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.

TPS25750: DRP role for sink and source application

Part Number: TPS25750
Other Parts Discussed in Thread: TPS2121

Hello, 

Explanation of design

Scenario 1) I am designing a board where I want to have data communication between the Samsung A7 Lite tab and my MCU over USB as well as power the MCU using power from the Tablet.

Scenario 2) When the A7 lite Adaptive charger will be plugged into my board then my design should charge the A7 lite Tablet as well as power the MCU. 

I am working on the design (Please check attached image). As per the guidance from Adam, I have been using the TPS25750 for my design, whereas I have some questions regarding the design. 

FYI, I am using ATMEGA32U2 as MCU, it have in built USB peripheral but unfortunately it does not have I2C peripheral. 

As per my information Samsung A7 lit tablet has USB Type-C 2.0 and supports fast charging 15W. 

Please find my questions below. 

1) I believe that I shall need to provide the I2C connector for externally configuring the TPS25750 into the DRP Role. 

2) Once Configured in DRP role, do I still need to have the EEPROM connected to TPS25750? in which scenario do I need to have EEPROM Connected. 

3)  The Samsung adaptive charger is a 5VDC 1.5A charger.  The Samsung website says Adaptive charger are QI compatible, dose this means that the VBUS voltage of charger will go above 5V?

4) Is it true that the PP5 pin is a unidirectional pin and we can use it to charge the Tablet up to 5V 3A? If yes then will it be safe to connect Charger VBUS pin to PP5 pin? 

5) As my power requirement for the MCU + Motor is not more than 800mA. Also I do not want to go for higher voltage PPHV during powering MCU in absense of charger Hence is there a provision in Application Customization tool where I can restrict PPHV signal to 5V only. 

Also I have made the schematic which I would like to share in private for your review and guidance. 

Thank you.

Piyush Saxena

  • Hi Piyush,

    I reckon the device you use is TPS25750D. And for TPS25750, FW is mandatary, thus I think you should add EEPROM on your board.

    So here are answers:

    1) -> TPS25750 need EEPROM to load its FW, and we configure all the setting in FW, so I think you cannot have TPS25750 work without EEPROM.

    2) -> As mentioned in 1), you may need a EEPROM.

    3) -> If you mean Qi compatible, I think it refers to the wireless charging standard; that will make no impact on the adaptor, it should be okay.

    4) -> Yes, PP5V should be connector to a DCDC, and when the port needs to supply power to the far end devices, TPS25750 will turn on the switch between PP5V & VBUS, then power can be sent out from DCDC to the far end power consumer device.

    5) -> If you don't need higher power contract than 5V, you can only enable Type-C state machine in application tool and disable PD. When the devices in both ends connected, they will do a handshake and determine the power current they need, but VBUS voltage will always keep at 5V.

    Thanks.

    Bing

  • Hello Bing, 

    Thank you for your feedback. Please find below my further questions. 

    1) -> TPS25750 need EEPROM to load its FW, and we configure all the setting in FW, so I think you cannot have TPS25750 work without EEPROM.

    --> Do you mean to say that once I have configured TPS25750 in DRP role from online Application Customization tool TPS25750 over I2C master port using Aadavark tool then TPS25750 will load this config file in EEPROM and will always work in DRP role. After the configuration, if I want to stick with this setting then I wont need to communicate with TPS25750 anywhere in the future using its I2C slave port. I am asking because my selected MCU doesn't have I2C port, so want to be sure of this point. 

    2) -> As mentioned in 1), you may need a EEPROM.

    --> Okay understood. 

    3) -> If you mean Qi compatible, I think it refers to the wireless charging standard; that will make no impact on the adaptor, it should be okay.

    --> Sorry my mistake, I mean to say that as per my finding on google the adaptive fast charger of samsung are QC compatible (They use Qualcomm chips) whereas there latest super fast chargers are USB PD compatible. But we are interested in their  earlier version adaptive fast charger as it is this charger which is shipped with the samsung A7 lite Tablet. The power rating of adapter is 5V 1.5A, based on your extensive experience do you think these will have the possibility of charging Tablet with higher voltage then 5V as I am not able to find this data on google anywhere?

