OK everything sounds really easy except for one item. A sunny day.
I hear we are supposed to get one Sunday.

That's OK - go ahead and take measurements when you can, no matter whether the sun is out or not - but that is an important item to record.

Preferably, around 10am, around 1pm-3pm, and maybe 5pm or so.

After all, you want this thing to work even if it's not really sunny out, right?

 

OK did some testing today and here are my findings.

It was a mostly sunny day, very few clouds.

Resistance of probes is .7 ohms. Probes + resistor is 10.6 ohms.

 

Your solar panel is woefully inadequate.

At it's best, it's outputting 218mA. You need 3 to 4 times that much.

 

Your solar panel is woefully inadequate.

At it's best, it's outputting 218mA. You need 3 to 4 times that much.

Thanks Sarge. Guess we will have to look at a replacement.


I don't want to sound like a dummy (actually as they say around here, you are only dumb when you don't know and don't ask), how did you come up with that figure?



I think in a previous post that I would need around 718ma to charge in eight hours. So you are right 218 would be way off.

 

OK, you changed it on me.

I thought you posted voltage measured across the resistor, like I'd asked.

Now I see you posted Amperes, measured as a short and across the load.

What was the voltage you were measuring across the resistor?

If the 11:00 current of 1.98A with the resistor inline was accurate, you should have measured about 19.6v across the resistor. That would be more than adequate.

 

OK, you changed it on me.

I thought you posted voltage measured across the resistor, like I'd asked.

Now I see you posted Amperes, measured as a short and across the load.

What was the voltage you were measuring across the resistor?

If the 11:00 current of 1.98A with the resistor inline was accurate, you should have measured about 19.6v across the resistor. That would be more than adequate.

My apologies. I may have misunderstood your instructions. To clarify, I measured the current with the resistor in line as if it were a load.

I have the day off on Friday and can make more measurements if needed.

If it sounds like the panel will do the job then I think for starters, I will need a voltage regulator.

I know in several previous posts (mine and posts by a few others) you have stated that the LM and similar regulators are fairly inefficient, dropping as much as 3V. So I guess they are out.

I have been doing some research and think I discovered that switching regulators are a bit more efficient?

 

Switching regulators are far more efficient than linear regulators.

However, the circuit you found is designed to power devices from a solar array, or a lead-acid battery - not as a solar powered battery charger.

You need a charge controller, as it appears that the existing charge controller is dead.

 

http://www.allelectronics.com/make-a-store/item/SCN-1/3A-SOLAR-PANEL-VOLTAGE-CONTROLLER/1.html

$17.00, not bad if you consider price of batteries. They have several others for higher power (not a bad idea), if you want to leave some expansion room...

 

Switching regulators are far more efficient than linear regulators.

However, the circuit you found is designed to power devices from a solar array, or a lead-acid battery - not as a solar powered battery charger.

You need a charge controller, as it appears that the existing charge controller is dead.

I read that paragraph and never saw that. I just reread it again about 3 times and finely see what you saw. I guess that is where experience pays off.

 

http://www.allelectronics.com/make-a-store/item/SCN-1/3A-SOLAR-PANEL-VOLTAGE-CONTROLLER/1.html

$17.00, not bad if you consider price of batteries. They have several others for higher power (not a bad idea), if you want to leave some expansion room...

Thanks. I think another poster (I think Bernard) suggested this one too. I am considering it. May be an easier way to go.

But what I am unsure of if it can turn on and off the light. There don't seem to be an specs on it but I guess I could always email AllElectronics.

Or I guess I could always add this feature. I was thinking about adding a Microcontroller anyway so that I can have settable delays and other features. Because most of these controllers just turn on the light when the ambient light has dropped to a point that the panel can no longer detect it even though there may still be plenty of light.

With my microcontroller, I could always turn the lights off after they have been on X number of hours.

 

If it sounds like the panel will do the job then I think for starters, I will need a voltage regulator.

You do not need a voltage regulator to regulate the output of the solar panel, since once connected to say a 12 volt SLA battery, the output will be pulled down to the battery voltage level anyways, and what you need is in fact what sarge suggested, a charge controller..... although you need to look for another type of controller to make use of the power output of the solar panel since Charge controllers intended for solar panels work by monitoring the battery voltage, and once it reaches full charge, the controller simply shorts the solar panel leads together. This doesn’t harm the panels, but it does waste any power they are generating, and this energy ends up heating up the transistors in the controller, which could lead to an early demise for the controller (which I suspect is what happened to yours).

You could easily design a circuit around a couple of op-amps to monitor the voltage levels and switch the output of the load automatically with a couple of mosfets when the battery has reached a full charge...

You can read more about my charge controller circuit here >> http://www.morse-code.com/id184.htm << and if interested I will dig up the schematic somewhere and post it for you...


Edit:
Actually I already posted my circuit here in this thread >> http://forum.allaboutcircuits.com/showthread.php?t=28152

and here is the circuit:

B. Morse

 

You need a charge controller that won't try to charge your battery at faster than a 5-hour rate. That's less than 1A. If it's charged too quickly, the battery won't last long at all. 5 hours is really too fast as it is.

