Buck circuit to solar panel
- Thread starter mrel
- Start date
You can get extra current but not extra power.
For instance, if you have a "12V" 1Amp panel for a 12V battery, a buck converter could double the current to 2Amps, but now it will not charge a 12V battery, but only a 6V one.
That is a bit simplified but you can get the idea.
A 12V 1Amp panel is 12Watts. Doubling the current to 2Amps with a buck converter now gives you a 6V 2Amp one, still 12Watts.
For instance, if you have a "12V" 1Amp panel for a 12V battery, a buck converter could double the current to 2Amps, but now it will not charge a 12V battery, but only a 6V one.
That is a bit simplified but you can get the idea.
A 12V 1Amp panel is 12Watts. Doubling the current to 2Amps with a buck converter now gives you a 6V 2Amp one, still 12Watts.
Unless the buck converter has added circuity all it will do is "collapse" the panel voltage to the battery voltage, with some voltage drop between that may or may not make things somewhat worse than direct connection to the battery. This is simply the nature of buck converter that is being operated with input power less than what is required at the output.
Buck converter:
Vout = Vin x duty cycle
power out = power in, assuming 100% efficiency
Let's say you have a very large battery that is at 11 volts and you want to charge it to 13.8 volts. You have a 100 watt panel that is specified to deliver 16 volts and 6.25 amperes at the maximum power point (MPP) with full "insolation." (I'm just making these numbers up for illustration, but they aren't far from what real numbers might be.) The power available at MPP will change with the amount of light, but until the light gets quite low the MPP voltage won't change much. MPP voltage is affected by temperature, but we'll ignore that here.
If you connect your array and battery directly together you will not get 100 W. The voltage of the PV panel will be pulled down to the battery voltage. The current is hard to determine without a graph of the panel characteristics, but it will be greater than 6.25 A. It will be less than the short circuit current but greater than the MPP current. Let's say it will rise to 7 A. The delivered power is now 7 A x 11 V = 77 W
We need to add another detail here to proceed. Let's assume that when the battery is at 11 volts the voltage will not immediately start to rise even if we put 15 A into the battery. It will, of course start to rise, but slowly. Let's assume that at 15 A it would take minutes to get to 11.1 volts. So, to put in 15 A at 11 V we would need 165 watts. We only have 100 watts from the PV array, and that's at MPP. At any other voltage we'll get less power.
An ordinary buck converter regulates the voltage at the output terminals. If the voltage the output begins to fall the converter raise the duty cycle and vice versa. For charging your battery, the output voltage would be set to 13.8 V, but your battery is deeply discharged and it won't let the voltage rise instantly even with a lot of current into it. The buck convert will "instantly" go to maximum duty cycle to try to get the voltage at its output up to the setpoint of 13.8 volts. It can't do it, but our equation Vout = Vin x duty cycle still applies. Let's say the buck converter can go to a maximum duty cycle of 97%. This time we work "backwards" from output to input, because we know the output voltage is "fixed" for the time being. Rewriting the equation
Vin = Vout / duty cycle = 11 V / 0.97 = 11.34 volts
For now, we're assuming the buck is 100% efficient and has no current limiting that would mess with our simple analysis.
We're almost right back where we started, with the panel forced to operate well below MPP voltage. We need to return to the graph we don't have, but we'll estimate that the current from the panel at 11.34 V will be a bit less than what we'd get at 11 V (we're working on the "upslope" away from short-circuit current to MPP current). We estimated 7 amps before, now we'll estimate 6.9 A. So
6.9 A x 11.34 V = just over 78 watts
We've gained next to nothing - and again we are assuming the buck converter is 100 % efficient. It isn't, so we'll actually wind up worse off than without it when all the details are counted.
If we add some circuitry to the buck converter to prevent it from pulling the input voltage down below the MPP point of the panel, then we get some gain. We are using 16 V for the MPP voltage for your panel. Now the buck is going to be forced to operate with 11 V at the output and 16 V at the input. We have 6.25 A available at 16 V (100 W) from the array. All switchmode converters are power converters, regardless of type and voltages. If we assume 100 % efficiency the power at the output is equal to the power at the input. So, now we get the full 100 W at the input into the battery, which is at 11 volts for this moment
100 W / 11 V = 9.09 A
We've made a big gain in the amount of charging current and power we're putting into the battery.
As you can see, the farther the battery voltage is from the PV MPP voltage, the more gain we'll get from a buck converter that is forced to operate at the MPP voltage of the array.
Once the battery voltage rises high enough, the power required to continue charging will fall below the power available from the array and the buck can make the transition from regulating the voltage at the input terminals to regulating at its output terminals as it normally would. If the battery could take it and there is even more control circuitry, we might let the battery voltage go above 14 volts for a while so we can slam in as much charge as possible in the shortest time.
Ordinary buck converter can't work this way. Some extra circuitry can be added (but not easily to an off-the-shelf board) to prevent the input voltage from going below the MPP voltage of the array. Typically you wind up with three control "loops" - one limits the maximum current at the output, one limits the maximum voltage at the output, one limits the minimum voltage at the input. The one that says "no more" at any given instant is the one that actually controls the power conversion.
Buck converter:
Vout = Vin x duty cycle
power out = power in, assuming 100% efficiency
Let's say you have a very large battery that is at 11 volts and you want to charge it to 13.8 volts. You have a 100 watt panel that is specified to deliver 16 volts and 6.25 amperes at the maximum power point (MPP) with full "insolation." (I'm just making these numbers up for illustration, but they aren't far from what real numbers might be.) The power available at MPP will change with the amount of light, but until the light gets quite low the MPP voltage won't change much. MPP voltage is affected by temperature, but we'll ignore that here.
