SMPS Tutorial (4): Boost Converters, Flyback Voltages, Switched Mode Power Supplies

[@theadmiralty9175]

This series may be one of the most clear and helpful things I've heard in the three years of comp-eng classes I've taken here at the university. Come to think of it, at LEAST two thirds of my actual functional knowledge has come from either here at TPAI or Dave over at the EEVBlog. Keep up the great work, guys!

[@nevermorefuzzy]

I've messed around w electronics for a few years and just treated some stuff ignorantly as "magic pixies" not inferring AvE as any kind of simpleton. I just didn't realize what was going on... Bigclivedotcom, AvE, and you have changed my world. I just hope at my age there's still time to be accomplished in your career fields. so many years wasted. I am humbled plz keep teaching this is mana to me.

[@drbun]

He does a better job explaining this topic better than my college professor. Awesome dude!

[@nagarajkumarambalgi5420]

Hello Sir, I'm from india. I was student of master degree. I always use to see your videos. Your videos helped me a lot to understand the basics of SMPS. Inductor design video was ultimate. All the explanation is easily understandable with your pleasant voice. Thank you so much sir such videos & my wishes to you sir.

[@techno_mesh]

Very good explanation related to simple and clear images. One thing that I think is important to stress or explain deeper, is that onece S1 is opened and the diode or S2 starts to conduct, the voltage output Vo (first without load) will depend on the resistance (impedance) the inductor will see in the circuit. In your schematic the capacitor will determine the voltage you can reach. If we forget for a while that capacitors have a break voltage, if the inductor current is driven into a capacitor with only a couple of pico-Farads (like an air gap) you will see a very high voltage. On the other side if the capacitor is of several farads (supercapacitor) it will take much more cycles to reach a high voltage especially in presence of a resistive load in parallel with the capacitor. The capacitor generates a voltage that is an integral of the current "pumped" into it and the slope of that integral is 1/C, therefore small capacitors will reach higher voltages. Once S1 is opened, the inductor will generate only a "flyback" voltage high enough to keep the current at the value when S1 was opened, and that current will decrease during the energy transfer from the inductor to the capacitor.

[@nosafetyswitch9378]

Simply A-MA-I-ZING tutorials. You explain it so well that even with little knowledge of electricity, one can make a simple Boost converter from scratch.Keep it up!

[@FFcossag]

It's good to finally be back on a usable Internet connection, so that I can watch your videos again. Excellent presentation, I'm looking forward to the next one. (By the way, I replied to your PM, a while back, in case you didn't notice).

[@ThePostApocalypticInventor]

Hey I read your message and I will reply. I just forgot it the other day, because there was so much to do. You will hear from me soon :)

[@RODALCO2007]

Brilliant video and great explanation and working examples. Keep up the great video's.

[@IdelcoEM]

My first time i see your videos, even with your strong german accent (btw im spanish with no very good english level lol) you talk very clear and easy to understand. Great video. Thanks for teaching!

[@privetteone1]

[@SovereignKnight74]

Fantastic description! This technology is fascinating! I'm using this tech in a mobile power bank right now. The guy who disliked this video is still living in the dark ages.

[@minoodeboo]

Sir, So nice of you to make such series of videos, very easy to understand and very informative too. Many many thanks for all your efforts.

[@scolic03]

Great videos. Understandable, clear and educational. Keep up the great work, there is something very different about these videos than any other on youtube.

[@Joru666]

Great series of videos, I am really interrested in power electronics and so far you did excellent work on covering the topic, very precise and in proper order. I can't wait to see new videos, especially about flyback converter and perhaps synchronous non-isolated converters.

[@ThePostApocalypticInventor]

Hello. It's nice to hear that you like my videos so far. I'm working on part 5 as we speak. Part 5 will maybe be the most complicated video yet. It deals witch the basics of inductive components, the magnetic circuit, calculating transients and other stuff and I will still need a few days to complete it. The buck-boost converter and flyback converter will (probably ) be delt with in part 6. After that I will explain the workings of the forward converter and push pull converter (and it's different variations like "half bridge" and "full bridge" converter (probably in part 7)). After that I guess I will make a special video about some real world SMPS, like ATX power supplies. It will also involve some flyback converter related information. After that I will start to make design videos for all the topologies, starting with the Buck Converter, then the Boost Converter and so on. I will have to see if there will be enough time to cover "synchronus rectification" in these videos. If not then there might be a special video about that issue later.