You also need for it to disconnect the solar panel when the panels' output drops below the battery voltage. Otherwise, the battery will be drained by the solar panel.

The circuit should then monitor the solar cell output, and switch on the light when the solar cell output falls to nearly 0v.

It should also disconnect the load when the battery falls to around 12v. If you don't mind buying a new battery every few months, wait until the battery is discharged to 11.5v.

If you want to use a uC, you will need an efficient way to step the battery voltage down to an acceptable level for the uC, usually around 5v. A linear regulator would waste a lot of power.

 

adding a blocking diode to the output of the panel will prevent the panels from drawing any power from the battery (but will also drop the voltage from the panel down), I'm sure there are alternates he could use besides some huge one like this...

 

You do not need a voltage regulator to regulate the output of the solar panel, since once connected to say a 12 volt SLA battery, the output will be pulled down to the battery voltage level anyways, and what you need is in fact what sarge suggested, a charge controller..... although you need to look for another type of controller to make use of the power output of the solar panel since Charge controllers intended for solar panels work by monitoring the battery voltage, and once it reaches full charge, the controller simply shorts the solar panel leads together. This doesn’t harm the panels, but it does waste any power they are generating, and this energy ends up heating up the transistors in the controller, which could lead to an early demise for the controller (which I suspect is what happened to yours).

You could easily design a circuit around a couple of op-amps to monitor the voltage levels and switch the output of the load automatically with a couple of mosfets when the battery has reached a full charge...

You can read more about my charge controller circuit here >> http://www.morse-code.com/id184.htm << and if interested I will dig up the schematic somewhere and post it for you...


Edit:
Actually I already posted my circuit here in this thread >> http://forum.allaboutcircuits.com/showthread.php?t=28152

and here is the circuit:

B. Morse

Wonderful article. This is what I am looking for. I am learning so much from you guys!

When I am replying to you I am seeing a schematic. Strange.

Do you have a larger version? Yes I know about the diode to prevent the battery from not powering the circuit. But how can I power my circuit when the panel can no longer provide power and it needs to come from the battery?

 

If you want to use a uC, you will need an efficient way to step the battery voltage down to an acceptable level for the uC, usually around 5v. A linear regulator would waste a lot of power.

And this is what I want to do. Or at least experiment with it.

So the question is, what is an efficient method to drop my voltage down to 5V?

And if I understand all of this correctly. I need a way to step doen the voltage. Then with Mosfets I can control how the battery is charged.

I also assume I can use the same type of switching to switch in and out my diode when required to power the uC? O just don't worry about it and power from battery always?

 

When I am replying to you I am seeing a schematic. Strange.

Sorry you must have posted as I was editing...

Do you have a larger version? Yes I know about the diode to prevent the battery from not powering the circuit. But how can I power my circuit when the panel can no longer provide power and it needs to come from the battery?

Yes the circuit will be powered off of the battery (which will drain some of the power ).... But the whole circuit does not use hardly any power when the battery is charging, only when the battery levels get up to 14.5 (or what ever trip points you set) will it activate the output relay to divert the solar power...

 

Sorry you must have posted as I was editing...



Yes the circuit will be powered off of the battery (which will drain some of the power ).... But the whole circuit does not use hardly any power when the battery is charging, only when the battery levels get up to 14.5 (or what ever trip points you set) will it activate the output relay to divert the solar power...

So the blocking diode just prevents the battery from trying to charge the panel?

 

I also assume I can use the same type of switching to switch in and out my diode when required to power the uC? O just don't worry about it and power from battery always?

The diode is always inline with the positive output of your solar panel, all of your circuits should always be connected to the battery at all times, in order to properly monitor your battery voltage levels... all you need to switch is the output of the solar panels when the battery reaches a full charge, basically you will just disconnect the output of the solar panel to the battery. You could use a relay that is properly rated to handle the load (like an automotive 30 to 40 amp relay) and use the mosfet to turn the relay on or off. If you have the solar panel wired through one of the normally closed contacts of the relay, then you will only need to power the relay once the battery reaches a full charge.

 

So the blocking diode just prevents the battery from trying to charge the panel?

The blocking diode prevents your solar panel from draining the battery when the sun goes down. It acts like a one way valve and only lets power go one way.

 

Another thing you could do (with out using a UC) is check out this IC from NATO (RT1072) it is a PIR sensor IC that can trigger an device with motion (unfortunately unless you have a chinese supplier, you can not readily get these), it has settings for how much light is present before it switches on, how long it stays on for, and sensitivity to motion adjustment.... this IC sounds like a good improvement to the circuit without having to program anything..... and I believe you should be able to get the same sensor already encased in a housing with all the functioning circuits at Harbor Freight or somewhere similar.. if you purchase the solar motion sensor light in the picture ..

The unit is cheap, but atleast the PIR sensor and circuitry is separate from the rest of the circuit, and only connects with 3 wires! (power, Ground and output) and it can either trigger a triac (for AC use) or a relay. which in your case should be a relay to switch the light on...