Video Tutorial, Analysis, Design, and Simulation of a flyback Part 7

[rbola35618]

Hi EEVbloggers,

Video Part 1
Here is tutorial video on the analysis, design and simulation of a flyback converter. I show a step by step in how to analyze the flyback converter by breaking into "TON" and "TOFF" circuits and then use the volt-second-rule to derive a transfer equation. Voltages and currents are calculated and will be compared to the results from SPICE simulations (coming soon in part 2).


Video Part 2.
Second video shows the simulation of the flyback. I compare the simulation with the calculated values.




Video Part 3
In this video, I introduced a second method and show how to model the flyback transformer. In part 2, I used the coupled inductor model. In part 3, I show how to use the ideal transformer model and show that both are equivament.


Video Part 4
In this video, I talked about the parasitic in a flyback converter and its effects on the supply. I show how to measure the parasitic and come up with a coupling coefficient which is then used in SPICE to model the leakage inductance. I then show how to snubber the effects of the leakage inductance by adding a snubber.





Video Part 5
In previous videos I have assumed ideal components and made assumption that efficiency is 100% and that the dutycyle of the PWM will be 50% in order to symplify the design and make the presentation easier to understand. In this video, I revisit the parasitic associated with non-ideal components and show how they effect the efficiency of the flyback. I also examine how to account for the PWM's minimum dutycyle when calculating the primary inductance.





Video Part 6
In this video, I show how to wind a simple flyback transformer, calculate the turns required to avoid saturating the transformer, I then measure the leakage inductance and calculate the coupling coefficient which I use in my SPICE model. I then measured the parasitic self resonance capacitance and also include it in my SPICE model.



Video Part 7
In part 7, I test the flyback transformer and compare the measurement results with simulation results. I show the breadboard techniques that I use that allows easy installation of the transformer and Mosfet (generally, the pins and long leads will give you more leakage inductance but because this is low power, I can get away with it). I present the ringing in the drain due to the parasitic leakage ringing with the mosfet's drain to source capacitance. And finally, I show how to damp the ringing by using a snubber.   



Comments and questions are welcomed.

Robert

[user_ivo]

I have been following your videos about the flyback and after that, a quick search i understood that, there are two different mode of operation, continuous and descontinuous, in the previous video of analysis of flyback this mode of operation is  take in account? or the flyback will operate in both modes ? 


Thank,
   
   Ivo.

[rbola35618]

Hi user ivo,

The example that I presented is for a discontinuous fly back. In the next video I will show that the current in the primary  (flux) actually goes to zero current. The is done by making sure the the transformers gets reset and setting the proper turn ratio. I am hoping to have the follow by this Thursday.

Thanks for the question.

Robert

[rbola35618]

Second video.

In this video, I briefly go over the design equations, values and then show how to simulate the flyback converter. I then compare the simulation results with the calculated values from my design example.

Questions and comments are welcomed.

Robert

[user_ivo]

  Hi,
   I tried simulate the flyback but using the LT Spice instead, and by mistake I loaded the output with 50 Ohm instead of 24 like  the video and I realize that the output value was bigger than 12V  , reducing  the load for 24 and i got the same result as the video. My questions is this behavior is normal, the transfer function Vo= N*(D/(1-D))*Vin only is valid if I  load the output in this case with 0.5A.



 Thanks,

    Ivo

[rbola35618]

Hi User_ivo

This is a very good questions.  Yes it is normal. It is called peak detecting. As the Rload resistance is increased, then the Vo=N*(D/(1-D) no longer applies. There is another equation that is in the Abraham Pressmen book that expresses Vo but in terms of Rload. The equation and will allow you predict the voltage. I will post the equation tomorrow if time permits.

A flyback is different in that it stores the energy and then releases it. But instead of the Vo being more  dependent on the equation (Vo=N*(D/(1-D)), it actually is more dependent or driven by the secondary current. It is a little complicated to explain here because there is a reset of the transformer that also comes into the picture.

Try running a simulation of a 24 ohms and then a 100 ohms Rload. Look at the voltage from Vin-drain.
During the time the mosfet is on, VLprimary = Vin-drain you will get Vin which is 24V. At Toff since the Vo for a 24 ohm load is and the Nr of the tranformer is 0.5 when have a reflected to the primary inductor VL=Vo/Nr which is -12V/(0.5) = -24V.  This satisfies the volts second rule. But now if you run a Rload of 100 ohms, Vo will increase to about 50V.  If you you now reflect this back to the primary, you should get Vo/Nr = 25/0.5 = -50 volts. Now the voltage second rule still applies. Let work this out, 

RL=100ohms
Vo=25V
Ton = 5us
Toff = I don't know at this time
VL_Ton = Vin = 24V
VL_reflected=Vo/Nr = 25V/Nr = -50v at is reflected back

VL-Ton*Ton = VLreflected*Toff,  Solve for Toff

Toff=2.4us,  This means that because you have a higher voltage refected back to the primary, the transformer gets reset which means that it released all of its energy sooner.
 