[@edgeeffect]

The title sequence / theme tune is one of the best on KZbin

[@NiHaoMike64]

The car ignition circuit (at least in older cars) is actually a flyback converter, and in fact is the first switching power supply used in a consumer product. The name apparently came from when a mechanic got zapped and promptly jumped backwards. :) Note that some newer cars like the 2011 and newer Hyundai Sonatas actually use some sort of step up DC/DC converter (most likely a push pull) to boost the 12V supply to 300-400V or so, then use that to run a forward converter to boost it up to the final 30kV or so. The reason why is because they can get higher peak spark power, which translates into better fuel economy and more horsepower.

[@ThePostApocalypticInventor]

:D Thats a pretty funny story ! You are right that most ignition systems are actually flyback converters. But the flyback converter and the boost converter are based on the same principle, only that with the flyback converter the energy stored in the magnetic field is discharged via the secondary winding rather than via the primary. In this video I used this only as a simplified example and I will say that it's actually a flyback converter in the video about flyback converters.

[@rickpercy87]

This really is a brilliant video. Great explanations and very thorough!

[@DrenImeraj]

I see the old Mercedes came in handy :p Thank you so much for the video, please, please, please keep them up :)

[@ThePostApocalypticInventor]

Somehow I like that old car. It would be a dream come true to see it run once again...But I guess to make that happen is a quest of a lifetime.

[@mortenrolsing7137]

Super good series, easily the best i have seen, on the subject. The series have helped me, to better understand coils and their effect. Looking forward to more videos. (Subscribed) Greetings from Denmark. Morten

[@sanketg10]

Excellent video! I realize that I am waiting for your videos now. I have spent many hours understanding and building these converters, and but still there are always many things to learn and pick up. I am trying to create my own video series on the fundamentals of electronics in daily life. Every day Electronics: Current, Voltage, Resistor, Capacitor, Inductor and more! Would love to hear from you all experienced circuit designers and tinkerers, what do you think should be covered in basic tutorial of electronics?

[@dennisaugustinea4]

very good tutorial sir,,,,,easy understantable,,,,my hearty congratz,,,,from india,,,

I think I love your videos. You are a big help for my thesis

[@alexrebega1289]

You`re videos all are very good.Thank you.

[@benjahnz]

A very thorough and detailed explanation. Please come and teach my power electronics course in NZ. Also nice job on the V's and W's. Your English is clear and easy to understand.

[@TelmoMonteiro]

Great channel I just found! Thank you so much! I will learn so much here!

[@peaceandlove5855]

Telmo Monteiro sure :D

[@marcillomarchell1605]

Really brilliant electronics the best you can study!!

[@nevermorefuzzy]

mind blown... need more input. Stephanie. .. reassemble. plz

[@arjun1363]

i am fan of your so simple instruction thank you for your videos i

[@BOROZ28]

Good job thanks for your time helping people that don't have the time to go and look inside books but still want to learn (students or serious hobbyists).There is a lot of material mostly garbage on You-tube (teenagers building plasma reactor with tuna cans) .But sometimes you can get quality material on You-tube although hard to find. No offence but when I saw the channel name "The Post Apocalyptic Inventor" i thought that another teen tries to build plasma reactor with duck tape and tuna cans.lol Thanks and keep on producing.

[@ThePostApocalypticInventor]

Thank you for your kind words. I'm still a big fan of reading books myself and I use them every day, but I think that videos can have a huge impact on peoples education too. KZbin could be a great infrastructure for the educational system of the future. I guess in some years politicians and teachers will recognize that too....btw: yes there is a lot of very cheap content on KZbin, but there is good stuff too. It just takes a long time to find it :D kind regards !

[@ThePostApocalypticInventor]

The Post Apocalyptic Inventor Hey I just saw that you send me a message a couple of days ago. I just saw it right now and I wrote a long answer. But I cannot send it to you, because youtube says, that you only accept messages, from people on your contact list... It doesn't work, even though i used the "answer - button"

[@BOROZ28]

The Post Apocalyptic Inventor janmartinbolstad

[@BOROZ28]

Janusz Gluszak @

[@BOROZ28]

Janusz Gluszak gmail

[@tolikvoron]

Interesting points, thank you.

[@Gy6HackPh]

How can you make a self pulse switching swtch? Great Video!

