• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 64 subscribers
  • Views 1016 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.

Strange: Vout of TPS62730 rises to 2.35V when CC2541 goes into PM2 mode. (ON/Bypass pin is fixed to 1)

FrankXu
Genius 3500 points
Other Parts Discussed in Thread: TPS62730, CC2541

I tried the SimpleBLEPeripheral example program in TI BLE1.4.0 stack. I change the P0_7 to P1_5 because I use P1_5 in my board.

In all the following experiments, HCI_EXT_MapPmIoPortCmd( HCI_EXT_PM_IO_PORT_P1, HCI_EXT_PM_IO_PORT_PIN5 ); is always disable, and I force P1_5=1; before osal_start_system(); which keeps TPS62730 always in DCDC mode.

And I am pretty sure the ON/Bypass is always 1 during the experiments.

1. With POWER_SAVING enable. 

    The Vout of TPS62730 should be stable 2.1V. But I get the following waveform.

      The Vin of TPS62730 is 3V, which is connect to a TEK power supply.

      The time interval is 1s, which is the ADVERTISING_INTERVAL.

      As can be seen in the figure, the Vout of TPS62730 rise to 2.35V when CC2541 goes to PM2 mode. That is quite strange.

2. With POWER_SAVING disable. 

    The Vout of TPS62730 is stable 2.1V. 

3. In the case of 1, if I connect Vout of TPS62730 to the PIN2 of CC-debugger(PIN2 is connected to 74AVC245 VCCB pin). The waveform is 2.1V stable.

4. In the case of 1, if I delete this line "ADCCFG |= HAL_ADC_CHANNEL_5;" in my program, which is executed before  P1_5=1. The waveform is 2.1V stable. I think maybe this change increase the Iout a little.

      I think in 2,3,4 shows that, the problem may have some relationship with Iout of TPS62730 on my board. It seems when the Iout is too small, the Vout rises to 2.35V. So I tried the 8 experiment.

5. I disconnect the Vout of TPS62730 to the power net on my board, so that Iout is nearly 0. (P1_5 is still forced to 1) I suppose to see the Vout be unstable. But the result is: 2.1V Vout is stable.

6. I tried 3 boards, they have the same phenomenon.

      I think there must be something wrong with my hardware. I've checked the schematic, I think the part of TPS62730 is the same as the KeyFob.

     And I've checked the value of caps and inductances, they are all right.

      According to the waveform in 4, does anybody have any idea on the cause of the problem?

Frank

  • 0
    TI__Guru** 106705 points
    Hi Frank,

    Could it be an issue with a light load (higher than 0, but lower than x uA)? Can you repeat what you did in point 5 above, but with a 1 Meg resistor on the regulator output?

    I see you have posted the same question in the DCDC forum. If you get any constuctive feedback there do let us know.

    Cheers,
    Fredrik
  • 0 in reply to Fredrik K
    Genius 3500 points

    Hi Fredrik,

            Thank you for your advice.

             I tried 1Meg resistor, but the regulator output is 2.1V stable. Then I measured the Iout of TPS62730, the Iout is 10.9uA. But I remember that when I use 3.3V to power CC2541 directly, it's 0.95uA.

            Finally, I found that when the R10 is removed, the 2.35V problem is solved. I made a mistake. I used to want to monitor the battery voltage(VIN) through P07, so I connect VIN with R10 to P07. But TPS62730 output 2.1V, so VIN is higher than VCC, which cause the problem.

            Although the 2.35V problem is solved. The original problem 

    https://e2e.ti.com/support/wireless_connectivity/f/538/t/440554 

    still exist. 

    3. With POWER_SAVING and HCI_EXT_MapPmIoPortCmd( HCI_EXT_PM_IO_PORT_P1, HCI_EXT_PM_IO_PORT_PIN5 ); both enable

        The connection is unstable, sometimes the connection can not be estlibalished, sometimes the connection keeps for several seconds.

        I used to think 3 maybe caused by the 2.35V problem. But now the 2.35Vproblem is solved, the phenomon still exists.

        The BLE connection is unstable. 

        While,

    1. With POWER_SAVING disable and HCI_EXT_MapPmIoPortCmd( HCI_EXT_PM_IO_PORT_P1, HCI_EXT_PM_IO_PORT_PIN5 ); enable

        The VCC is keeping 2.1V. The device can be connected and connection is stable.

    2. With POWER_SAVING enable and HCI_EXT_MapPmIoPortCmd( HCI_EXT_PM_IO_PORT_P1, HCI_EXT_PM_IO_PORT_PIN5 ); disable

        The VCC is keeping 3.3V. The device can be connected and connection is stable.

        Do you have any advice？

    Frank

  • 0 in reply to Fredrik K
    Genius 3500 points
    By the way, this is DCDC forum. So I deleted the same question in BLE forum.
  • 0 in reply to FrankXu
    TI__Guru** 106705 points
    Hi Frank,

    Good to hear that you resolved the 2.35 V issue. For battery monitoring you can put the regulator in bypass and then measure the supply voltage directly with the ADC.

    For the other problem, let us stick to the original thread.

    Cheers,
    Fredrik
  • 0 in reply to FrankXu
    TI__Expert 7110 points
    Frank,
    I went through your info's above. From what I understood, the issue from power prospective is that TPS62730 regulates to 2.35V instead of 2.1V, right? You posted, however, that this problem was solved by software?
    Can you quickly summarize the DC/DC issues you face? Can you pleas post your power layout as well?
  • 0 in reply to Florian Feckl
    Genius 3500 points
    Yes, the problem is the Vout of TPS62730 is 2.35V when the CC2541 get into PM2 mode.
    I made a mistake in the schematic, I connect VIN which is 3V to R10. That makes P07 has a higer voltage than Vout of TPS62730, and that cause the 2.35V problem.
    I connect Vout of TPS62730 to R10, and then the problem is solved.