Hi guys,
I'm designing a device powered by a USB-C port, but I have some questions due to my limited experience.
I would like to draw between 1A and 3A at 5V from the USB-C port without Power Delivery.
I understand that depending on the pull-up resistance Rp (power supply or host side), the device can determine the maximum current available from the power supply/host itself.
I also want to protect VBUS from overcurrents, so I believe the right way to do this is to implement an eFuse that can read the analog voltage of pins CC1 and CC2 and change the set current limit accordingly.

Are there any reference designs or specialized ICs?

 

There's a very good article here: https://www.allaboutcircuits.com/ne...-power-exchanges-with-without-power-delivery/ that should give you a better insight.

I also want to protect VBUS from overcurrents, so I believe the right way to do this is to implement an eFuse that can read the analog voltage of pins CC1 and CC2 and change the set current limit accordingly.

I'm not sure this makes sense. The USB-C source end is responsible for overcurrent protection, to limit the current supplied to 500mA,1A or 3A as applicable to the capabilities of the supply. The device drawing current does not, conventionally, implement overcurrent protection as it draws whatever current it needs to draw. It can, optionally, read the CC voltage to determine current availability and alter its functionality accordingly.

 

I have not seen any USB wiring that I would use for 3 amps current, if reliability mattered.

 

I have not seen any USB wiring that I would use for 3 amps current, if reliability mattered.

Still stuck in the past? USB-C cables rated for 3 amps and even 5 amps are available in accordance with the USB-C PD standard. I have an Apple charger rated at 140 watts – 5 amps at 28 volts.

 

It is not the first time that I have been lied to.
On many occasions that is because others assign different meanings to words, either in ignorance or for their own purposes.

I do regard trusting every conductor as being adequate as a poor choice.

And besides that, consider that the past is often more accurately known than the current or the future..