• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 14 subscribers
  • Views 1109 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.

LM98640QML-SP: Power up time

Bradley Harden
Prodigy 160 points
Part Number: LM98640QML-SP


Hi,

The LM98640QML-SP datasheet indicates that the part can be powered down, in whole or in part, by setting a register value. However, it does not indicate how long it takes for the part to return to full operation once it has been powered back up. We are considering the part for use in a low power imaging system. To save power, we would like to power down the part during the image integration time and power it back up when reading the image. Would that be possible? Does TI have any data on the time needed to power up the entire system and/or the various, individual components?

Thank you,

Bradley

  • 0
    TI__Genius 17125 points

    LM98640 POWERDOWN RECOVERY TIME.pdfHello Bradley

    The LM98640 was specifically designed for low power applications and the powerdown features were added for addiational power savings.

    See attached for typical recovery time from power up for the various functions.

  • 0 in reply to Kirby Kruckmeyer96848
    Prodigy 215 points
    Hello,

    I am a bit confused by some data in this file, especially the first and last picture.
    I am guessing the INCLK has been running for awhile and that the shown SCLK is the data transfer to bring the device to life.
    So I have the following questions:
    - How long before (or how many periods) do we need to apply the INCLK for the SPI communication and the device to wake up and to work properly?
    - On the first graph the cursors are giving a deltaT of 30 µs but the text above gives a delay of 800 µs, so why is there a difference in values?
    - The last picture is showing two signals, one being SCLK but what is the other signal and how is it linked to the REF buffer?
    Regards.
  • 0 in reply to PRrrr
    TI__Genius 17125 points
    To wake up the part from power down, you just need to write Powerdown register.
    The time it takes for the part to wake up should be measured from when SENb goes high after the Powerdown register information has been clocked in.
    The person who prepared this file is no longer with TI. I am not sure why he included the plots showing SCKL.
  • 0 in reply to Kirby Kruckmeyer96848
    Prodigy 215 points
    Thank you.

    Would you have an answer for the first question which was:
    - How long before (or how many periods) do we need to apply the INCLK before any SPI communication for this communication and the device to work properly?
    Regards.
  • 0 in reply to PRrrr
    TI__Genius 17125 points

    I am not sure why the datasheet states that SCLK frequency must be less than the INCLK frequency. I was able to write and read the registers with no INCLK.
    With an INCLK signal, I put the part in Powerdown mode. I shut off the signal to the INCLK and then wrote the registers to take it out of Powerdown mode. I turend on the INCLK source and the part operated in the full powered up mode.

    UPDATE Sept 10, 2018:

    I stated that I was able to read and write to the registers with no INCLK signal present.   When I did this, SCLK was at 41 kHz.   A customer reported that when SCLK was at 1 MHz, he needed an INCLK signal to communicate with the part, as stated in the datasheet.

  • 0 in reply to Kirby Kruckmeyer96848
    Prodigy 215 points
    The fact that INCLK is present means that we are sampling the input data and using power, something we might not want when starting the device and configuring it. Perhaps some commands need to have the INCLK to actually have a visible effect but the configuration should be independent.
    I am not sure you have tested in your testing above the device start-up without INCLK and then configuring the device with the SPI to see if it samples correctly when applying the first INCLK periods. I am guessing the DLL needs a number of INCLK periods to correctly stabilize. Before that we also need to configure it properly, perhaps reset it too and check whether it is in False Lock Detect state. Maybe that's the catch..... I will create a new thread about how to properly start and check the DLL state and reset.

    I guess it is worthwhile to check why the datasheet is stating the necessity of SCLK <= INCLK.

    Regards.