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.

LMR14020-Q1: LMR14020-Q1

Part Number: LMR14020-Q1
Other Parts Discussed in Thread: TLC5941-Q1, , TPS92661-Q1, LM3489-Q1, TLC5941, LMR14020, LMR14020SEVM

Hello,

We are currently developping a LED driver for our customer ; final application is an automotive dashboard, planned for a mass production (about 500k / year).

We built an evaluation board made of :

- 5 RGB LEDs

- one LED driver : TLC5941-Q1

- one buck controller : LMR14020-Q1

- all automotive components

The board is supplied by :

- 5V power supply for LED driver (Vcc)

- 12V power supply for VIN buck controller (12V automotive battery)

The principle is that buck controller generates 5V at its output to supply the LEDs ("V_LED", on anode side), while the LED driver outputs connected at the LED cathodes allow to control LED current (dot correction + grey scale).

When "V_LED" is directly supplied by a 5V power supply, the LED driver controls precisely all the LED currents in all cases (whatever VBAT, dot correction or grey scale values are).

When "V_LED" is supplied by the buck controller output (set at 5V), in some cases LED currents are not correct because the buck output voltage falls below 3 or 4V ; this depends on VBAT/dot correction/grey scale values : sometimes its correct, sometimes its not. Currents in LEDs have a square wave shape, I suppose that buck controller has difficulty to control voltage during too high load transients.

Looking at other LED drivers datasheet, I read that "high-side sensing hysteretic buck current source" are preferred for LED supply because of their higher current control bandwith (example : TPS92661-Q1, §8.1 page 38), while LMR14020-Q1 is a "step-down converter".

Does that explain why voltage regulation is not correct in certain cases ?

Shall I better use for example the LM3489-Q1, which is an hysteretic buck controller ?

Or should I better :

- avoid using a buck controller which is not directly included in the LED driver as it is the case for TLC5941 ?

- ... and connect directly "V_LED" to VBAT or use a linear power supply ?

Final system requirements are :

- 15 led outputs used on TLC5941 driver (LED driver reference will not change)

- 20mA nominal current for each LED ; up to 40mA max

- system supplied from 12V battery

Thank you,

Best regards,

Julien

  • Hi Julien,

    The total load current is 15x40=600mA-max. LMR14020 output can handle this load transient current by proper circuit design.
    So it's 12V to 5V/600mA step down converison
    What's your prefer switching frequency of LMR14020?
    What's input 12V VBAT variation range ? 8V to 16V?
    What's your requirement on output undershoot at load transient?
    I can help to design application circuit based on above application requirements. you can apply one LMR14020SEVM and modify circuit for your application.

    B R
    Andy
  • Hi Andy,
    Thanks for your feedback.
    To answer your question :
    - I have no prefer switching frequency, something between 1M and 2MHz would be ok
    - 12V VBAT variation range : from 7 to 16V
    - I must be able to control my LED current even in worst case condition ; the VF max of my LED is around 3.7V ; so for output undershoot at load transient I would say something like +/-0.5V max

    To precise my problem : It does not always occur exactly when current goes from 50mA (for example) to 0A ; its like sometimes the buck controller stops regulating the 5V, then it restarts to regulate. As LM14020 is an asynchronous buck (no internal low side mosfet), is it possible the bootstrap capacitor it not charged when no load condition occurs (I_LED = 0A), leading to some kind of deactivation of LM14020 high-side mosfet ? Then when LED current is not null, it allows the bootstrap cap to charge and the buck restarts ?

    I dont have any oscilloscope screen to show you, but I can get one if you want.

    Best regards,
    Julien
  • Hi Julien,

    Pls send me your application circirt of LMR14020 and some waveforms like SW node, output riple if you have.

    The circuit should work abnormally as you mentioned the Vout may falls to 3V or 4V, noramlly the ripple below 5% at no load condition.

    Pls check actual voltage on LMR14020 VIN pin by Oscilloscope during operation, and check current limit setting of 12V power supply.

    Pls feel free to send email to me for more detail discussion.

    B R
    Andy
    andy.chen@ti.com