how to discharge caps and inductors at powerdown?

Jul 23, 2003 5 Replies

Hi guys, I noticed a problem on a board I designed: when I power off the board, the inductor and the capacitor of the LC filter (470 uH and 1000uF) keeps powering the voltage regulator for a certain amount of time. The bad thing is that during power down I noticed an oscillatio of the regulated +5V



connector --> LC filter --> voltage regulator --> ICs



How can I get rid of all the reactive enrgy stored in the LC filter, at power down? I don't want this energy go to re-power the voltage regulator.



In principle I need a resistor that is active just for the time of power down.



thanks Enrico


Hmm, what about a parallel reverse diode?

Bernhard Roessmann Don't Fear The Penguins!

"Spehro Pefhany" schreef in bericht news: snipped-for-privacy@4ax.com...

Good idea, but I would like to know why the regulator oscillates.

-- Thanks, Frank Bemelman (remove 'x' & .invalid when sending email)

Enrico,

What is the frequency of the oscillation we taking about here? and is the

1000uF the only C on the input side?

Normally you would also add some smaller caps (e.g. 1* 220n and 1*22p) for the higher frequencies. (This is also the case at the output side ). Large C's have a relatively large internal inductance.

Furthermore, do you need such large C on the input side if it's powering the system for a while? The system is rather "high impedance" (no much current draining) otherwise it couldn't be powered by the C for a while. You may consider a smaller buffer cap.

gerard.

formatting link

You may want to use a regulator with SHTDN input. Many of this kind are available from any major semiconductor vendor. Once the input power is low, assert the SHTDN signal to switch the regulator off instantly.

Vladimir Vassilevsky, Ph.D.

DSP and Mixed Signal Design Consultant

formatting link

Enrico Migliore wrote:

There's one more thing I should say about this:

Many voltage regulators do not like having the output voltage being higher than the input voltage. The circuit I described above can cause this situation. So the solution is to add a diode in parallel with the input and output terminals of the voltage regulator. For a positive voltage regulator, the diode's cathode would go to the input terminal of the voltage regulator.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required