3
\$\begingroup\$

I'm designing a board with a USB-C upstream-facing port (UFP) for power only (no data). I've included a common-mode choke, as well as a ferrite bead filter in pi configuration. Since I'm presenting as a UFP, I need 5.1 kΩ pulldown resistors on CC1 and CC2.

Should these resistors go before the common-mode choke or after?

In other words, this?

enter image description here

Or this?

enter image description here

One possible consideration is that I will attach CC1 and CC2 to ADC inputs of an MCU (in order to sense the DFP's current capability advertisement).

I've tried to think through this. If there's no common-mode noise, I think the two options behave basically identically, since the CMC appears as approximately a zero-ohm resistor. If there is common-mode noise, I can't quite wrap my head around which option is less bad, both for the DFP and the MCU.

(Not part of the question, but extra credit for pointing out anything else I might improve with this circuit. I'm considering adding ESD shunt diodes for example.)

\$\endgroup\$
4
  • 1
    \$\begingroup\$ ESD is very high frequency, I dont think it would pass through the filter, but it may kill it \$\endgroup\$
    – Anas Malas
    Sep 29, 2022 at 6:24
  • \$\begingroup\$ @AnasMalas True, but strikes on CC1 or CC2 as drawn might kill the MCU. If I spring for protection it'll be on all connector pins so that peak pulse current on the bead isn't an issue. Thanks for the input! \$\endgroup\$
    – TypeIA
    Sep 29, 2022 at 6:28
  • \$\begingroup\$ Oh, I has not realized CC1 and CC2 are connected to the MCU, if you want to do PD, are you sure there need to be pulldowns? My mind went to 5V 2A mode when I saw those \$\endgroup\$
    – Anas Malas
    Sep 29, 2022 at 6:41
  • \$\begingroup\$ @AnasMalas No, I don't want PD, and yes, I'm sure they need to be pulldowns to indicate a UFP. USB-C DFPs advertise their current capability with the strength of their pullup (or current source) forming a voltage divider with the UFP's pulldowns, so we check voltage on CC1 and CC2 to see what the current capability is. Above a certain threshold means 3 A capable. \$\endgroup\$
    – TypeIA
    Sep 29, 2022 at 6:43

1 Answer 1

Reset to default
-1
\$\begingroup\$

The two schematics presented are the same, the choke is not connected on the CC lines and will have no impact.

If there is no need for any other option than having the port configured as a 5V sink, the CC lines can only be connected to the pull down resistors and nowhere else.

\$\endgroup\$
3
  • \$\begingroup\$ Hi, thanks for your answer. 1) They're not the same: in one the return path for the CC pulldowns goes through the choke. In the other the return path is direct to the connector. 2) I will connect CC1 and CC2 to an MCU as described in the question (for the purpose of detecting the DFP's current capability advertisement). \$\endgroup\$
    – TypeIA
    Sep 30, 2022 at 9:35
  • \$\begingroup\$ 1) Oh indeed sorry for that. For protection purposes I would add TVS on the CC lines more than a choke, and connect the resistors directly to the connector pins. 2) Why would you need to measure the DFP current capability ? You have a mean to adjust your power consumption if DFP capability is too low ? \$\endgroup\$
    – Augustin
    Sep 30, 2022 at 10:03
  • \$\begingroup\$ Yes, I control a load switch, and switch on a higher-current (about 1 A on the 5 V side) analog portion of the board only if the MCU confirms the DFP is advertising 3 A capability. // I did mention I'm already considering TVS, but I also really do want a CMC because the board has a non-isolated analog input which could pick up noise via a ground loop. \$\endgroup\$
    – TypeIA
    Sep 30, 2022 at 10:11

Your Answer

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

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