Newbie - Help with making LED circuits

jgray152

Newly Enlightened
Joined
Aug 3, 2010
Messages
16
Hi all,

I am new to the forum and the LED world. I have looked at how LEDs work but a little lost when it comes to finding out how much power LEDs need and how to calculate how many LEDs I can run when running off small batteries.

My business is StoneCrete Tekniques, I evolved from landscaping to designing and forming and molding artificial stone like structures. Now I want to create light posts that are roughly 3' high that can emit light from the top of the post on all 4 sides.

Some short quick needs;

I need them to run of batteries and have the batteries charged by solar power panels which I want located on the top of the posts.

I want the LEDs to be bright enough to light up a walkway possibly 5' away from the post with decent brightness. Nothing like these "slight glow" LED lights you buy at the stores for $20 which only produce enough light for you to know you bought some lights....if you know what I mean.

I guess I want to know how to calculate the amount of time the LEDs can run with a certain amount of power driving them and how many LEDs I can run of a single power source.

Its late right now so tomorrow I will add some more details and possibly some links to some LEDs and things.

Any "beginner" or Starting Guide information will help.

Thanks all!
Justin
 
What you want to do is going to be difficult to do, especially if you want this to be inexpensive. The problem comes from your demands. To illuminate the area you require, you are going to have to use higher power LEDs, which leads into the next issue (but has a number of issues by itself to deal with, like heat). The current draw for the LEDs will most likely be much greater than a small solar panel will be able to deal with. To combat that, you would need larger, or more efficient (read: expensive) solar cells, and may get too large for the structure you are trying to build. Why not just use low voltage hookups like traditional architectural lighting?
 
Hi all,

I want the LEDs to be bright enough to light up a walkway possibly 5' away from the post with decent brightness. Nothing like these "slight glow" LED lights you buy at the stores for $20 which only produce enough light for you to know you bought some lights....if you know what I mean.

hi Justin,

I applaude your desire to create something new, but I think you may be disappointed with what can be done. The "slight glow" LED lights are pretty much state of the art when dealing with the constraints you mentioned.

The biggest issue is the desire to power the LEDs from a solar panel. A small panel of a few square inches is just not going to produce much power... maybe a watt at best. The best you might accomplish is to produce 1 watt of light for 4 hours or so, and this falls far short of lighting a walkway.

You'll have to compromise on some part of your requirements. Maybe use a 12v power supply like most other outdoor lighting, maybe increase the cost and size of the solar panel, etc.

It might even be worthwhile to contact a company that makes those little yard lights. If you explain the business idea, they might be able to tell you what they could make. I'm assuming that you don't want to become an electronics assembly business, and would rather just add an electronics package to your stone structures.

regards,
Steve K.
 
Thanks for the replies guys.

My first post was very vague but I have done some more research and found out that it would be hard to do.

I have been thinking of going with low voltage systems since that would cure all my needs but then eliminate the one need for "quick easy installation". Now you would need to run wires, although low voltage systems are easy to hook up it may scare others away.

So maybe what I will do is create small rock with the LEDs which will be close to the ground which means I don't need them to be as bright as if they were lighting from the top of a post.

So now that I have scaled down the initial project this might be a bit more do-able.

Now I know there are already "light rocks" out there but they usually consist of a larger light removing the natural look of the rock. I would be able to hide the smaller LEDs better.

So here is what I was thinking and please share your knowledge.

Now I do want them bright so I am looking for the highest MCD I can find but I notice that when converting that figure to Lumens, adding in the projection angle can make a huge difference.

These 5mm LEDs here...http://www.superbrightleds.com/cgi-bin/store/index.cgi?action=DispPage&Page2Disp=/specs/W18030.htm

18000 Mcd, 30 Degree Angle = 3.8 lumens (roughly) Lumens - MCD Calculator
3.4v / 30mA

So if I wanted 5 of these LEDs it should equal roughly 19 lumens right?

5 LEDs * 30mA = 150mA

Using three rechargeable AA batteries which may be 1.2v / 2000mAh would equal 3.6v. LEDs would be the same color and should be all the same specs so I could run them parallel correct?

2000mAh / 150mA = 13.3 Hours of life??

Next issue would be to figure out how to allow a solar panel to safely charge the batteries. Have not found a detailed right up when it comes to know if I need resistors or any other electronic parts.
 
Your math is fine, IMHO.

However, I think you're going to want at least 350mA with neutral white....if you use good emitters.

The solar circuit is what's going to be tougher.
 
Using three rechargeable AA batteries which may be 1.2v / 2000mAh would equal 3.6v. LEDs would be the same color and should be all the same specs so I could run them parallel correct?

2000mAh / 150mA = 13.3 Hours of life??

Next issue would be to figure out how to allow a solar panel to safely charge the batteries. Have not found a detailed right up when it comes to know if I need resistors or any other electronic parts.

13.3 hours, but you probably won't want to fully discharge the battery. Maybe only use 70% of the capacity? I think you'll need some sort of low-voltage shut down circuit to avoid discharging the battery too far.

