TUSB4020BI: Pulse width

Yuta Nishie

Part Number: TUSB4020BI

Hi,

In the following post, I told you that the voltage of USB_DP_DN1 rises due to the influence of BATEN.

https://e2e.ti.com/support/interface-group/interface/f/interface-forum/1043553/tusb4020bi-usb_dp_dn1-output-voltage/3859695

There were cases where this pulse width became wider.
Why is the pulse width wider?

Best Regards

Nishie

over 3 years ago

0 Malik Barton57 over 3 years ago

TI__Guru 64326 points

Hi Nishie,

This pulse with BATEN pulled high could be the battery charging negotiation. What is the state of BATEN for this port. Do you have a device connected?

0 Yuta Nishie over 3 years ago in reply to Malik Barton57

Genius 5015 points

Hi Malik,

Thank you for your reply.

I'm sorry I didn't fully understand the intent of your question.

Is the following answer correct?

The BATEN1 pin is connected with a 3.3V pull-up.

Best Regards,

Nishie

0 Yuta Nishie over 3 years ago in reply to Yuta Nishie

Genius 5015 points

Hi Malik,

I have an additional question.

Is there min value and max value for the output timing of DP / DM pulse wave?

Best Regards,

Nishie

0 JMMN over 3 years ago in reply to Yuta Nishie

TI__Guru 60870 points

Hi Nishie,

Can you respond in the other thread about the timing of the DP/DM pulse with respect to upstream port connection and if the downstream ports are connected to devices or not?

Regards,

JMMN

0 Yuta Nishie over 3 years ago in reply to JMMN

Genius 5015 points

Hi JMMN,

I sent the schematic to Brian.

Is it possible to share schematic with Bryant?

If not, is it possible to send by private email?

Best Regards,

Nishie

0 JMMN over 3 years ago in reply to Yuta Nishie

TI__Guru 60870 points

Hi Nishie,

Yes, Brian has shared the schematic. For your application, there should not be a pullup on PWRCTL_BATEN for the port that has a permanently connected device. When the hub has battery charging enabled and the USB_VBUS is low, DP and DM are shorted by the hub downstream port (DCP mode). When the permanently attached device powers on it will have a DP pullup, which will be reflected on the DM line in DCP mode. This is causing the pulse that is seen.

Regards,

JMMN

0 Yuta Nishie over 3 years ago in reply to JMMN

Genius 5015 points

Hi JMMN,

Thank you for answering.

The waveform measurement was performed by opening the USB_DP / DM_DN1 line of the circuit diagram.
A pulse similar to the previously sent waveform is output on this line.
Is it correct to output a pulse even if the Downstream port is opened?
In addition, the pulse output timing and pulse width are random for each device.
Do you know the cause of the random output?

Best Regards,

Nishie

0 JMMN over 3 years ago in reply to Yuta Nishie

TI__Guru 60870 points

I will confirm the behavior in the lab.

Regards,

JMMN

0 Yuta Nishie over 3 years ago in reply to JMMN

Genius 5015 points

Hi JMMN,

Thank you for your support.

Please let me know if there is anything to check with the customer for operation in the lab.

Best Regards,

Nishie

0 JMMN over 3 years ago in reply to Yuta Nishie

TI__Guru 60870 points

I will have an update by end of week.

Regards,

JMMN

+1 JMMN over 3 years ago in reply to JMMN

TI__Guru 60870 points

Hi Nishie,

I was able to confirm this behavior is related to the battery charging feature enabling on in the downstream ports of the TUSB4020B. The pulse on DP/DM is related to the time between TUSB4020B powering on/exiting reset and when it detects USB_VBUS on. Even if VBUS is on at power on, it will not be detected immediately by the hub.

My recommendation is to disable battery charging by removing the pullup on PWRCTLx to prevent the pulse in your system on the internal ports.

Regards,

JMMN

0 Yuta Nishie over 3 years ago in reply to JMMN

Genius 5015 points

Hi JMMN-san,

Thank you for conducting the evaluation.
Regarding the cause of the long pulse, I am not well understood the relationship between VBUS detection and DP / DM pulse.
Is it possible to present the evaluation results and timing charts?

Best Regards,

Nishie

+1 JMMN over 3 years ago in reply to Yuta Nishie

TI__Guru 60870 points

Hi Nishie,

For the TUSB4020B, when battery charging is enabled on a downstream port and it is in DCP mode (no upstream connection, USB_VBUS off) there will be the voltages on DP/DM lines. In the customer system, when the TUSB4020B powers on, battery charging is enabled and by default USB_VBUS is considered off until it has been sampled and debounced. This time for sampling and debounce of USB_VBUS is the time of the DP/DM pulse on the downstream port. For this short time, the downstream port is in DCP mode. The timing will vary depending on the customer system. To remove the pulse, they should disable battery charging on the port entirely.

Regards,

JMMN

0 Yuta Nishie over 3 years ago in reply to JMMN

Genius 5015 points

Hi JMMN,

Thank you for your reply.

In the customer system, DM / DP pulse will be 5ms or 100us like the first posted waveform.

Is there such a change depending on the customer system?

Also, is there any effect due to variations in device specifications?

Best Regards,

Nishie

0 JMMN over 3 years ago in reply to Yuta Nishie

TI__Guru 60870 points

Hi Nishie,

I would expect the pulse to vary with USB_VBUS timing in the system.

Regards,

JMMN

0 Yuta Nishie over 3 years ago in reply to JMMN

Genius 5015 points

Hi JMMN-san,

I understood that it depends on the system.

Is it possible to disclose what kind of waveform was obtained in yours measurement environment?

Best Regards,

Nishie

0 JMMN over 3 years ago in reply to Yuta Nishie

TI__Guru 60870 points

I will take waveform in the lab tomorrow to share.

0 Yuta Nishie over 3 years ago in reply to JMMN

Genius 5015 points

JMMN-san,

Thank you.

Nishie

0 JMMN over 3 years ago in reply to Yuta Nishie

TI__Guru 60870 points

Please see the attached waveform.

+1 JMMN over 3 years ago in reply to JMMN

TI__Guru 60870 points

Typical case above, below is the similar case but zoomed out to show the downstream VBUS behavior (if controlled by PWRCTL).

Regards,

JMMN

0 Yuta Nishie over 3 years ago in reply to JMMN

Genius 5015 points

Hi JMMN,

I am very grateful for your immense support.

I was able to confirm waveforms with different pulse widths.

I also confirmed that the output voltage of DP / DM is about 2V.

Best Regards,

Nishie

Interface

Interface forum

TUSB4020BI: Pulse width