[@shepard750]

great video

[@hassanomer7777]

Thank you for sHaring this information with us

[@pwfcpwfc]

Thanks for the videos , well done. One side question , I notice you have transistor and LCR tester from Peak manufacturer , is it a good to have testers ? I am debating myself to get its or not for a long time. Thanks.

[@ThePostApocalypticInventor]

I absolutely recommend both the LCR and the Semiconductor-Tester (DCA) from Peak. The LCR is great because it can measure even very small and very big inductors and very small and very big capacitors too. Its a good addition to a DMM. The DCA is also extremely usefull, especially if you are repairing a lot of stuff. It can test bipolars, Mosfets, Diodes, LEDs and it always tells you the pinout too.

[@pwfcpwfc]

The Post Apocalyptic Inventor Thanks for the reply , will get both for my tool box.

[@marshmallowpone]

If I built a boost converter using a MOSFET what exactly is limiting my current output? Is it the resistance of the inductor?

[@jukees3658]

Everything basically, the size of the inductor, the type of mosfet or wires in the circuit ( although the resistance is so small that it can be neglided.)

[@arrshithrg5359]

Hi can u explain the working of flyback converter

[@davesmith679]

can you share the circuit for that boost converter.

[@mahmoudmetwally1920]

I have a 6v lead acid battery and a boost converter to power a dsl modem when there is a power outage. what is the best way to switch it automatically with the wall adapter? thank you.

[@ThePostApocalypticInventor]

The easiest way: Take an Ac adapter that directly feeds a relay. Once there is a power outage the relay will turn off, the switch will change position and connects the boost converter (+battery) to your dsl modem. But there are other possibilities. You could for example use a comparator that constantly compares the grid voltage to a reference voltage. Once the grid fails, the comparator will pull its output signal to "high" which can then activate a relay, MOSFET etc...that could connect battery, converter and modem. Problem: The comparator needs a supply voltage to operate. Maybe the best way: The Battery must be constantly connected via a charger to the grid. Connected to the battery is a comparator. Then a small transformer is connected to the grid, its output (around 5 V or so) is rectified and smothed out by an LC filter. That DC is fed to the inverting input of the comparator, while the other input of the comparator is connected to a voltage reference circuit. The output pin of the comparator leeds (via a transistor) to a relay which disconnects the modem from the load, while it connects the boost converter to the battery. The relay can be configured in such a way that it will consume no energy in the "on" state, when the grid is failing. Instead of a relay, power transistors could be used as well . + Connect a large capacity to the input of the modem, so that it can "bridge" the time needed to switch the relay

[@DiePloegskaar]

Firstly, thank you for this very informative series of videos! Would it be wrong to assume that that humble "Joule Thief" is actually a type of switching boost converter ? (a very simple SMPS)

[@ThePostApocalypticInventor]

The so called "Joule Thief" is an unregulated "self exciting" (or "self oscillating") Boost Converter.

[@DiePloegskaar]

The Post Apocalyptic Inventor Thank you, that makes sense. But on that topic, do you have any suggestions or tips for construct the optimal and most efficient self exciting Boost Converter ? I am especially interested in the maths behind the commonly used toroidal inductor, for example is bigger better or just overkill ? Is a 1:1 turn ration the best approach ?

[@ThePostApocalypticInventor]

Die Ploegskaar I cannot answer this question properly in this comment, nor could I do that in a mail of 5 pages :D Why? Well, building super efficient self oscillating switching converters is one of the most difficult things in power supply design. It requires a lot of knowledge, calculations, simulations and experiments. But some short advice: To increase the efficiency, the power dissipated by the transistor and the ferrite core must be minimized. You would have to try out different types of transistors and minimize the input current that is necessary to allow proper switching action. Furthermore you would have to use a ferrite material that disspates as little power during magnetization and demagnetization as possible. The turns ratio between the two windings must be just big enough to activate the transistor. I don't see how a turns ratio of 1:1 would increase efficiency since one of the windings doesn't carry any of the load current. It simply activates the transistor. The volume of the ferrite core limits the maximum reactive power that can be conveyed by the circuit. A bigger core allows more power, and would also allow the same inductance with fewer windings. But: Ferrite cores have many different properties like the Permeability, Maximum Flux Density, Residual Flux Density, Coercitive Field Strength, Electric Conductivity of the core material and many more. So simply picking another (bigger) core would maybe also have several other consequences, unless you know it is made of the same kind of ferrite. Two ferrites that look very similar can have totally different properties ! Furthermore: The circuit looks simple, but the math behind it, is surely not ! Oscillating systems are hard to predict and to calculate. The behaviour of the converter will change when different loads and or sources will be attached. The oscialltion frequency will change depending on the load, which will make the design of input and output filters more difficult. etc etc etc

