I cannot believe the amount of work you put into your videos. The animations must take forever to make. And then you actually built an active load to properly show how it behaves instead of just connecting some resistive loads. Simply amazing and much appreciated!
The animations do take a good while to make. Partially because I'm not great at python 😅 Comments like this make it all worth it! Thanks so much for watching :)
@@sto2779 you must be the town idiot xD you don't learn every aspect of every field in school. I did mostly digital electronics, not analog. Also, who spends time replying to a year old comment 😅
You deserve millions of subs, for such good beginner videos! :) I am a retired electronics engineer and I always struggled to explain in simple terms to juniors, and you do it GREAT!!!!!!!!!!!!!!!!! :)
I appreciate and respect you for showing us the "how & why" that goes into the power supply design. I'd really, really like to see that "Proper lab bench power supply with all the bells & whistles" that you mention at 1:42 !
I've seen probably a dozen videos building the wall-wart Eurorack power supply. Yours is the first I've seen that gets it right - everyone else makes the assumption that you can draw some large fraction of the transformer's rated current as output current. Bravo for setting matters straight! Most of the clones of this design I've seen in other videos would have exploding caps or a smoking power transformer if operated even close to the claimed load. With a linear regulator downstream, you can treat the load as a constant-current sink for a slightly simpler calculation. Linear regulators don't consume more than a tiny amount of current for housekeeping, so input current=output current is a decent approximation. Also, you have to watch out for the capacitor's ripple current rating - the capacitive-doubler design is often pushing close to the limits for cheap caps. (Your two-cap design is much safer, good job!) At some point I'll probably make my own video about this topic (and cite yours in hopes of sending a minuscule amount of traffic your way) and put up a Python script for designing such a beast. (I've got one right now, but it's "user friendly" in the sense of "if you're my friend I might let you use it"). I'd get a little more into what numbers to look for on the data sheet, I think. My back of the envelope for a quick-n-dirty +/12V power supply started with a 14VAC, 40VA wall wart, designing for an output current of a third of an amp. Designed with Schottky rectifiers (SB2405 is what was on hand, if memory serves) on the input, but 1N4004's for the protection diodes. Calculation showed 920 µF to maintain regulation, so went with two 680 µF on each rail (didn´t bother with that 1R resistor in between them, it doesn´t actually accomplish all that much). Calculation showed about a 37° (0.65 rad) conduction angle, a reasonably gentle ripple current of 565 mA per capacitor, and an input current of 1.36 Arms per rail (2.72 Arms total), just under the 40 VA rating of the transformer. All of these were for a 60 Hz power line because I live in Yankeeland. Ludicrously inefficient, but functional. The wall wart gets kind of toasty. The capacitive doubler has a deservedly bad reputation. Conclusion: Buy a switcher. Designing your own switching power supply is a PITA. I know. I've done it. Also spend the $30 or so on an RCD-protected line cord because I can't figure out how to put a proper safety interlock on the rack and there will be musicians' fingers inside, wiring modules. Use a commercial IEC power entry module, and insulated spade connectors, and a safety cover on the terminal block, and all *should* be finger-safe, but I still don't trust musicians with mains voltage - or to remove the mains voltage when working on the rack. Plus, on gigs, there's too much chance that someone will spill a drink into the skiff.
what you teach our college professors can't explain at all with such a ease ,Thankyou so much for sharing your knowledge with all of us , love Respect from INDIA
So this is basically very similar to my Eurorack Powersupply I built a year ago. I'd like to share some optimizations I did, as I wanted to gather as much current from it as possible. -I've started with an 11.5V halogen transformer 50VA which is not ideal at all, but due to optimisations, I can draw 500mA per rail continuously! -The rectifier diodes are beefy Schottky diodes. -I used no serial resistors, so there was no further voltage drop, but this means the transformer and diodes are more stressed. -For the linearregulators, use LDO (low dropout regulators)! So you don't need as much headroom, 2V vs 50mV makes a BIG difference. -Adjust the filtering capacitors to optimize between ripple voltage and inrush current. -Use a large Heat sink. At full load, you have to dissipate quite some power in the regulators.
