Solar panel charge controller

[mike_mike]

Dear Forum,

I have an 12V 3W solar panel and an 6V 1.3Ah lead acid battery. I need a schematic that charge the battery during the day and during the night it lights up a few LEDs.
The power consumption of the LEDs is about 70-100mA in total.
I also need a circuit that protects the lead acid battery from overdischarge.
I tried to design a schematic using some information from web but I am not sure if the schematic is correct. Please have a look at the attached schematic.
Can someone please help me with a schematic ?

[metrologist]

I looked at your circuit. I think there are better ways. Where did you see these plans?

The relay will engage the solar panel at 9V and release at about 0.6V. The voltage regulator looks to be adjusted to around 7v up to ~11V with that 5k pot arrangement. I think you can achieve the same result with just the pot alone or a couple of the right size resistors. I have modeled this circuit in LTspice.

The green trace shows your daily solar panel cycle (pretend it's a sine wave).

The blue is the output of the regulator. Red is the half voltage output of the battery, which I have modeled as a sine wave to illustrate it's nightly power cycle. Turqouise is the collector of the transistor and pink is its base.

BTW I am not suggesting this circuit...

[Seekonk]

You don't have any power to waste.  Buy a lm2596 type regulator board for $2 and that will convert the 17-18V power point of the 12V panel to the voltage you want.  These can easily be made to maintain power point, but with size it hardly seems worth it. Try it and if the panel drops below 15V we can discuss further options.

[Rog520]

I kind of wonder if your tiny 12v solar panel would overcharge your 6v battery, even with a direct connection. The panel is essentially a constant current source and its voltage is going to track that of the battery. Even so, just to be safe, you could hang a 7.5v 5W zener off of the panel (where D1 is at in your schematic) which would serve to keep the voltage within the limit of the 6v battery. Make sure you keep D2 in series on the battery side of this junction to prevent the battery backfeeding into the solar panel (at night) or the zener. You could then eliminate U1, etc.

[mike_mike]

BTW I am not suggesting this circuit...

I designed this circuit starting from an charger schematic and an overdischarge schematic which I combined resulting the schematic from the first post.

Can you please recommend a good circuit ?

[metrologist]

BTW I am not suggesting this circuit...

I designed this circuit starting from an charger schematic and an overdischarge schematic which I combined resulting the schematic from the first post.

Can you please recommend a good circuit ?

I meant that in context of my LTspice model. The topic interests me. For example, I've modeled a circuit based on this website: http://www.clarenceho.net/2006/02/

[mike_mike]

I made that circuit on a piece of test board and it seems to work. I used a 5.6V zener diode.
I used as load a white LED with a 470 R resistor in series.
The LED lights up at about 5.4V input voltage and goes off at about 5.3V input voltage. Between 5.4 and 5.3V, the LED is lighting just a little (it is dim). When the LED is off there is a small voltage across the LED and the series resistor (I measured about 0.8-1V). Is that ok ?
If the protection circuit is ok, can we now focus our attention on the charging circuit and on the relay ?

[mike_mike]

Regarding the red marked diode (D2), it is correctly placed for blocking the current to flow back into the LM317 if there is no voltage at the output of the LM317 or if the voltage at the output of LM317 is lower than the battery voltage ?

[Rog520]

Regarding the red marked diode (D2), it is correctly placed for blocking the current to flow back into the LM317 if there is no voltage at the output of the LM317 or if the voltage at the output of LM317 is lower than the battery voltage ?

Yes, but again, you really need to get rid of U1 because it's not going to work properly when connected to a solar panel. Have a look at a solar panel's I/V curve and you'll understand why. All is well with no load or minimal load, because the panel will be operating in its (relatively) constant voltage region (the vertical part of the curve). As your load increases, however, U1 us going to conduct more and eventually draw itself into the constant current region (horizontal part of the curve) where it will be unable to regulate properly. Voltage input to U1 will now start to drop, power will decrease, and the cycle repeats until U1 is in/near saturation and at the far left of the curve -- exactly where it needs not to be. You'll be at or near 0v, which of course, means zero power delivered to a load. What you'll probably find if you try to implement this is that it works sometimes (if you can manage a light enough load) and other times it doesn't.