If you connect your array and battery directly together you will not get 100 W. The voltage of the PV panel will be pulled down to the battery voltage. The current is hard to determine without a graph of the panel characteristics, but it will be greater than 6.25 A. It will be less than the short circuit current but greater than the MPP current. Let's say it will rise to 7 A. The delivered power is now 7 A x 11 V = 77 W
We need to add another detail here to proceed. Let's assume that when the battery is at 11 volts the voltage will not immediately start to rise even if we put 15 A into the battery. It will, of course start to rise, but slowly. Let's assume that at 15 A it would take minutes to get to 11.1 volts. So, to put in 15 A at 11 V we would need 165 watts. We only have 100 watts from the PV array, and that's at MPP. At any other voltage we'll get less power.
An ordinary buck converter regulates the voltage at the output terminals. If the voltage the output begins to fall the converter raise the duty cycle and vice versa. For charging your battery, the output voltage would be set to 13.8 V, but your battery is deeply discharged and it won't let the voltage rise instantly even with a lot of current into it. The buck convert will "instantly" go to maximum duty cycle to try to get the voltage at its output up to the setpoint of 13.8 volts. It can't do it, but our equation Vout = Vin x duty cycle still applies. Let's say the buck converter can go to a maximum duty cycle of 97%. This time we work "backwards" from output to input, because we know the output voltage is "fixed" for the time being. Rewriting the equation
Vin = Vout / duty cycle = 11 V / 0.97 = 11.34 volts
For now, we're assuming the buck is 100% efficient and has no current limiting that would mess with our simple analysis.
We're almost right back where we started, with the panel forced to operate well below MPP voltage. We need to return to the graph we don't have, but we'll estimate that the current from the panel at 11.34 V will be a bit less than what we'd get at 11 V (we're working on the "upslope" away from short-circuit current to MPP current). We estimated 7 amps before, now we'll estimate 6.9 A. So
6.9 A x 11.34 V = just over 78 watts
We've gained next to nothing - and again we are assuming the buck converter is 100 % efficient. It isn't, so we'll actually wind up worse off than without it when all the details are counted.
If we add some circuitry to the buck converter to prevent it from pulling the input voltage down below the MPP point of the panel, then we get some gain. We are using 16 V for the MPP voltage for your panel. Now the buck is going to be forced to operate with 11 V at the output and 16 V at the input. We have 6.25 A available at 16 V (100 W) from the array. All switchmode converters are power converters, regardless of type and voltages. If we assume 100 % efficiency the power at the output is equal to the power at the input. So, now we get the full 100 W at the input into the battery, which is at 11 volts for this moment
100 W / 11 V = 9.09 A
We've made a big gain in the amount of charging current and power we're putting into the battery.
As you can see, the farther the battery voltage is from the PV MPP voltage, the more gain we'll get from a buck converter that is forced to operate at the MPP voltage of the array.
Once the battery voltage rises high enough, the power required to continue charging will fall below the power available from the array and the buck can make the transition from regulating the voltage at the input terminals to regulating at its output terminals as it normally would. If the battery could take it and there is even more control circuitry, we might let the battery voltage go above 14 volts for a while so we can slam in as much charge as possible in the shortest time.
Ordinary buck converter can't work this way. Some extra circuitry can be added (but not easily to an off-the-shelf board) to prevent the input voltage from going below the MPP voltage of the array. Typically you wind up with three control "loops" - one limits the maximum current at the output, one limits the maximum voltage at the output, one limits the minimum voltage at the input. The one that says "no more" at any given instant is the one that actually controls the power conversion.
Only if the maximum power point (MPP) of the solar panel output is at a panel voltage greater than the battery charging voltage.
The maximum power point, where the equivalent load resistance equals the panel resistance, is below the maximum open-circuit panel voltage. It is somewhere around 80% of the open circuit voltage, depending on the panel design.
Last edited:
Here's a data sheet for a Kyocera PV module that would typically be used connected directly to a battery with a simple on-off charging controller.
It would yield substantially higher power with an MPP converter. The curves show how the MPP point does not move much with illumination but does move with temperature. I think I have some power plots I made for this module, but I don't know where I've left them if I do. If I find them, I'll post them. The MMPs are close to the knees of the curves. This same sort of curve is also typical of thermoelectric generators.
The Spectrolab TASC is quite cool and just tossed in for interest. They are basically engineered off-cuts from space solar cells. Instead of just wasting the area that is cut off of round cells to allow higher packing density, some clever person decided there would be a market for the scrap if it were actually made into conveniently usable pieces. I did some circuit boards for these once. I don't know what they were used for - one of those "if we tell you we'll have to kill you" military things.
It would yield substantially higher power with an MPP converter. The curves show how the MPP point does not move much with illumination but does move with temperature. I think I have some power plots I made for this module, but I don't know where I've left them if I do. If I find them, I'll post them. The MMPs are close to the knees of the curves. This same sort of curve is also typical of thermoelectric generators.
The Spectrolab TASC is quite cool and just tossed in for interest. They are basically engineered off-cuts from space solar cells. Instead of just wasting the area that is cut off of round cells to allow higher packing density, some clever person decided there would be a market for the scrap if it were actually made into conveniently usable pieces. I did some circuit boards for these once. I don't know what they were used for - one of those "if we tell you we'll have to kill you" military things.
There is a saying. You can't get blood out of a stone. If you increase the current then you decrease the voltage. And vice versa. If charging a batery is your goal then you can't do better than to purchase a charge controller. Morningstar has an excellent line of controllers.
