AM335x the VDDA3P3V_USB0 be pulled up when only the USB0_VBUS is supplied

Daisuke Maeda

Guru 20205 points

Other Parts Discussed in Thread: TPS65910

The VDDA3P3V_USB0 pulls up to 0.836V when only the USB0_VBUS is supplied in the customer boards, also it can show in the EVM. In the EVM, the VDDA3P3V_USB0 (J31_2) pulls up to 0.627V when only the VUSB_VBUS0(J36_1) is supplied 5V by an external supply.

Is there the internal leakage current path from the USB0_VBUS to the VDDA3P3V_USB0?

Best regards,

Daisuke

over 11 years ago

0 Biser Gatchev-XID over 11 years ago

TI__Guru**** 393215 points

Hi Daisuke,

I will report this to the factory team.

0 peaves over 11 years ago in reply to Biser Gatchev-XID

TI__Guru 62995 points

We are aware of this leakage path. The USB PHY design team has evaluated the condition and determined it does not cause any issue for AM335x.

Regards,
Paul

0 Daisuke Maeda over 11 years ago in reply to peaves

Guru 20205 points

Hi Paul and Biser,

Thank you for your reply.

Our customer is concerned that the pulled up 3.3V rail affects other devices before it is powered up.

Is there a recommended workaround?

Best regards,

Daisuke

0 peaves over 11 years ago in reply to Daisuke Maeda

TI__Guru 62995 points

This has not been an issue for any other customers. I only recall having one customer ask about this and once we explained there was no risk to AM335x, they were okay.

The only workaround I can think of would be to add a separate 3.3 volt power source for the USB 3.3 volt analog supply. However, this power source must not have a reverse path. Many LDO have a body diode in the series-pass transistor that will forward bias when the output voltage rises a few hundred millivolts above the input voltage.

Regards,
Paul

0 Daisuke Maeda over 11 years ago in reply to peaves

Guru 20205 points

Hi Paul,

Thank you for your reply.

I will recommend that the customer uses a separate 3.3 volt power source.

Best regards,

Daisuke

0 Daisuke Maeda over 11 years ago in reply to peaves

Guru 20205 points

Hi Paul,

Can a method like the workarounds for Advisory 1.0.32 in the errata be not applied?

Best regards,

Daisuke

0 peaves over 11 years ago in reply to Daisuke Maeda

TI__Guru 62995 points

No, you cannot insert a resistor in series with the power supply pin. Any series resistor large enough to limit the back-bias current would also limit the power supply current to the internal USB PHY.

Regards,
Paul

0 Daisuke Maeda over 11 years ago in reply to peaves

Guru 20205 points

Hi Paul,

Thank you for your reply.

The VAUX33 of TPS65910 connects to VDDA3P3V_USB0 and VDDA3P3V_USB1 on the customer boards. This connection is the same as the application note: http://www.tij.co.jp/jp/lit/ug/swcu093c/swcu093c.pdf

Does this condition not cause any issue for VDDA3P3V_USB1?

The VAUX33 pulls up to 0.836V before VCC3 of TPS65910 is supplied. Can this condition be allowed?

Best regards,

Daisuke

0 Daisuke Maeda over 11 years ago in reply to peaves

Guru 20205 points

Hi,

Daisuke Maeda said:

The VAUX33 pulls up to 0.836V before VCC3 of TPS65910 is supplied. Can this condition be allowed?

For TPS65910, here is a new post: http://e2e.ti.com/support/power_management/pmu/f/43/t/331467.aspx

Best regards,

Daisuke

0 Daisuke Maeda over 11 years ago in reply to peaves

Guru 20205 points

Hi Paul,

For the workaround, I suggest to our customer that VBUS is not connected to USB0_VBUS until VAUX33 supplies 3.3V.

I have one more question. Does USB1_VBUS not have the same problem? Just in case, I want to confirm that.

Best regards,

Daisuke

0 peaves over 11 years ago in reply to Daisuke Maeda

TI__Guru 62995 points

Both ports implement the same PHY design, so I would expect the same behaviour on the other port.

Regards,
Paul

0 Daisuke Maeda over 11 years ago in reply to peaves

Guru 20205 points

Hi Paul,

Thank you for your reply.

The same workaround will be used on the other port.

Best regards,

Daisuke

Processors

Processors forum

AM335x the VDDA3P3V_USB0 be pulled up when only the USB0_VBUS is supplied