0
\$\begingroup\$

I have a system that is powered with 3-5 VDC but there are 50ms gaps where power is interrupted and I would like the system to continue running normally when that happens. At 3.3V, the current may have very brief peaks of 1A, but usually is a quarter of that.

How can I prevent this power interrupt from crashing my system? Would a capacitor in parallel with the power source work? If so, what size cap? Other ideas? I don't want to bridge much more than a 50ms gap because I want it to sometimes shut off purposefully

\$\endgroup\$
1
  • 3
    \$\begingroup\$ Cap won't work for 1A with such a long disconnect. Where is the 3.3V coming from? It is always better to put a hold-up capacitor on the higher and unregulated voltage than on the regulated voltage. If the 3.3V is coming from a regulator, it may be better to put a capacitor on the input to that regulator. \$\endgroup\$ Commented Jul 29, 2021 at 4:07

2 Answers 2

Reset to default
3
\$\begingroup\$

You can calculate voltage drop versus time as i = C dv/dt. If i=1A, dt = 0.05seconds, and you didn't want more than, say, dv = 2V (voltage drop during cutout). Then C = i * dt / dv = 1 * 0.05/2 = 0.025 farads. That's a very large capacitor. Depending on what you are up to, it might be easier to diode-OR in an auxiliary power supply. 50ms is a long down time if there's an appreciable amount of current draw.

\$\endgroup\$
5
  • \$\begingroup\$ 3.3 V supercaps and ultracaps are quite reasonably priced and sized for this sort of application. \$\endgroup\$ Commented Jul 29, 2021 at 10:26
  • 1
    \$\begingroup\$ @DaveTweed Indeed so. I was working with them a decade ago. Typical part: csicapacitors.com/3000f-ultracapacitor \$\endgroup\$ Commented Jul 29, 2021 at 11:01
  • \$\begingroup\$ @PeterSmith: That might be a bit of overkill here :-) I was thinking more in terms of devices in the range of 0.2 to 1.0 F that are about the size and shape of a coin cell. \$\endgroup\$ Commented Jul 29, 2021 at 11:51
  • 1
    \$\begingroup\$ @DaveTweed I had to test some to destruction to assess the potential use case for power holdup in avionics. I could have sold tickets :) \$\endgroup\$ Commented Jul 29, 2021 at 11:55
  • \$\begingroup\$ @Peter Smith would that more of electronics were such a spectator sport. \$\endgroup\$ Commented Aug 14, 2021 at 1:32
1
\$\begingroup\$

You don't specify how low the supply can go before your circuit resets, or how your supply is structured, but one thing you can do is use a small battery in parallel with your normal supply both feeding the power rail via diodes.

You may be able to get a modest size cap in parallel with two CR2023 in series as your backup supply. The CR2032 will only put out about 0.1A but in parallel with a capacitor, they may just do the trick.

Alternatively, you may consider regular primary batteries such as AAA or 1/2AA which would more than cover the current but you need to make sure your circuit works at the voltage produced by these batteries minus the diode drop.

\$\endgroup\$

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.