• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 32 subscribers
  • Views 271 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TUSB8042A: Questions on TUSB8042A/43A

Kunal Bhadane
Prodigy 20 points
Part Number: TUSB8042A
Other Parts Discussed in Thread: TUSB8043A

Hello, 

We have an application in which we use TUSB8042A/43A. We have three USB 3.0 type-A downstream ports (ports 1 to 3) which are removable and one non-removable USB 2.0 downstream port (port 4) connected to our micro-controller. We configure the hub chip using SMBus. In hardware, AUTOENz is disabled by pulling the pin high, FULLAUTOEN and GANGED pins are also disabled by pulling them down. In software, we configure the hub chip to set removable ports as mentioned above, enable all the ports and set manufacturing and product strings. We have used pullups for all three removable downstream ports for BAT_EN and power to all three ports is controlled by the 0micro-controller instead of the hub-chip.

 

Question 1: We are using 0 ohm resistors to pull up and pull down the pins in above mentioned configuration. Will that be a problem? We aren't using them as status pins.

Question 2: If we configure this micro-controller port as a non-removable port and if we reset the micro-controller without resetting the hub chip, will it have any impact on the hub chip?

Question 3: Since the hub chip is configurable only once after hub reset assertion, will it have any issue after resetting the micro-controller, if micro-controller sends above mentioned commands over SMBus?

Questions 4: Recently, one of our TUSB8043A hub chip had an issue. The hub chip was functioning properly but for some reason it started pulling extra 200mA current approximately. After investigation we found out the impedance measured on pin 52 which is one of the 3.3V pin to GND was very low approximately 16 ohms. We also measured impedance of all the pins and they all measured as expected. Could you please explain how pin 52 is used internally and aren't all 3.3V pins tied together? Also, it would be helpful to know what are the possible reasons that could short pin 52? Even though this pin is shorted, the hub is functioning properly without affecting the communication.

Question 5: We also had an issue with one of the TUB8042A. One of the USB 3.0 ports (port 1) was not detecting any 3.0 device connected to it. So, we measured impedance of all the pins and we found out that pin 4 which is  USB_SSTXM_DN1 measured a very low impedance. We are unclear if this chip was damaged already or our testing led to this failure but we would be interested in knowing the possible causes of this failure as well.  

 

Since our design is confidential, I won't be able to share it with you on public forum. If you would like to take a look at the schematic, I can share it with you over email.

Thank you.

  • 0
    TI__Guru 64326 points

    Hi,

    1. For configuration pins please use 4.7k. 

    2. It should not, USB Host will see a USB 2 endpoint reset. The configuration for non-removable is to edit the descriptors reported to the USB Host.

    3. After the SMBUS host programs the CFG_ACTIVE bit correctly. all SMUS function will stop and USB Hub functions will start. The hub does not continue to access the SMBUS interface after this bit has been programmed.

    4. What is the current consumed here? There is no specific function to pin 52 versus all other power pins. I would ensure that thermal pad has ample solder connecting the part to GND. This can cause elevated current consumption to be seen. Also I would ensure that all pins have good contact with the board. If you swap this device with a new device do you see the same short?

    5. A damaged pin would prevent the USB 3 termination form being present and is most likely the cause of the issue. One possible cause could be improper reset of the part or incorrect routing of the pin. I would double check the physical connection between USB 3 connector and USB hubs pin pad.  

  • 0 in reply to Malik Barton57
    Prodigy 20 points

    Hi Malik, 

    Thank you for the prompt response. I have few additional questions based on your reply. 

    1. For configuration pins please use 4.7k. 

    - Is there any particular reason why you are recommending to use 4.7K resistors? We recently referred to the EVAL board schematic and it has 1K resistors. I am just trying to understand if it would have resulted into the failures we have seen. 

    2. It should not, USB Host will see a USB 2 endpoint reset. The configuration for non-removable is to edit the descriptors reported to the USB Host.

    - Great. That's what we expected. 

    3. After the SMBUS host programs the CFG_ACTIVE bit correctly. all SMUS function will stop and USB Hub functions will start. The hub does not continue to access the SMBUS interface after this bit has been programmed.

    - Great. That's what we expected. 

    4. What is the current consumed here? There is no specific function to pin 52 versus all other power pins. I would ensure that thermal pad has ample solder connecting the part to GND. This can cause elevated current consumption to be seen. Also I would ensure that all pins have good contact with the board. If you swap this device with a new device do you see the same short?

    - Our quiescent current on the board is 200mA on 12V line and it increased to 300mA on this particular board. We have other boards working fine and this particular board also worked fine and suddenly it started pulling more current. We ensured that all the pins have good contacts as well as the thermal pad has ample solder connecting the part to GND. We observed that all 3.3V lines are isolated from each other since this short affects only one pin. If you could find out where that pin is used that would be really helpful and would allow us to find what caused it to fail. After replacing the hub chip we don't see any problem. So, this issue is related to one chip only. 

    5. A damaged pin would prevent the USB 3 termination form being present and is most likely the cause of the issue. One possible cause could be improper reset of the part or incorrect routing of the pin. I would double check the physical connection between USB 3 connector and USB hubs pin pad.  

    - As I mentioned above, we have few other boards working as expected. So, I don't think the problem is related to design and we ensured that the connections are solid. Could you please elaborate more on the improper reset of the part? Is it improper resetting of the hub chip you're talking about? If it is the improper reset of the hub chip, what's improper reset? 

    Thank you.  

  • 0 in reply to Kunal Bhadane
    TI__Guru 64326 points

    1. I do not think this would contribute to your issues. 1K is fine as well. I recommend 4.7K for weaker connection to pins that have some output indicator. 

    4. It is possible this part was damaged internally during assembly. I can check to see if there are some additional details here. I will follow-up

    5. I am referring to the reset of the hub. The proper reset can be seen under section 7.6 Timing Requirements in the datasheet. This ensures that the hub is brought up in the correct state after power-up. I suspect reset is okay form your description however. Do you have ESD protection on the USB 3 lines? 

  • 0 in reply to Malik Barton57
    Prodigy 20 points

    Hey Malik, 

    1. We will follow the EVAL board schematic and will use 1K resistors. 

    4. Please let me know when you have information on pin 52. 

    5. We checked the timing requirements and verified that hub reset is behaving properly. We don't have ESD protection on USB 3.0 lines since we didn't see it on Eval schematic and we have been following the same design. 

  • 0 in reply to Kunal Bhadane
    TI__Guru 64326 points

    In system design I recommend placing some external ESD to handle ESD event above what TUSB8042A can handle. SSTX/SSRX pins can be damaged in this kind of scenario.

    Checked the team here, Pin 52 does not have a special function here. I suspect that the part was damaged somehow during assembly.