• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 98 subscribers
  • Views 724 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.

OMAP-L138 Power-on sequence

Victor Kviat
Prodigy 100 points
Other Parts Discussed in Thread: OMAP-L138, TPS65070

Hi,

I am designing the board with OMAP-L138 and have a question related to the power-on sequencing of the processor.

Section 6.3.1 of OMAP-L138 datasheet defines the following power-on sequence:

  • 2a) All variable 1.3V - 1.0V core logic supplies (CVDD)
  • 2b) All static 1.2V logic supplies (RVDD, VDDA_12_PLL0, VDDA_12_PLL1, USB_CVDD , SATA_VDD).

At the same time it says that - "If voltage scaling is not used on the device, groups 2a) and 2b) can be controlled from the same power supply and powered up together."

The board is being designed for 375MHz processor without frequency scaling and I am going to use 1.2V for both groups 2a) and 2b).  At the same time the board should support future upgrade to 456MHz processor that requires 1.3V for the core and internal RAM.

This means that I need to have 3 groups of devices:

  • A) CVDD (powered from 1.2V today and from 1.3V in the future) - ex-2a) group
  • B) RVDD (powered from 1.2V today and from 1.3V in the future) - ex-2b) group
  • C) VDDA_PLL0, VDDA_PLL1, USB_CVDD, SATA_VDD (always powered from 1.2V) - ex-2b) group

What should be the power-on sequence in my case?

Do I need to have a special power-on sequence like A (core) -> B (RAM) -> C (peripherals)?

Or may I combine groups A and B and have the power-on sequence A+B (core+RAM) -> C (peripherals)?

Thanks.

  • 0
    TI__Guru 62975 points

    Victor,

    If you never plan on implementing voltage scaling, you can combine CVDD and RVDD together with the following:

    • (A) CVDD and (B) RVDD powered together to either 1.2V or 1.3V
    • (C) All other static 1.2V

    Note that you can still do frequency scaling (running slower) while leaving the voltage at the higher level.

    -Tommy

  • 0 in reply to tlee
    Prodigy 100 points

    Thanks, Tommy!

  • 0 in reply to tlee
    Expert 1410 points

    Dear Tommy:

    We use Omap-L138 in our avionics products. I need to power Omap-L138 using TPS65070. The Omap-L138 will run at 320MHz

    1.  I don't use RTC and will not use "Voltage scaling". According to DS SMOMAPL138-HiRel (SLVSQ9) - "the groups 2a and 2b can be controlled from the same PS and powered up together."

    2.  I will use DC-DC3 which starts first at 1.2V. I will not use VLDO2 = 1.2V. Is this OK?

    3.  The next according to DS must be "all static 1.8V IO supplies". I use only DVDD18 and DDR_DVDD18 - USB and SATA are not used. I will use DC-DC2 at 1.8V to power DVDD18 and DDR_DVDD18. I will not use VLDO1 = 1.8V. Is this configuration OK?

    4. The next according to DS must be "all analog 3.3V PHY supplies (USB0, USB1) which I don't use and "LVCMOS IO - DVDD3318_A, DVDD3318_B, DVDD3318_C which are all = 3.3V in my application. I use DC-DC1 at 3.3V with external enable (TPS65070-Pin14). The external EN_DCDC1 has voltage detector which rises to "high" after 1.8V on DC-DC2 stabilizes.

    5.  I use also DC-DC2 = 1.8V and DC-DC1 = 3.3V in the range of 50mA to power the external devices.

    6.  According to Omap-l138 DS - there is no violation in power sequencing. Is this correct?

    7.  According to TPS65070 DS - there are very moderate load on DC-DC3, DC-DC2 and DC-DC1. Do you agree?

    Will my system be powered up and work OK?

    I don't use LDO2 = 1.2V and LDO1 = 1.8V

    Thank you very much for your help,

    Boris Ruvinsky

    802-877-4978

    boris.ruvinsky@goodrich.com

     

  • 0 in reply to Boris Ruvinsky
    TI__Genius 14545 points

    Boris,

    Please start a new thread with your questions rather than posting it on the back of another thread. If you post on the back of another thread - we may miss your post. I just happen to see this one by chance.