battery-Solar charger

Thread Starter

captoro

Joined Jun 21, 2009
207
Hello

I am using a solar to directly charge a battery through a diode from solar panel and battery. The voltage is 14v from solar panel to 13.3 to battery. This is not the issue.
FROM the schematic you can see the ground from the battery is not being referenced by R2 and R3. Which sould give the voltage indication of the battery to OPAMP pin2. Pin3 voltage is set so that OPAMP output is 0 when battery desired voltage is reach.
The ground from the battery is not being referenced by R2 and R3, there I am not getting good voltage division. I cannot short both grounds for obvious reasons. Anyone has an Idea to get around it ?
SB_Charger.jpg
ken
 

Ian0

Joined Aug 7, 2020
9,670
Power the control circuit from the battery, not the solar panel, then use the MOSFET to switch the solar panel.
I would suggest:
1) Use a comparator, not an op-amp. Op-amps don't always make good comparators.
2) Switch the solar panel on at 13.5V and off at 14.5V by adding some hysteresis.
You won't need the 7812.
 

Thread Starter

captoro

Joined Jun 21, 2009
207
Power the control circuit from the battery, not the solar panel, then use the MOSFET to switch the solar panel.
I would suggest:
1) Use a comparator, not an op-amp. Op-amps don't always make good comparators.
2) Switch the solar panel on at 13.5V and off at 14.5V by adding some hysteresis.
You won't need the 7812.
If I power from the battery then I am using energy to make the comparator circuit all the time instead just when there is day light.
And I'm still going to have the issue with no common ground.
Solution :
I think by removing the FR diode and placing a PMOS instead would work. Also removing the N-MOSFET and grounding J2-2 .
what you think ?
 

Ian0

Joined Aug 7, 2020
9,670
A comparator such as TLV2302 uses 1.8uA. During the night on average it will consume 21.6uAh of charge from your battery, that's 93mAh over 6 months. Some batteries will self-discharge half their capacity in 6 months. If the battery is bigger than 1Ah, you'd never notice the current used by the comparator.
Why does the solar panel need a common ground with the battery?
 

Thread Starter

captoro

Joined Jun 21, 2009
207
If I dont have a common ground I would either have floating voltage on the battery or solar panel ( my reference).
What you think about adding this PMOS ?
 

Ian0

Joined Aug 7, 2020
9,670
P-channel MOSFETs are less efficient than N-channel, and cost more money. Usually worth avoiding. You can switch either terminal of the solar panel - it doesn't matter.
If you had a switch, and you turned it off, wouldn't you have floating voltage on the solar panel?
Assuming it's in a box in your shed because you haven't built your circuit yet - at this very moment it's got floating voltage on it!
Assuming that neither side is connected to a metal rod driven into the ground - you have floating voltage on it.
 

Thread Starter

captoro

Joined Jun 21, 2009
207
I get you on the floating voltage.. He is my issue
The comparator compares the value from the battery with a reference voltage, now the reference is coming from the solar panel LM7812.., I can transfer that to the battery side, but thats another IC that will continously sink out energy from the battery.
Yes once the MOSFET is turned on to ground, THEN you have a common reference and no more floating, BUt in the other case floating voltage I get ramdon values in input of the comparator, for me thats unstable.
 

Thread Starter

captoro

Joined Jun 21, 2009
207
Thanks for the LM4040, looks like an improved zener diode for voltage reference. I will highly consider it.
I still dont know where I can add that N-MOSFET with no common ground.
 

Thread Starter

captoro

Joined Jun 21, 2009
207
OOO, I think If I power everything fromn the battery like you said, then Put the NMOSFET on J1-2 to ground.
what you think ?
 

Ian0

Joined Aug 7, 2020
9,670
Apologies for the untidy sketch, try this. Reference and op-amp on the solar side, sense resistors on the battery side. The schottky stops current flowing back into the panel when it is dark.
0ECD7014-4D0D-420C-B14D-5AE068FCB3CC.jpeg
 

bassbindevil

Joined Jan 23, 2014
824
Have you seen how cheap a PWM solar charge controller is on ebay? Under $10 for one that includes an LCD voltage display and two USB charge ports.
If I was going to build one, I'd aim for a more sophisticated (and more efficient) MPPT charger controller.
 

Thread Starter

captoro

Joined Jun 21, 2009
207
Apologies for the untidy sketch, try this. Reference and op-amp on the solar side, sense resistors on the battery side. The schottky stops current flowing back into the panel when it is dark.
View attachment 273825
Apologies for the untidy sketch, try this. Reference and op-amp on the solar side, sense resistors on the battery side. The schottky stops current flowing back into the panel when it is dark.
View attachment 273825
Thanks Ian. I was on the right track but now use the battery to power all the IC instead of the solar panel. Cheers
 

Thread Starter

captoro

Joined Jun 21, 2009
207
Hi,

I dont think that circuit will work. The battery on the right has no common ground with the Comparator,
 

Ian0

Joined Aug 7, 2020
9,670
Hi,

I dont think that circuit will work. The battery on the right has no common ground with the Comparator,
It doesn't need to. Everything is referenced to the positive supply so that you can use an N-channel FET. You can turn it upside down to use a P-channel FET if you want to, but all that will happen is that it will cost you more.
 

Thread Starter

captoro

Joined Jun 21, 2009
207
It doesn't need to. Everything is referenced to the positive supply so that you can use an N-channel FET. You can turn it upside down to use a P-channel FET if you want to, but all that will happen is that it will cost you more.
Hi,
I have not much experience with P-FET, but from what I understand, I would need a negative voltage to fully activate it, which I dont have in the circuit ? Am I wrong ?
 

Ian0

Joined Aug 7, 2020
9,670
Hi,
I have not much experience with P-FET, but from what I understand, I would need a negative voltage to fully activate it, which I dont have in the circuit ? Am I wrong ?
Yes. It just has to be more negative than the source terminal, which would be at 12V. 0V on the gate would switch it on.
Just take my drawing, and reflect it about a horizontal line, reverse the direction of the diodes, and you have the correct circuit for a P-channel FET.
Everything works the same, but you are £2 worse off.
 

Thread Starter

captoro

Joined Jun 21, 2009
207
Yes. It just has to be more negative than the source terminal, which would be at 12V. 0V on the gate would switch it on.
Just take my drawing, and reflect it about a horizontal line, reverse the direction of the diodes, and you have the correct circuit for a P-channel FET.
Everything works the same, but you are £2 worse off.
HI,

I would put the P-MOSFET between the solar panel and battery. Both solar panel and battery grounded together.
 

Thread Starter

captoro

Joined Jun 21, 2009
207
HI,

I would put the P-MOSFET between the solar panel and battery. Both solar panel and battery grounded together.
How about that ?

Also, In my circuit, seems I have a floating voltage on the solar panel. On your circuit you have swapped the solar panel and Battery. It seems you now have floating voltage on the battery. Why would your circuit work better then mine ?
I am not saying your circuit would not work, I am trying to understand why.
 
Top