Loop compensation of a multiple output flyback part 4

[rbola35618]

Here is part 4 in the series of loop compensation and using Jack Alexander AC model.


Part 4, Loop compensation of a multiple output flyback



Part 1, Introduction of Jack Alexander's model



Part 2 Verification of Jack's model


Part 3 Simulation of a Voltage Mode flyback using Jack's model

[user_ivo]

Hi Robert,

  Bringing back this topic , I have been following your videos about the flyback and already put this in practice building a power supply 28V input and 5V output, I got good results in line with  what you already have shown us  . Now I'm wondering to build a more complex power supply  with maybe 3 output and use one of the output to provide bias and regulate power (feedback) to the PWM IC ( resuming 2 isolated output  in 1 not isolated(bias and feedback)). My question is and taking in account of what was told about compensate the multioutput, should I reflect the capacitances and the loads to this output used to provide bias and feedback voltage, or  since this output is isolated from the others two consider only his load and capacitance. I will not be able to measure the gain a phase of the power supply and check which will be the best approach to compensate the loop   .
Best Regards 
  Ivo

[rbola35618]

Hi User_ivo

I posted video 8 where I show actual compensation performed. Since my last video, I bought the Cleverscope with has the capabilities of a FRA (Frequency response Analyzer). I show that Jack's simple model works and was able to get almost the same results from SPICE and the Cleverscope.



I agree the next progression would be to add a couple of extra windings and show how to setup the model and then confirm the simulations with the Cleverscope.

I have been a little busy in tutoring my son in MOSFET small signal analysis and been posting video for him to watch. He learns by repetition; I do too. As soon as he takes his next exam, I should have more time to wind another transformer.

By the way. I am planning to put another series of videos on how to design a two switch forward converter. I am working on the proto at this moment. Is this something you think people might be interested?   

Jack's model also works for a forward converter.


Robert Bolanos
rbola35618@aol.com
"May your simulations always converge"

[user_ivo]

Hi Robert,

Thank by the reply, actually I haven´t seen this video, yet, it´s good match between simulation and measurement. Regarding the forward with two switch, for people like me that have interest in this field, is a totally different topology and of course will be interesting .
I'm looking forward for the next videos and I wish good luck for your son but with a good mentor will be easy  task  
 

[rbola35618]

User_ivo,

If you don't have an FRA, and if the ripple in the output of the flyback is low enough, you might be able to use an audio transformer to inject a signal with a function generator and at least verify the crossover frequency and the phase margin.

Here is a video showing how to perform the measurement.



Robert
rbola35618@aol.com

[rbola35618]

User_ivo,

Here is a picture of the simple forward that I will use to do the forward converter videos as soon as I get a chance. I am using the me-square, I think they are called that. I think this is called Manhattan style prototype. This only shows the power stage. I need to use a PWM and close the loop.

I will probably start the forward video series in about a week.

Robert