091}}
{{gwi:2130092}}

4. here we can go 4 batteries for 6.4v which is 2 leds
{{gwi:2130093}}
{{gwi:2130094}}