• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 100 subscribers
  • Views 127 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

AM62L: SPI Voltages

Blake Carpenter
Expert 6865 points
Part Number: AM62L

Tool/software:

Hi Support, 

The Assigned Focus customer asks:

In the table below, we have note (1) next to the CAP_VDDSHV_MMC signal name. I’m having trouble understanding what the note is trying to say.

Specifically, where is says “when the SDIO_LDO is being used to source VDDSHV3 or VDDSHV4”.

For example, if I am using MMC2 and I want to use it at 3.3V, then I would supply the VDDSHV4 pin with 3.3V.

If I want to use MMC2 at 1.8V, then I would need to supply the VDDSHV4 pin with 1.8V.

What SDIO_LDO is the datasheet talking about? Is there a way to use MMC2 with out supplying the voltage to the VDDSHV4 pin using an external supply?

Best Regards, 

Blake

  • 0
    TI__Guru 62215 points

    The AM62Lx device has an internal LDO circuit called SDIO_LDO that is controlled by MMCSD1 host controller. SDIO_LDO is used to source VDDSHV3 when connecting an UHS-I SD Card to the MMC1 port, where the initial IO operating voltage of an UHS-I SD Card is 1.8V and changes to 3.3V when switching to one of the higher speed modes. This internal SDIO_LDO allows the MMCSD1 host controller to change the VDDSHV3 operating voltage associated with the MMC1 signals at the same time it tells the SD Card to change its operating voltage. However, the system designed must connect the internal SDIO_LDO output pin named "CAP_VDDSHV_MMC" to the appropriate decoupling capacitor and the VDDSHV3 power pins.

    I need to change the note and remove the reference to VDDSHV4. When creating the note, I assumed the internal SDIO_LDO could be controlled by MMC1 or MMC2 similar to how it was done in AM65x. However, I recently confirmed the internal SDIO_LDO implemented in AM62Lx is only controlled by MMCSD1 host controller. So, there would never be a case where "CAP_VDDSHV_MMC" is connected to the VDDSHV4 power pins.

    Note: The SDIO_LDO input pin (VDDA_3P3_SDIO) should be connected to same 3.3V power rail that is used to power the SD Card, and that 3.3V power rail needs to have a power switch in the path so the SD Card power and SDIO_LDO power can be turned on/off at the same time when needing to reset the SD Card. The switch needs to be controlled by the AND function of power-on reset, warm-reset, and a GPIO so the SD Card can be power cycled anytime it needs to be reset because cycling power to the SD Card is the only way for it to be reset. The VDDSHV3 power rail needs to cycle on/off at the same time as the SD Card since the IOs associated with both devices are not fail-safe. This is why the SDIO_LDO that powers VDDSHV3 need to be connected to the same 3.3V power rail that sources the SD Card.

    Regards,
    Paul

  • 0 in reply to peaves
    TI__Expert 6865 points

    Hi Paul, 

    Thanks!  

    Two clarification questions:

    1) Understood, but please clarify. Are you saying that the SDIO_LDO output voltage can be changed by the MMCSD1 host controller from 1.8V to 3.3V and from 3.3V to 1.8V?

    2) Understood. However, then why is the design such that two of the MMCSD1 host controller’s interface pins are under the VDDSHV1 domain (MMC1_SDCD and MMC1_SDWP)? Any change in voltage from the SDIO_LDO when it’s connected to VDDSHV3 will not be reflected on these two input pins. Please advise

    Best Regards, 

    Blake

  • 0 in reply to Blake Carpenter
    TI__Guru 62215 points

    1) Yes, there is an internal signal sourced by the MMCSD1 host controller that is connected to the SDIO_LDO, where it controls the output voltage of SDIO_LDO.

    SDIO_LDO will pass its 3.3V input to its output by default. Its output will be regulated down to 1.8V at the same time the MMCSD1 host controller sends a command to SD Card to changes its IO voltage from 3.3V to 1.8V. The SD Card has a similar IO voltage control circuit. Once the SD Card is placed in 1.8V IO mode, the only way for it to return to 3.3V IO mode is cycle the card's 3.3V power source. Power cycle is required since the SD Card doesn't have a reset pin. It is important that the VDDSHV3 power rail is power cycled at the same time as the SD Card power is cycled. Therefore, the SDIO_LDO 3.3V input and the SD Card 3.3V power needs to be sourced via a common 3.3V power switch that is controlled by the AND function of power-up reset, warm reset, and a GPIO. This power switch and AND function allows the SD Card power and the SDIO_LDO to be cycled for any of these three inputs, which will reset the card and return it to 3.3V IO mode.

    2) The SDCD and SDWP signals only connect to the card detect and write protect switches built into in the SD Card connector. These signals are pulled up to VDDSHV1 with external pull resistors and the card detect and write protect switches will connect these signals to VSS to indicate a card is detected or write protect is set. We purposefully put these signals in an always-on IO voltage domain because the MMCSD1 host controller needs to know the state of these signal even when the VDDSHV3 power rail is off or changing voltage.

    Regards,
    Paul

  • +1 in reply to peaves
    TI__Guru 62215 points

    I just realized I had a mistake in my initial reply. I accidently swapped the voltages in the statement describing the SD Card IO voltages. The statement should have said "the initial IO operating voltage of an UHS-I SD Card is 3.3V and changes to 1.8V when switching to one of the higher speed modes."

    Regards,
    Paul

  • 0 in reply to peaves
    TI__Expert 6865 points

    Got it and passed along the data.  Thanks Much!

    Blake