I will make another video where I will explain and show how the transformer get reset and also the effects on the output voltage

Stay tune for the next video.



Robert

[Marco]

A flyback is different in that it stores the energy and then releases it. But instead of the Vo being more  dependent on the equation (Vo=N*(D/(1-D)), it actually is more dependent or driven by the secondary current. It is a little complicated to explain here because there is a reset of the transformer that also comes into the picture.

I don't see the value in thinking about flybacks in this way, needlessly confusing ... rarely are you designing power supplies for known loads.

The basics of a discontinuous flyback transfers are simply that the energy stored in the primary inductance at cut off is transferred to the secondary, with secondary current starting at Np:Ns times the primary current at cut off and the primary and secondary acting as plain inductors at any other time. Output voltage is far less fundamental, best to not dwell on Vo calculations for known loads too much. You should be able to calculate them, but making a point of that early on distracts from the fundamental difference between normal transformer action and flyback. Better to just go straight to control loops.

[rbola35618]

Hi Marco,

I agree, it can be confusing. You are correct that a discontinous flyback strores the energy in its primary and then releases it thru the turn ratio thru its secondary. I also agree with the rest of your feedback, I appreciate your feeback because it indicate to give me what I need to explain better.

In my opinion, you need to know what is your maximum loads, input voltage range, ie minimum input voltage,and capacitance load so that you can compensate it properly. 

In the next video, I will explain it the way you mentioned and show the simulation. I like to show the effects of the leakage inductance. I am building this flyback (circuit, and winding my transformer) in my lab and I would like to compare calculated values, simulations and prototype measurements

Keep those comments-feedback coming


Robert

[rbola35618]

Video Part 3
In this video, I introduced a second method in how to model the flyback transformer. In part 2, I used the coupled inductor model. In part 3, I show how to used the ideal transformer model and show that both are equivament.



Comments and questions are welcomed.

Robert

[rbola35618]

Video Part 4
In this video, I talked about the parasitic in a flyback converter and its effects on the supply. I show how to measure the parasitic and come up with a coupling coefficient which is then used in SPICE to model the leakage inductance. I then show how to snubber the effects of the leakage inductance by adding a snubber.

[rbola35618]

In time I will. These video were originally intended for my fellow junior engineers as a way of mentoring them. After thr flyback, the next series will be a two swtich forward converter. After that soft switching techniques.

Thanks for the comments.

Robert

[rbola35618]

Video Part 5

In previous videos I have assumed ideal components and made assumption that efficiency is 100% and that the dutycyle of the PWM will be 50% in order to symplify the design and make the presentation easier to understand. In this video, I revisit the parasitic associated with non-ideal components and show how they effect the efficiency of the flyback. I also examine how to account for the PWM's minimum dutycyle when calculating the primary inductance.



Comments and questions are welcomed.

Robert

[rbola35618]

Video Part 6

In this video, I show how to wind a simple flyback transformer, calculate the turns required to avoid saturating the transformer, I then measure the leakage inductance and calculate the coupling coefficient which I use in my SPICE model. I then measured the parasitic self resonance capacitance and also include it in my SPICE model.

[user_ivo]

Hi Robert.

  Thank for the videos. Just a little comment,  in Part4,  in the formula to determinate the ringing frequency you forgot to include the sqrt. In the rest of the video you include the sqrt in the formula, so is easy for us to notice i guess, its just for  put  you aware of this, in the case you haven´t  notice yet .

 I 'm looking forward  for the next videos. 

[rbola35618]

Hi User_ivo

Thanks you for catching that error .  I guess I had a senior moment again 
 
I will try to see if I can make an annotation on the Youtube video. I'm lucky to have smart people like yourself checking my math. 

Hoping to get Part 7 out next week. I suffered a micro-stroke about 2 months ago that affected my motor skills, speech and balance. I am slowly recovering.

Again, many thanks for spotting the error and pointing it out to me!

Sincerely,
Robert Bolanos

[rbola35618]

Video Part 7
In part 7, I test the flyback transformer and compare the measurement results with simulation results. I show the breadboard techniques that I use that allows easy installation of the transformer and Mosfet. I present the ringing in the drain due to the parasitic leakage ringing with the mosfet's drain to source capacitance. And finally, I show how to damp the ringing by using a snubber.