    4) -> Yes, PP5V should be connector to a DCDC, and when the port needs to supply power to the far end devices, TPS25750 will turn on the switch between PP5V & VBUS, then power can be sent out from DCDC to the far end power consumer device.

    --> Once configured in DRP role does TPS25750 switches from sinking / sourcing modes automatically based on voltage levels at the input pins? Can you please let me know which are the pins which are monitored for switching the profile from sinking to sourcing? 

    5) -> If you don't need higher power contract than 5V, you can only enable Type-C state machine in application tool and disable PD. When the devices in both ends connected, they will do a handshake and determine the power current they need, but VBUS voltage will always keep at 5V.

    --> If we configure the device as per your above suggestion, What will be the maximum current we will be able to draw from Tablet as well as what will be the maximum current at which the TPS25750 will be able to charge the Tablet when adaptive fast charger in plugged into my board? Also I hope this setting will allow us to have the data communication right?

    6) As a safety precaution, I was looking for having some kind of reverse current protection IC, to avoid the current from PPHV to flow back into the Adaptive charger port. Can you suggest any such extreme low dropout reverse protection IC which can also sustain the 20VDC(not sure since adaptive charger may generate 20V for fast charging the device ) from adaptive charger port?

    7) Can you please let me know how will I be able to share my schematic with you in private. 

    Thank you. 

    Regards

    Piyush Saxena 

  • Hi Piyush,

    Please notice that the Sink/ Source power role is determined by Type-C  CC state machine, which means that all the behavior of PD controller's will be determined by CC's state.

    1) -> TPS25750 need EEPROM to load its FW, and we configure all the setting in FW, so I think you cannot have TPS25750 work without EEPROM.

    --> Do you mean to say that once I have configured TPS25750 in DRP role from online Application Customization tool TPS25750 over I2C master port using Aadavark tool then TPS25750 will load this config file in EEPROM and will always work in DRP role. After the configuration, if I want to stick with this setting then I wont need to communicate with TPS25750 anywhere in the future using its I2C slave port. I am asking because my selected MCU doesn't have I2C port, so want to be sure of this point. 

    -> Yes, you can configure FW in the application tool, and generate bin file. Then you can flash the bin file into EEPROM. When every time TPS25750 is powered on, it will load FW from EEPROM.

    2) -> As mentioned in 1), you may need a EEPROM.

    --> Okay understood. 

    3) -> If you mean Qi compatible, I think it refers to the wireless charging standard; that will make no impact on the adaptor, it should be okay.

    --> Sorry my mistake, I mean to say that as per my finding on google the adaptive fast charger of samsung are QC compatible (They use Qualcomm chips) whereas there latest super fast chargers are USB PD compatible. But we are interested in their  earlier version adaptive fast charger as it is this charger which is shipped with the samsung A7 lite Tablet. The power rating of adapter is 5V 1.5A, based on your extensive experience do you think these will have the possibility of charging Tablet with higher voltage then 5V as I am not able to find this data on google anywhere?

    -> You mean the earlier version adapter and A7 lite tablet is in one production set. TPS25750 cannot support QC, but it can support Type-C protocol and PD protocol. I reckon even if the earlier version of this adapter, it can also support Type-C protocol as well as its a Type-C port. But in only 5V is available with Type-C protocol and three level of current: 0.9A, 1.5A, 3A. And notice that there three values are the maximum current limited by Type-C protocol, but the actual current drawn by load is determined by the device. For example, if you set Type-C at 0.9A, the max current can send out by PD is 0.9A, if you draw 1A, TPS25750 will do an over-current protection.

    4) -> Yes, PP5V should be connector to a DCDC, and when the port needs to supply power to the far end devices, TPS25750 will turn on the switch between PP5V & VBUS, then power can be sent out from DCDC to the far end power consumer device.