[mike_mike]

Thanks for the reply.
1. What can I do to replace U1 (LM317) ?
2. What should I use in the place of U1 ?
3. If I will get rid of U1, the charge current will be higher ? I read somewhere that the charge current should be 0.1*C. Is there any solution to limit the current ?
4. If I get rid of U1, the charging voltage I think will be highter and I think that this is not good for the battery. Is there any solution to limit the voltage ?

[Seekonk]

Sorry Mike, you didn't listen to me.  You ain't got enough power to run a relay and charge a battery! And use a TL431 to turn LED on and off.

[mike_mike]

I thought that a 3W 12V solar panel has about 3W/12V=0.25A... But I think that it depends on the light conditions.
The relay power consumtion is about 30mA, and the battery charging current is about 100mA.
I do not want to upset anyone...
Please correct me if I am wrong.

[james_s]

The ratings on solar panels tend to be highly optimistic, even for a reputable supplier it will be direct sunlight around noon on a clear day with the panel aimed optimally.

Don't worry about upsetting people, just be polite and you'll be fine. Any forum this big will have a few abrasive folks but overall I think you'll find people here are quite willing to help.

[Rog520]

Thanks for the reply.
1. What can I do to replace U1 (LM317) ?
2. What should I use in the place of U1 ?
3. If I will get rid of U1, the charge current will be higher ? I read somewhere that the charge current should be 0.1*C. Is there any solution to limit the current ?
4. If I get rid of U1, the charging voltage I think will be highter and I think that this is not good for the battery. Is there any solution to limit the voltage ?

You'll save yourself a lot of trouble if you use a 12v battery, which would be directly compatible with your solar panel, and which would allow charging more closely to the Vmp of the panel (the "knee" in the I/V curve where power output is at its highest). Then you wouldn't need to limit the panel's voltage or disconnect it at a preset voltage to avoid overcharging the battery.

[mike_mike]

1. If I already bought the 6V battery, can I use a 6V solar panel for this battery ? It is the same as using a 12V solar panel for a 12V battery ?
2. Were can I find a schematic for this circuit ?
3. For example, if the panel has 350mA maximum output, and the battery has 1.3Ah, the charging current in direct sunlight won't be too high ?

[Rog520]

I thought that a 3W 12V solar panel has about 3W/12V=0.25A... But I think that it depends on the light conditions.
The relay power consumtion is about 30mA, and the battery charging current is about 100mA.
I do not want to upset anyone...
Please correct me if I am wrong.

A relay might be OK, depending on what solar panel you use and how much sun you tend to get. You could also set up a simple (and very efficient) comparator on the output of the panel (on the solar panel side of D2) which would turn on your lights when the panel dropped below a certain voltage (you could make it adjustable so as to tailor it to a certain light level). It would then turn them back off when the panel's voltage rose the next morning.

[Seekonk]

That wattage is at 19V, not 12V. It is also very optimistic based on ideal conditions like it not getting hot in the sun, optimum angle and those conditions will only last a short time daily.  I don't feel you can waste 30ma and have reliable operation in poor conditions.

[mike_mike]

Are there any solutions for this problem ?
What can I do to make the circuit correctly functional ?

[metrologist]

I know there can be a very simple circuit to control this switching. I have a similar solar panel that I was planning to use for a similar purpose. I could look into this sooner, but was planning to use a microcontroller for some additional features.

This is another similar approach to yours: https://www.circuitsdiy.com/solar-to-6v-batt-to-led-emergency-light-circuit/

There are a lot of designs that use the LM317 though.

Here is another idea: https://www.electroschematics.com/9503/6v-solar-charge-shunt-regulator/

[Rog520]

Are there any solutions for this problem ?
What can I do to make the circuit correctly functional ?

I'll get something drawn up later today when I have some time.

[Seekonk]

Use a 220K/100k voltage divider from panel voltage to REF the ref pin of TL431. A to common -.  C to + battery thru 10K resistor.  That C feeds to gate of a FET through a with a 100K pull down.  Look up the data sheet and see what 431 pulls down to and figure out what ? could be.  You are here to learn aren't you?