0
\$\begingroup\$

I have a circuit taken from an FTDI app note that takes USB +5V VBus and filters it into 5V Vcc. This circuit works well. I now want to modify this circuit to output +3.3V and I am unsure exactly what to do about the filtering. Should the filtering be placed before or after the 3.3V regulator? Is it best to filtering the 3.3V after the regulator (bottom drawing)? Or is it better to filter the 5V and then run that through the 3.3V regulator (top drawing)? Below are circuits for both configurations. enter image description here

\$\endgroup\$
  • \$\begingroup\$ Can you share a link to the FTDI app note you mentioned? \$\endgroup\$ – The Photon Jun 18 '19 at 18:17
  • 1
    \$\begingroup\$ Two questions to ask yourself: 1. What does the LM3940 datasheet say about input (and output) capacitors? 2. What does the USB spec say about inrush current on the VBUS line? \$\endgroup\$ – The Photon Jun 18 '19 at 18:17
  • \$\begingroup\$ Here's the FT230SX datasheet link: ftdichip.com/Support/Documents/DataSheets/ICs/DS_FT230X.pdf \$\endgroup\$ – Doug12745 Jun 18 '19 at 19:07
  • \$\begingroup\$ Photon, thanks for pointing me there. The LM3940 indicates that the input should "A 0.47-μF capacitor should be connected to the input, and A 33-μF low ESR capacitor should be connected to the output. So it looks like these two caps are going to be necessary. So I'm not sure about whether or not to eliminate the inductor and other three caps. I'm going to see if I can model this circuit. Will post results if I am able to do so. \$\endgroup\$ – Doug12745 Jun 18 '19 at 19:20
  • 1
    \$\begingroup\$ A ferrite bead is probably not meant for inrush current limiting but for noise suppression. But you should be aware there are inrush current limits in USB, and they will limit how much capacitance you can connect directly to the VBUS line. \$\endgroup\$ – The Photon Jun 18 '19 at 20:14
1
\$\begingroup\$

You likely can do away with the filtering all together, as the linear regulator will filter out the noise sufficiently. If you want to be conservative, though, you should do it as shown in your first circuit. (Probably with a small electrolytic on the output for stability).

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

The approach i would use is to put an LC filter before the regulator with generous capacitance lets say 1-10uF with a cut off frequency in which the regulator has a really good psrr (datasheet for that) and then in the output side a couple of ceramic capacitors, one around 4.7uf and the other one 100nf, so you have low impedance over a wide frequency range

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

Since I can't see how to post a picture in a comment, I will do it this way. diegogmx, I incorporated your suggestions. The TI datasheet indicates that the highlighted is needed. So I placed a 10UF and ferrite as the incoming filter, connected that through TI's required circuit, and added a 100nF to the output. Diegopmx, is that similar to what you were suggesting? And thanks to all who contributed.

enter image description here

\$\endgroup\$

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.