    --> Once configured in DRP role does TPS25750 switches from sinking / sourcing modes automatically based on voltage levels at the input pins? Can you please let me know which are the pins which are monitored for switching the profile from sinking to sourcing? 

    -> As mentioned before, Sink/ Source is determined by CC pin's status, and TPS25750's switch will take action according to the power role. When the first time device is connected, the power role is determined.

        And I need more details here; TPS25750 only has one port, so how the adapter connected to your board? If the tablet has connected to the port?

    5) -> If you don't need higher power contract than 5V, you can only enable Type-C state machine in application tool and disable PD. When the devices in both ends connected, they will do a handshake and determine the power current they need, but VBUS voltage will always keep at 5V.

    --> If we configure the device as per your above suggestion, What will be the maximum current we will be able to draw from Tablet as well as what will be the maximum current at which the TPS25750 will be able to charge the Tablet when adaptive fast charger in plugged into my board? Also I hope this setting will allow us to have the data communication right?

    ->As mentioned in 3), it is determined by how you configure the CC state machine. 

    6) As a safety precaution, I was looking for having some kind of reverse current protection IC, to avoid the current from PPHV to flow back into the Adaptive charger port. Can you suggest any such extreme low dropout reverse protection IC which can also sustain the 20VDC(not sure since adaptive charger may generate 20V for fast charging the device ) from adaptive charger port?

    ->I think TPS25750 has integrated RCP, you can find it in the datasheet. 

    7) Can you please let me know how will I be able to share my schematic with you in private. 

    -> Not sure if there is an FAE that you can contact? I think you can try to contact FAE in the first step.

    Thanks.

    Bing

  • Hello Bing, 

    Thanks for your feedback. Please find my reply below. 

    1) Okay understood. 

    2) Okay understood. 

    3) In order to understand the connection I would like to request you to have a look at the image I uploaded at the starting post of this thread. FYI, I am designing a board which will have a usb type C to wired input at one end which will be used for connection between the A7 lite Tablet's usb data,C1 and C2, Vbus and GND connection and the board. On the other end of my board there will be a USB type C receptacle where the original A7 lite tablet adaptive charger's type C cable will be connected. 

    I am not sure if the A7 lite tablet adaptive charger has C1 and C2 pin, even if it has I do not where should I connect them, as I have already connected the Tablet's C1 and C2 pins to the TPS25750. I am only using Vbus and GND pin of the A7 lite tablet adaptive charger's pins for power, Do you think by using only this pin it shall be able to deliver 5V 1.5A ?

    4) As mentioned in point 3, 

    Samsung A7 lite tablet ----(Type c to wired connection)-----------> My Board ---------(Type C receptacle )-----------> Samsung Adaptive Charger type c cable. 

    In my Board the C1 and C2 pins, VBUS, GND and Data pins from Samsung A7 lite tablet are connected to TPS25750. 

    I am only using VBUS and GND pin of the samsung adaptive charger type C cable. 

    My Board has a MCU and a Motor driver IC for driving motor, I have a 5V to 3.3V regulator for powering the MCU and the motor is Powered by 5V. I want the MCU to be powered by the A7 lite Tablets power when there is not samsung adaptive charger connected to the other end of my board. If the Adaptive charger is connected then I want to chager the A7 lite Tablet as well as power the MCU from adapter's power. 

    Now in order to achieve this, I will Initially configure the TPS2750 in DRP role, but my question is that will this sink/ source mode will be switched automatically by the TPS25750 based on power level at its input/output? Or I will have to switch the role by digitally communicating with TPS25750 with my MCU. 

    5)  okay understood. 

    6) Okay, understood. 

    7) Do you how can I contact FAE for sharing my schematic. 

