24V 10A 240W power supply - schematic & halfbridge explained

In this video I explain in detail how does a half-bridge switching power supply work. I use my 24V 10A 240W SMPS for demonstration, but most AT and ATX computer power supplies and many other high-powered power supplies work the same way.
The first part:
• 24V 10A 240W power sup...
You can support me on Patreon:
/ diodegonewild

[@TazkoHoZabit]

Your channel has the potential to teach people more than the university.

[@BillAnt]

And also a magically appearing 26Khz above the transformer at 6:11 ;)

[@thewld2]

It does already. I’m 4 classes away from my EE degree and already have a CE degree. Have not learn this much practical material at the university, only on my own and videos like this

[@BillAnt]

@Thewld < Formal learning will teach you the necessary theories which are very important, while these videos provide you practical experience. In my opinion both are equally important to become a knowledgeable engineer. ;)

[@Basement-Science]

Very good explanation! Especially about the feedback circuitry. It´s by far the most complex part of these power supplies. When I started learning about switching converters, I originally wanted to build one with all _discrete_ circuitry. Then I learned about the insanity called feedback circuits.

[@sovietrepublic38]

Feedback circuits are really unfathomable. Those people who designed them are genius.

[@BogdanSerban]

Feedback itself is quite easy to understand and quite fun to design also, but compensation is simply insane.

[@cagmito76]

im 5 mins in and never have a learned so much off one video and hes only just warming up

[@nikiamz6501]

You explain everything very easy, so it's very easy to learn from you!👍😀

[@craign8ca]

This is the best in-depth explanation of a switching power supply that I've seen. Thanks for sharing your knowledge.

[@Hellenios]

Wow, looking forward to the upgrades... One of the best videos on this topic which I have ever seen...

[@gordonwedman3179]

Just came across your videos recently and I think they are very informative and well produced. This was probably the most complete description of a fairly complex SMPS that I have seen and I have never seen anyone describe the feedback mechanisms in such detail. Really appreciate that.

[@AlexanderBukh]

Mega video. So much information. Obsolete scheme or not, it provides so much insight. Thanks a bunch! I will rewatch it several times in coming weeks.

[@stevenyamada70]

These are my favorite youtube video's. The schematics with an explanation. They really help to understand the way the circuits work. THANKYOU!

[@carldavis8228]

Fantastic review of the circuit. Best in-depth explanation I have ever heard. I can't believe you remember all the information without reference material. I guess that is why you have so many subscribers.

[@nevilleborg5031]

Calling it excellent is an understatement. Amazing explaination.

[@RedmiNote-nd6hm]

Výborný kanál o "component level" elektronice, díky za tu práci a gratuluju k úspěchu videí.

[@KGE64]

The startup circuit is especially clever. And they save an optocoupler and still have isolation between primary and secondary. It also makes troubleshooting easier because you connect the oscilloscope to the isolated part.

[@savinogiovannimuscio2953]

This is one of the best explenation on the HB converter that i saw, congratulation

[@DrHouse-zs9eb]

This scematic-drawing is quite genius! How the hell you know so much about power supplys?

[@electroniquepassion]

Dr. House Dr House !!! Super Docteur en France !!

[@cristonysolomon9439]

Probobly he is an electronic entusiast and lovingly dedicated to electronics

[@electroniquepassion]

cristony solomon I think yes

[@krisu9127]

Jeżeli jest geniuszem to ty jesteś debilem

[@ekuvie]

Can one draw a schematic just by looking at a circuit. If yes how do I learn.

[@richardnanis]

Thanks for this great review and in detail explanation of switching power supplies in half-bridge configuration! This is probably one of the most in-depth explanations i have come across yet! Looking really forward to your modification to a lab bench power supply! There are so many powerfull server PSUs out there very well worth beeing converted to bench PSUs :-) Some of them reach several kW of power and can deliver well over 100 Amps at 50 V e.g. But they are even much more complicated than this example ;-)

[@6siqueira]

I arrived to your channel by accident and I'm enjoying it a lot. Your videos are the best!

[@kapegede]

Well explained. At the first moment, I thought I'll never understand that. But now it's better. I'm an educated electonics specialist, but it was 30 years ago, so my knowledge is a bit rusty now. Thanks for brushing it up!

[@MegaSunRise3]

I'm currently learning on how to use the TL494 and I want to try to build my own power supply with it. This is very useful, thank you.

[@1pcfred]

To make it worthwhile I don't think the complicated design of this one is particularly useful for me to try to make in small quantities. YMMV

[@CliveChamberlain946]

Don't forget its twin - the KA7500 is pin compatible and runs up to 300kHz

[@timh2870]

Excellent video! Extremely informative and educational. Thank you and keep making more videos like this!

