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.

  • TI Thinks Resolved

TLC5944: Optimizing power consumption (VCC), with all LED's off

Part Number: TLC5944

Hi, I'd like some input on how to read the supply current specs for the TLC5944 led driver.

  • Can I assume that Icc3/Icc4 in the datasheet (picture below) is without any LED's "on"?
  • Why does Typ/Max increase together with Riref?
    • it seems to have a relation with the constant current setting
    • please clarify what can be done to reduce the current consumption unrelated to actual sink led-current.

 replied (in now locked thread):

Icc3/Icc4 is with LEDs all on of full greyscale while Icc1/Icc2 is without any LED's ON since BLANK is high.

Rief changes the Reference Current in IC and will cause Typ/Max's cahnge. You can refer to the function block diagram.

Follow-up questions:

- Icc3/Icc4, please clarify why the numbers are different as only the Riref resistor has changed (the numbers in the table I assume excludes the actual LED's)

- If we lower DCn (dot correction current register) to say half of the configured Riref, will this also lower the current consumption of the led-driver itself?

- As Icc1/Icc2 will show an increase in quiescent current when led's are off (blank is high), is there a way to reduce the quiescent purely in software? (i.e. I don't want to switch the supply off, nor trying to actively increase the resistance of Riref)

  • Hi,

    Riref changes the reference current of internal circuits, and the Icc changes accordingly. The Icc is just the current that IC consumes, excluding LEDs.
    Yes, lower DC will also result lower Icc.
    What do you mean by "Icc1/Icc2 will show an increase in quiescent current when led's are off (blank is high)"?
    As you can see, when OUTs(LEDs) are off, Icc is lower. And if you don't want to change BLANK, you can try higher duty cycle and lower Ioutmax(higher Riref) to achieve same average current. But if you use RGB LEDs, Ioutmax impacts the color temperature.

    Thanks.
    Kenneth
  • In reply to Kenneth Du:

    Hi Kenneth,

    • re: Icc1/Icc2, I'm basing this off the datasheet, I was expecting Blank to perhaps turn off the current-mirror to save power when the led-driver is not active, or is there an other way?

    please also answer:

    • If we lower DCn (dot correction current register) to say half of the configured Riref, will this also lower the current consumption of the led-driver itself?

    thx david

  • In reply to Dav01:

    Hi, David,

    As you can see from datasheet: "The counter is reset to zero when BLANK is high. The counter value is held at zero while BLANK is high, even if the GSCLK input is toggled high and low.", that's how the outputs are off.
    Lower DC will lower the current consumption of the led-driver itself.

    Thanks.
    Regards,
    Kenneth
  • In reply to Kenneth Du:

    Ok, thanks Kenneth,

    So, any other ways to reduce the standby current of this part (without cutting the power that is)!?

    i.e. something we can do via the SPI/Serial pins to reduce the unwanted 5-10mA Icc2 current...

    ps. our application is battery-powered by a Li-Po battery.

  • In reply to Dav01:

    Hi, David,

    Understand your requirement but there is no other good way to reduce the Icc current since that's the need of internal circuit.

    Thanks.

    Regards,

    Kenneth

  • In reply to Kenneth Du:

    Thanks for clarifying, so the answer to saving power is to lower all DC-registers to zero before stopping the clocks to the TLC5944.


    Can you please also help unlock and reply this thread:

    e2e.ti.com/.../544761
  • In reply to Dav01:

    Hi, David,

    Sorry but I failed to unlock that thread.

    If you want to disable LOD you can try TLC5945 or TLC59461.

    These are PIN compatible devices and won't turn LED off even LOD is detected.

    Thanks.
    Regards,
    Kenneth
  • In reply to Kenneth Du:

    Interesting, I didn't see these before,

    - are these new parts?

    - identical in function!?

    related:

    - for low-power applications, which led-driver from TI would you say is most suitable (still offering pwm + dot-correction, 16-24ch)

    thanks
    david

  • In reply to Dav01:

    Hi, David,

    These are not new parts and have similar function with TLC5944.

    TLC59461 has lowest Icc among these 3 devices.

    Thanks.
    Regards,
    Kenneth
  • In reply to Kenneth Du:

    perfect, sounds like we should switch from 5944 to 59461 then, still a bit confused what exactly are the differences of these two.

    does any of the TLC5x family come with 1.8V I/O option?

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.