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.

TPS25750EVM: + BQ25731EVM: Configuration and GUI

Part Number: TPS25750EVM
Other Parts Discussed in Thread: BQ25731, TPS25750, BQ25731EVM,

Hello,

I have a customer working on a BQ25731 + TPS25750 design using the included GUI, they are running into some setup issues.

  1. Do we have a setup guide to attaching the 2 EVMs. ie jumper settings and connection points?
  2. They are not able to get the BQ25731 to pass power to the TPS25750. They shouldn't have to program the BQ25731 separately (ie the TPS25750 should control it out of the box), correct?

Here is their setup on the EVM boards:

  • All jumpers default except
    • BQ25731.J7->set to 5 cell
    • TPS25750.J1->jumper removed and pin 3 attached to 3V3 on BQ25731(TP32)
    • IC2 on BQ25731 (TP30,31) attached to TPS25750.J7 (pins10,8)
    • Bench supply attached to BQ25731.VBAT screw terminals, set to 20VDC.
    • BQ25731.VIN screw terminals attached to the TPS25750.SYS_POWER and GND TPs

They get the 3V3 from the BQ25731 over to the TPS25750 but the BQ25731.VIN screw terminals don't pass the 20VDC, even when I plug in a USB-C device to the TPS25750 that requests 20VDC. I'm wondering if they’re missing something on how the wiring and Vregs are connected or some sort of programming or boot-up they need to do to the BQ25731?

Best,
Nick

  • Hello,

    I've assigned this issue to the relevant expert who will get back to you soon

  • Hi Nick,

    Do you have a block diagram or anything of their current setup? 

    We don't have a document at the moment but we do have this battery charger supported already on the GUI. What you described in your second point is correct, they shouldn't have to configure the BQ25731 as the PD should configure it via I2C.

    Did the customer flash the TPS25750 using the most updated GUI, one of the questions they have to select is the BQ25731 when they are prompted on which battery charger they are using. 

    Also what is the jumper J3 set to on the TPS25750 EVM? When you are referring to the VIN on BQ25731, is that the same as VBUS input on the battery charger as well? If so, they could connect this pin to the PPHV on the TPS25750 EVM, as this would be the direct power path for the PD controller which would output to the VBUS side of the TPS25750 in source mode.

    Thank you,

    Hari

  • Hi Hari,

    1. Yes I believe it is the latest (GUI Version :  6.5.4C4), I do select the BQ25731.
    2. J3 is on pins 2-3 (Source) (however I've tried 1-3, see answer 3 below)

    I had TPS25750EVM.TP3(SYS_PWR) attached to BQ25731EVM.J1(VIN) (which with BQ25731EVM.JP1 set on 1-2 results in it being attached to BQ25731.VBUS) but... I also had TPS25750EVM.J3 set to 2-3 which means PPHV is attached to VAR_DCDC

    It does seem like having TPS25750EVM.J3 set to 1-2 makes more sense as this would attach PPHV to SYS_PWR I just tried changing that so; BQ25731.VBUS is attached to TPS25750.PPHV (which I believe is what you are recommending, makes sense to me)
    However still no luck.

    We've confirmed that when the TPS25750 is powered up there is some initial communication on the I2C line, but there is never any I2C commands sent after that. Even when a USB source or sink is plugged into the TPS25750, there is no I2C communication. What should we expect to see on the I2C line during startup?

    Image of the current configuration:

    One additional question, which I2C bus should we be using? We currently have it tied to I2C3.

    Thanks,
    Nick

  • Dear Nick,

    Please remove JP6 on the BQ25731EVM in order to enable OTG. The OTG/VAP pin needs to be high in order to enable OTG.

    Thanks,

    Mike Emanuel

    Please click  "This resolved my issue" button if this post answers your question.

  • That did not address it. We still do not see any voltage on the "output" of the BQ25731. Although unless properly initialized by the TPS25750, I'm not if I would expect there to be one.

    Nick

  • Hello,

    This is a default configuration for these two EVMs and GUI, I'm sure someone has done this already, can we get them to chime in?

    Can you provide a block diagram or jumper configuration for this EVM/GUI setup?
    Is there a step by step initialization?
    if you have this configuration working on your end, can you send us a picture of your configuration?

