So, basically, using high voltage series solar-solar panel arrangements and a buck-converter to step down to 2.7V for a supercap array.


Using a parallel solarpanel arra with a voltage close to a series supercap array and using a buck-boost converter.

  • 1
    \$\begingroup\$ You haven't asked a clear question. \$\endgroup\$
    – Transistor
    Apr 1, 2020 at 21:08
  • \$\begingroup\$ You have to be careful here. You mention using solar array (SA) connection close to the supercap max voltage, but have you studied the I V characteristic of the SA? Even if you just compare the max power point of the array to the open circuit voltage, there is a wide voltage variation. Spacecraft solar power systems have many approaches from Shunt Regulators that throw away unneeded power, to switched array segments, to SMPS peak power trackers, just to name a few. In other words, you will need a regulator in any approach. \$\endgroup\$
    – xstack
    Apr 2, 2020 at 2:09
  • \$\begingroup\$ @xstack if we take a generous 0.75V maximum open circuit voltage per single cell, we could use 4 Cells in series and with basic cap protections that burns off anything above rated voltage we should be so far at the end of the iv curve that what needs to be burned off isn't that much. \$\endgroup\$
    – Horst
    Apr 2, 2020 at 6:58

1 Answer 1


For a hobby application I agree with you. There are though, quite a few variations with cell type, temperature, cell internal resistance that could significantly change your performance.

enter image description here

Looking at the above graph, the Voc is roughly 1.19 x Vmp (max power voltage) So sizing the array for max voltage will mean that you are very roughly 19% lower voltage at max charging rate.

That's not so bad, but then take temperature into account.

enter image description here

If you scan the Net, you will see manufacturers ratings of -0.3%/deg C to -0.9% deg C. Take the middle ground and look at a +25 deg C change; -0.6%/deg C

-0.006 x 25 = - 0.150 ( - 15 % )

Summary so far; -19 % max power to Voc -15 % for temperature -34 % overall

0.34 x 0.75 volts = 0.25 volts less

So this is what you are left with;

0.75 volts - 0.25 volts - 0.5 volts.

It is important to understand that all these numbers given depend heavily upon the many other factors like solar array material, architecture, percent of full illuminatioon, etc.

See next curve.

enter image description here

We haven't even talked about internal cell losses ......

Anyway, at the end of the day, you can experiment as you have mentioned, but do put SOME PROTECTION for the supercap. At least have a dumb zerner diode to clip excess SA voltage and be sure to heat sink it well. Put the diode directly across the incoming array connections.

Hope this helps.

  • \$\begingroup\$ Thanks for taking the time! Considering supercap protection I think I will go with the TL431 approach as it seems to fill my needs quite nicely. Thinking further along those lines of efficency, a PWM control targeting a fixed voltage of a known iv curve should come pretty close to optimal before venturing into mppt, which I think will not add that much benefit since I have some space constraints but don't need to aim for abolut maximum efficency. \$\endgroup\$
    – Horst
    Apr 2, 2020 at 16:47
  • \$\begingroup\$ There isn't too much info how to design a supercap charge controler with a solar panel as input, but from the data you showed, choosing a panel with two times VOC of the Supercap array should be a good bet for cloudy days. A PWM controller could keep the panel at it's desired Voltage (~80% VOC seems to be close). When the supercap-array reaches it's target voltage the pwm controller can shut the panel off since it's already in place and there is no need to burn the access energy off close to the supercaps all the time. \$\endgroup\$
    – Horst
    Apr 2, 2020 at 16:53

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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