TPS25751: development problems

Hi Jan,

I read through the previous thread, and am aware that you will want answers and a solution urgently. I will prioritize this issue and begin by doing the following:

1. Reviewing the firmware implementation for the active RDO.
2. Reviewing the feature requests for this device.

I will update you on Monday with my findings and next steps.

Kind regards,

Conner Gillette

Hi Conner,

I made an error on Friday, I forget to cycle the power after new patch data generated from USBCPD Application Customization Tool, so I was evaluating the system with patch data from previous version of USBCPD Application Customization Tool.

So when I do the power cycle now in order to patch the TPS25751 with correct data, the system is not working neither.

In Sink mode, it is working correctly, but in Source mode, connected device (sink - e.g. mobile phone or our development board from STM) is not able to establish stable PD contract with the source.

Could you help please with this? It was working better with PTPS25751...

Thank you

Hi Conner,

I attach json configuration file.

To describe better our device, it will be 4s smart battery charger with USBC PD input and output, like smart power bank with replaceable battery. It will be possible to charge battery from PD source connected with 20V 3,25A power supply (or weaker if not available this powerful one) and if battery is charged, it will be possible to power external system with our device with one of the following profiles - 5V/3A, 9V/3A, 12V/3A, 15V/3A or 20V/5A.

What was functional with previous version of USBCPD Application Customization Tool and PTPS25751 is:

1) When battery is not connected and PD power source supporting 5V/3A, 9V/3A, 12V/3A, 15V/3A or 20V/3,25A is connected, then TPS25751 correctly establish PD contract using highest power profile available - 20V/3,25A, this is correct.

2) When battery is connected (system is powered up including TPS25751) and PD power source supporting 5V/3A, 9V/3A, 12V/3A, 15V/3A or 20V/3,25A is connected, then TPS25751 surprisingly choose 5V/3A profile and uses it for battery charging. This is incorrect, we want the TPS25751 to chose also highest power profile available - 20V/3,25A.

3) Battery is connected (system is powered up including TPS25751) and external device (PD sink) is connected no PD contract is reached, this was partly working with previous version of USBCPD Application Customization Tool and PTPS25751, usually lowest power profile was established using STM P-NUCLEO-USB002 device.

could you help us with this please?

Thank you

{
  "questionnaire": {
    "device": "TPS25751",
    "answers": [
      null,
      0,
      4,
      4,
      0,
      0,
      0,
      0,
      1,
      1,
      1,
      1,
      2,
      1,
      16.8,
      4,
      0,
      0,
      2.88
    ],
    "vendorId": "0000",
    "productId": "0000",
    "version": "1.0.0.2"
  },
  "configuration": {
    "data": {
      "selected_ace": [
        {
          "register": 6,
          "data": [
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0
          ]
        },
        {
          "register": 22,
          "data": [
            10,
            56,
            0,
            140,
            0,
            4,
            0,
            12,
            0,
            0,
            3
          ]
        },
        {
          "register": 40,
          "data": [
            2,
            0,
            47,
            4,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0
          ]
        },
        {
          "register": 41,
          "data": [
            82,
            80,
            128,
            0
          ]
        },
        {
          "register": 50,
          "data": [
            5,
            168,
            42,
            44,
            145,
            1,
            16,
            44,
            209,
            2,
            0,
            44,
            193,
            3,
            0,
            44,
            177,
            4,
            0,
            244,
            65,
            6,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0
          ]
        },
        {
          "register": 51,
          "data": [
            5,
            44,
            145,
            129,
            0,
            44,
            209,
            2,
            0,
            44,
            193,
            3,
            0,
            44,
            177,
            4,
            0,
            69,
            65,
            6,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0
          ]
        },
        {
          "register": 55,
          "data": [
            63,
            64,
            31,
            65,
            144,
            145,
            1,
            0,
            6,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0
          ]
        },
        {
          "register": 92,
          "data": [
            13,
            12,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            5,
            0,
            0,
            0,
            48,
            4,
            0,
            0,
            0,
            0,
            0,
            0,
            5,
            12,
            0,
            0,
            0,
            0,
            0,
            0,
            5,
            8,
            0,
            0,
            73,
            0,
            76,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            1,
            0
          ]
        },
        {
          "register": 112,
          "data": [
            0
          ]
        },
        {
          "register": 152,
          "data": [
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0,
            0
          ]
        }
      ]
    }
  }
}

Thank you for the additional detail and the json file. I will review it and root cause your issues and provide an update tomorrow.

Kind regards,

Conner

Hello Conner,

do you have any updates please? We would really like to make it work quickly.

thank you

My apologies Jan. I need at least one more day to investigate and I will get back to you with a meaningful update.

Regards,

Conner

Hello Conner,

I just wanted to let you know our design is derived from TPS25751EVM and we use BQ25731 battery charge controller with USB type-C PD support and external microcontroller for communication over I2Ct interface with TPS25751 and we control BQ25731 from mcu too.

Thank you for any news

My apologies for not having a meaningful update yet Jan. I am actively consulting with my peers to help find the root cause of your issue.

