Sorry for typos. Here again: Good point. You might know that some journals (e.g. IEEE Transactions on Power Electronics) encourage authors to add a video as supplement to papers.
Hi professor, I understand that switched inductor model tried to include the inductor voltage in the model and your comment in TAES 92 claimed that it is essential in DCM, while Dr. Vorperian responded the average inductor voltage is always zero in one switching cycle, which also seems right to me. However later reported literature also claimed that average switch model will fail at high frequency in DCM, can you explain more why inductor dynamic is important to be included for high frequency prediction in DCM?
Its a long story. The main benefit of including the inductor is that you end up with one model for CCM and DCM and it moves smoothly from one to the other. AS for one cycle averaging, you may be interested in www.ee.bgu.ac.il/~pel/pdf-files/jour90.pdf
@@sambenyaakov Thanks professor, my question is well explained in your attached paper, It is very interesting to me. I hope these different averaging issues can be discussed in mainstream textbook so we don't take it granted.
Hi Sam. When you use the combined CCM/DCM model, does the DON equation stay the same for a current mode controller when in DCM mode? In this video you only modify Doff, Gb, Gc. However in your paper - www.ee.bgu.ac.il/~pel/pdf-files/jour92.pdf, you specify that for DCM mode, DON must change equations for a current mode controller.
In voltage mode the output of compensator is the duty cycle control . IN PCM the sensing is of the peak current and there is a need for translating it into the duty cycle command.
@@sambenyaakov Hi, thanks for the response. I understand the different translations from the compensator output to the duty cycle in CCM for peak current mode control. But I am asking about how DON is calculated in peak current mode when using the combined DCM and CCM model that you discuss in the video. Does the DON equation stay the same as on CCM?
Very useful for my works , Thanks . I have a question .Your technic permit to have a totally linear model so how can i obtain a transfer fonction of a converter using this method if it is possible ? We can take the exemple of a buck converter. Thanks
@@sambenyaakov I'm sorry. I just wanted to know how can i obtain the transfert function of the DC DC converter using your average model ? If possible. Thanks
Thank you very much for presenting this average model method. Your method combing the inductor with the diode an the switch and seems to be smoothly recognized CCM or DCM within one model. How do I apply this average model for Flyback , Push-Pull or Forward converter ?
Sorry,I have failed to notice your comment. In all the cases you mentioned you just scale the sources. E.g. forward is like Buck with a scaled current source to output and scaled reflected voltage source.
I suppose that the small-signal model equivalent for coupled inductors is a model of two reversed voltage source - inductor equivalent. (as that you use for inductor equivalent) ?
Dear Professor, I've had the opportunity to watch your presentation and I think that is very well explained. and very clear. I have a couple of questions: how can I close the loop (both voltage and current) and how can I match the average model with the slope compensation? Thank you in advance. Best wishes.
@@sambenyaakov Good afternoon Professor. I am not sure to have catched your answer. There is some link missing? It would be useful if you could indicate some paper or some example containing both the average model for the power cell and the control loop. Thanks a lot.
Thank you so much for the well presented video, it's the best one among others. Actually, I watched the video motivated by a question related to small-signal modelling of four-switch buck-boost converter (noninverting buck-boost). I have tried modelling it as a cascaded stages of buck followed by boost converters with the inductor placed in between, however the open-loop gain wasn't correct as expected. Would you help in let me know if this method of modelling explianed in the video is applicable to 4-switch topologies?!
Thanks. We have found extremely good matching between simulation and experimental results. The trick is to put in the parasitics. This may require someone with experience. However for the general behavior anyone can prepare simulation model which is in general excellent.
Yes, I can imagine the different parameters of components to be the trouble for less accurate results. Usually it's also the experimental components that do not comply 100% with the data, leading to a different result than expected, isn't it? Is that a big problem in commercial devices? So far, I've only used LTSpice for simulation, which I think is a very accurate programme, I just do not have enough experience to do the values exactly right as you said.
