My friend and I have started this project. We are simply going to use a solar panel (max 16V output at 140 to 180 ma) to maintain a 12V lead acid battery's charge. I am planning on using a circuit based on an LM3914 bar graph driver IC. I found a simple circuit that I plan on modifying for my use (http://www.uoguelph.ca/~antoon/circ/batmon12.htm). I'm going to chage some resistor values around to make it sensitive from 11 to 14 volts instead of the values given in the diagram. To signal whether to charge the battery or not, I'm planning on using a transistor triggered by LED 10 (100% charged) turning off. However, this is where my problem arises.

From what I know, the battery will float between the voltage required to light LED 9 (90% charge) and LED 10 (100% charged). As soon as LED 10 turns off, the power from the panel is directed to the battery, and the battery's voltage will rise. Eventually, this rise will reach the the voltage required to turn of LED 10, stopping the charge. As the battery charges more and more, this process will repeat at a fast rate (charge on, charge off) thus causing the switching transistor to get hot. I know I'm not switching much current (less than 200ma) but at a fast switching rate, won't it heat up the transistor? Will this be a problem with the design if I choose to use a high rated transistor (like 5 amps?)? What other possibilities do I have to remedy this?

Sorry for typing so much...I tried to summarize it as much as possible. Hopefully you get the gist of whats going on.

Also, we're on a somewhat limited budget, so thats why no microcontrollers are involved. Otherwise, I would have done it that way and it would have been MUCH easier.

Thanks in advance,

Dan

 

I think you're over engineering your solution. 16v 180ma peak output is not going to harm a lead acid battery, unless it is an unusually small battery.

using a diode to prevent discharge through the cell, you'll be dropping nearly 1v right there. add in losses from the wiring, resistance of the battery terminals, etc, and I think you'll find you are well within spec for the battery. if you build the monitoring circuit, and figure on 8ma per led, that's 80ma eaten up right there.

In other words, just wire the cell to the battery (through a diode) and call it done

__________________
If you don't have a planet, what good are gold bars?

want to contact me directly? gmail gordonthree
check out my project website: http://projects.dimension-x.net
Favorite numbers:
09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0

 

Hi Ductapemaster,

I go along with justDIY's statement.

However you may want to use more solar panels after a while, observing that your battery is barely being charged. To keep the power consumption low you should wire the LM3914 as a dot indicator tying pin9 to pin11 instead of V+.

The LM3914 is an active low IC, which means you must use negative logic to control any circuit connected to the output pins. Fast switching does not do any damage to transistors. They are made for fast switching.

In your particular case I would speak of occasional switching.

Here is a circuit matching your requirements. Note that a P-Channel MosFet is used for switching. The input circuitry will light up all 10 LEDs at an input voltage >4V. Therefore you must adapt it to your requirements.

The MosFet used (IRF4905) does not require a heatsink at that low current.

Boncuk

 

Your link doesn't work. Please attach the schematic here.

I have one of those little solar panels. It produces 150mA at noon in the Sahara desert. You might be lucky to get 80mA at noon in summer through a window. Nearly nothing in winter.
It might already have a diode in series with it. Check its datasheet or measure for reverse current when in the dark.

The current is so low that a charger circuit is not required.
Add the LED voltmeter if you want, but it draws more than the solar panel provides.

__________________
Uncle $crooge

 

Thanks for the suggestions guys! I may consider just connecting the battery to the panel. Maybe I'll use a couple of diodes in series to drop the maximum voltage down to 13.8 (suggested for maintenance voltage). I plan on taking some measurements (hmmmm, maybe i'll build a datalogger...heh, just another project for me to do) of the output throughout the day. It is going to be placed on the roof in some sort of waterproof box, so I should be able to get a little more output than if it were placed in the window.

ALso, Boncuk, thanks for those schematics. I am eventually going to make a somewhat more sophisticated one of these for myself (this one is for my friends house) and I will use those for that.

 

Please excuse my double post, but here is the schematic you were requesting, audioguru.

Attached Images

batmon12.gif (6.4 KB, 23 views)

 

the link works fine for me ... I think what throws off those most is the extra ) at the end of the url; most likely appended by the forum software or a typo compliments of the OP

__________________
If you don't have a planet, what good are gold bars?

want to contact me directly? gmail gordonthree
check out my project website: http://projects.dimension-x.net
Favorite numbers:
09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0