• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 3 answers
  • Subscribers 33 subscribers
  • Views 643 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TPS65983: Vbus current limit OCP working not correctly

steven hone
Prodigy 180 points

Part Number: TPS65983

Hi 

Our design is TBT 3 device use Intel Alpine ridge chip  ( 1 USB-C port only and 1 PDO power source 5V , 900mA use PP_5V0 path , Peak Current setting is 100% )  , configuration tool version 3.10 , firmware version 4.45. At USB-PD VBUS test the over current test should cut off at 1.8A according test specification. But Tps65983 cut off at 3.4A.  Attach the .pjt and .bin  file .tps65983_pjt.rar 

  • 0
    TI__Genius 10370 points

    Hi Steven,

    What are you building (PC, hub, charger, etc.)?

    For thunderbolt 3 ecosystems, please use 5V/3A. One must advertise 15W on Thunderbolt 3 applications per Intel specs. I believe our firmware sets the current limit to 3A (regardless of the set current limit because this is a Thunderbolt application) and hence why you see the current limit begin right above 3A.

    Also, we have a new GUI config tool update for TPS65983 (version 3.13). Please update to the latest version. I hope this helps and If this answers your question, PLEASE select  This resolved my issue 

  • 0 in reply to Aramis P. Alvarez
    Prodigy 180 points

    Hi Aramis

    Our design is RAID subsystem that support totally 9 x single port TBT device. Because it is wall power device. So we need supply power but don't support daisy chain. It is quite difficult to support all 9 port with 16.5w power each. So we discuss with Intel this special case. We low down specification to 5V/0.9A( for Apple TBT3 to TBT 2 cable support  ) . It work ok  in all usage case. But at certify lab. they need VBUS cut off before 1.8A. Any thing we can do to change the default firmware set at 3A?

    Thank you

  • 0 in reply to steven hone
    TI__Genius 10370 points

    Hi Steven,

    If you have a 9 Port Thunderbolt System that means that you have a total of 5 Alpine Ridge TBT controllers and 9 TPS65983. Is this correct?

    Also, which team did you speak with at Intel that stated this was ok to do?

    You mentioned that you are making a RAID system that does not daisy chain which implies that you will only have one TBT port that will connect you a host or a daisy chainable device. What are the other 8 ports functioning as?

    The TBT spec recommends that only a host can have source policy manager for 15W (5V @ 3A), but it requires that the first 4 ports support 5V @ 3A, and the other may support 5V @ 900mA minimum. However, the 5V @ 900mA will not be sufficient for most bus powered TBT devices (for example the TBT3 to TBT2 adapter, which supports 10W).

    Please check the 6.4.1 Self Powered Device section of the TBT3 spec which defines that all ports must support 15W @ 5V on VBUS to every port.

    You must follow the Intel Reference Design when it comes to the Thunderbolt ports on your system. We provide the firmware to all of the Thunderbolt reference designs and to make sure there is consistent interoperability. Which means that we cannot change the default firmware.

    If this answers your question, PLEASE select  This resolved my issue 

  • 0 in reply to Aramis P. Alvarez
    Prodigy 180 points

    Hi Aramis

    Design used 9 Alpine Ridge SP TBT controllers. All 9 port is single port TBT downstream device only support to host connection and all 9 ports are symmetrical. So no daisy chain support. It also means it can  connect to 9 host computer( PC or MAC ) simultaneously.     

    Our Intel PAE window is Zohar Shauly.

    We don't need support the bus power device( single port , no daisy chain ). And no intention to support host charge. 5V/900mA is supply  TBT3 to TBT2 cable for TBT2 host connection.

    Yes, We did violate the self power device spec. That is why we need discuss with Intel for this special case.

    Thank you 

  • 0 in reply to steven hone
    TI__Genius 10370 points

    Hi Steven,

    Thank you for your answering those questions. I'm going to go ahead and close this thread, but if any other issues come around feel free to post a new thread so that we can try to resolve it :-)

    If you can, could you PLEASE select  This resolved my issue  

    Thanks in advance!