• State TI Thinks Resolved
  • Locked Locked
  • Replies 14 replies
  • Answers 2 answers
  • Subscribers 64 subscribers
  • Views 742 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.

TPS2546: Is the behaviour of TPS2546 normal?

Nic Lin
Intellectual 2020 points
Part Number: TPS2546

Team,

Customer is testing TPS2546 in notebook system and got below waveforms. 

1. When system in S0, CTRL1/2/3=111; When system in S5, CTRL1/2/3=011.

2. When system in S0, drive current 2.1A，there is no VBUS drop. 

3. When system shut down, change from S0 to S5, CTRL1 change from 1 to 0, VBUS will drop to 0 and then back to 5V, is it normal?

4. When system in S5, drive current 2.1A, VBUS will drop to 0 and then back to 5V as well, is it normal?

Could you let me know where I can find the explanation in datasheet? Thanks.

Note: in below waveform, 抽载means drive current 2.1A, 关机means system shut down (change to S5)

  • 0
    TI__Intellectual 2020 points

    Team, could anyone help on this? Thanks. 

  • 0 in reply to Nic Lin
    TI__Mastermind 32641 points
    3. When system shut down, change from S0 to S5, CTRL1 change from 1 to 0, VBUS will drop to 0 and then back to 5V, is it normal?

    This is normal behavior because when the change of mode, the bus is discharged and then the BC1.2 state is renegotiated.


    4. When system in S5, drive current 2.1A, VBUS will drop to 0 and then back to 5V as well, is it normal?

    This is not necessarily normal, but I need to know what it happening on the D+/D- bus to debug what is happening.

    Please ask your customer if they have the D+/D- bus captures available. There is a negotiation that happens between the device and the TUSB2546 that needs to be observed to describe the behavior.
  • 0 in reply to Chuck Branch
    TI__Intellectual 1350 points

    Hi Chuck, 

     

                I’m FAE of Lenovo team in Beijing. Currently we meet some problem tuning TPS2546. It appears that the OUT voltage drops and raise back when add a 2.1A load, as the waveforms below shows.

                     I have check with customer and get the following information:

    1. S0 in the waveform corresponds to CTL1, CTL2, CTL3, ILIM_SEL of 1111, while S5 in the waveform corresponds to 0111(DCP_Auto in datasheet)
    2. There is only DCDC converter as the 5V source so they don’t transfer from LDO to DCDC as the datasheet claims. And they confirm that the EN pin keeps high during the whole period. (Which in my opinion, indicates that there should not be a 2s drop as the holding time in the waveform)
    3. The ILIM_HI resistor is selected to be 22.1kohm and ILIM_LO resistor is selected to be 2.7Mohm.

     

    Is the waveform satisfy with the design? Could you please help to explain the process?

     

                 And if not, could you please give some comments to help solve this issue?

     

                 Thanks for your help.

  • 0 in reply to Max Guan
    TI__Intellectual 1350 points
    It is the same issue that Nic proposed and please check the detailed information. Look forward to your comments.
  • 0 in reply to Chuck Branch
    TI__Intellectual 1350 points

    Hi team,

              Anyone can help on this?

  • 0 in reply to Max Guan
    TI__Mastermind 32641 points
    Max,

    Here is the expected discharge behavior:

    Any of the following charges initiates a discharge:

    SDP1 to DCP_Auto
    SDP2 to DCP_Auto
    CDP to DCP_Auto
    DCP_Auto to SDP1
    DCP_Auto to SDP2
    DCP_Auto to CDP


    no Discharge:

    CDP to SDP2
    SDP2 to CDP



    Based on these rules, I see that all of the discharges in your scope captures are correct.
  • 0 in reply to Chuck Branch
    TI__Intellectual 1350 points

    Hi Chuck,

          Thanks for your information. But I still get confused on some parts. 

          As the table below from datasheet shows, S0(1111) in the waveform is the CDP mode and S5(0111) is the DCP_Auto mode. So the first discharge happens when switching from S0 to S5. But the second and third discharges both happen within DCP_Auto mode without mode switching. And as I claimed in previous message, the customer are not using LDO here (just DCDC) and the enable pin keeps high during the process, so I expect there should be no mode change for these two discarges in S5. Could you please explain more detailedly about the reasons for these discharges?

          Thanks for your help.

  • 0 in reply to Max Guan
    TI__Mastermind 32641 points

    The TPS2546 will definitely cause one of the discharges when the mode change occurs, but I cannot say why the other two are happening without seeing the D+/D- lines.

    Both sides of the connection can cause a discharge to occur.

    Is there any way to gain access to a capture of the D+/D- lines when this event is occuring?

  • 0 in reply to Chuck Branch
    TI__Intellectual 1350 points

    Hi Chuck,

           A few more tests are done to provide more evidence for this phenomenon. The following is the loading test under S5(CTL1-3, ILIM: 0111).It appears that the discharges happen sometimes  after loading or released for 2s. And regularly, the discharge happens if the loading period is short.

     1

    Also here provide the D+ D- wave forms when discharges happens under S5. 

     2

    Could you please help to explain the reason for each discharge based on these waveforms? Thanks very much.

    Down here are the rising edge and falling edge captures for the discharge in case they are helpful.

    34

  • 0 in reply to Max Guan
    TI__Mastermind 32641 points
    The D- state changes that I see in this diagram is not caused by the TPS2546. It appears to me that a change in D+/D- is causing the state change and the discharge.

    I suggest that the the customer look into why the connected device is changing state on the D+/D- lines.
  • 0 in reply to Chuck Branch
    TI__Intellectual 1350 points
    Chuck,

    My apology that didn't explain it clear in the very beginning. Actually the customer did the loading test by directly adding 2.1A load between Vbus and GND and left open the D+/D- wires. It means that no load detection and communication on D+D- is finished before adding a 2.1A load on Vbus. Will this influence the mode detection and discharge of Vbus?

    And under this condition that D+/D- load detection and communication are not finished and directly adding load on Vbus, could you help to figure out the reason of the discharges under S5 in the waveform?

    Thanks for your help.
  • 0 in reply to Chuck Branch
    TI__Intellectual 1350 points
    Hi Chuck,
    Any comments on this?
  • 0 in reply to Max Guan
    TI__Mastermind 32641 points
    This looks very similar to another ticket.

    It appears that the device is doing a primary and secondary detection to differentiated between SDP and CDP. The first detection fails, then the device discharges the bus and the attempts a second bus detection with 1.2V short mode and this also fails and the device ends up in SDP mode at the end of the cycle.

    I cannot be sure because I cannot see the D+/D- lines.

    If this does turn out to be a redudant thread, then please close one as resolved and I will track the other.

    Regard,s
    Chuck
  • 0 in reply to Chuck Branch
    TI__Mastermind 32641 points
    Max,

    I believe that this is a redundant thread to 753035. I am going to mark this as TI Thinks resolved. Please close out this thread if you agree.