[@neteagle2k9]

Very good video and explaining. The idea to convert it to a bench power supply is very good too !

[@alejandroperezkeogan8281]

It's the best explanation of an SMPS that I ever heard, Congratulations 👏👏👏👏👏. No matter the language I understood everything. Very good 💯💯💯

[@electroniquepassion]

you have a very beautiful cat !!!

[@Zebra_Paw]

Hey Electronic Passion! Love your channel, and continue watching DiodeGoneWild, he has a very good electronic knowledge! French: salut Electronic Passion, j'aime beaucoup ta ghaîne, continue de regarder DiodeGoneWild, il a un savoir incroyable en électronique

[@Electester]

Superior explanation... that is really how it is... It’s worth to mention that half-bridge and full-bridge SMPS are more efficient, more powerful derivatives from push-pull conception - 2 alternately switching transistors, divided windings, 2 diodes which means double the power charging the coil which is collecting energy. Best wishes from Poland.

[@gordonrobertson102]

This was very interesting. I hadn't seen this topology before, was impressed with the start up arrangement. Your cat is very cool!

[@Kostanj42]

very good explanation! i love the black magic that power supply is made of..

[@grahampinkerton2091]

Oh your english is such a great inspiration to others learning it!

[@lkahfi]

Thank you so much for this explanation! It's help me to understand how it's work, especially the PID of the feedback system

[@Electester]

Your explanation is superior... I can not get out of admiration.

[@kapsabet3]

I am new to Circuits but your explanation though complex is very captivating. I will watch again and again it awesome how you put it all together. Thank YOU Thank YOU so much

[@PeterMilanovski]

Anything is possible when you understand the operation of something and what all the components do in the circuit. I would definitely love to see you convert a SMPS into a bench top power supply! Especially if there's two of them to get both positive and negative rails. A retail dual rail power supply is way to expensive but it's a necessary tool for electronic work...

[@henriksyvertsen3998]

Fantastic video. Please keep making videos about SMPSs. Thank you very much!

[@sleepwalkerbg1]

And that is pure knowledge with very much experience in background. I am really suprised with this wonderful and educational video. Great job and thank you !!!! Cheers

[@dedecekhribecek3049]

Super video, perfect explanation of the controll loop stability.

[@mortenlund1418]

I just found your channel. This is very very informative. I have not as I remember seen any such detailed walk tru a complex diagram. Lucky me I found your channel. Thaks for your effort. Wish I knew just a fraction of what you know of engineering electronics.

[@gnif]

Awesome! Thanks, this helped me to identify the difference between the 12V and 24V model and alter mine into the 24V model. If anyone else want's to do this on the S-250-12 model, change R33 from 10K to 22K, and remove one of the two shunts to prevent over current at the higher voltage. Also if you just want to raise the voltage a bit, replace R33 with 12K and you then get 12V - 16.8V.

[@beatrute2677]

This is one of your best videos man, if not the best

[@prathikprashanth2932]

a phenomenal job by you and your cat! you guys never disappoint!

[@CliveChamberlain946]

Yansi (at EEVblog forum) made some observations that may help mods for the conversion to a bench supply): ".. A diode is probably missing in the center tap of the NPN driving transformer. Without it, the current can get partially shunted back to the power supply. Not sure if mistake when the schematic was drawn or intention to shave every cent possible from the BOM (bill of materials - production terminology). I see two more mistakes (either be it schematic drawing, or the Chinese designer guy was on drugs): The base-to-ground resistors of the C1815 driving transistors are connected a bit weird and the auxiliary power winding should in fact be referenced with the center tap AFTER the current sense shunt. The schematic also contains some other interesting circuit nuances I'd consider at least strange and not much reason for them (for example the parallel RC in the pin 16 of the TL494)." As for circuit review: ".. explanation of the current limit feedback loop I find kind of strange ... The current is not sensed by pin 16, that's just GND reference. In fact, it is sensed in the node on the pin 15, where the reference current (5V ref divided by the 100k resistor) should equalize the node voltage to that of GND potential with the current flowing out of the node through the 1k resistor, towards the shunt - yes, the current is sensed as a negative voltage with respect to circuit GND. (That is because this way the burden voltage is compensated for in the voltage feedback loop, as the 5V reference is referenced to control circuit GND, which happens to be the output V- terminal). " ".. very well executed explanation, at least for beginners or those who are looking for more circuit specifics of these power supplies"

[@DiodeGoneWild]