I think the big problem is finding solar panels large or efficient enough to charge that battery under nominal or worst case situations. Were you planning on mounting the solar panel on the rock or post? That really will reduce the ability of the panel to charge the battery (unless the rock is a foot in diameter and completely covered by the solar panel).

I'm betting that you can find data to allow you to properly size the solar panel. You'll need to do an energy calculation.... find out what amount of sunlight falls at your location (watts per square meter), what the conversion efficiency of the solar panel is, length of the day, charge efficiency of the battery, and that should be about all you need to figure out how big the panel will need to be. Or just go buy a panel, play with it, and see how well it charges the battery. Charging a battery from a panel is usually very simple... just add a series diode between the battery and panel to prevent the panel from discharging the battery when it's dark. I don't think you'll need to worry about overcharging the battery.

regards,
Steve K.
 
I was going to have the solar panel mounted on the rock but now that you guys mention the difficulty of charging the battery, it may be best to mount a larger solar panel on a post connected to one or more rock lights.
 
If I use three batteries which are 2000 mAh and wire them in series, this would increase the voltage but wouldn't this also add up the mAh? 3 2000mAh batteries last longer than 1 or is this only if its weird in parallel
 
Last edited:
Voltage is cumulative in series, current is cumulative in parallel.

Also, my first suggestion on where to start would be to choose and buy some LEDs, and build some simple battery powered mock-ups to see how much light you need and work out some basic kinks. Without knowing how much light you need you don't know how much power you need and hence can't really design anything. What works on paper might not look good in final design.

You may find 150mA works, or maybe 350mA as I suggest. 5mm / 10mm LEDs are cheap to play with.

Just one suggestion - stick to neutral white LEDs no higher than 4100k. The few yard lights I've seen using neutrals rather than the fugly cool-whites look a lot classier.
 
You could take a more scientific approach:

- How much area do you want to light up and how bright do you want it? Not sure, go buy a $30 light meter and find an equivalent lighting level and measure it. When you have the lighting level and area, lumens pops right out

- Assuming losses in covering the area

- Is this mission critical? I.e. does the light have to be on? If so, your math is way out.

- 2000mA = Day 1. If you discharge completely, that battery is toast in less than a year

- 2000mA at what temperature?

- What about the days when you have effectively no sun? .... or more accurate days and days...?
 
Given that a battery can only hold so much energy you might want to consider letting the lights run for only a certain amount of time after it gets dark. If you need lighting all night long you are going to find it extremely difficult to balance the size of solar panels and battery requirements for free standing LED lights.

I have two solar lights (made and sold only in Japan) that have a larger than average solar panel, decent quality construction and uses three AA batteries.

They drive the single Cree LED at around 150mA and that is sufficient for my front path lighting requirements.

The 3 AA NICD batteries receive good charge and run most of the night during the summer, but, during winter months they will only run for short times before going dark.

I have since modded them with newer LEDs that generate a little more light for the same amount of drive and added a little microprocessor timer board that is configured for 4 hours. After 4 hours the light is turned off saving the light from fully draining the batteries. Even during the winter time I get full brighness now for the full timer duration.

Having the batteries get fully drained they don't receive much of charge every day and thus discharge faster. Turning them off allows the batteries to remain partially charge and will recharge to full every day even in the winter.

Wayne
 
Wow, thanks for all this information. There is certainly a lot more to this than I quickly thought. Thats ok though, makes it interesting. I'll buy some LEDs and make a simple battery circuit to see how much light I really need.

Are there circuits out there which turn the LEDs off when the batteries drain to a rough specific point?

Also, I am pretty sure there is a way, I may just have to get some sort of small converter but I could potentially offer this in two options, Solar, or low voltage and potentially use the same LED circuits and lights for both? The low voltage would just need some conversion to get the voltage right. This would be really interesting too. Does anyone know if a place online which carries low voltage parts? I can only find whole kits which doesn't help me much.
 
Last edited:
Are there circuits out there which turn the LEDs off when the batteries drain to a rough specific point?

I don't know how much this will help, but I'd use something called a reset controller or voltage detector. This is the device used to warn a microprocessor when the power supply voltage is getting dangerously low. The OnSemi NCP304 is an example:
http://www.onsemi.com/PowerSolutions/product.do?id=NCP304

As an electrical engineer, I could figure out a way to incorporate this into a circuit that could shut off the leds and still not draw much power when it was shut down. It would work very well as an "enable" or "shutdown" signal for a little converter circuit like a Zetex ZXSC310 boost converter. For someone who doesn't do any electronics work, well, this might just be a foreign language.

regards,
Steve K.
 
For someone who doesn't do any electronics work, well, this might just be a foreign language.

Haha. You got that right. This might be something I may want an professional to design and build. You do this type of work on the side? If so, when I am ready and know exactly what I want I could have you build one or two for starters for testing.
 
I'm busy trying to find time to build up a couple of my own home projects, so I will have to pass on this opportunity. :)
It's not very complicated, though, and it shouldn't be hard to find someone who could do it.

Steve K.
 

Latest posts

Top