Solar Powered Circuit

[gravymister]

So how would I easily measure the voltage drops and group them?
Is it as simple as using a multimeter to measure the voltage drop from one terminal to another? How accurate does it need to be?

 

[audioguru]

If you are a manufacturer then you can measure thousands of LEDs and sort them into groups with the same forward voltage. But You are not buying thousands of LEDs, you get whatever is available.

Cheap Chinese flashlights have LEDs connected directly in parallel. Some last for half an hour or less before the LEDs begin to burn out and others last for a few hours.

Good quality LED flashlights connect LEDs in series or each LED has its own current-limiting resistor and the LEDs last for years.

 

[gravymister]

Ok, so correct me if I'm wrong but, I don't have enough voltage even with 8 panels to power more than really 2 white LED's in series. So I need to put them in parallel, which is fine if I use a limiting resistor before.

I only 3 of my solar panels have the NiCd batterys they came with, so I was going to use them to charge old phone Li-on batterys, but from what I've read this can be quite dangerous if I use too high a voltage and I can't regulate the voltage from the panels easily, if its cloudy 4 in series might produce 4V but if its sunny they might give off 8V so I don't think thats wise! So its back to the drawing board.

 

[gravymister]

On a side note, I have another problem, I have a lot of resistor that are green and have the bands: brown, black, brown, gold. Which I guess means 100 ohm resistors. But when I test them on my multimeter it reads 2 ohms!? Whats going on?

 

[carbonzit]

1. Brown-black-brown = 100Ω, so must be something wrong with your multimeter. Try another one.

2. To add current, put sources (solar panels, battery cells) in parallel.

3. To add voltage, put sources in series.

4. When charging batteries, charging voltage must be higher than battery voltage--but not by too much. For quick-and-dirty regulation, if you're concerned about overvoltage when direct sun hits solar panel, put a [x]-volt zener diode across the feed to the battery under charge to limit the voltage. (I say "[x]" volts because I don't know anything about your specific batteries or what charging voltage they should properly have.)

5. At some point you're going to have to stop dicking around with all your various bits and pieces, and sit down with pencil and paper and work this damn thing out, instead of just speculating and firing off random questions. As I've stated several times already, you need to budget your power. You need to know:

• how much current your solar panels produce
• how much voltage your panels produce
• how much voltage your various loads require
• ditto for the current of each load

Then you need to match your sources to your loads. Only then can you come up with some kind of sensible arrangement that will actually work. Otherwise, we're just playing a lot of word games here.

 

[gravymister]

Ok Carbonzit, I've taken your advice and come up with two circuits that I'll run with the materials I have, I've attached the .Jpegs.

The Radio schematic is powering a small radio which includes an amplifier and speaker, just hooking up directly to the battery terminals. I'm using this instead of the Pringles speaker. I might hook some batteries up to it as well, but currently its just going to be powered in sunlight. Not sure if the capacitor is needed? I was thinking it might help smooth small fluctuations in sun light.

The LED schematic is powering a 9 LED light, to be used only at night. I only have 3 decent batteries totalling 3.6v in series, so I've included some worse batteries at the bottom to boost power, the switch can choose between high and low voltage (wasn't sure what the resistance of the drained batteries would be.

Any feedback/improvements would be greatly appreciated!

 

[audioguru]

Your radio circuit shows both wires of the capacitor shorted together so it does nothing. It should be connected in parallel with the power supply wires of the radio.

Your LED circuit has all battery cells connected with backwards polarity.

You should learn about electronics before one of your wrongly wired circuits burns down your home.

 

[gravymister]

Yeah I'll admit the capacitor error is pretty bad - I'm not sure I really need a capacitor there anyway? The Batteries I just forgot to rotate, I know which way they're supposed to be wired when doing it!

 

[gravymister]

Here are the corrected schematics, I've added in a battery feature so the radio can be operated after dark.

My main concern is that the Ni-Cd batteries will be charged too heavily 1.2v being charged at between 1.2-2.5v and between 5-100mA

 

[carbonzit]

The capacitor ain't gonna buy you much; maybe a few seconds of playing time if the sun goes behind a cloud. If, as you say, this is mainly for amusement's sake, just let it fade in and out. You'd need a much bigger cap (probably a supercap) to do any real good there.

And as Audio McDuck pointed out, the capacitor goes in parallel with the batteries.

By the way, your JPEGs are a huge improvement over your PDFs!

So far as your overly-complicated solar light circuit goes, it's a real mess. Obviously you're limiting yourself here to stuff you already have on hand. But if you were designing this from the ground up, you'd want all the solar cells grouped together as one source, and all the battery cells grouped together as storage. As it is, you're trying to run heterogeneous stuff in series, which while it's not dangerous or anything, it's certainly not efficient and you'll end up throwing away precious power.

But hey, it might just work ... actually, on second look, there's a major error in your design. You have several current sources (batteries of differing capacities) placed in series. That's no good. The "stronger" cells (the ones with higher mAh ratings) will be trying to force excess current through the "weaker" ones.

You really, really want all your batteries (with the same voltage rating, obviously) to be placed in parallel to form a bank of storage. Is this so hard to do? Since you're going to all this trouble, how hard would it be to just take everything apart and reassemble it in a better configuration?

 

[gravymister]

T
But hey, it might just work ... actually, on second look, there's a major error in your design. You have several current sources (batteries of differing capacities) placed in series. That's no good. The "stronger" cells (the ones with higher mAh ratings) will be trying to force excess current through the "weaker" ones.