The diode in the centre tap is really not there. (I know that this diode is present in AT/ATX power supplies, but they also use the auxiliary transformer for current sensing.) But it works anyway. The current from the auxiliary transformer probably backfeeds into the auxiliary power supply, but it doesn't seem to be a problem. It may actually help to power the chip. ... The base to ground resistors of the C1815s are really connected this way (each of them a different way), no matter how odd it looks. It's not my drawing mistake. I was also surprised by this and I tripple checked it. Maybe this gives the circuit a starting position on power up. ... The current sensing shunt drops only less than 50mV, so it's not that important where the control circuitry ground is connected. But you're right, the ripple of the shunt goes into the auxiliary power. I was also a bit surprised by this. Maybe it's done like this so the transformer has one pin less. The centre taps of the main and auxiliary secondaries are on the same pin. Maybe the auxiliary winding is actually not a winding, but only taps on the main secondary! This may be the explanation. ... The RC on the pin 16 is there for a good reason. Both inputs of an OpAmp should be connected via the same resistance (here it is 1k). This is a common practice with OpAmps. Using the same resistance in both inputs helps to cancel the voltage drop on the resistors caused by the leakage current of the inputs. Ideally the inputs draw or supply no current, but in real life there's always some tiny current. ... Where the current is sensed on the shunt, depends on the perspective :). It depends whether your reference point is before or after the shunt. You're in fact right, but it was easier to explain it saying that the voltage of the shunt is compared with a 50mV reference voltage. This is true if our reference point is before the shunt. Of course, the GND of the chip is connected after the shunt, but this doesn't change the behaviour of the OpAmp.

[@andymouse]

Thanks to a great explanation I understand more than I thought I would, certainly more than basic operation but goodness no's how you guys design these from scratch !

[@zaprodk]

I like seeing your pets in your videos. They are a good addition :)

[@omsingharjit]

Your videos are extremely Knowledgeable and informative thanks :)

[@rejvs]

I could understand every bit of information effortlessly. You are the ideal teacher Sir. Please convey my regards to the cat too...

[@sm6fie]

This video and explanations is very professional. Top ranked.

[@alparslanyakupoglu8421]

As an engineer, i learned a lot from you. Thanks a lot.

[@mannyp3237]

exceptional intellectual or creative power or other natural ability- genius, now its 100k subs,good job bro

[@andrewsmith3081]

Another brilliant walkthrough, thanks!

[@homemade83]

amazing keep going we need the push pull topology too

[@giubaca]

Very keen to see your conversion into a power supply 😁

[@armunro]

Fantastic. I fixed one of these yesterday. I thought it was complicated but not this complicated!

[@electronic7979]

Very useful video. Thank you so much

[@melplishka5978]

Great Scott and you both e plain everything perfectly.

[@thekaiser4333]

Exzellent, exzellent, exzellent! Zhis channel actually DOES, wat all zhe other elektronik channels like EEVblog etc. should do! I have zhis PSU and zhis iz exactly wat I vant to know! Very well done!

[@guilhemedemassenaladario]

Another amazing video!

[@TestTest-fs6eh]

Congratulations. Good job.

[@zolatanaffa87]

What an explanation! a real marathon! congratulations very alive, although, I am convinced, without the help of your cat you would not have done for sure! Lol Replace the aluminum of the main inductor should not make changes if you keep the same section and number of turns: you could only raise the maximum current bearable by the inductor (not the ouput current) Differently, if you change the number of turns you open a world of possibilities .... but in this case, the teacher is you! To the next video!

[@sohail0275]

Your videos are very informative, thanxforyourteaching

[@kevinmills5293]

Fantastic explanation.

[@mbender8653]

You do great videos, you sound like you design switching power supplies for a living. Of course I noticed that all your videos receive a cat scan before they are released. That must be a great help!

[@grigoreman7198]

You are a very good teacher. Thanks.

[@mostafashahwan9864]

Wonderful and great work

[@bm830810]

this was very very informative, I guess half-bridge is not black magic anymore thanks

[@ComoGT00]

Congratulations for 100K subs!

[@1fertube]

Excellent explanation!

[@AmitabhAnkur]

Please do a QNA on 100k subscribers.

[@FaysalKhalashi]

103k now..

[@aakasoto]

@@FaysalKhalashi 104k now...

[@cisarvialpando7412]

@@aakasoto 154k now

[@barnabasnagy9595]

That's a good idea

[@Tom-90210]

165k

[@Flapjackbatter]

Top notch video. Examples are crucual for effective and complete learning. I think.

[@jp040759]

Fantastic analysis. Thank you.