[@rafatomczyk3622]

Great video:) i can't wait next videos:) I hope you will show some practical circuits in the future:) and some methods to design own SMPS:) thank's for sharing your knowledge:) Can you reccommend some literature that will help me design my first SMPS? I knowe there is a lot of chip's which are designed to this but i would like to build something from single parts:) I'll appreciate that:) Greetings from Poland:)

[@ThePostApocalypticInventor]

Yes I will do all that. I will explain how you can build SMPS from "single" discrete parts. The boost converter which I used in this video has for example no integrated parts at all. I will give circuits and ideas to make own circuits. But first I will explain all the basic topologies. Many people want to build mains powered converters and the topologies needed (flyback, forward and push-pull converters) must be explained first.The problem with most books is, that they are not suited as a manuals for building power supplies, because they are too theoretical in nature. If you want to start building switching converters you should start by building a square wave generator, that can produce a PWM with variable Duty Cycle. You can do that with the NE 555 Timer or with an astable multivibrator. Then all you need is a Mosfet, a Shottky Diode, a choke (for example 100 uH) and an electrolytic capacitor. ( Most of those parts can be salvaged from ATX-power supplies). With those parts you can build Buck, Boost, and Buck-Boost converters. You should experiment with those simpler topologies first. Power them with a 9V-battery or similar. Later, when you have gained more knowledge you can start to build a flyback converter that is mains powered. But I will explain all that in detail in my videos. I hope that I can make them faster in the future.

[@rafatomczyk3622]

The Post Apocalyptic Inventor I've bought some inductors recently. Is this coil: www.tme.eu/de/details/coil0810-0.1/vertikaldrosseln/ferrocore/# good enaugh fo SMPS? I'll try to do something with NE555 timer and mosfet. thank's for reply:) it's nice of you:)

[@ThePostApocalypticInventor]

Hello. You can theoretically use a cylindrical coil like that for a switching conveter. The maximum current is also important. ( 1,7 A is enough for many applications). I however would recommend to buy toroidal chokes like this one : www.reichelt.de/TLC-10A-100-/3/index.html?&ACTION=3&LA=446&ARTICLE=105605&artnr=TLC+10A-100%C2%B5&SEARCH=ringerkerndrossel The value of 100 uH I mentioned is just a typical value. I would also buy or salvage some coils with different values for experimentation purposes. Like 47 uH, 100 uH, 220 uH, 470 uH, 1mH.

[@rafatomczyk3622]

The Post Apocalyptic Inventor ok, thank's a lot. I've got some toroidal chokes from old TV's and my old ATX. I must buy or build inductance meter because i don't know their values. My multimeter was cheap ( about 15 Euros ) so it does not have this function. But i've got an osciloscope so i can measure time to determinate a current value in a simple RL circuit. I hope you understand what I'm writing because my english is not as good as i whish:) what do you think about measuring time base of the circuit and calculate the L value from t=L/R equation?

[@ThePostApocalypticInventor]

I do understand you well :) If you want to buy a meter for measuring inductances I recommend this meter: www.reichelt.de/ATLAS-LCR40/3/index.html?&ACTION=3&LA=446&ARTICLE=81767&artnr=ATLAS+LCR40&SEARCH=peak+LCR It is very good for this job because it can measure small inductances (down to 1uH). But it would also be possible to measure the inductance in the way you are suggesting. But it depends on what kind of oscilloscope you have there. If you simply connect a resistor to an inductor in series and switch it to a voltage source the transient in which the current is rising to 99 % of the maximum current will probably be very short. To make that visible your scope needs to have a "storage" function.

[@terrylee6319]

What a great teacher you are !! d^^b

[@marcosborges8]

Excelente tutorial amigo, lástima que antes de aprender electronica debo aprender ingles... xD +Like

[@hitmanLis]

7:40 shouldn't the current go the other way?

