TUSB4041I: TI Hub TUSB4041PAPR compatibility issue

Hello Gareth,

We do not see this issue with the Microsoft mouse.

Please provide information on the entire test setup:

• Host - USB 2.0 only port or USB 3.0?  Vendor?  Operating System?  Driver?
• Hub - TUSB4041
• Device - Microsoft mouse  - any other devices connected to the hub?

Regards,

JMMN

Hi JMMN

• Host-USB 2.0 only port or USB 3.0? Vendor? Operating System? Driver?

There are 2.0 and 3.0 WIN10 64bit. Driver built-in

• Hub-TUSB4041

• Device-Microsoft mouse-any other devices connected to the hub?

At present, both the USB keyboard and mouse are connected, and it will be confirmed. Is there any problem with only connecting the USB mouse?

Can they install usbview.exe or USB Device Tree Viewer.exe and share the report before (good case) and after reboot (bad case)?

I am not able to duplicate behavior in the lab.

Regards,

JMMN

Hi JMMN

Please refer to the attachment. thank you. The customer uses the same port as the mouse and restarts the PC. Have seen the symptoms of failure

Can you share a schematic?  I cannot duplicate this on an EVM.

Regards,

JMMN

Hi JMMN

After disable auto charging mode, the mouse can work normally.

Do you have any idea about when to enable auto charging mode? Thank you.

Hi Gareth,

Is battery charging enabled on the downstream ports of the hub?  Auto charging mode should only be enabled when there is no upstream port connection (USB_VBUS = low), but it is possible that the mouse is getting confused by one of the charging modes and not reconnecting properly on reboot.  Does the customer need to use the auto charging mode or can they just use the normal battery charging modes instead?

Regards,

JMMN

Hi JMMN

Customers can do without charging. Also, do you have the performance report of this hub?

At present, it is actually worse than the competing Genesys. Thank you

Passmark USB test for USB2.0 port,

TI reading is 295Mbs, writing is 143Mbs.

Genesys reading 321Mbs/Write 262Mbs)

Hi Gareth,

Performance depends on the setup.  Does the customer have USB 2.0 L1 power state enabled?  That would impact the throughput, otherwise throughput should be determined by the host and device unless there are retries due to signal quality.

Regards,

JMMN

Hi JMMN

The customer's eye diagram test PASS, is to run the following Passmark USB test, the efficiency is poor, do you have any comments?

PassMark Software - USB 2.0 Loopback Plugs and USB Test Software

Other than confirming that USB 2.0 LPM power states are not enabled and checking for signal quality issues that could cause retries I can't think of anything that would impact the USB 2.0 throughput.

Regards,

JMMN

Hi JMMN

USB 2.0 L1 power state enabled? Where is this function?

Thank you

USB L1 would be enabled by the USB host controller driver.  It will not be enabled if the test devices do not support it.

Regards,

JMMN