1
\$\begingroup\$

This is somewhat a duplicate.

I'm looking for a way to limit the inrush current of a small supercapacitor circuit. Input is regulated 5V (USB), output feeds a 3.3V regulator. Goal is a 10s runtime at 100mA, so the supercap is 5V, around 500mF if my math are correct.

The simplest inrush limiter is well presented in the question above, but I'd like to find something that charges the supercap faster, without loosing the ability to charge it to (almost) the supply voltage. Something like constant current, around 300mA.

The other typical answer to this question is the LTC4425, or another supercap charger from Linear. From what I read, those are

  1. expensive and
  2. designed to handle and equalize two 2.5/2.7V supercap, which isn't warranted given the large availability of 5V supercaps.

What's the intermediate solution?
The system isn't battery powered, so efficiency is not a major consideration.

\$\endgroup\$
  • 2
    \$\begingroup\$ Ic=C*dV/dt applies to both charge and discharge times. so 0.1A=0.5F(5-3.3)/dt if using 0V dropout LDO then dt=8.5s. Thus requires a precision CC current regulator using a 50mV shunt R and Vref=50mV and comparator to FET \$\endgroup\$ – Sunnyskyguy EE75 Dec 2 '16 at 0:06
  • \$\begingroup\$ @TonyStewart.EEsince'75 Thanks for confirming. \$\endgroup\$ – jmr Dec 2 '16 at 0:12
3
\$\begingroup\$

Answering my own question.

Using a P-channel FET as a current limiter seems to be a quite common technique. A good introduction can be found in this EETimes article.

The technique has been packaged in multiple $1-$2 ICs, including: https://www.fairchildsemi.com/datasheets/FP/FPF2123.pdf http://www.diodes.com/_files/datasheets/AP255x.pdf http://ww1.microchip.com/downloads/en/DeviceDoc/mic20xx.pdf

This page is an example of someone using the Fairchild chip in a supercap charging application.

Using LTSpice, I compared the "limit the current with a resistor" approach (presented in the first link of the question) with the PNP+FET approach (presented in the EETimes article above). The PNP+FET circuit is roughly 3x faster.

\$\endgroup\$
0
\$\begingroup\$

Actually the LTC4425 is a great solution to this type of problem and the chip is only 5 dollars from Digikey and the supporting components can be had for less than 2 dollars. Other solution could easily costs 5 dollars in supporting components. This chip will work with a single 5v capacitor with the addition of two small capacitors. Check out Linears web site and check out the datasheets. There are numerous other chips design for this purpose and some of them may also deal with your other problem which is routing power to the load from either the capacitor or the USB power source.

Most of my designs start out with looking for something that already exists that does all or as much possible of what I need it to do. I am currently work on project that does not use supercapacitors but instead using large conventional electrolytic capacitor to provide power to a load during startup in an automobile. I simply used the capacitor with nothing else before and it worked well but the inrush current is a problem.

My first design was focused on just dealing with the inrush current. My latest design takes a more comprehensive approach. In an automotive environment you also need to take into consideration the much higher than 12V can be supplied to your circuit at times and with high current levels available to fry things up crispy like. My new design also take into consideration protected the load and the capacitor for over-voltage events.

Take advantage of the work other very smart engineers have put-in to make your task easier. There is still plenty to learn from building something that someone else put the time and money into to figure out how it do it right. And it gives you time to thing of what else you might want to circuit to do.

\$\endgroup\$
  • 2
    \$\begingroup\$ There's nothing "only" with an IC that costs $5+$2, that's a very expensive current limiter. \$\endgroup\$ – pipe Apr 22 '17 at 4:44
  • \$\begingroup\$ There is more than just current limiting to be done here. The USB port can provide up 5.25V by USB spec which if applied to the supercapacitor will damage it once it has charged enough to not drag the voltage down. \$\endgroup\$ – Brian Nebeker Apr 26 '17 at 2:04

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.