    Regards

    Piyush Saxena 

  • Hi Piyush,

    If the A7 adapter is in the physical form of Type-C, it shall follow all Type-C standard. I don't think it is a valid device if it is in Type-C form, but violates Type-C standard. That means, if you only connect VBUS and GND to the A7 adapter's Type-C port, you cannot also get the power, because Type-C Specification asserts that no voltage should present on VBUS unless CC1 and CC2 have negotiated its Type-C state machine. So I assume maybe you need two PD controllers, or, for the A7 adapter port, you can only have a Type-C controller on it.

    So if the port is in the physical form of Type-C, no matter how many power charging protocols it supports, it shall follow Type-C Specification firstly.

    And for "a usb type C to wired input at one end", I think you mean the Type-C captive cable?

    Thanks.

    Bing

  • Hello Bing, 

    Please find the above image which shows the actual adapter and cables which shall be used in my device. Also I have drawn a block diagram of internal connections of my propose board. 

    I hope this answers your question. 
    Please let me know what do you think about my point 3 and point 4 of my previously asked question. 

    Regards

    Piyush Saxena 

  • Hi Piyush,

    Got it. The port connected to A7 adapter is the Sink only port, thus you need to put a 5.1 kOhms resistor on its CC1 and CC2 pins. After the adapter detects the resistor, it will send out 5V on VBUS then the board is powered from A7 adapter. If you want to see more details for this kind of adapter (Type-A plug to Type-C plug), you can take Type-C Specification as reference.

    After board is powered from the port connected to A7 adapter, TPS25750 will start to work. So yes, you don't need to worry about TPS25750's behavior, it can choose the power role automatically if you set FW right. (DRP, prefer source).

    Hope these will answer your questions.

    Thanks.

    Bing

  • Hello Bing, 

    I think with the last Image you were much more clear with my requirement. Anyways thank you for your answer.

    Now just to summaries our entire discussion which we have in this thread. Please find below my queries. 

    After connecting 5.1KOhms resistor on C1 and C2 line of A7 charger's Adpater port connector. Will it be safe to assume that the VBUS voltage will not go above 5V?

    Now once I have configured the TPS25750 in DRP role and will customize it to 5V 15W maximum source and sink power configuration with data role of USB2.0 speed. So doing so will it ensure me that the PPHV voltage of the TPS25750 will also never go above 5V?

    Similarly with the above configuration I shall also be able to charge the A7 lite Tablet via PP5 line of TPS25750 with 5V 1.5A  A7 Charger adapter plugged into my board, am I right?

    Once the TPS25750D is configured with the DRP role and I want to stick to this role with the above mentioned customization setting and loaded into the EEPROM connected to TPS25750 then I do not need to have any I2C communication between the MCU and TPS25750. I believe that TPS25750 will continue to work as required with the need to any I2C communication for MCU, am I right?

    One last thing, I have made the schematic as per our discussion, can you please suggest me how can I get some expert comments like your but in private as I do not wish to share it in public domain?

    Thanks

    Regards

    Piyush Saxena 

  • Hi Piyush,

    1.For Type-C standard, 5V is the only voltage level for VBUS. So the answer is yes, the port connected to A7 adapter will be not more than 5V.

    2. If the hardware is good, then a EEPROM, and the setting with "DRP, prefer source" of TPS25750 can meet all your requirements.

    3. If you can contact with FAE, you should be able to submit thread to internal E2E, then we can review your SCH.

    Thanks.

    Bing

  • Hello Bing, 

    Thank you for your support. 

    Regards

    Piyush Saxena 

  • Hello Bing,

    Hope you are doing good!

    I have made the schematic for my requirement. I am sharing you only one page of the schematic which has the TPS25750 section, Power switching section and 2 connectors one for the charger adapter and other for the tablet connection. 

    FYI. In power switching section there are 2 redundant circuit, one solution is using only mosfets and other solution is using TPS2121 IC. 

    I would like to request you to go through it and let me know if you find everything fine. 

