TUSB321: TUSB321 not detecting cable inversion

Alexander Zaprudsky

Prodigy 120 points

Part Number: TUSB321
Other Parts Discussed in Thread: TUSB542,

Hi.

I have exactly the same issue that was discussed in "TUSB321 not detecting cable inversion" thread, i.e.:

I'm using the TUSB321 as UFP configured according the spec.

I'm using Type-A to Type-C certified USB3.0 cable.

Once the Type-C plug is connected a little bit slowly the TUSB321 fails to detect cable inversion.

Currently the "TUSB321 not detecting cable inversion" thread is locked, so I'll be more than happy to know if there is a solution for this issue?

Thanks.

Alexander.

over 7 years ago

0 Michael.Walker over 7 years ago

TI__Mastermind 19720 points

Can you please post your schematic?

0 Alexander Zaprudsky over 7 years ago in reply to Michael.Walker

Prodigy 120 points

Yes, see below:

0 Michael.Walker over 7 years ago in reply to Alexander Zaprudsky

TI__Mastermind 19720 points

Are you using the same test procedure as the post that you linked to?

0 Alexander Zaprudsky over 7 years ago in reply to Michael.Walker

Prodigy 120 points

Yes, I'm using the same test procedure.

And my issue is exactly the same that is described in "CASE A" paragraph "e" in the linked post.

0 Michael.Walker over 7 years ago in reply to Alexander Zaprudsky

TI__Mastermind 19720 points

Are there any CC log files? can you give me scope shots of the VBUS, and the CC lines?

0 Alexander Zaprudsky over 7 years ago in reply to Michael.Walker

Prodigy 120 points

The VBUS, CC1 and DIR lines were captured in 2 cases:

Correct detection:

Wrong detection:

As can be seen on the 2nd capture the 0.4v on the CC1 line arrives with delay (>177ms) due to "slow" plug-in of the Type-C connector.

The DIR line stays HIGH = Incorrect state for this case, i.e. according the TUSB321 datasheet the DIR line should be pulled LOW when voltage level within the proper threshold is detected on CC1 line:

0 Michael.Walker over 7 years ago in reply to Alexander Zaprudsky

TI__Mastermind 19720 points

Can you also capture CC2 in the scope captures so that I have a complete picture of what is happening?

0 Alexander Zaprudsky over 7 years ago in reply to Michael.Walker

Prodigy 120 points

Hello and sorry for delay.

According to your request to capture the CC2 signal, see below:

Correct detection (CC1 marked in Red, CC2 marked in Green):

Wrong detection (CC1 marked in Red, CC2 marked in Green):

Regards,

Alexander.

0 Michael.Walker over 7 years ago in reply to Alexander Zaprudsky

TI__Mastermind 19720 points

How are you powering the TUSB321? Are you powering it by VBUS?

0 Alexander Zaprudsky over 7 years ago in reply to Michael.Walker

Prodigy 120 points

The TUSB321 is powered by VBUS through load switch (TCK112G).

In addition, the TUSB321 was powered directly by VBUS, but no improvement / change were inspected, i.e. the issue still happens when the Type-C plug inserted slowly.

0 Michael.Walker over 7 years ago in reply to Alexander Zaprudsky

TI__Mastermind 19720 points

This problem could be that the TUSB321 VDD is not powered all the way to 0V to allow the device to reset properly. To test if this is the issue you can try waiting a long time(30s) between plugging into the port both slowly and quickly and see if it detects the inversion.

0 Michael.Walker over 7 years ago in reply to Alexander Zaprudsky

TI__Mastermind 19720 points

Also can you tell me the state of the ID and the VCONN_FAULT pin when the flip is working and not working?

0 Alexander Zaprudsky over 7 years ago in reply to Michael.Walker

Prodigy 120 points

I took this scenario into a count, and did the experiments after waiting even more then 30s, no difference, the issue still exist.

0 Michael.Walker over 7 years ago in reply to Alexander Zaprudsky

TI__Mastermind 19720 points

Any details on the ID pin and the VCONN_FAULT pin?

0 Alexander Zaprudsky over 7 years ago in reply to Michael.Walker

Prodigy 120 points

In our design the VCON_FAULT pin isn't connected (floating), anyway I've tested both ID and VCON_FAULT pins when the flip is working and not working.

In both cases the ID pin is 3.3V and the VCON_FAULT pin is ~0.92V.

0 Michael.Walker over 7 years ago in reply to Alexander Zaprudsky

TI__Mastermind 19720 points

I think your issue might be that you are some how back powering the TUSB321 through the ID pin. Can you replace the pull up resistor on ID to 200K ohms?

0 Michael.Walker over 7 years ago in reply to Alexander Zaprudsky

TI__Mastermind 19720 points

Is there any updates on this issue?

0 Alexander Zaprudsky over 7 years ago in reply to Michael.Walker

Prodigy 120 points

Yes, the Pull Up resistor on ID pin was replaced with 220k (No 200k in stock), and re-tested again.

The TUSB321 still fails to detect the correct side in case the plug inserted slowly, and the ID pin voltage is 3.27v in both cases.

0 Michael.Walker over 7 years ago in reply to Alexander Zaprudsky

TI__Mastermind 19720 points

Can you please NC the ID pin so that there is noway it can be back powered. Also how are you connecting the DIR pin? what is the pullup value? what else is it connected to?

0 Michael.Walker over 7 years ago in reply to Michael.Walker

TI__Mastermind 19720 points

Are there any updates?

0 Alexander Zaprudsky over 7 years ago in reply to Michael.Walker

Prodigy 120 points

The ID pin was disconnected, but the issue still exist.
The DIR pin is connected to TUSB542 pin SEL, with 30k pull up resistor. Nothing else is connected to DIR pin.