Deciphering Flyback converters active clamps

An intuitive explanation of the operation of Flyback converters active clamps.

[@topphemlig1191]

Thank you professor, I really appreciate the effort you make in these videos. The material really helps my understanding of these topics. Sometimes even having used these techniques for years it helps to calibrate the mind every now and again to keep a sharp understanding. Thank you

[@sambenyaakov]

Thanks for comment. Yes indeed. I have found that even highly experienced engineers fall into a routine and may loose the understanding of the gist of matters.

[@longwang3271]

Thanks so much, at the first line, the drawings of the current/voltage curves scared me a little. But with your explanation, everything can be understood properly. Really nice work.

[@sambenyaakov]

👍😊

[@amineoufella4953]

You are an amazing teacher. Thank you for your time, for your patience for your teaching methodology. I appreciate all your videos. I wanna be like you. Cordially, K.A

[@sambenyaakov]

Thanks Amine for warm words. Comments like yours keeps me going.

[@sadunnakipoglu9471]

شرح مفصل وجميل تحياتي لك من تركيا.

[@sambenyaakov]

Many thanks

[@RazorRadios]

Excellent video, very much enjoyed watching this!

[@sambenyaakov]

Thanks

[@slip1959]

Very well explained video, thank you very much for sharing your knoledge. Now i think i know how to address a problem.

[@sambenyaakov]

Thanks for note.

[@xuexi1137]

To understand what I learned from your video, I built a flyback converter based on your teaching. I tried out different sizes of inductors using a toroid ring, with various ON time and frequency. The initial voltage waveforms at the drain of the switching NMOS looked similar to your video at 7:14. but the ringing after the collapse of the magnetic flux was quite different. The voltage swing was much smaller, with high frequency ringing and fast died down. I reduced the period such that the NMOS turned on before all the magnetic fluxes collapsed. The waveform at the NMOS drain changed to a square wave. At this state, It delivered more power to the load. The conversion efficiency was higher than 80% which was better than any other setups which waited till the complete collapse of the magnetic field.Few of my NMOSs blew up due to the inductor gone into magnetic saturation with run away current. To solve this issue, I varied the on pulse width and monitored the drain voltage until the drain voltage started bending upward, Then I set the maximum ON pulse width to 90% and added a fuse in series of the inductor. I ran my converter for a week, It worked fine and the fuse was not blown.I was using an Arduino nano as the controller and ran the switching frequency at 250KHz. Is there a reason why I shouldn't turn on the NMOS before the inductor magnetic field is totally gone ? Put it in other words. Why Is it not a good practise to operate an inductor with the magnetic field not dropped to zero? The reason I prefer this way is because the toroid size can be smaller.I am not worried much about the runaway effect now. My reason is if there is residue magnetic flux in the inductor. The residue magnetic flux will accumulate every period and more magnetic energy is accumulated every cycle. During the NMOS off period, more power will be pumped out from the secondary winding and reducing the residue flux. At a certain point, the residue flux will be at equilibrium. Is there a problem with my reasoning? Your lecture is clear and precise. I learned a lot from it. Thanks very much and keep up the good work. Xue.

[@sambenyaakov]

Thanks for sharing

[@sanjayagrawal6143]

Sir, Thanks for your presentation. It is very intuitive. Best Regards

[@sambenyaakov]

😊🙏

[@sanjayagrawal6143]

@@sambenyaakov Sir, Is it possible to get you whatsapp no.. Warm Regards

[@petegaslondon]

This is great stuff, I never quite got these before Gotta say I was getting impatient, saying come ON show me the Clamps! When I got there I had to play it twice before I got it though ;) Nice exposition !

[@sambenyaakov]

Thanks for comment.

