• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Subscribers 34 subscribers
  • Views 783 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.

TPS65982: Seeing three 0V@0A PDOs coming from TPS65987

Jon Key
Intellectual 920 points
Part Number: TPS65982
Other Parts Discussed in Thread: TPS65987, TPS65981EVM, TPS65987D, TPS65981

Hi Eric,

My customer is using a TPS65987 as a source that should be advertising 3 PDOs,, 5V@3A, 10.4V@3A, and 20V@5A, but their sink (TPS65982) reports 0V@0A for all 3 PDOs. Is there something in the configuration files sent via email that could be causing this?

Thank you,

Jon

  • 0
    TI__Mastermind 24075 points
    Hi Jon,
    I am evaluating the sent config files on our EVM and will get back to you as soon as possible.
  • 0 in reply to Kedar Manishankar
    TI__Intellectual 920 points
    Hi Kedar,

    Do you have any updates you can share?

    Thanks,
    Jon
  • 0 in reply to Jon Key
    TI__Mastermind 24075 points

    Hi Jon,

        In the TPS65982, if it is to be configured as a sink you do not need try.Src. In port information(bits 0..2) of register 0x28, have the setting to toggle Rp/Rd on CC (no trySrc). You want to double check the resistor setting on ADCIN1 in your schematic, and verify the boot setting of BUSPOWERZ with the d.s. If you are able to recreate this issue consistently, please send PD message logs when the error occurs. Is this tested on the customer's specific schematic? If so please recreate setup on EVM to help us isolate if the issue is with the config, PD controller or with schematic.

  • 0 in reply to Kedar Manishankar
    TI__Intellectual 920 points
    Hi Kedar,

    The customer would like the TPS65982 to act as a source in the event that a USB C accessory is plugged into the main board, or as a sink if it is connected to their source board. Since this should be a DRP, should they still remove try.Src?

    I will follow up with the schematics via email. Can you clarify what is meant by the "PD message logs?"

    Thank you,
    Jon
  • 0 in reply to Jon Key
    TI__Mastermind 24075 points
    Schematic Review Completed offline. Awaiting results from EVM testing
  • 0 in reply to Kedar Manishankar
    TI__Intellectual 920 points

    Hi Kedar,

    Following the rework you recommended, the following tests were performed and the corresponding results were observed:

    1. TPS65987D Customer Board was connected to TPS65981EVM (default EVM FW configured to be a sink) Result: PD contract is achieved with correct 20V contract chosen. PD message log attached as TPS65987D_to_TPS65981EVM.tdc (same test case was performed in customer_to_customer.tdc index 1 - 46)

    2. TPS65987D Customer Board was connected to TPS65982 Customer Board w/ Customer's FW. Result: No PD contract is achieved. TPS65987D Board sends its Source_Cap the Sink CRCs and requests the correct PDO, the source accepts, the sink CRCs, but the source board toggles the incorrect to control the variable DCDC converter and never sends a PS_RDY signal. PD Message log shown in customer_to_customer.tdc from index 47 to index 82. (NOTE: the earlier indexes show the first test case)

    In the first test case the TPS65987D sets the proper GPIO (GPIO17) to set the variable DCDC Converter to output 20V, but when the sink changes in the second test case the TPS65987D sets GPIO14 high generating 5V. Since both PD controllers are expecting 20V but only get 5V, I assume both controllers time out and try to negotiate the contract again.

    Do you have any idea what could be causing this, or any other experiments we could try to get to the bottom of this?

    Thank you,

    Jon

    PS e2e wont let me attach the .tdc so I will follow up with an email.

  • 0 in reply to Jon Key
    TI__Mastermind 30000 points
    Hi Jon,

    Kedar is no longer within the PSIL-PI team and has moved over to our battery charger group. I'll be taking over in supporting this issue. Would you be able to send over both project configuration files for the TPS65981 and TPS65982 from the above tests? From what it sounds like, the problem is somehow related to either the TPS65982 project configuration, or a potential issue with the customer schematic and how the TPS65982 is implemented. Does not seem like the issue is related to the TPS65987D as it is able to negotiate correctly with a TPS65981EVM.

    In addition to sending both of these configuration files, would you be able to share with me the customer schematic for the system using the TPS65981? You can share it with the "TPS65981 ACS005-B Schematic" as Kedar has added me to that email chain.

    Regards,
    Adam
  • 0 in reply to Adam Mc Gaffin
    TI__Intellectual 920 points

    Hi Adam,

    Sent you a note with the requested schematics.

    Thanks,

    Jon

  • 0 in reply to Jon Key
    TI__Intellectual 920 points
    Hi Adam,

    We solved the issue of the two customer boards not communicating. The customer had the variable DC-DC converter that connects to PP_HV1 at 0V until a PD contract was negotiated. Reworking the board such that 5V was always present on the PP_HV1 pin has fixed the issue for now.

    However the TPS65987D board needs to be in an ultra-low power mode when there is no cable inserted, is there a way to completely turn off the variable DC-DC converter until a PD contract is negotiated (i.e. have 0V on PP_HV1 until a contract is negotiated)?

    Second question, the customer would like the source to be capable of sourcing 5A through the TPS65987D, but the utilities tool will only let him transmit a source capability of 3A. Is there a way to configure the 65987D to transmit a source cap of 5A?

    Thank you,
    Jon
  • 0 in reply to Jon Key
    TI__Mastermind 30000 points
    Hello Jon,

    That makes sense as to why the board was failing, glad you and the customer were able to figure that out.

    As far as having the DC-DC output 0V until a contract is negotiated, you could use a GPIO event such as "PlugEvent_Port1" (described in Host Interface TRM). My thinking is that you could have this GPIO event connected to the gate of a FET controlling the enable of the DC-DC. When a cable is plugged in to the Type-C port, this event will go high enabling the DC-DC and put 5V on PP_HV1. You can change the final implementation the customer decides to use, but this is the basic operation I am thinking.

    What was the customer using before to have 0V presented on PP_HV1 until a contract was negotiated?

    For advertising 5A, this is contingent on the cable that is plugged into the Type-C port. The TPS65987D will detect the cable it is connected to and determine the current capabilities it is capable of supporting. If the 87D is programed to advertise 5A, but the cable attached can only handle 3A, the 87D will change its source capabilities to advertise only 3A.