TPD4E001: VBUS line overshoot caused by adding a capacitor to TPD4E001

Hi Sugimoto-san, 

Do you require protection for the VBUS line? If so, I would recommend leaving the TPD4E001 VCC pin floating or unconnected and place a single channel TVS on the VBUS line for protection. 

If you can provide the nominal operating voltage, then I can provide a solution for you. 

Best,

McKenzie

Hi McKenzie,

Yes, protection for the VBUS line is necessary. Currently, the VBUS line of the USB Type-C connector (pins A4, A9, B4, B9) is protected by connecting it to the VCC pin of the TPD4E001. The VBUS voltage is 5V, and the current during normal operation is less than 1A. As I mentioned, since VBUS is connected to the input of the TPS2116, it needs to be protected from all transient voltages so that it does not exceed 6V. I would appreciate any suggestions for a solution. Also, if the VCC pin is left open, will the external capacitor still be necessary for the protection of the other I/O pins?

Best regards,

Yasuaki Sugimoto

Hi Sugimoto-san, 

I would recommend ESD441DPLR for this. It has a working voltage of 5.5V and can clamp at ~6.3V for a 1A strike. 

The external capacitor is still recommended. See the snippet below. 

Best,

McKenzie

Hi McKenzie,

I understand the need for a capacitor.
Regarding the ESD441DPLR you suggested, you mentioned clamping at ~6.3V, but since the absolute maximum voltage rating of the TPS2116 is 6V, wouldn't this fail to protect the device?

Best regards,

Yasuaki Sugimoto

Hi Sugimoto-san, 

It wouldn't necessarily fail to protect the device. The clamping voltage for our diodes is a transient voltage while the TPS2116 has an absolute maximum with a DC voltage, so there is some wiggle room. 

I tried to see if a 3.6V diode could work for this, but most of them have a minimum breakdown below 5V which means the diode will be leaking excess current during normal operation. 

Best,

McKenzie

Hi MaKenzie,

While testing prototypes with the current design, several TPS2116 chips failed. Investigation revealed that a 7-8V, 0.5μs strike occurred when the USB cable was connected from the AC adapter. We suspect that repeated plugging and unplugging of the USB cable accumulated damage, leading to failure.

If we use the ESD441DPLR you suggested to protect the VBUS line, and this peak is reduced to approximately 6.3V, would that put it within the "some wiggle room" range and prevent failure? As this is a charging port for mobile devices, considering the use case, this voltage will be applied almost daily.

As you mentioned, we also need to prevent leakage current during normal operation, and we understand that we cannot use a TVS value that is too small.

Best regards,

Yasuaki Sugimoto

Hi Sugimoto-san, 

Do you know the current during the 0.5us strike? This diode will clamp at lower voltages if the current is lower. 

Best,

McKenzie

Hi McKenzie,

It varies depending on the USB cable, but based on my testing with the 15 cables I currently own, the current is in the range of 0.4 to 1.5A. We cannot control the cables used by users, so it's possible that some cables might reach up to around 2.0A.

Best regards,

Yasuaki Sugimoto

Hi McKenzie,

I searched for a Spice model for the ESD441 but couldn't find one. I contacted your support representative in Japan, and they said it could be created but would take a considerable amount of time. Could you please suggest any devices with similar specifications that have existing Spice models? I would like to evaluate them through simulation first.

Best regards,

Yasuaki Sugimoto

Hi McKenzie,

Thank you for your prompt reply. This will be helpful, and I will immediately use this model to verify it.

Best regards,

Yasuaki Sugimoto