• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 64 subscribers
  • Views 778 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

LM3409 Low Duty Cycle Ringing

Dominic Leone
Prodigy 120 points
Other Parts Discussed in Thread: LM3409EVAL

I am using the LM3409EVAL board and have connected it up to a PWM for dimming control. It works except for low duty cycle.

As you can see at 5.9us ontime ~25% duty cycle i experience no ringing. the yellow/green line is the PWM at the EN pin. the blue/purple is drain of the FET at the cathode of the recirculating diode. At 25% duty cycle up to 100% i experience no ringing. as soon as i drop below 5.9us on time, i begin to experience ringing on the recirculating diode. 

What could be the cause of this? my output is running at 300kHz fromt he eval board. The dimming PWM is running at 42kHZ. the rise and fall times and well within the capabilities of this setup as they require on times of ns and i am in the us range. 

Is this a result of the dimming frequency being too fast and i should drop it to like 1kHz?

[Edit] Did more testing. Changed the PWM from 42kHz to 1kHz. Ringing seems to be constant under the circumstance of t_off > 18.5us. Is this an occurrence that just is inherit in the controller for t_off > 18.5us? would i be able to just add a cap and resistor filter in parallel with the reciprocating diode as a filter to reduce this effect? It seems to probably be the effect of the inductor and diode flipping as the inductor power drops below the Vf of the diode, so i would see this at a normal operation when i just turn the thing off at 18.5us i am assume and it has nothing to do with the switching PWM at all. 

  • 0
    Prodigy 120 points
    Did some more testing - tested at 100kHz, in hopes to switch faster then ringing can take effect if it is indeed related to off-time > 18.5us. now the ringing begins at 7.4us off time. which is about 25% duty cycle. which goes back to issue number 1 where it seems at below 25% duty cycle i get ringing.
  • 0 in reply to Dominic Leone
    TI__Mastermind 25500 points

    Hello Dominic,

    Is there any way you can measure the inductor current? If you haven't added an output capacitor it will be the same as the LED current. The likely cause is the inductor current falling to zero and it will ring due to parasitic inductances and capacitances in the LED load and the diode. That is normal operation. It looks like the switch remains on for the duration of the duty cycle also so you are probably not even reaching the peak current during the PWM/EN on time at such high PWM frequencies. That is fine if it works the way you like, it will just have a different profile for LED current versus duty cycle than you would get at much lower frequencies (it will be less linear versus duty cycle). This would also explain the shortening of time at the higher PWM frequency since the inductor will charge to a lower current with the shorter on time.

    Thanks,

    Clint

  • 0 in reply to Clinton Jensen
    Prodigy 120 points
    Hi Clint

    That was my assumption. Is there any way to reduce the effect? under final design i am going to switch to the 1kHz PWM/EN signal because the switcher can turn on and i get the same overall effect with output light conditions. I just feel better if the system is actually doing something, like switching, to control the output.
  • 0 in reply to Dominic Leone
    TI__Mastermind 25500 points

    Hi Dominic,

    The only real way to reduce it is to reduce parasitics (low parasitic cap diode and short leads to the LEDs). If you use some output capacitance it will change the profile (longer ring time but lower frequency). But no output cap tends to result in the least amount of ringing overall. It is normal with any inductive switcher and it will not hurt anything, you will also not see it in the LEDs. At 1kHz it will only happen at the turn off point and you will likely not even be able to see it in the waveform at that frequency.

    Best regards,

    Clint