HD3SS3220: Cable orientation detection query

KennyMillar

Intellectual 385 points

Part Number: HD3SS3220

I have a board with HD3SS3220 on board, configured in DFP mode.

I am having difficulty in that peripheral devices only work when their cable is inserted in one direction.

Looking at CC1, CC2 inputs and the DIR output I see the following:

First orientation:
CC1 : 0.48 V, CC2 : 4.96V, DIR 2.8 V, This orientation works.

2nd Orientation:
CC1: 4.96 V, CC2 : 0.48 V, DIR 2.8V, this orientation does not work.

So it looks like CC1 and CC2 are swapping roles OK, but the DIR output doesn't change, and therefore the MUX inside the HD3SS3220 doesn't flip either.
(DIR is pulled up using a 200K resistor.)

All cables are good quality parts.
As I understand things, there should be a 56K resistor inside the cable, and when combined with the 5.1K inside the HD3SS3220, the voltage readings make sense.

What could be causing the cable orientation to not work?

Have I missed something?

over 6 years ago

0 David (ASIC) Liu over 6 years ago

TI__Guru**** 177301 points

Kenny

How is the VBUS_DET pin being connected? For DFP mode, per the Type-C spec, VBUS can only go high after CC pin detection and debounce.

Thanks

David

0 KennyMillar over 6 years ago in reply to David (ASIC) Liu

Intellectual 385 points

Hi David,
Thanks for the response.

VBUS_DET is pulled up via 900K to VBUS (+5V)

0 David (ASIC) Liu over 6 years ago in reply to KennyMillar

TI__Guru**** 177301 points

Kenny

Is the VBUS present when the initial connection is being made?

Thanks

David

0 KennyMillar over 6 years ago in reply to David (ASIC) Liu

Intellectual 385 points

Hi David

Thanks again for the reply. Sorry for the delay I was away on business yesterday.

Yeah the VBUS is present when the peripheral device connects.

Since the device is in DFP mode, the VBUS has to be present when the peripheral connection is made, because my board is providing power to the peripheral.
The board is based on the reference design in Figure 11 of the datasheet.

0 David (ASIC) Liu over 6 years ago in reply to KennyMillar

TI__Guru**** 177301 points

Kenny

Per the Type-C spec

1. The Source uses a FET to enable/disable power delivery across VBUS and initially the Source has VBUS disabled.

2. The Source supplies pull-up resistors (Rp) on CC1 and CC2 and monitors both to detect a Sink. The presence of an Rd pull-down resistor on either pin indicates that a Sink is being attached. The value of Rp indicates the initial USB Type-C Current level supported by the host.

3. The Source uses the CC pin pull-down characteristic to detect and establish the correct routing for the USB SuperSpeed data path and determine which CC pin is intended for supplying VCONN.

4. Once a Sink is detected, the Source enables VBUS and VCONN.

5. The Source can dynamically adjust the value of Rp to indicate a change in available USB Type-C Current to a Sink.

You can do a quick experiment by removing the 900k resistor on VBUS_DET and see if the voltage of DIR toggles base on the CC1/2 connection.

Thanks

David

0 KennyMillar over 6 years ago in reply to David (ASIC) Liu

Intellectual 385 points

Thanks for that detailed reply!

I will try the test you mentioned and see what happens.

-Kenny

0 KennyMillar over 6 years ago in reply to David (ASIC) Liu

Intellectual 385 points

Hi David

So I removed the pull up from VBUS to VBUS_DET and now the port works fully in both orientations.

Thanks again.

0 David (ASIC) Liu over 6 years ago in reply to KennyMillar

TI__Guru**** 177301 points

Kenny

Thanks for the update. Removing the 900k is a debugging step. To be fully Type-C compliant, you need to make design modification such that VBUS will be off when connection has been made, and then gets enabled after connection has been made.

Thanks

David

Interface

Interface forum

HD3SS3220: Cable orientation detection query