• State Resolved
  • Replies 10 replies
  • Answers 1 answer
  • Subscribers 67 subscribers
  • Views 168 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

TPS25751: Proper device configuration for dead battery configuration and EEPROM requirements

William Fahrenkrog
Prodigy 20 points
Part Number: TPS25751
Other Parts Discussed in Thread: TPS25730, TPS26750

Tool/software:

Overview:

As background, the application our team is working on uses the TPS25751S USB PD Controller.  The goal is to have the TPS25751 and attached hardware powered only by USB power going to the TP25751.  There are no other power sources in this design.  Furthermore, there are no batteries, nor battery chargers included in this design either.  I believe this is considered a dead battery configuration.  A microcontroller is also connected to the I2C interface on the TP25751 chip as well.  Based on that, our team has come up with a few questions based on review of the documentation.

1. Looking at table 2-1 of the Common TPS25751 Use Cases and Setting Using EC document, our team is looking at two cases:  ‘AlwaysEnableSink’ and ‘NegotiateHighVoltage’  I placed the text for each mode below:

AlwaysEnableSink: The device always enables the sink path regardless of the amount of current the attached source is offering. USB PD is disabled until configuration is loaded. This configuration is used with an external embedded controller. The embedded controller manages the battery charger in the system when present.

Negotiate High Voltage: The device always enables the sink path during the initial implicit contract regardless of the amount of current the attached source is offering. The PD controller enters the APP' mode, enable USB PD PHY and negotiate a contract for the highest power contract that is offered up to 20V. The configuration cannot be used when a patch is loaded from EEPROM. This option is not recommended for systems that can boot from 5V. This configuration is not valid to use with any supported battery chargers.

Based on this description, it appears that AlwaysEnableSink mode enables the sink path immediately without PD negotiation and allows for USB PD negotiations to be enabled by the EC afterwards. 

The Negotiate High Voltage mode also appears to enables the sink path but performs PD negotiations using an initial contract.  some sort of default contract.  The contract that then be updated later through the embedded controller.

 

The questions for this part are:

Is the initial contract a default contract built into the PD controller?  If not, where does it come from or how is it configured?

Does Negotiate High Voltage mode require an EEPROM for proper operation?  Or does this mode require an EC to configure and operate the PD controller?

What is the primary objective for these two modes for how they are intended to be used?

 

2. The documentation is not very clear.  Is EEPROM required to load contracts onto the TPS25751?  Documentation appears to indicate that contracts can be loaded via EEPROM or an attached embedded controller.

 

3. EEPROM initial load.  If using EEPROM, is it possible to start off with a blank EEPROM?  Or is some sort of initial load required?

  • 0
    TI__Mastermind 24322 points

    Hi William, 

    Thank you for reaching out!

    See my comments below: 

    • The device is shipped empty and would need to be configured either by loading the configuration from an external EEPROM or configuring the PD using MCU/EC over I2C.
    • TI's GUI Tool creates a configuration image (binary file) for the user's application based on a few simple questions.

    • When configuring the PD controller using an MCU, there is a particular process called Patch Burst Mode (PBM) that would need to be implemented. 

      Essentially, the LowRegion Binary file (generated from the GUI Tool), would be pushed over I2Ct using 4CC commands. See Section 5 in the Technical Reference Manual for the PBM flow. 

    • The dead-battery configuration options describe the different behavior of the sink path before the configuration is loaded onto the device (see the following from the datasheet)

    • Typically, the recommended dead-battery configuration will depend on how the configuration is being loaded onto the device (i.e. EEPROM or MCU/EC)
      • AlwaysEnableSink is recommended when using an EC to load the configuration in order to enable the sink path and provide power to the system while loading the patch bundle over I2C. 
      • The NegotiateHighVoltage dead battery configuration will load a "default" configuration onto the device in which it will enter APP mode and negotiate a PD contract before the EC loads the patch bundle.

