TUSB319-Q1: Bulk Capacitor Placement – Before or After Load Switch?

Hi Nir,

We recommend placing the 150uF capacitor after the VBUS switch, like how it is on TUSB319EVM. We also have it like this on our new CC controllers. Here's the TUSB320, just as an example:

I'll look at having this updated on this datasheet.

As for your concern with placing it after the VBUS switch, this bulk capacitor for any DFP/DRP-DFP connection is part of the USB-C spec, so it is recommended to be implemented, and shouldn't cause any issues, as far as I am aware. We haven't seen any issues with it previously.

Please let me know if you have any other questions.

Thanks,

Ryan

Hi Ryan,

Regarding the inrush concern: I understand that the 150 µF capacitor is required by the USB-C specification and is expected for DFP or DRP-to-DFP roles. My concern is specifically about scenarios where a USB Type-A to Type-C cable is used—since the Type-A side provides VBUS immediately, this could potentially result in a large inrush current into the capacitor when it’s placed directly at the connector (i.e., after the load switch).

Do TI’s designs account for this situation?

Personally, I believe that placing the capacitor at the input of the load switch could help isolate it during hot-plug events, such as in the example above. Once the switch closes safely, the capacitor can still fulfill its role—even if it's located slightly farther from the connector's VBUS.

Appreciate your help!

Best regards,
Nir

Hi Nir,

In the situation where the TUSB319 is acting as the DFP, which is most cases since I believe its a DFP only device, then it should be supplying VBUS through the type-C port, so VBUS coming from the A to C plug should not be a concern. Are you worried about the scenarios where VBUS comes into the device from the type-C port, or just that with a C to A cable, VBUS will be immediately enabled?

Thanks,

Ryan

Hi Ryan,

Thanks for your reply!

To clarify — yes, I'm referring to a scenario where the system is a DFP, and is expected to supply VBUS. However, I’m concerned about what happens if a user mistakenly plugs in a USB Type-A to Type-C cable into this DFP port.

In this case, the Type-A side is always hot — it sources 5V VBUS immediately upon connection. If our design places the 150 µF bulk capacitor after the load switch, directly at the connector, then the moment the Type-A cable is inserted, VBUS from the cable will be connected directly to the bulk cap, which could result in a large inrush current.

This is different from a normal USB-C to USB-C negotiation scenario where power delivery is coordinated and the switch is enabled in a controlled manner.

So my concern is: does placing the capacitor after the switch pose a risk in this hot-plug (Type-A to Type-C) case? And would it be safer to place the capacitor before the switch, so it’s not immediately connected during a mistaken hot plug event?

Thanks again,
Nir

Hi Nir,

The way the VBUS switch performs when a type-C to type-C connection is made should be identical when a type-C to type-A connection is made, so I don't believe placing the capacitor after the switch should be an issue. The voltage/current supplied from the VBUS switch should be able to be supported by the capacitor in the case of a type-A to type-C.

I believe you can place it after the switch without any worry of capacitor being damaged, as long as the capacitor is rated to be able to handle the expected voltage.

Thanks,

Ryan

Hi Ryan,

I appreciate your continued responses, but I have to be honest — I don’t think the core issue is being understood.

This is not a question about capacitor ratings or whether the VBUS switch can supply current to the capacitor. The concern is about uncontrolled inrush current from an external VBUS source, specifically in the case where someone mistakenly connects a USB Type-A to Type-C cable into the port.

In that scenario, the Type-A host immediately sources 5V to the Type-C connector. If the 150 µF bulk capacitor is placed after the load switch (i.e., directly at the connector), it will be exposed to this VBUS immediately — before our system has had a chance to enable the switch. That leads to a large inrush current coming from the external source, not through the controlled VBUS path. This is a potential reliability issue and could violate USB inrush limits or cause backfeeding problems.

If TI has a reason why this is not a concern — or why the EVM design ignores this case — I'd appreciate a clear explanation. Otherwise, could someone else at TI who is more familiar with this particular issue take a look?

Thanks,
Nir

Hi Nir,

Ah, okay, I understand now I think. Sorry for the confusion on my side. You specifically are worried about an EXTERNAL VBUS coming into the type-C from a type-A that is supplying VBUS, I understand now.

This isn't a concern we encounter often when discussing setup of our CC controllers, but I can understand why you would be worried about this, especially if this is something your customers will be able to do.

Personally, I haven't ever seen any issues with having the bulk cap on this side of the VBUS switch, and I believe it is required to be on the output side of the VBUS switch, but I will confirm this. We have used our EVM's in the lab with type-C to type-A cables where the type-C supplies VBUS, but I will try testing this in the lab to see if there is any adverse effect.

I can reach out to some of our other team members that have a bit more experience with this device, and I will let you know if they believe there is any cause for concern in this specific scenario.

Thanks,

Ryan

Hi Ryan,

Thanks for the thoughtful reply — and I really appreciate you taking a fresh look at the issue.

You now understand the scenario I’m concerned about — an external VBUS (from a Type-A host) being applied to the Type-C receptacle before our local load switch is enabled, exposing the 150 µF bulk capacitor to an uncontrolled inrush event.

This might seem like an edge case, but it's not unrealistic in consumer-facing products, where end users often plug in whatever cable they have lying around — including Type-A to Type-C.

As a reference, this same concern was addressed in the USB 2.0 spec, section 7.2.4.1, which limits VBUS input capacitance on the upstream port to 10 µF. The reason? Because Type-A ports are always hot, and an excessive capacitor on the VBUS pin would draw a large inrush current when plugged in — exactly the situation we’re talking about here. The only difference is that back then, USB-A was always a DFP, so the spec explicitly protected against this case. With USB-C’s role flexibility, the burden shifts to the system designer.

So while I understand that the USB-C spec calls for a 150 µF bulk cap on the DFP side, it doesn’t explicitly account for this backpowering scenario — which is now a real risk in mixed-cable environments. That’s why I’m suggesting that placing the bulk capacitor before the load switch (rather than at the connector) might be safer in practice. It allows us to meet the spec requirement after we’ve enabled VBUS in a controlled way, while protecting against inrush when a Type-A host is mistakenly connected.

Thanks again for offering to test this and to loop in others with more experience. I look forward to hearing what you find.

Best regards,
Nir

Hi Ryan,

Thanks a lot for taking the time to investigate this and for confirming the capacitor placement — that’s very helpful.

This topology is definitely preferred in our case. From a functional perspective, the physical placement of the 150 µF bulk capacitor doesn’t need to be right at the connector — it’s not as critical as a high-frequency decoupling cap in that regard. So placing it slightly further upstream doesn’t impact normal operation or compliance with the USB-C spec, but it greatly improves protection in scenarios where an external VBUS source might be connected.

Thanks again for your help — really appreciate the support!

Best regards,
Nir