    Nick

  • Dear Nick,

    Can you please confirm that the J3 connection is on source on the TPS board (it looks like it is on Sink in the picture) and the JP6 is removed on the BQ EVM? In addition, can you please capture the commands on the I2C line and decode it to see what is being broadcasted?

    Thanks,

    Mike Emanuel

    Please click  "This resolved my issue" button if this post answers your question.

  • Hi Mike,

    We will try that and we will get a clearer scope shot of the I2C lines. I do want to ask a more broad question for these boards and the GUI.

    The ultimate goal would be to setup the TPS25750 to communicate with a device plugged into the USB C port, determine if the external device is a source (UFP) or a sink (DFP) and then program the BQ25731 appropriately to either receive or provide power.

    Given the need to hard set the TPS25750 and BQ25731 jumpers for either source or sink rather than on-the-fly determination has me concerned that this isn't possible on these EVMs, is that correct?

    If it is possible, what is the proper configuration to do so?

    Thank you,
    Nick

  • Hi Nick,

    Yes from your description this is the intended configuration for the device. The BQ25731 does require external GPIO settings which will have to be done externally however, it is possible to do this on the EVM. 

    After discussing with my team, I believe you will have to change a few things around that I've went over below and I've also shown a diagram as a reference.

    1. To power the board, please use the battery output, VBAT and connect that to SYS_PWR of the TPS25750 EVM. This will power the board and both the 5V and 3.3V rail will be up. At the moment, you are only supplying 3.3V, via VIN_3V3. With this change, you can also go ahead and connect pins 2 and 3 of J1

    Per the PD Specification, you need 5V bus ready for any type of Source contract to be negotiated, therefore, if this is not there, the PD controller may not be sending the right I2C commands to enable OTG mode.

    2. You will also need to remove J3 completely and directly connect your BQ25731 VIN to PPHV of the TPS25750D EVM. This will ensure that the battery charger only connects to this power path, since SYS_PWR is also connected to other circuitry which could also cause issues and we do not want to connect SYS_PWR to PPHV in this scenario.

    3. For OTG mode, you can keep the OTG pin pulled high on the BQ25731 EVM because it will still require the I2C command to EN OTG before it is fully enabled. 

    Could you please test it out in this configuration?

    Thank you,

    Hari

  • Hi Hari,

    Thank you for the configuration breakdown! We will get working on this right away, only immediate question based on the image above:

    • Do we use I2C1 or I2C3 for the BQ25731 communication?

    Thanks,
    Nick

  • Hi Nick,

    You will use I2C3, which I what I believe you mentioned you are already using if I'm not mistaken. 

    I should also add when you flash the project configurations to the PD controller, you will need to power cycle the PD controller for the new configurations to load. You may already be doing this but just wanted to mention this so you don't have any power to the PD when you reset.

    Thank you,

    Hari

  • Hi Hari,

    So we got the setup almost working, we are running into a couple errors. 

    When we plug in our USB-C sink (which requests 20Volts) the initial PD negotiation seems to be working. But then for some reason the voltage goes to 5V and then a hard reset occurs and the VBUS voltage drops. I've attached the screen grab if that helps.

    Attached is another screen capture of a different USB-C sink device that has a similar problem to the last one (this one seems to auto reset though). Both of these devices work fine with other USB-C sources.

    The TPS25750 seems to be getting programed correctly because I can see the PD profiles I change in the GUI. I've also attached a screen shot and event table for the I2C bus during initial power on. I don't see data on the bus when I plug in the USB-C device though so I'm guessing something in initialization may not have worked?

    I'll send you a .csv of the i2c line over email.

    Best,
    Nick

  • Any update?

    Hari, I sent you the I2C capture data over email in parallel to this post.

  • Hi Nick,

    I looked at the I2C messages but it seems like there’s lines for addresses but many of the items do not show the data values. Are these the BQ25731 register addresses, I assume it’s not different I2C addresses correct? Also, I’m not sure if I pointed this out earlier, but you have to make sure the BQ25731 is powered on before the TPS25750. This way when the PD sends the commands over the battery charger can read and configure the settings as so.

    I have some questions though which could help narrow down the issue. Are you using the TPS25750 GUI? If so, could you send over the .pjt file, this will help me to make sure there aren't any settings that could be preventing sourcing power. Also, are you just trying to negotiate a 5V contract? I couldn't see which contract was requested by the port partner. 

    Thank you,

    Hari