Subscribed to you like a year ago and never got around to watching anything til now. This is fantastic!!! I'll be binging your videos this week, you are really great at using accessible language without leaving out critical info. People like you are great resources for folks like me, been into it for a couple years but don't have much education and havent been in school for a decade so it's not easy plowing through textbooks. Love the animations too, I really appreciate the work you put into this. Keep it up :)
I did this the other day for a extemely hard to find analogue video power supply!!!! I combined 3 wall warts, two 7v and one 5v. Connected the grounds together of 5v and 7v and the positive of the other 7v to the ground to make +7, -7v, and 5v. I then probed the connector and made an educated guess about which one connection was which and it worked!
Solid Job on that. Looking forward to more videos Some stats: The spoken duration of the video is: 1263 seconds - 50.85463 of silence 4826 total words spoken in HMS: 00:20:12 of total time: 00:21:03 excluding 00:00:50 silence 238 de-silenced wpm
Excellent video!! I have an associates degree in EE and remember understanding ripple voltage, but didn’t quite understand why it was occurring. You cleared it up for me in this video. The draw of the load. Thank you!!
Thank you, I learned some new things! A note about the 2 volts that you specified for headroom. Two volts is the minimum to prevent drop out. Three volts is the minimum required to allow the regulator to achieve its minimum ripple factor. Read the fine print in the data sheet and application notes. Also, read the testing perimeters that the manufacture uses to achieve the best ripple results, for marketing bragging rights. You will find that they use a minimum of 3 volts of extra headroom. HTH. Cheers!
THE BEST video on power supplies I've ever seen. I will say... you mentioned that this video is meant for a total beginner. I am a total beginner, and more than 50% of this was all WAY over my head. BUT... I have kind of learned to expect this. I realize that "beginner" can be very relative to someone who already knows electronics. Maybe you meant... "Not new to electronics, but... New to power mgmt." or... "...new to audio power..." Either way... Brilliant video. I'm going to support you on Patreon!!!
Wow! Thanks!! Yeah beginner friendly is always a relative term I guess. Feel free to shoot me a message on Patreon if you need any help on anything specific! My advice as a beginner is that you will rarely understand everything first time. You have to gradually build up your knowledge through repeated attempts at understanding various things and slowly more.and more of it will make sense. At least that's how I found it!
@@TheAudioPhool Yes. I agree. Being comfortable enough with the idea of not understanding everything, but not giving up... is a tool that I've used all my life for all my endeavors. I work in computers, and you'd better believe, I have daily experiences where I am talking with someone about something that I barely grasp myself. I will definitely reach out to you on Patreon if (when) I come to something that is specific. I have also had to learn, however, that there is an art to asking good questions, and even though I WANT to send you 10 questions per video... I realize that my $7/month or so USD doesn't exactly translate into you becoming my own personal tutor. :-) Ha ha. But... I do think that maybe a "One question per month" limit is a good place for me to shoot for. Ha ha ha... I am just incredibly grateful that you are making these videos. They are basically EXACTLY what I've always wanted to see on TH-cam.
This is a concise and intuitive lesson in every aspect of building a power supply. Amazing visual explanation of the diode's and capacitor's mysterious actions in this circuit. Great work!💯💯
Your channel is soo good I hope you can get back to posting some new videos soon, I believe you are busy but seeing that "proper" power supply video one day would be amazing, cheers! :)
It's worth noting that it is _disadvantageous_ to use larger filter capacitors than is necessary. This is because the power dissipated in the linear regulators is the product of the average voltage across the regulator and the current drawn. By using capacitors that are only a little larger than the minimum needed (allowing for tolerances, of course), you maximise the ripple voltage, which consequently minimises the average voltage across the regulator and hence minimises the power dissipated in it. It also minimises the inrush current on startup and maximises the conduction angle, so this is a good example of not over-engineering a design.
