TCPP01-M12: How to read CC lines with ADC in 5V application

[sb_st]

Hi!

 

I’m slowly working my way through a design in which I plan to use the TCPP01-M12; my device is a USB-C sink, and I don’t intend to utilize any USB-PD negotiation. I have a conflict with my chosen MCU and the UCPD peripheral (the pins of this peripheral need to be allocated to another use), so I am interested in being able to read the CC lines of the USB connection to understand the power capabilities of the upstream source, without dealing with any actual CC communication. 

 

The datasheet for the TCPP01-M12 shows this exact use, on Page. 18, Figure 20. As I understand it, I can use my ADCs to read the voltages on the CC pins to calculate the current delivery capability of the source. What confuses me, however, is that apparently the source will also try to start communicating over the CC lines on connection. As I understand it, this communication happens at a logic level of between 0-1V or so, regardless of the actual current carrying capabilities. 

The obvious problem is that if I sample the CC lines with my ADCs at the wrong time, I’d read invalid/incorrect voltage values, and have a false understanding of the source.

 

I’m finding it tricky to find resources that describe how I might be able to do this properly. I’m curious if anyone else has done this, and if so, how did the CC reading part of things work? 

 

Thanks!

[mƎALLEm]

Hello ​@sb_st ,

You are talking about TCPP01-M12, could you please tell why are you posting that question on STM32 MCU products forum board?

[sb_st]

Hi;

 

Well, I suppose: I found this example in the TCPP01 datasheet, but my question is not actually about the TCPP01 - it’s about how to sample the CC lines via an ADC. My apologies that the subject line is misleading (which I realize only in hindsight). It’s not evident to me how I can go about modifying/correcting the subject...

[Andrew Neil]

It’s not evident to me how I can go about modifying/correcting the subject...

See:

 

[FBL]

Hi ​@sb_st 

 

A PD capable source may send BMC signaling on CC after detecting Rd, so a direct ADC sample can be misleading. I assume X-CUBE-TCPP includes the exact application you need. I think this thread should be moved to Embedded Software forum.

[sb_st]

I understand,  yes - that’s sort of the basis of my question. And I’m curious how much this intersects with my hardware design - like, would lowpass filtering be applicable/useful in this situation?

Trying to muddle through this with Claude - it is suggesting that a lowpass filter is unlikely to mitigate the BMC signaling, as that signaling happens at a different voltage level than some of the default contract voltages. But it still recommends SOME filtering, and it suggests taking multiple samples with an ADC, then using a median function to identify a “good” sample from the BMC messaging ones. 

 

I was hoping to get the perspective of a human rather than AI, however, for obvious reasons. The example you mention does seem like it might hold some answers for me - thanks so much! 

I don’t really know where the best forum is to ask this question.

[LCE]

Why do you want to use the TCPP in the first place?

If you don’t communicate, I’m pretty sure you get the minimum 5V / 1.5A from USB-C.

 

I have a design with a USB-C connector only using USB 2.0 communication, and I only need the above minimum power.

To find out what the USB host can deliver, the CC pins are pulled down with 5k1, these signals go to the STM32H7 ADC.

In my schematics I find this note:

CC pull-ups to 5V at host:
56k -> Vucc < 0.45V -> 0.5 A
22k -> Vucc < 0.95V -> 1.5 A
10k -> Vucc < 1.69V -> 3.0 A

… and this is what I find in real life when connecting to either a USB-C or USB 2 port.

I have the H7 ADC running all the time in the background with circular DMA for monitoring about 8 supply voltages, about every 300 ms I check the DMA buffer and do some comparisons, including the CC pins.