Thank you for a quick response. The paper is very helpful I must say. However, I am still finding challenges in modifying the paper to operate in closed loop boost mode. My actual controller design is as exactly as described here www.researchgate.net/publication/323649952_Control_of_a_ChargerDischarger_DCDC_Converter_with_Improved_Disturbance_Rejection_for_Bus_Regulation. I have achieved this in circuit based model which is very slow during simulation. Since I needed a fast simulation to achieve my objectives, I was wondering if you could send any lead/advice for a dynamic average model of a closed loop boost converter in MATLAB/Simulink or as described in the above link. I truly appreciate your kind respond to ezennayasamuel@gmail.com.
Dear professor, thank you for your video. I m trying to implement one cycle control of switching converter. Please professor, could you help me on that ?
Thanks a lot professor. I wish every paper I read had its explicative video, jajaja.
Good point. You might know that some journalc (e. IEEE on Power Electronics) encourage authors to add as supplement to paper.
Sorry for typos. Here again:
Good point. You might know that some journals (e.g. IEEE Transactions on Power Electronics) encourage authors to add a video as supplement to papers.
Hi professor, I understand that switched inductor model tried to include the inductor voltage in the model and your comment in TAES 92 claimed that it is essential in DCM, while Dr. Vorperian responded the average inductor voltage is always zero in one switching cycle, which also seems right to me. However later reported literature also claimed that average switch model will fail at high frequency in DCM, can you explain more why inductor dynamic is important to be included for high frequency prediction in DCM?
Its a long story. The main benefit of including the inductor is that you end up with one model for CCM and DCM and it moves smoothly from one to the other. AS for one cycle averaging, you may be interested in www.ee.bgu.ac.il/~pel/pdf-files/jour90.pdf
@@sambenyaakov Thanks professor, my question is well explained in your attached paper, It is very interesting to me. I hope these different averaging issues can be discussed in mainstream textbook so we don't take it granted.
Excellent video tutorial Doctor. Please whats the objective behind modeling in DC (steady state) and AC small and large signal analysis ?
Hi Sam. When you use the combined CCM/DCM model, does the DON equation stay the same for a current mode controller when in DCM mode? In this video you only modify Doff, Gb, Gc. However in your paper - www.ee.bgu.ac.il/~pel/pdf-files/jour92.pdf, you specify that for DCM mode, DON must change equations for a current mode controller.
In voltage mode the output of compensator is the duty cycle control . IN PCM the sensing is of the peak current and there is a need for translating it into the duty cycle command.
@@sambenyaakov Hi, thanks for the response. I understand the different translations from the compensator output to the duty cycle in CCM for peak current mode control. But I am asking about how DON is calculated in peak current mode when using the combined DCM and CCM model that you discuss in the video. Does the DON equation stay the same as on CCM?
Thanks professor for the video. Why did you change the sens of Ga into the buck model implementation ?
Please indicate minute of video you are referring to
@@sambenyaakov 11:02
Yes, are Gb & Gc arrows pointed in the right direction at 11:02, what am I missing? The previous Ga, Gb, Gc diagram 10:16 was of a boost?
At 18:30, relation for Doff , i found a sign change sir. the RHS part should be multiplied by -1. Please correct me if i am wrong sir. Thank you.
Can you please write the expression as you think it should be?
Very useful for my works , Thanks .
I have a question .Your technic permit to have a totally linear model so how can i obtain a transfer fonction of a converter using this method if it is possible ?
We can take the exemple of a buck converter.
Thanks
Nor sure I follow your question.
@@sambenyaakov I'm sorry. I just wanted to know how can i obtain the transfert function of the DC DC converter using your average model ?
If possible.
Thanks
Thank you very much for presenting this average model method. Your method combing the inductor with the diode an the switch and seems to be smoothly recognized CCM or DCM within one model. How do I apply this average model for Flyback , Push-Pull or Forward converter ?
