• Locked Locked
  • Replies 2 replies
  • Subscribers 38 subscribers
  • Views 591 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.

Need help compensating my dc dc converter

Nick Wernicke
Intellectual 775 points

I am looking to make a simple step down converter with the T swift series.  I have been playing with the switcher pro tool, unfortunately I only understand control loops and feedback in terms of stable/unstable-- which doesn't seem to be a problem, I tried for 2 hours and still haven't been able to make an unstable one (as long as I am correct in assuming that unstable means a + gain margin).

It seems to me there should be an ideal transfer function for my dc/dc converter, but I don't know how to define what an "ideal" transfer function would look like beyond 50degrees phase margin and a negative gain margin.  I read a document that explained the K factor, which was defined from things like boost and transient response, but what is the required transient response or boost in a dc power supply? Its input is dc, its output is dc, as far as I can tell, other than power on, there is no time that any of that stuff matters.

  • TI__Mastermind 26205 points

    Nick,

    Sorry for taking so long to reply to you, we are addressing this and will return asap.

    regards

    rogerio

  • TI__Intellectual 1280 points

    hello Nick,


    Well, you have some great questions here, and I'm happy that you are exploring the world of DC - DC switchers.  I'll do my best to keep it simple.

    First off, to make a design unstable (or conditionally stable) you'll need to get a phase margin less than 30 degrees and or a gain margin more than -10db.   When phase margin is less than 30 degrees the supply will oscillate under certain transient conditions.  The gain Margin really is an indicator of the response time.  So if you have adequate phase but inadequate gain you won't oscillate but it might take twice as long to stabilize the output.

    Another good indicator of stability is the slope of the gain at the crossover point.  if the total gain is crossing at a 20db/ decade you're in good shape.  if it's crossing at 40db/decade you probably won't have enough phase there to give you adequate phase margin.  Phase drops off very quickly when gain is dropping at 40db/decade.

    You might check out the training modules on the analog elab design center page  http:www.ti.com/analogelab to get some more information about switching power stability.  You might also review some of the power seminars we've done at power.ti.com/seminars

    also look at application note SLVA057 Understanding Buck Power Stages.