[@petegaslondon]

@@sambenyaakov Oh Hi Professor - nice that you're there. 'course the reason i never understood these was the same reason I never really trusted active rectifiers - it always troubled me that they could go 'out of whack' and snatch current back from the reservoir THIS circuit RELIES on snatch-back, which of course, you can't get with a diode - though I found the initial exposition a bit long, I really need the latter part - I now get it, so thank you! ... Oh and if you're still there, i have a question - many years ago before mosfets were commonplace, I'd read articles about advanced high efficiency 'Glass Ferrites' I think they were called, that were supposed to usher in tiny highly efficient megahertz range SMPS - these never really materialised (or maybe they did - in niche applications?) Perhaps the HF switching losses in the silicon lost the efficiency gain? Have you ever heard of these? And how far HAVE people managed to push the smallest size/highest frequency envelope? (Perhaps by tolerating greater 'iron loss' ? I've seen at least one design with the core heatsinked, so I was wondering about the tradeoffs. Presumably GaN's now raise the frequency one can use) Meanwhile I'll be playing a bunch more of yours tonight - I know my basic topologies, and can usually manage to fix the things, but there's quite a few subtleties I probably miss! Yours from Lockdown, Pete & William StereoCat Gas Electronics London

[@jackywang1717]

very valuable analysis! thanks very much! and take care of yourself!

[@sambenyaakov]

😊

[@aqteaching]

Thank you, Professor. Your lectures about flyback are excellent. What could be some good material to learn more about designing this snubber for an unfolding flyback? Once again, thank you.

[@MahdiMotahari-o9n]

Hello Professor. Can we use an active clamp circuit in the secondary side of the fly-back converter?

[@sambenyaakov]

No. This will not protect the transistor at the primary.

[@dinhbkvn]

Very nice lecture. Thank you very much.

[@sambenyaakov]

Thanks

[@jcoejfjfnathan]

Hi Professor, I wanted to ask what controller IC's are you referring to? Thanks!

[@sambenyaakov]

The presentation is discussing the generic approach. There a number of ICs on the market.

[@siamak1246]

Thanks for nice topic and detailed explanations

[@sambenyaakov]

Thanks for comment

[@rakeshpanda2116]

Hello Professor, thank you for the video series. I have gone through lot of your video and they are very helpful. But unfortunately the explanation for ACF is not well understood. Infact its not difficult as it seems from this explanation. Will be helpful if you can make a simplified video on the same topic. Basically how the current in the Primary FET is negative. That explanation is not at all understod from this video.

[@sambenyaakov]

Yes, I am not happy myself with this video. I may post a revised version. Thanks for alerting me to that.

[@hoofheartedicemelted296]

Does a bidirectional flyback converter always use the inductor to store energy or can the energy be transferred to a secondary source ie; a battery bank? Thank you sir.

[@sambenyaakov]

Not sure I follow. The storage is temporarily. In the operation, energy is stored and discharged.

[@hoofheartedicemelted296]

@@sambenyaakov The switching of electrons between source and ouput using tvs diodes and power mosfets supposedly interferes with electtron behaviour, at least that's what I've been told. Knowledgeable engineers everywhere can sometimes be aloof and have little time for questions that are not the normal garden variety. Even if you don't understand tthe question you replied, that speaks well of you sir. Thank you for time sir.

[@tr_voyages]

Hello professor thank you for the detailed presentation, my question is can we use active clamp as PFC and controlled 400v at output

[@sambenyaakov]

I guess it can be done but you may loose soft switching at some input voltage range

[@jacobs7281]

Thank you, there is not not many active clamp videos. @41:44 With Low CC, Just before M1 turns on in ZVS mode, Will the polarity on Snubber Cap Cc(Vc) go reverse(Negative) with reference to Vin+?If o which equation determines the peak negative volatge?Any design tips fr this particular region of operation?

[@sambenyaakov]

Cn not go negative there are two diode in series to ground.

[@solewalk]

At 18:18, in the third graph from the top among the the collection of graphs on the right hand side, the green colored graph describes Id (the current through the diode). I guess that, the shape of the graph of Id during the time M1 is open as a switch, should very closely resemble the shape of the graph describing the voltage change across the inductor during the same time period. Is my guess correct?

[@lifepobatteries8671]

very informative.

[@sambenyaakov]

Thanks

[@MrBubblegumx]

Hello professor i have a question: If the load changes and we need way less ouput power, can we still achieve ZVS for M1 without changing our switching frequency? Recycling of the leakage Energy should still be possible, but we wont be able to use the negative current to achieve ZVS, because the voltage over M1 will just build up again, right? Or is this design not suitable for deep DCM and changing loads? EDIT: I just saw the part about the modified implementation, that might actually be the answer already, right?