You really, really want all your batteries (with the same voltage rating, obviously) to be placed in parallel to form a bank of storage. Is this so hard to do? Since you're going to all this trouble, how hard would it be to just take everything apart and reassemble it in a better configuration?

Are you talking about the LED light circuit here? I was going to run it like that so that there was 4.8V @ 20mA until the bottom batteries run out then I'd flick the switch to the left so it bypassed them. Its because I need 4.8v (ideally) to run the LED's but I don't have enough good batteries.

I would be more than happy to reassemble it, not a problem, if you let me know how it would run better?

Also I found some 4x AAA 1.2v 650mAh NiCd or even & 2x1.2v 1.5Ah batteries which could be utilised?

 

[carbonzit]

Sheesh, what a mess.

OK, referring to your latest drawing for the LED lighting circuit, we have, over there on the left, a series string of three solar lights:

• 1.5-2.5V @ 5-50mA
• 1.2V @ 900mAh
• ??? (unknown voltage and current)

So adding all this up, we come up with a grand total of

2.7-3.7V + X @ 5-50mA

("X" being the unknown quantity)

Remember: you can add voltages in series, but not currents. Currents add in parallel.

Here's what you actually want (generalized circuit, no values):

**broken link removed**

Nice and clean. The solar cells are arranged to give you the voltage you need to supply your load and charge your batteries. The batteries are arranged so that they're all the same voltage and in parallel to sum their current.

You cannot achieve this by connecting your little solar lights together as you indicate.

 

[gravymister]

Ah right, thanks for your feedback!

Bearing in mind I would only be powering these at night so the solar panels are purely there to charge the batteries. The idea was the batteries would run @4.8v as using around 200mA there are 4x 1.2v in series for 300mAh or until the unknown capacity batteries ran out at which point I'd throw the switch and change to run off 3.6v for the remaining 600mAh (assuming they are fully charged, which they probably won't be).

Your diagram is much nicer and I'm not using the circuit boards from the solar lights anymore, just the solar panels and batteries so I could try and rearrange it more like the schematic you supplied. So I'll have a look at desiging one with what I've got an repost.

 

[gravymister]

Ok here's a new schematic, only thing that concerns me is the charging of the battery, I'm not sure of the acceptable limits for NiCd charging? I've removed all the extra stuff from the solar garden light circuit board so its really what you see is what you get in this schematic.

 

[audioguru]

Gravymister,
I wish you learn about voltage and current soon because your latest schematic has serious errors:
1) An old-fashioned Ni-Cad cell is 1.4V to 1.5V when fully charged. You have four in series so they should have a total of 5.6V to 6.0V.
But they will not charge when the solar panels produce only 0.8V to 1.8V after the diode. Also without a voltage higher than 3.2V the LEDs will not light.

2) If you change the circuit so it produces 6.0V then the 50 ohm current-limiting resistors for the LEDs limits the current too high at 56mA so they will burn out.

 

[gravymister]

Hang on, I'm pretty sure that schematic is wrong because the charging voltage isn't high enough. I don't think with the components I have I can feasibly make a schematic similar to yours.

 

[carbonzit]

Basic stuff to keep in mind:

To charge a battery, you need a voltage higher than the battery voltage. You cannot charge a battery with a lower voltage.

You can achieve this by putting your solar cells (they're not really solar panels) in series/parallel, as I suggested in my schematic. Since they're only being used to charge the batteries, current is less of a concern than voltage.

 

[gravymister]

Ok, I see what I did wrong audioguru, but how do I correct it? If I reduce the batteries, then I can have a max voltage post charge of 4.5v which works with the resistors.

In full sunlight though the charging voltage looks to be too high.

 

[carbonzit]

How are you figuring current for your solar array?

On your schematic, you indicate 5-50mA for the solar cells. (I ASS-U-ME that means for 1 cell?)

Up above, though, you have two current ranges, 10-100 mA and 20-100 mA. Where do you get these values from?

Remember, if you have two sources in parallel, you add their currents. In this case you have two series strings of cells in parallel, so you double their current output. So shouldn't the current be 10-100 mA?

Your LED resistors are now a little too high. Since you're "wasting" 0.4 volts (3.6-.2) @ let's say 20mA, you need a 20 ohm resistor. (You could increase it to 40 oihms if you want to limit the current to 10mA, but the LEDs then may not be bright enough.)

Regarding overcharging our batteries, keep in mind that if you're only imposing a trickle current, you don't have to worry about it. Now, I don't know exactly what would be considered trickle current for your particular nicads. I'm sure someone else here does, though.

 

[gravymister]

Oh thats a typo I think, you're right its 10-100ma range. With regards to the trickle is it not 1/10C is considered a trickle so since the batteries being used are 900mAh and 600mAh trickle currents would be 60mA and 90mA and the max either would get would be 50mA so that would be ok?

All in all though If I obtained one more solar panel, perhaps this design would be more effective? I also thought about adding an LDR so that in the evening, when the lights are not being used the LDR will block any discharge from the batterys around the circuit through the panels, I could be wrong in that assertion though.

Also, where am I wasting .4 volts?, I could use 4 batteries and have a higher R value if its cutting it too fine with too little voltage? Ideally I think I want 20mA through the LED's I would like them to be quite bright.