    Please let me know if you have additional questions/concerns. 

    Best Regards, 

    Aya Khedr

  • 0 in reply to Aya Khedr
    Prodigy 20 points

    Aya,

    Thank you for your response.  This does not quite answer the questions I had though.  Just to clarify, we are currently NOT planning to include a EEPROM in our design.

    You reference the "default" configuration when talking about the NegotiateHighVoltage dead battery configuration.  Where does this default configuration come from?  Is this default configuration native to the TPS25751S USB PD Controller?  What exactly is the behavior of this default image?

    The table explains that the in the NegotiateHighVoltage dead battery configuration, the PD controller will negotiate the highest power up to 20V.  Does this mean it will automatically accept anything below that 20V?  such as 5V at 3A?  Why is this option not recommended for systems that can boot from 5V?

    Regards,

    William

  • 0 in reply to William Fahrenkrog
    TI__Expert 6520 points

    The default is something inside the ROM. When there is no EEPROM or no MCU to load the patch then depending on the ADCIN configuration, PD controller will drop to default settings. 

    If this is sink only application and you have no EEPROM plans and looking only for defaults then TPS25730 is a similar device with more options for defaults. Please refer to that device.

  • 0 in reply to Ghouse Mohiuddin23
    Prodigy 20 points

    Thank you Ghouse,

    Can you tell me what this default actually is? your statement that this is something inside the ROM implies that this is native to every TPS25751S USB PD Controller.  What is its behavior while in default mode? For instance, if the PD controller is in Negotiate high voltage mode, what happens when 5V @1A is applied?  What happens if 20V @3A is applied?

    Unfortunately, this is not a sink-only application.  We are planning to use two TPS25751S USB PD Controllers in this design.

    Regards,

    William

  • 0 in reply to William Fahrenkrog
    TI__Mastermind 27995 points

    Hi William,

    Let me describe the device behavior of the TPS25751 to hopefully clear some things up.

    Snapshot taken from section 8.1 of the datasheet.

    As Ghouse mentioned, the device has ROM that contains "default" configurations. These configurations are only for "dead battery" behavior and are not intended to be the final configuration of the PD controller, merely a "passing" state to allow different behavior when loading in a dead battery state. 

    The device also has volatile memory that will hold patches and configuration settings(fw image), which is expected to be loaded on boot up. There are two options here: 1) An EEPROM connected over I2C is automatically loaded from at boot or 2) a I2C host/mcu loads the image on boot. 

    The TPS25751 can be powered two ways: 1) internal(to the system) power from a 3.3-V rail that feeds VIN3V3, or 2) bus powered, from VBUS in "dead battery mode".

    When unpowered, the TPS25751 exposes Rd pulldown resistors on the CC pins of the IC. This enables it to advertise as a sink device in an unpowered state, which allows power to be supplied by an attached USB-C Source device. With 5-V on VBUS, the TPS25751 can boot in the "dead battery state", and it's following behavior will follow one of the three "dead battery configs" defined in the datasheet.

    The reason for these different options is to allow for some flexibility for the customer to boot their system from dead battery, but does not change the fact that a fw image is expected to be loaded after this point from an external device.

    William Fahrenkrog said:
    Can you tell me what this default actually is?

    Default refers to the dead battery behavior described in the Datasheet. On boot in dead battery, the PD controller will have one of theses three "default" behaviors on boot.

    William Fahrenkrog said:
    For instance, if the PD controller is in Negotiate high voltage mode, what happens when 5V @1A is applied?  What happens if 20V @3A is applied?

    All three will start with a normal USB-C sink contract, which is the standard 5-V contract.

    "Safe mode" does not close the sink path and waits in that configuration. It stays in PTCH mode

    "Always enable" will close the path and not negotiate any other contract in the dead battery and unprogrammed state. It stays in PTCH mode

    "Negotiate High Voltage" enters APP mode and will negotiate a PD contract up to 20-V.

