TUSB321: please help to check if the design is reasonable

TUSB322i is better it  due to temeprature.

regards

Brian

Hi Brian；

thanks，in  addition，  Is there a problem with my circuit design?After replacement, is the connection circuit for each PIN correct?

did they use ID to control vbus switch?

apply：elevator control panels, replacing NXP with TI. For the TYPE-C interface, there was a problem in the test, the TYPE-C interface supports double-sided plugging, but the customer cannot identify one side when testing, so the customer wants to ask whether there is any abnormality in the schematic

i can't see how did they control  vbus switch with ID. also ID pin need 200k pullup.

regadrs

brian

The ID pin is Pull high on the opposite side, so the resistance on this side of the TUSB321 is removed by NC

They feel that the ID is the output of the TUSB321, and its state should not affect the output of the DIR pin.

is this DRP mode?

Can customer take waveform for CC1/CC2/VBus/ID?

Regadrs

Brian

After they disconnect the VBUS_DET pin, the IC can correctly identify the front and back connections of the TYPE-C cable, please help consult the original factory why this phenomenon is caused, thank you.

the cable is type-c to A? did they add 5.1k pd on CC1?

The test cable we used is a standard C-port to A-port cable, which already includes this 5.1KΩ Rd.，At the same time, this cable can be used normally on the PC。what  cause  this，Could  you  please  help to  check  it？Thanka！

does it has issue for type-C to type-C cable?

We  use  type-C to type-A，The test cable we used is a standard C-port to A-port cable, which already includes this 5.1KΩ Rd.。。At the same time, this cable can be used normally on the PC.

Can you ask customer take waveform for  Vbus, ID, CC1, CC2 and DIR for both direction?

Hi；

as follow, please help to check it ,  thanks,

Figure 1

Figure 2

Figure 1 and Figure 12 Not connected cable 。

Figure 3

figure 4

figure 3 and figure4 are Waveform graph of frontal access。

figure 5

figure 6 

figure 5 and figure 6 are  Waveform graph of reverse access。

do you have ID pullup? ID should not always low.

Vbus_det should be low until cable connected.

Hi Brian；

Our motherboard has two USB-C ports, one is OTG design and the other is Host design. However, the current USB controller only supports host mode, so both USB-C ports actually work in Host mode, so Vbus will be 5.2V as long as it is turned on, and will not be controlled by the ID signal.
In order to simplify the analysis, another Host designed the circuit as follows (ID signal NC), the failure phenomenon is the same, it is recommended that you use this circuit analysis to avoid interference by the OTG description.

its same either OTG (drp) or host (dfp), need to use ID to control vbus.

when id goes low and turn on Vbus.

Regards

Brian

Below is the waveform of the ID signal after adding the 200K ohm pull-up resistor, and the DIR output still does not correctly identify forward and reverse insertion.

ID pull up not enough. I didn't see ID goes low. Is Vbus always high?

VBus should be low until ID goes low and turn on Vbus.

Regards

Brian

The current USB controller only supports host mode, so both USB-C ports actually work in Host mode, so Vbus will be 5.2V as long as it is turned on, and will not be controlled by the ID signal.  So Vbus always high

To simplify the analysis, another Host designed the circuit as follows (ID signal NC), and the failure phenomenon is the same,

Is it clearly stated in the datasheet that the ID pin must be low?

You  said ID pull up not enough that mean Pull-up resistor of  ID pin  value is not enough?

I mean id not just pulled high, also need ID tot control Vbus .For DFP or host, Vbus should be low until Id goes low and turn on Vbus.

This is for CC spec.

Regards

Brian

ID is high when no device connected, it will be low once an uFP device is attached successfully

Hi  Brian；

Really  thanks  for you  help, my  customer already  understand it, but  they  have  another  problem,  please  help to confirm it.  

The SOC module currently in use does not support OTG, so VBUS cannot be turned on or off based on the ID signal.
In this case, if we want to satisfy the Type-C Specification, we need to add an ID signal to turn VBUS circuitry on or off.

So if it is set to DFP, according to the description of the spec, can the VBUS_DET be unconnected?