Regards,

Conner

Hello Conner,

thank you very much for your replies, but could you give us at least some more details please? Do you thing it will work in the end as desired in dual role power applications, please?

We are couple of engineers here waiting for it and would very much prefer to wait for successful end than to have to start again from scratch with different solution.

We have cross fingers that you will be able to figure it out how to get it working.

Thank you

Hello Conner,

I tried to be polite and not to push too hard, but it seems this is not the right approach. We have had serious problems with your products over the past 6 months. Firstly, with the active product TPS25750, where the declared functions are not working correctly and you admitted that. (No solution came from your support.) You forwarded us to your new TPS25751, which at that time was in the preview state PTPS25751, and it was supposed to finally work fine. Our development team waited a few months for this chip, which cost us a fortune. PTPS25751 maybe solved some of the previous bugs of TPS25750, but it had new bugs, which we reported. You (TI) wrote that you knew about these bugs we reported, but we were assured that these bugs would be solved in the final version of TPS25751, which was supposed to be on the market in the next 3 months. Our team waited again, and the opportunity cost we spent on that cannot be easily described. After COVID-19 times, it is very hard for smaller manufacturers like us to wait such a long time. When the final version of TPS25751 came (with about a 3-week delay), it still had the same bugs, which you confirmed as known bugs and bugs that would be solved in the final version. To be exact, these bugs are making this part useless in the way you describe on the web and in datasheets. When we reported that to you, we got an understanding response, but no solution for almost 3 weeks. This product we are working on is very important to us, and we are in the very final stage of development, just waiting for a solution from you that would make this part work as declared in your datasheets and on the web.

We really need to be more involved in the steps you are taking right now. We need to know if there is an issue with the current batch of TPS25751 or if you are going to discontinue the part because of this behavior, which makes this part useless for everyone (because it cannot offer sourcing power role). Or if you found the issue and are working on some workaround that would make the functionality of the part right.

We are very frustrated from waiting and waiting and we need to be informed accordingly.

Since there is no other support channel than this forum, we are forced to write it here.

Thank you.

Jan

Hi Jan,

I am not a member of TI group, just a user to use TPS25751 and BQ25756. In our design, TPS25751 and BQ25756 are also controlled separately by MCU.

The same issue of TPS25751 in PREVIEW stage which mentioned in your last post also occured on our desgin, and it seemed to be solved on the new TPS25751 in ACTIVE stage(It was not confirmed by official TI, so I used ‘seemed to be solved’ here).

The key was thay we found that there's Reverse Mode in the charger BQ25756, and this mode had to be enabled when the TPS25751 worked as Source, and you had to set REG0x0C to a wider limit(The default limit of this register is 5V).

You could see my last post for reference, hope it could solve your problem.

Thanks,

Alex

Hello Alex,

thank you a lot for your message, unfortunately this did not resolve our issue, as we have more serious issues with TPS25751, not only original issue as described at the beginning. Could you email me at "pa.vavra (at) volny.cz" I would like to ask you some additional information, but I don't want to write it here as I would like to keep the thread clear and empty for direct TI support.

We are still waiting for TI support, please, Conner, could you give us some information?

Jan Zastera said:

2) When battery is connected (system is powered up including TPS25751) and PD power source supporting 5V/3A, 9V/3A, 12V/3A, 15V/3A or 20V/3,25A is connected, then TPS25751 surprisingly choose 5V/3A profile and uses it for battery charging. This is incorrect, we want the TPS25751 to chose also highest power profile available - 20V/3,25A.

Are you able to provide the USB-PD logs of this behavior? There is a possibility that a 20V contract is being requested, and this will help clarify the behavior.

Jan Zastera said:

3) Battery is connected (system is powered up including TPS25751) and external device (PD sink) is connected no PD contract is reached, this was partly working with previous version of USBCPD Application Customization Tool and PTPS25751, usually lowest power profile was established using STM P-NUCLEO-USB002 device.

Do you have the USB-PD logs for this situation as well?

---

I feel like the above two issues are related to communication to the BQ25731 device, and that voltage transitions are not occuring (circled in figure below) after tSrcTransition but before the PD controller sends the PS_RDY message, causing a hard reset in question 3 and a default to an implicit Type-C contract in question 2 above.

What values are you reading in the Active RDO register right after the SinkTransitionCompleted interrupt?

Can you provide logs that elucidate the values communicated and timing of the communication, as well as whether or not the voltage transition is occuring?

Hello Conner,

thank you for getting back to us,

add 2) There are two scenarios of TPS25751 SINK,
- case1 - not fully charged battery is inserted, then current draw occurs right after connection to SOURCE and in fact, no contract is reached. See picture bellow:

- case2 - fully charged battery is inserted, then after connection to SOURCE also no contract is reached. See picture bellow:

- As I stated above, this interrupt is never triggered as PD contract is not reached.

Conner Gillette said:

Can you provide logs that elucidate the values communicated and timing of the communication, as well as whether or not the voltage transition is occuring?

- again I think I cannot do it as no PD contract is reached.