    So for your questions, in negotiate high voltage, the PD controller will negotiate the highest SPR (<100-W(20V 5A)) contract offered, whether it is 5V or 20V.

    This is not a recommended final operational configuration of the PD controller. It is expected to use this config to allow power into your system to boot an MCU/EC that will then program the PD controller with a FW image.

    William Fahrenkrog said:
    Unfortunately, this is not a sink-only application.  We are planning to use two TPS25751S USB PD Controllers in this design.

    understood, yes, then the TPS25730 can't be used in this case.

    Let us know if you have any more questions or if any are still unanswered.

    Thanks and Regards,

    Chris

  • 0 in reply to Christopher Lim1
    Prodigy 20 points

    Chris,

    Thank you for the clear response.  I am beginning to get a clearer picture of how this works.  I have one further clarification question.

    I am reviewing the TPS25751 Technical Reference Manual.  The manual references a legacy mode in table 3-10.  Would this imply that when a non-PD compliant source is connected to the TPS25751 when in NegotiateHighVoltage mode, the sink path will still be closed?

    Is there any documentation of this default behavior I can review?

    Regards,

    William

  • +1 in reply to William Fahrenkrog
    TI__Mastermind 27995 points

    Hi William,

    William Fahrenkrog said:
    I am reviewing the TPS25751 Technical Reference Manual.  The manual references a legacy mode in table 3-10.  Would this imply that when a non-PD compliant source is connected to the TPS25751 when in NegotiateHighVoltage mode, the sink path will still be closed?

    Yes, in fact, the normal type C contract, which is a 5-V contract initiated by the CC line resistive dividers is considered a non-pd contract and falls under legacy here. Yes, the sink path will be closed.

    Unfortunately, there is no additional documentation outside of the Datasheet regarding Dead Battery behavior at this time.

    Thanks and Regards,

    Chris

  • 0 in reply to Christopher Lim1
    Prodigy 20 points

    Thank you, Chris,

    I am going to go ahead and mark this issue as resolved since my primary questions have been answered at this point.  thank you to everyone who took the time to assist me on this.

    Regards,

    William

  • 0 in reply to Christopher Lim1
    Prodigy 40 points

    Hi I've got a related basic question/ clarification that I think is appropriate to add here: 

    What is meant by "sink path" ?

    Looking at TPS26750 ds fig 8-1 full system block diagram for example, is the "sink path" the load switch between VBUS and PPEXT to the battery charger enabled by POWER_PATH_EN ? So to "close the sink path" means that POWER_PATH_EN drives high to enable a load switch able to sink current from VBUS ? Or if not then can you please show me where that term is defined and what the TPS26750 does when closing the sink path? 

  • 0 in reply to Steve Lyon
    TI__Mastermind 27995 points

    Hi Steve,

    Thanks for reaching out. In the future, please use the "Ask a related question" to keep separate threads.

    Steve Lyon said:
    What is meant by "sink path" ?

    In general, sink path can designate any "power path" that is used to transmit power when the PD controller is acting as a switch. Typically it refers to the switch being used. In the case of the TPS26750, we utilize an external "sink path" that requires a separate load switch or gate driver + FET circuit. The PD controller manages the switch control and effectively controls when the sink path is open or closed.

    In the case of some of our other ICs, like the TPS25751D, there is an internal high voltage power path. In this situation, the internal power path can be configured to be the sink path.

    Steve Lyon said:
    Looking at TPS26750 ds fig 8-1 full system block diagram for example, is the "sink path" the load switch between VBUS and PPEXT to the battery charger enabled by POWER_PATH_EN ? So to "close the sink path" means that POWER_PATH_EN drives high to enable a load switch able to sink current from VBUS ? Or if not then can you please show me where that term is defined and what the TPS26750 does when closing the sink path? 

    Yes, your understanding is correct here.

    Thanks and Regards,

    Chris