    Thanks

    Regards

    Piyush Saxena

    7673.Sheet1.pdf

  • Hi Piyush,

    It reminds me that, do you need to support a kind of situation, where A7 lite tab is firstly plugged to the board, (here the tab will charge the board), then A7 adapter is connected to the board and the a dynamic power flow direction transition occurs, and A7 adapter starts to power the board and A7 lite tab?

  • Hello Bing, 

    Greetings of the day!

    Yes your understanding is correct, based on our past discussion in this post I have prepared a schematic design and share with you in my last post. I would like to have your expert comments on my schematic design. 

    Thanks 

    Piyush Saxena

  • Piyush, Bing is out of office today. Please give him time to check and get back on your request. 

  • Hi Prem 

    Okay, 
    I will wait for his feedback until he returns. 

    Regards

    Piyush Saxena 

  • Thanks! please also note that April 15th is holiday for TI.

  • Hi Piyush, the schematic looks good. And your Power Switching Section is used to lock the voltage? -> My understand is that when the adapter is plugged in, the voltage from PPHV will be cut off.

    I think there is a point, as mentioned before, " then a dynamic power flow direction transition occurs". I think you may also need connect I2C1 to the MCU which instructs PD to do the transition. PD cannot do that by itself. But let me check what is the command for TPS25750.

  • Hello Bing, 
    Its good to hear from you back. 

    Yes the Power Switching section is used to restrict the reverse flow of the power and also to give the charger highest priority when connected. 

    FYI, as per our earlier conversation on this thread I have mentioned that my controller doesn't have I2C peripheral. Hence I asked specifically if the DRP sink or source transition is automatically taken care by the TPS25750 as per its configuration from the EEPROM without any need of the external I2C communication from the MCU. 

    Since my MCU doesn't have any I2C hence we discussed this at the starting of this thread, so that I will look forward to use this MCU.

    1) Can you please check if you are sure that TPS25750 will not switch the DRP role automatically ? Can you please double check and let me know if there is any other option?

    2) Is it because of my Power switching section which blocks the reverse voltage of PPHV line?

    3) I also want to ask you, I am not sure how the ADCIN1 and ADCIN2 should be handled for my application. What should be voltage level  and the signal to be give for the ADCIN1and ADCIN2 pin.

    Thanks 

    Piyush Saxena 

  • Hi Piyush,

    1) Can you please check if you are sure that TPS25750 will not switch the DRP role automatically ? Can you please double check and let me know if there is any other option?

    I need to try it on EVM. I don't think TPS25750 can cover the situation where dynamic power flow change is needed. (application scenario is that tab is plugged in first and charge the board, then adapter is plugged in from another port.)

    2) Is it because of my Power switching section which blocks the reverse voltage of PPHV line?

    Power switching section should be acceptable. But TPS25750 may not have the automatic mechanism to change the power flow direction.

    We should have transferred the power flow control to the host, that is, an MCU or other controller. Because PD controller doesn't know the whole power situation of the system, in some scenarios, power flow change is not needed, thus we didn't set the power change feature in TPS25750. But I can try if there are other ways to trigger the power flow transfer.

    3) I also want to ask you, I am not sure how the ADCIN1 and ADCIN2 should be handled for my application. What should be voltage level  and the signal to be give for the ADCIN1and ADCIN2 pin.

    Based on your application, I recommend tie ADCIN1 & ADCIN2 to LDO_1V5. With this setting, TPS25750 will always enable Sink Path when the Tab is connected to the port.

  • Hi Bing, 
    Hope you are doing fine!

    Do you have any feedback for the 1st and 2nd point of my previous post. 

    Thanks 

    Piyush Saxena 

  • Hi Piyush,

    1) The behavior of TPS25750 is based on whether it will clear its Dead battery flag; Because the power from adapter side is connected to TPS2570's 3.3V power, before the adapter is connected in, it is in Dead battery mode. It should not change the power flow direction automatically.

    2) Power switching section is good, I don't see abnormal.

    Thanks.

    Bing