[@ThePostApocalypticInventor]

Glad to Be Normal No. An inductor is not a resistor. Current through and voltage across and inductor is governed by Faraday's Law of Induction and not by Ohm's Law. If you check any book about SMPS you will see, that the current flows in the same direction and does not reverse, just like in this video. Also check these two graphs from the Wikipedia-article about boost converters: Shows the current arrows: en.wikipedia.org/wiki/File:Boost_operating.svg Shows a graph of I_L, depicting that it never goes negative (doesn't reverse direction): en.wikipedia.org/wiki/File:Boost_chronogram.svg

[@TheHunChem]

10:17 Is that really a disadvantage? Like I mean you need the dc to dc conversion so that you don't have to deal with ac at all. If I look at it like this, the things that comes to my mind are: a) Why would you need thousands of volts that you stepped up from the mains? b) Why can't you just simply have 2/more coils on the inductor core if you really need the insulation just like in the car or in the HV generator? (and looking at you, SMPSUs ._.)

[@ThePostApocalypticInventor]

HunChem a.)There are plenty of applications, where you need high voltages. Most of them are related to tube-based technologies: magnetrons, cold cathode tubes, neon tubes, tubes for power transmitters etc.... b.) It is possible to connect one or more secondary windings to the storage choke of a boost converter. But once you have done that, you call it a "flyback converter", even though it is nothing else but just an an isolated boost converter. Flyback converters will be the topic of other videos that I will make in the future. c.) "insulation" is absoluetely not the same as "isolation" and you should learn the difference. If you want to know what "electric isolation" is, watch my video: "The Power Grid, Isolation Transformers and Earthing"

[@TheHunChem]

The Post Apocalyptic Inventor Thanks for the reply, it really made a lot clear. And my bad for the last one, I meant isolation, I just screwed up :/ My other problem was that I didn't know that flyback converters are a completely different thing.

[@EETechs]

The Post Apocalyptic Inventor Don't you need to talk about type 2 and 3 op amp compensators when working switch mode power supplies? Else, how do you plan to keep a stable feedback loop?

[@misium]

The fuel in a combustion engine cylinder is not exploding, but combusting. I don't think there is such a thing as a "controlled explosion" really, unless you only mean its force. The combustion of fuel is carefully shaped and timed to achieve best total combustion. You can't do that with an explosion. /Just hair-splitting.

[@ThePostApocalypticInventor]

misium I don't know the exact definition of an explosion. So you might be absolutely right, What I meant is the rapid increase of the volume of the gas inside the chamber that occurs during point 3 and 4 of the p-v diagram of the Otto Cycle, with point 3, being the moment of highest pressure, where as far as my knowledge goes, the spark is created. Maybe it is a mere "expansion" rather than an explosion. I called it"controlled" because it is triggered and harnessed inside a controlled environment.

[@researcher4good]

When it truly explodes, it's bad. Called pinging or detonation.

[@TelmoMonteiro]

It's called explosion, yes. Combustion is what happens inside a diesel engine, resulting from high pressure and temperature (no sparks). Explosion is the result of ignition sparks inside a gasoline engine, resulting in a controlled explosion, at least while the only thing that moves there is the piston ;)

[@TelmoMonteiro]

At least this are the definitions I learned from the car mechanic's theory...

[@misium]

The Post Apocalyptic Inventor I had to look it up, yes explosions apparently can be slow and fast. Fast explosions, i.e. detonations are characterized by a high pressure and temperature gradient in the medium through which they spread (a shock wave). In slow explosions, that is deflagrations, heat transfer becomes more dominant as the means of spreading the energy and it does so more uniformly (e.g. pressure increases in pretty much the whole volume of the cylinder, after the combustion started). In other words supersonic explosions are detonations and subsonic explosions are more or less fast cases of combustion. Fuel in a cylinder (be it diesel or gasoline) does burn and although it does it rather fast it does not detonate. So yes, you were correct in calling the combustion of fuel explosion, and I mixed up explosion with detonation when writing my comment.

[@Mr.Exquisite]

4:41 shouldn't it be "increased"?

[@ThePostApocalypticInventor]

+MrExquisite Yes it should be "increased". English is not my first language. Things like that happen sometimes. But thanks for telling me I will make an annotation.

[@StanislavJochman]

can yo give some schematics

[@nebnebnosam]

is that oglethorpe? sounds way too competent to be a plutonian lol

[@songhothe6991]

cam on vi deo nay,