0
\$\begingroup\$

Working on a project with a source USB type C port. Below are the criteria:

  • ESD protected D+/D- and VBUS
  • Source only port, will not accept power input
  • 3A power capabilities, with the ability of the microcontroller to adjust the power output
  • Low cost application without the need to implement a USB PD controller
  • Phones/devices will recognize this port as an available USB power source and sink the desired current (up to 3A)

Below is my schematic thus far. From left to right:

  • The TPS2557DRBR regulates the current based on the resistance on ILIM. NTZD3154NT1G is a dual logic level n-channel MOSFET controlled by the MCU to adjust the resistance.
  • The DMG2305UX and DMMT5401-7-F create an ideal diode and only allow power to source from the USB port
  • USBLC6-2P6 provides ESD protection on the D+/D- and VBUS (Note, the symbol is incorrect, the specific component uses bidirectional diodes. Also, TPS2557DRBR provides ESD protection on the VBUS line)
  • USB-7070C is a 16 pin USB type C port. The 10K resistors on the CC lines should indicate to sink devices that the port can supply 3A.

Schematic of the USB-C power output

Since my experience is somewhat limited, I would appreciate any feedback on this design and if it would achieve the goal and if there are any critical faults.

Thanks, Snoop

Edit:

Thank you all for the helpful feedback on the initial schematic. Below is a revised version to address the issues raised. The main issue was due to VBUS always being powered. With the added P-FETs the CC lines will first need to be pulled low to power VBUS. The reverse polarity protection was removed as well.

I was considering a dedicated USB-C PD controller but from the options I was looking at (TUSB320, WUSB3801, Fusb302mpx) they all had limitations without using the I2C interface and didn't add any functionality to the design.

Something else to note, I could connect a GPIO from the microcontroller to each of the cc lines and monitor if they're pulled low. Then connect the EN pin from the TPS2557 to another GPIO to active the power output. This would then give more control over the port but at the cost of 3 GPIO and more passives. Not sure if there's really any use cases for this.

enter image description here

\$\endgroup\$
10
  • \$\begingroup\$ For what purpose are the data pins? Is this a USB host, which connects to a USB device and gives power to it? \$\endgroup\$ Commented Nov 20, 2024 at 5:27
  • \$\begingroup\$ What stops 5 V being applied to the USB connector when nothing is plugged in with pulldown resistors? That's what the EN input of the TPS2557 is designed for. \$\endgroup\$ Commented Nov 20, 2024 at 9:51
  • \$\begingroup\$ Also, "the symbol is incorrect, the specific component uses bidirectional diodes" - no, it doesn't. That would short out the 5 V supply. \$\endgroup\$ Commented Nov 20, 2024 at 10:11
  • \$\begingroup\$ Finally, given the maximum reverse leakage current of 1 uA for the TPS2557 I don't see the need for Q6/Q7 at all. They'll also negate the ESD protection inside the TPS2557 by getting in the way. \$\endgroup\$ Commented Nov 20, 2024 at 10:31
  • 1
    \$\begingroup\$ @SnoopDogg No USB device will output 20V when you plug it in. Unless it is damaged/faulty. It needs PD communication to change from 5V to 20V. Basically you are trying to protect from scenario that should not be possible or improperly made device, but still making an improper device yourself by outputing always 5V on a receptacle. \$\endgroup\$ Commented Nov 20, 2024 at 17:25

1 Answer 1

Reset to default
1
\$\begingroup\$

The receptacle is not allowed to output 5V unless a device that requires power is detected to be plugged in.

While you don't say how you intend to repurpose the USB pins, you either leave them disconnected or be host on this connector or provide some means of USB BC or QuickCharge communication.

A phone plugged it may not draw much current at 5V only, even if your port advertizes 3A.

Anyway, with the resistors, you set the role of the port to be DFP which means it provides power and is USB host role. There is no role swap mechanism to keep sending power but change the data role to be USB device.

\$\endgroup\$
6
  • \$\begingroup\$ Is this part of the USB-PD spec to not supply power unless a device is connected? From my understanding USB hosts always supply 5V even without a device connected. Older style USB connectors have lack pins to determine if a device is connected and PD is backwards compatible. Think of the data pins as I2C. The idea is the user can either connect the add-on accessory that uses the data pins or just connect their own device for power. \$\endgroup\$ Commented Nov 20, 2024 at 18:37
  • \$\begingroup\$ @SnoopDogg Yes it's in the specs. And Type-C is nothing like Type-A because you can have Type-C receptacle on both UFPs and DFPs and that's why you have Type-A and Type-B receptacles for device roles. No, Type-C receptacles are not allowed to output 5V unless a device is detected, because then it means a short circuit if you plug in a charger, as plugs on chargers and A-to-C adapters are allowed to always have 5V on them. No, PD is not even used on your design, PD is data packets between devices. No, the data pins should not be used as I2C. \$\endgroup\$ Commented Nov 20, 2024 at 18:53
  • \$\begingroup\$ I think my confusion here is between host, device, receptacle and plug. This port is a plug and should act accordingly and provide power if a USB device is plugged in. What I could do is, add a P-FET between VBUS and 5V and connect the gate to the CC pins (on FET for each CC and have them in parallel). Then VBUS will only be powered if a device is connected. But if they use a type-C to micro cable then they wouldn't have power? It appears my pull up resistors need to be adjusted to better reflect a plug design. And yes, I know PD is not in the design, I thought you were referring to PD spec. \$\endgroup\$ Commented Nov 20, 2024 at 20:00
  • \$\begingroup\$ @SnoopDogg That USB connector on your design is a socket/receptacle. It is not a plug. The CC lines are pulled up. Therefore, the role of your gadget is a "DFP" or "Host" or "Charger" or "Powerbank" etc. If you connect a Type-C to MicroUSB cable, the cable plug will have a resistor to pull CC low to signal it needs power. Another device with Type-C receptacle like SSD or plug like flash drive will also have a resistor to pull CC low. \$\endgroup\$ Commented Nov 20, 2024 at 22:44
  • \$\begingroup\$ "Is this part of the USB-PD spec to not supply power unless a device is connected?" It's part of the USB-C spec, not USB-PD. And it's not just a device being connected, it's a device being connected AND requesting 5 V using a pulldown resistor. \$\endgroup\$ Commented Nov 21, 2024 at 13:04

Your Answer

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

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.