I have a 21V DC solar panel power source, that feeds a DC-DC boost converter. I set the output voltage of the boost converter to 19.5V to charge my laptop, but since the input voltage is higher than that, the output voltage is also 21V. I cannot use a buck converter, because when I draw current from the boost converter, input voltage drops below 19.5V, so it must be boosted to stay at 19.5V. How can I prevent the initial (when first plugged in) voltage to be below 19.5V?


On mains power, the laptop draws about 1.5A of current at 19.5V, and that is about 30W. My solar panel is 50W under direct full sunlight.

The PV panel's Voc is 21.46V, Isc is 3.21A,
Vmp is 17.46V, Imp is 2.98A.
Is does not have a part number.

  • 5
    \$\begingroup\$ You need a buck-boost converter. \$\endgroup\$ – John D Jul 2 '20 at 15:08
  • 1
    \$\begingroup\$ @Ilya Sadly, what you say is correct in dome but not in all cases. Some laptops have buck-boost internal circuitry and can accept a range of voltages. Others - and I have seen a number of these - will not accept less than a voltage quite close to the formal spec. For example, with a 4S LiIon battery, Vmax at battery = 4.2V x 4 = 16.8V. The charger needs some headroom so maybe 18V in some cases and maybe somewhat more for some. If the system CANNOT boost the input then it will not work if Vin is too low. ... \$\endgroup\$ – Russell McMahon Jul 4 '20 at 1:04
  • 1
    \$\begingroup\$ ... On the other hand I have seen an HP laptop that had 3S and 4S battery packs available and it would happily charge either from a 19V adaptor - HP are masters of power management, unlike some. || Also, if a charger is loaded below its related voltage it may attempt to supply more than rated current - this may cause shutdown and may cause damage depending on design. If say an adaptor is rated at 19V 3A and is loaded to draw 5A it may drop to 16V. The laptop MAY reject the voltage, the psu may shut down, it may fail or it all may just work. \$\endgroup\$ – Russell McMahon Jul 4 '20 at 1:07
  • 1
    \$\begingroup\$ ... In the case of a PV panel - if it works at lower voltage then no damage is likely BUT you will usually not get more than about an extra 5 to 10% under heavy load below specified voltage. \$\endgroup\$ – Russell McMahon Jul 4 '20 at 1:08
  • 1
    \$\begingroup\$ @EmreTapcı Thanks. You MAY be able to get this to work by adding a large capacitor across the panel output - but a boost converter with Vout ~=19V out is more likely to work. || Those figures match what I and Tony suggested reasonably well. Vmp/Voc = 81% and Isc/Imp= 8% more current. || The panel has enough current and power capability but probably not enough voltage capability. | IF you have a suitable variable power supply avail;able you can find where the laptop's cutout point is - supply 19V at enough amps and then reduce Vpsu until the laptop stops charging. \$\endgroup\$ – Russell McMahon Jul 5 '20 at 22:28

No this will never work well.

A Solar Panel is an optical-powered current-source that starts at a high impedance with no power. It has an open-circuit voltage, Voc and a short circuit current, Isc. In order to capture maximum power, there is a law that states the source & load impedances must be matched. There are many algorithms to maximize this power transfer or MPPT, e.g. hunting, and setting the approximate voltage based on a solar input sensor etc.

At max. solar input the Vmax is ~ 82% for Pmax out. This also defines the MPPT impedance R = V/I characteristic of the active source to the load.

However, a mobile smart battery charger expects a constant voltage and draws current according to its State of Charge (SoC). e.g. 19.5V +/-10%. Usually starting at constant current, CC then CV then cutoff at 10 to 5% of CC.

So you need to regulate both the MPT impedance to get maximum power and supply a fixed voltage within a reasonable range of 19.5V. This DCDC smart charger will have its own battery charger and step down regulators for logic and analog voltages simultaneously so it does not have to be exact, yet to get max power from the panel, it does need to be more precise..

You can choose an MPPT dual regulator for your battery and Solar Panel chemistry or something that matches the V/I characteristic your Solar Panel and conditions and charger voltage range of say 10%.

The panel needs to have Voc=24V to make a simple efficient regulator and have enough power to start the mobile charger so it must be sized greater than your maximum charger power rating to get started.


As soon as you plug in your laptop and it starts charging, the voltage will drop down to 19.5V and everything will be okay. There's also a decent chance that 21V is within tolerance for your laptop anyway and it won't be harmed. But, just to remove the risk, you want to make sure it never ever goes higher than 19.5V. Fair enough.

You don't want to drop the input voltage when it's below 19.5V, because then you're needlessly wasting power.

With that in mind, I suggest that you can connect a shunt regulator to the input. A shunt regulator is a device which, when the voltage is too high, draws current to bring the voltage back down. When the voltage is low enough, it doesn't do anything.

The simplest type of shunt regulator is a Zener diode, but remember this has to absorb some of the power from the solar panel when the laptop isn't plugged in, and I don't think they make 50-watt Zener diodes.

The basic design is to put a transistor in parallel with the input, and then adjust the transistor's base current (or gate voltage) until it's dissipating enough power to bring the voltage down to where you want it - something you can do with an op-amp. Putting an extra resistor in series will shift some of the power dissipation to the resistor, which might be more tolerant of high temperatures. Here's a rough draft:


simulate this circuit – Schematic created using CircuitLab

OA1 adjusts the current through R1 and M1 (for some reason I couldn't rename it to Q1) until the + input voltage (1/3 of the supply voltage) matches the Zener voltage (1/3 of 19.5V). Both R1 and M1 will need heatsinks.

  • \$\begingroup\$ Sadly, it seems that while the PV panel Voc is above spec, the Vmp is probably BELOW the laptop charging voltage and the system tends to cycle in and out of charge . \$\endgroup\$ – Russell McMahon Jul 6 '20 at 10:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.