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.

LM3429: Pulse mode driver

Part Number: LM3429
Other Parts Discussed in Thread: LM5022

I am looking for a boost converter to convert 24V (BAT) =>  80V to over-drive a LED array (Vf ~ 80V - If ~ 4.9A - rD = 3.56 Ohm) at pulse width of about 200us (f ~ 20Hz).

I have run some tests using LM3429, injecting the pulse at nDIM (pin-7) but I cannot get a good response time from the driver at 200us pulse width. It works pretty well down to 1-2ms range pulse width but seems slow for 200us pulse.

I am wondering what would you recommend for over-driving a LED array (Vf ~ 80V - If ~ 4.9A - rD = 3.56 Ohm - pulse width ~ 200us - f ~ 20Hz)?

#1 - Using the nDIM of an LED driver such as LM3429 (using PWM input for pulse mode operation)

#2 - Using a boost converter such as LM5022 at about 0.6A, charging a big cap and then using a switch to dump the big cap charge into giant LED. Probably this method needs a minimum load, true?

#3 - If I use a LED driver (i.e. LM3429) and keep the LED driver in continuous mode (always running nDIM=Hi) and control the big LED array (load) using an NMOS low side switch (100us, short pulse controls the NMOS gate) then does it make a difference to use a boost converter (i.e. LM5022) instead of LED driver (i.e. LM3429)?

In other words what is the difference between using a LED driver (i.e. LM3429) vs a boost (i.e.LM5022) about what I described above, would it be the same?

Please let me know your thoughts.


  • Hi Reza,

    May we ask what is this application for so we can understand some of the tradeoffs to better help you. The LM3429 is only specified for 75V max, so converting it to VLED of 80V with high side sensing exceed the rating of the part.

    In general, if you use a voltage boost converter to boost up the voltage and discharge it to an LED with a switch like a FET then the current is not well controlled. An LED driver like the LM3429 would control the current for PWM'ing but your requirement of 200uS width might be too narrow depending on switching frequency, output caps and inductor.

    Thanks Tuan
  • Hi Tuan,

    I need to make a big burst of light (200us pulse width) in sync with camera sensors to take high quality/high resolution images.

    This specific giant LED has an over-drive voltage of about  vf=80V and for the narrow pulse of 200us, I think I don't need to control the current, only will try to control the Vf and then LED will draw about less than 4.9A or so.

    But based on LED curve I believe 80V is going to push about 4.9A current thru the LED ad I'll get the light burst in the short pulse period. 

    What voltage boost would you recommend to charge the capacitors at about 80V? I see most of TI boosts are current mode not voltage. Your suggestions?



  • Hi Reza,

    There are voltage regulators that are voltage mode controlled and also current mode control.  The term current mode control here just means that there is an inner control loop that gets the current feedback from the inductor and swiitching transistor.  But they are in essense voltage regulator.  You can find more articles on to understand more.

    For your application where the current does not need to be well controlled in the PWM'ing fashion, it probably best to use a boost votlage regulator to charge up caps and use a FET switch to dump that energy into the LED.  It will need to be a controller with external FET switch and you can search at

    Thanks Tuan