[@rmjnm]

Great explanation.

[@blake301987]

Great video, love the cat transitions.

[@sethuraj5717]

Awesome explanation. Thanks man!

[@baylorx6436]

You, sir, are the hero I need.

[@andiyladdie3188]

Outstanding explanations and schematic, very nice, try to explain in another video, boost pfc circuit.

[@sathyapalreddy2157]

I want that schematic Great explanation

[@anirbanchakraborty3081]

Best explanation...

[@mohammadr797]

You are great, perfect. always waiting for new video, please do some analysis in detail for your projects on the website specially induction heater and welder. Thanks

[@paranoidandroid9511]

Fantadtic video. Great job at explaining. Only thing i would like is to have pictures of the components on screen as you talk about them in the diagram. Thank you again.

[@anandgaurav1980]

My salute to you. Genius you are

[@supul32]

Well explained. Love your cat and the dog

[@isoguy.]

Awsome video. Thank you for sharing.👍👍👍

[@dipubiswas8520]

very well explained

[@dardosordi]

I love this videos, you make this hard topics very easy to understand. Please, change the color balance of your camera from 'auto' to a fixed (hopefully calibrated) value. Thanks!

[@AgentOffice]

Beautiful ripple

[@josifvissarionovich5320]

Come on! Tell the truth (kápni božskou). Your cat is whispering most of that knowledge to you. LOL But for real nice and simple explanation. Half an hour video from you equals like 10 hours buried in books. Keep it up. Congrats to 100K!

[@recyclechannel1556]

Your english is becaming better and better 👍👍👍

[@mcflapper7591]

thank you very much. great content.

[@savinogiovannimuscio2953]

great video and congratulations for the cat :)

[@ats89117]

Great video. That's pretty strange with the output filter inductor. You say it may have been from a 12V supply. Maybe it was in a common mode choke configuration on the 12V supply? Then repurposed with the two sides paralleled to work with 5V, but if the inductor were being specified, it's not clear why they wouldn't just use one with a single winding... Anyway, your power supply knowledge is really impressive and it's amazing that you are making these videos rather than doing high-price consulting.

[@mrjohhhnnnyyy5797]

CM chokes use high-permeability cores, like ferrite. In normal operation the fields cancel each other and all you see is active resistance of the coil. But for CM noise it presents a high impedance, since then you have two inductors in series effectively and 4 times the inductance of the single winding.

[@ivannepomuceno4757]

Waiting for part 3!

[@sanjeevkumargupta6551]

Awasome job.Very very thanks

[@adamlipsky8010]

We want Q+A!!! 100k subscribers have been achieved!

[@kunalraut9509]

Even your cat is gone totally wild.,.really nice video,explanation n cat too,...lucky to be ur channel subscriber,,,, love n regards from india

[@yoyobom1130]

11:17 "out of fu'ck you's " made me laugh so hard

[@mohitsorout6669]

Amazing explanation about power supply as always, 👍👍👍👍👍, similar as pc atx supply

[@ivolol]

Hi, it would be very cool if you could try to build your own power supply of the various topologies, going simplest to more complex, showing your considerations in the design, and then building it to see how well it works. You could even take parts from other supplies so you built your own 'franken-supplies' :)

[@alejandroperezkeogan8281]

Another Excellent 👍

[@hendriagustian7082]

Thank's for your upload this video. Very interesting, on GDT dead time can be passed.

[@longrunner258]

Regarding that arrangement error with the NTC thermistor, I think I've now seen the correct implementation of what they've apparently butchered-down: One of the old ATX power supplies I have, a Hipro HP-300SN (probably fairly "high-end" in 1998 when it was designed according to the PCB, though underwhelming by the standards of modern reviewers; now salvaged for parts), used a *pair* of NTC thermistors (each 2R5 nominal; it's common enough in other units to use just one such thermistor which I consider marginal on 230V) in the + and − terminals of the diode bridge; so that the 100~127V input range charges the capacitors through the 2.5Ω on their respective sides, while the 200~250V range charges the pair through the total 5Ω, thereby allowing a more forgiving compromise between maximum inrush and dissipation under load. So perhaps someone later copied such a design, then removed one thermistor without understanding the reason behind the arrangement.

[@Peltio]

I'm watching this while drinking Becherovka and digesting a Medovnik honey cake I ate five weeks ago :-). Wonderful explanation, I wonder if you could also show the noise in the output in some other video...

[@adrianvonino2574]

Thanks alot for the video :)

[@mcflapper7591]

still eager to see the SMPS to Lab PS conversion ;-)