Something that may not be obvious for a beginner: A linear regulator produces heat according to the excess voltage that it has to "burn". E.g. with a capacitor voltage of 19V and a output voltage of 15V, running at 400mA, we will burn (19V-15V)*0.4A = 1.6W of heat. Your regulator will need a small heatsink attached to it to be able to dissipate that heat.
I have never seen a diode arrangement like that for a dual voltage output linear regulated supply. I always have used full wave bridge rectification if the transformer has a center tap. However, the more popular general power transformer output winding design for dual voltage output is to have two separate equal voltage output windings. Some transformers are wound so you can actually have two separate full wave bridges one dedicated to the positive supply and one for the negative supply. This has the advantage of spreading the heat out at the cost of two smaller bridges versus one large bridge. With the two "floating" supplies approach you can choose where to locate the single point star ground relative to the loads you are driving, like an audio power amplifier. This will prevent mixing of different rail currents through a long common ground impedance.
Glad you're enjoying it! The animations are a fair bit of work but I think have really taken the videos to a higher level. People seem to be enjoying them!
Wow, what a truly educational video. Learnt so much that others just dont teach. Perhaps we should rename you "Carlsberg". If Carlsberg did Electronics videos.......
There are variable 78 and 79 series regulators. They have a GU(1C) suffix in a 4 terminal power package. The output voltage specifications between the 2 polarity types do not match however with the 79GU regualators offering a slightly greater voltage range at the low end. The 78GU1C is often used in dc servo motor control for a constant speed. Come across them in audio applications in tape decks etc. They have been somewhat superceeded by the LM317 and LM337.
Sir awesome video explained in a simple way , when is your next video on same topic i.e. how to design power supply in detail , explaining the working of every components , why it is used. Waiting for your next video . Thanks
It was really cool to see the oscilloscope traces for this. I've seen the diagrams with ripple current etc. loads of times, but somehow it makes it a lot more real to see the traces as you adjusted the load in real time. Very nice to see how in practice it exactly matched the theoretical behaviour!
You have done a super job. Ask utube to provide a soft button so viewers can make a donation without said donation becoming monthly. Some channels have this set up to buy the creator a beer, coffee etc.. Some viewers can not make regular Patreon payments but, can make occasional expressions $ to reward your work.
I would describe an diode in forward bias as a 0.7 V voltage source, not a short. A short means that both terminals have the same voltage. What's important is that current is not limited (like in an voltage source).
Please, make a lab power supply design. It could even be a course like some Phil's Lab videos on KiCAD with more detail on paid course or for Patreon. I have seen few channels playing bringing up the idea, but it's never done or left not finished. There is plenty of videos with simple power supplies - from just chopping of leads of a computer ATX power supply and attaching banana plugs one way or another, maybe adding some Chinese voltmeter ir DC-DC converter module. EEVblog Dave started a design years ago, but it tracked different way to a pocket supply, that got wery advanced but wasn't finished. It is nowhere else to be found - how to make custom LCDs and heatsinks. Interesting. But it would be more practical to understand the necessary principles, what to be cautious about and build a real bench lab power supply. With some bells and whistles added. But I believe I'm not the only one who has done a fair share of simple L78xx/L79xx and L317 circuits, but would like to make a proper lab/bench power supply but lack the skills. Chinese modules are useful, but you don't really learn much by wiring them together and they have significant EMC and ripple drawbacks. Thank you for this video, circuits shown are simple and known, but you shared a good amount of knowledge and details to be aware of, many others skip over or probably don't know themselves. I think microreps and Signal path had idea of a lab power supply, but it hasn't came to life.
This is a fantastic video, so thanks for that. But I've got a question that occurred to me that I'm hoping you can offer an opinion on: Is there anything to be gained as far as maximum current output by using a transformer that outputs a voltage higher than than the regulated output, e.g. using a 15vac xformer to produce +/-12v? Using the same assumption that you did for 12V, i.e. minimum ripple V = 12+2+.5+1=15.5V, and assuming the max ripple V for a 15V xformer to be 20.5V gives an allowable ripple = 5V. Assuming a capacitance of 2000uF and rearranging the equation @ 12:35 to solve for Idc gives 600ma. (I used f=60Hz for US power...) Is this a reasonable assumption? Or is there something really basic that I'm overlooking? Google seems to be pretty useless when it comes to the specifics of charging and discharging of filter caps in linear power supplies...