[@sambenyaakov]

There seem to be some ways to still get ZVS but changings frequency is probably the best.

[@torbjornandersson5997]

Thank you for making these fantastic videos! There is one thing I've noticed while experimenting with ACF converters. Adding an active clamp to a converter running in DCM will drastically change the transfer function of the converter. The Bode plot will start to look like a poorly compensated CCM converter. I guess this has something to do with the extra transfer of energy. Do you know if there is any literature covering this topic in detail? Best regards!

[@sambenyaakov]

Interesting , Thanks for sharing. No, I was not aware of this.

[@rahanchowdhury1865]

Hello professor. Thank you for describing in detail. I was going through one of active clamp circuit. I was unable to design it or extracting the parameters. Is it possible to give a video tutorial how to design active clamp? One more thing if there is no Cdss across than how would I calculate time delay? Also if in the output side if two diode and capacitor connected before the Cout how output will work? Is is possible apply flyback and forward concept by adding diodes and capacitors at the output circuit? I think I will get some help in this regard. Thank you.

[@sambenyaakov]

Will try. Thanks for comment.

[@salehpaz]

Thank you for your great video. I have a question related to Active Clamp Flyback converter. How can I find an equation for the on_time of the mains switch to be a function of the clamp switch, magnetizing & leakage inductance, the phase-shift of these switches and the input current? Could you please guide me through it?

[@sambenyaakov]

The on time is not a function the parameters you mentioned. It is independent. Otherwise you would not be able to control the stage.

[@shreelokhande]

Nice video Sir.

[@sambenyaakov]

Thanks for comment

[@pappaflammyboi5799]

Do you have a link to the article "Boost-Type Power Factor Corrector With Power Semiconductor Filter for Input Current Shaping", or any equivalent design and/or simulation for a similar PFT control loop article? Great presentation...

[@sambenyaakov]

ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7913719&tag=1

[@pappaflammyboi5799]

@@sambenyaakov I'm aware of that link... can't read it though

[@sambenyaakov]

@@pappaflammyboi5799 This is protected by copyright so I can't send you. Write to one of the authors and ask for copy.

[@kkh8623]

Thanks for your great lecture. Can I ask you the real simuation circuit in LTspice? I want to duplicate it but it is little bit diffcult to make same waveforms.😅

[@sambenyaakov]

If you will be more specific I might be able to help. [e-mail]

[@kkh8623]

@@sambenyaakov Thanks Professor, Can you please check the mail?

[@wakeupsiddy6907]

Can you tell us how can we design particular RCD Clamp network ?? I mean the values or RCD.

[@sambenyaakov]

see kzbin.info/www/bejne/sHu0fmtqZ8ifma8

[@wakeupsiddy6907]

Thank you so much I have found the solution from your video. This video helps me a lot in My Master Thesis.

[@solewalk]

what is the role of the capacitor Cdss across the drain and source of M1 in the diagram at 18:18? Could anyone explain? thanks

[@sambenyaakov]

This is an inherent parasitic capacitance of the device

[@solewalk]

@@sambenyaakov oh i see. thanks

[@joshuathomas6326]

Dear sir, can you share the simulation file please?

[@sambenyaakov]

To which video/circuit?

[@joshuathomas6326]

Flyback converter with active clamp

[@joshuathomas6326]

I am trying to build a hardware for this converter and i am finding it difficult to simulate in Matlab

[@PSPS437]

Dr. Sam: It would have been better if you could have explained the circuit as a flyback rather than drawing it as a buck - boost type of configuration. The current directions are little confusing and hard to analyze mentally.

[@sambenyaakov]

Hi Swaraj, Sorry for confusion and many thanks for comment that can help improve future videos.

[@willylagwinski9749]

Unfortunately this is a lamentable chaos confusing rather than explaining.

[@sambenyaakov]

Indeed, admittedly, this is not one of my best videos. Only for the brave. I may redo it. Thanks for taking the time to write the comment.