• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 62 subscribers
  • Views 374 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

LM3481: Minimum duty cycle operation (<1% duty)

Albin Vinu30
Prodigy 210 points
Part Number: LM3481


I am using LM3481 for flyback operation. In my application, there is a chance of no-load at 32V input. What will be the effect at this point of time ? 

LM3481 operation: Freq: 140 kHz (7.14us); Tmin as per datasheet : 250ns -> Duty: 3.5%

Requirement : Low load and Max. input voltage - 32V -> Duty: ~1% or less

During this light load at max. input voltage duty goes much lower than min. on time. What will be effect at that time ? Will output voltage shoot or even with 1% demand from system, controller will operate at 3.5% duty or will it go for pulse skipping mode ? 

Thanks,

Albin

  • 0
    TI__Guru* 78665 points
    Hi Albin,

    Thank you for using the LM3481. In your case, the circuit may skip the switching cycles because the min duty is greater than needed and you may see slightly increased output ripple voltages owing to the min duty issue.

    Also, the 250ns is at the driver output unloaded. In real circuit the effective duty that is what you really care about would be smaller than your calculated 3.5%.

    If you really don;t like the slightly increased output ripple voltage at no load, the simplest way is to add a dummy load. Yes it appears to be a permanent loss but that loss may not be prohibitive in reality.

    Thanks,
    Youhao Xi, Applications Engineering
  • 0 in reply to Youhao Xi
    Prodigy 210 points

    Hi Youhao,

    Thank you for addressing fast.

    Will i be able to calculate actual min. duty by considering 250ns ? Is there a formula or something ?

    Also, will i be able to estimate the increase in ripple voltage ? Is there a way ?

    Regarding min. load -> Already there is a minimum load resistor. Putting load resistor  to meet the min. duty cycle (<1%) is actually increasing the bleeding current and reaches almost equal to my load current as my actual load current is not huge. So, not really interested in increasing the fixed load heavily.

    Thanks,
    Albin

  • 0 in reply to Albin Vinu30
    TI__Guru* 78665 points
    Hi Albin,

    You need to check your MOSFET gate voltage pulse waveform, and estimate the effective duty, namely the pulse with at about the Vth level.

    The Vout ripple also depends on your loop response time. Note that at no load the loop gain drops a lot. I suggest you to build the board and get a real test. It may not be as bad as you expected.

    Thanks,
    Youhao