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.

LM3409EVAL: How to lower response time to few nano seconds ?

Part Number: LM3409EVAL
Other Parts Discussed in Thread: LM3409

Hello,

I plan to use the LM3409 for our new fast drivers. With the eval board, I cannot go faster than 1µs of response time using the PWM input. In the datasheet, you have screenshots with extremely low response time (less than 100ns). Can you explain your complete setup and the modifications to do on the board if any ?

Thank you.

  • Hello Julien,

    Can you reference which screenshots you are referring to?  What are you driving the PWM input with?  What is the load besides the PWM pin (what other components are there)?

    Best Regards,

  • Hello,

    I refer to the picture below (figure 18 from datasheet). Vin = 24V. Vout = 15V. I out set to 1.5A.

    When I drive Q3 directly from signal generator, I have 3 µs of response time (from generator edge to 90% of photodiode signal). When I use Q2 or an external couple of MOS, I have 0.9 or 1 µs. I would like to save 200 to 300 ns more.

    Thank you,

    Julien

  • Hello Julien,

    Figure 18, as above, is external FET PWM dimming.  This is Shunt or Parallel FET dimming.  The speed of this has to do with the external circuit design and can be extremely fast.  It is shunting the LEDs with an external 'parallel' MOSFET.  If you are looking for extremely fast response time you shouldn't be using a photodiode (which I assume is an opto-coupler).  If you want to have fast current response you will not be able to use the PWM pin since that enables the inductor current.  it will be limited by V = L*(di/dt).

    Best Regards,

  • Hello Irwin,

    Yes, shunt dimming is what I want to use. The photodiode I was speaking is the equipment I use to monitor the load LEDs, it is not an optocoupler in the path of Q3 (the shunt MOSFET) but I didn't think about the delay it added. I measured the voltage at LED+, assuming I have no delay between voltage and current. Response time of my photodiode is about 200 ns. I improved the external circuit to drive Q3 and I saved 100 ns. So I reached my target.

    Thanks to have driven me with the photodiode delay !

    Julien