Sorry,I have failed to notice your comment. In all the cases you mentioned you just scale the sources. E.g. forward is like Buck with a scaled current source to output and scaled reflected voltage source.
Thank you for your response
I suppose that the small-signal model equivalent for coupled inductors is a model of two reversed voltage source - inductor equivalent. (as that you use for inductor equivalent)
?
Thank. Have you seen ua-cam.com/video/qLoXOlrqZCU/v-deo.html ?
Dear Professor,
I've had the opportunity to watch your presentation and I think that is very well explained. and very clear.
I have a couple of questions: how can I close the loop (both voltage and current) and how can I match the average model with the slope compensation?
Thank you in advance.
Best wishes.
Perhaps this will be of Help
@@sambenyaakov Good afternoon Professor. I am not sure to have catched your answer. There is some link missing? It would be useful if you could indicate some paper or some example containing both the average model for the power cell and the control loop.
Thanks a lot.
Thank you so much for the well presented video, it's the best one among others. Actually, I watched the video motivated by a question related to small-signal modelling of four-switch buck-boost converter (noninverting buck-boost). I have tried modelling it as a cascaded stages of buck followed by boost converters with the inductor placed in between, however the open-loop gain wasn't correct as expected.
Would you help in let me know if this method of modelling explianed in the video is applicable to 4-switch topologies?!
Thanks for kind note and sharing. Follow the method of simulating a Flyback which is basically the same. Look for examples in my UA-cam channel.
@@sambenyaakov Thanks for the reply, I'll watch the flyback simulation video hopefully it addresses the doubts
Great video! Dear Professor, Where can I practice those techniques? Solve some problems?
Thanks. Try to locate recent power electronics textbooks.
@@sambenyaakov There are so many books. Can you give an example?
impressive, nice and thanks
Thanks
As always, very useful video! Thank you. I have one question, what is usually the error margin between simulation and real application?
Thanks. We have found extremely good matching between simulation and experimental results. The trick is to put in the parasitics. This may require someone with experience. However for the general behavior anyone can prepare simulation model which is in general excellent.
Yes, I can imagine the different parameters of components to be the trouble for less accurate results.
Usually it's also the experimental components that do not comply 100% with the data, leading to a different result than expected, isn't it? Is that a big problem in commercial devices?
So far, I've only used LTSpice for simulation, which I think is a very accurate programme, I just do not have enough experience to do the values exactly right as you said.
Hello professor, It is really interesting. Could you please provide some link to get to know about the average modeling of CLLC dc dc converter.
Average modelling is tricky. Best to use time domain simulation to get the transfer function
Thanks for the great video. I truly appreciate. Any hint on how to implement same in simulink/MATLAB? Thanks in advance
Yes no problem. Hereis how we did it with a somewhat older version of the average model www.ee.bgu.ac.il/~pel/pdf-files/conf39.pdf
Thank you for a quick response. The paper is very helpful I must say. However, I am still finding challenges in modifying the paper to operate in closed loop boost mode. My actual controller design is as exactly as described here
www.researchgate.net/publication/323649952_Control_of_a_ChargerDischarger_DCDC_Converter_with_Improved_Disturbance_Rejection_for_Bus_Regulation.
I have achieved this in circuit based model which is very slow during simulation. Since I needed a fast simulation to achieve my objectives, I was wondering if you could send any lead/advice for a dynamic average model of a closed loop boost converter in MATLAB/Simulink or as described in the above link. I truly appreciate your kind respond to ezennayasamuel@gmail.com.
Thanks professor for the video ! How I tell LTSpice to do an average simulation instead of a time step one ?
in the simulation window set to ac analysis
Thank you again
Thankyou very much
Awesome Video
Thanks for comment.
Dear professor, thank you for your video. I m trying to implement one cycle control of switching converter. Please professor, could you help me on that ?
Sir your video is very useful I request to you pl. Your video translate for hindi legwege