Small Lighting Project

Surfpath

I need to run a 10 watt light for 5 hrs a night in a remote area (ie no grid). I am thinking about using these locally available 12v 12 amp/hr AGM batteries (advertized as up to 1200 cycles).

I get an average of 5 hrs of sunlight a day, year round (25-30 degrees c temps).

From the following formula I realize that two of the proposed batteries in series needs at least a 40 watt panel:
12 volts * 24 AH * 1/0.77 panel+controller derate * 0.10 rate of charge = minimum ~40 watts array for a 10% rate of charge.

I am thinking about one of these 50 watt 12v panels (17.3 VMP)

Will this be a balanced relationship?

24 amp hr battery 'bank' at 12v = 288 watts potential.

10 watts of lighting turned on average of 5 hours a night = 110 watt/hr in loads.

So my nightly draw down should be < 50%

I will use a very basic PWM controller. Are there other efficiencies I need to consider?
Cheers,
SP

BB.

Re: Small Lighting Project

~50% cycling on the battery is pretty deep for daily use... You may run out of hours in a day to recharge (2-4+ hours of absorb time per day).

You may need to double the AH capacity of your battery bank so you only get 25% discharge instead (shorter absorb time--closer to 2 hours).

And then there is the question about what happens on your rare cloudy days... Will there be somebody to turn off the lights, start up a small genset or something to make sure that battery does not get taken dead during bad weather?

There is some good with keeping the battery "small"--Small is cheap and if there is deeper cycling than planned, it does not take much money to replace an early death battery.

Not that your design is "wrong" or unbalanced--Just questions on my side.

-Bill

Surfpath

Re: Small Lighting Project

BB. wrote: »

~50% cycling on the battery is pretty deep for daily use... You may run out of hours in a day to recharge (2-4+ hours of absorb time per day).

You may need to double the AH capacity of your battery bank so you only get 25% discharge instead (shorter absorb time--closer to 2 hours).

And then there is the question about what happens on your rare cloudy days... Will there be somebody to turn off the lights, start up a small genset or something to make sure that battery does not get taken dead during bad weather?

There is some good with keeping the battery "small"--Small is cheap and if there is deeper cycling than planned, it does not take much money to replace an early death battery.

Not that your design is "wrong" or unbalanced--Just questions on my side.

-Bill

You are right Bill,
It'll be enough but it will reduce battery life to 1.5 years under ideal isolation (BTW, I will be there to turn it off on cloudy days).

110 watts draw down/288 watts bank potential = 38% battery capacity nightly draw down (not 50%, but still quite a lot).

Using the battery's specs chart on p2. that means about 500 cycles of estimated use.

The next sized up 12volt deep-ish cycle battery available to me is a 80amp lifeline ($$ and would need a much bigger panel).

I guess I could add one more of these little AGM's (and more panel) for a 36 amp-hr bank (that should get my draw down to ~20-25% (off the top of my head) and therefore a more respectable 1000 cycles.

Still I'm realizing this is a pretty expensive little project just to run one 10 watt light for 5 hrs.:p.

BB.

Re: Small Lighting Project

I always ask, would a motion detector work for your need?

Can save lots of energy, and better (in my humble opinion] for security. Light turning on with motion draws more attention.

-Bill

zoneblue

Re: Small Lighting Project

And... find the most efficient 10W light bulb you can. Duh, obviously.

LED. You want something with an effciency of 100 lumens per watt or better.

Surfpath

Re: Small Lighting Project

zoneblue wrote: »

And... find the most efficient 10W light bulb you can. Duh, obviously.

LED. You want something with an efficiency of 100 lumens per watt or better.

Good ideas ZB and BB,
Yes, it is a security project. The 10w flood light I had my eye on was rated at 1050 lumens. There are even similar floodlight models with motion detectors incorporated, like this one.

But you got me searching and I found perhaps a better option. It's not a 10w flood light, but a weatherproof 46 LED light motion detector unit with battery/panel: $30.

I can get 2-3 of them. Beats the $200 I was going to spend on panel, wiring, 3 batteries and a small controller. If someone steals one I won't be as put off.

I already have the smaller 16 LED version by my gate and (after 12 months) it works fine (cept that one has an annoying 'dim' mode). This one only comes on when motion is detected which, as BB pointed out, is better.

BB.

Re: Small Lighting Project

Always be careful with white "T series" LEDs... A large number of devices drive lots of current into the LED to get bright light, but the plastic case with a pair of leads is a very poor heat sink--And heat kills the phosphors (and possibly the blue LED). It is (was) not unusual for the LEDs to turn brown/black over 100-500 hours of use.

Cree is well know for their (more less) flat LED mounted to a large heat sink. Usually a much more stable light source.

-Bill