Go into the maths and theory! There are too many simple and dumbed down channels that shy away from showing the actual physics and engineering --- don't fall into the same trap!
Had to sub again. Satan needs to keep his Meat Gazers off my buttons, and let me control the flow of information I receive. On a plus note. I like the illustrations you’ve added to your explanations. God Bless.
Thank you very much . I have a question please. From your experience, what are the factors that damage the double rail operational amplifier IC? And I have another question. For example, I forgot to provide it with the supply voltages, and at the same time I provided it with voltage at the two inputs. Will it be damaged?
I was actually hoping that this was about designing actual transformers, since there seems to be a lot of money to be saved for many electronics projects if you wind your own transformers. Also, single voltage supplies are very very common these days. I've got a ton of 5V, 7.2V, 9V and 12V powersupplies lying around, just from all of my electronics gadgets. It would be nice to power my audio electronics projects with those, but these projects usually require a positive and negative rail, so what is the optimal way of creating a split dual supply from a single supply? A voltage regulator seems to be the easiest solution, but not exactly the most efficient solution. Would be nice with something more efficient, but still very stable. Buck converters perhaps?
What kind of power supply is this? Linear power supply? The content is excellent, please make a video on how to make a SMPS (Switch Mode Power Supply).
Thank you, wish I'd seen this years ago. I've built a couple of these--more crudely than you described, though they seem to work just fine. More recently I used a charge pump (icl7662) and it seems to sound fantastic also (used with preamps for a guitar and a spring reverb driver). Is there any reason one might want to choose between methods for split rail?
Glad you're enjoying it! Mostly it's whatever works for you. Any project will have a certain input voltage, certain current draw and requirements for how "noisy" the supply can be and that dictates what topology you use for your supply. Definitely an interesting video in there somewhere comparing different supply designs!
i love your videos and am slowley begining to actually understand themm.... i am beggining to repair ... audio amplifiers ...you know how sometimes you need these symetric powers uspllies ...and you dont havea center tap transformer so iwas looking for an alternative circuit givin only a normal transformer ....thisis very usfual and the math is cool to ...but can they handle a serious current draw? because its happend to me that i i overheat the voltage regulator when in a audio circuit
This one can on paper handle up to an amp depending on what the load looks like. If your voltage regulators are overheading then you just need to put a bigger heatsink on them. If you look in the datasheet for the regulator there should besome calculations to show you how much heat you need to dissipate for every watt of power dropped across the regulator (assuming you're using a linear regulator)
I want to know how to build a +/- 12V PSU powered by a lithium battery... i.e. battery powered synths... But I've not found any good videos on the subject. I know that it involves a floating ground but, I'm certain there are lots of little gotcha's involved you need to plan for and am just not sure how to proceed.
I'm a bit late to this party with this one, but was wondering about the capacitors on each side of the 7815/7915 regulators. Your schematic shows all 0.1uF, but all the 7815 data sheets I have seen show 0.33uF and 0.1uF for the 7815 and 2.2uF and 1uF for the 7915. Clearly this has not caused any problems for you, but why would the chip manufacturers specify four different values of capacitor if one is enough? Does it make that much difference?
Great question! It's basically to do with frequency response. If you have calculated that you need a certain amount of capacitance, you can choose to implement that however you like. If you use different sized capacitors then you are effectively covering different areas of the frequency spectrum because they have differing reactances. Spreading it out is a 'safer' option. Since we are mostly filtering low frequency information (50Hz into DC) the HF end of the spectrum doesn't really need much coverage in this instance, which is what the smaller values cover. You could probably omit those smaller capacitors entirely since we have SO MUCH capacitance before and after it, but i always prefer to have a little bit of high freqeuency coverage so that we don't get little blips in the output from the diodes switching on & off and any noise that mightbe floating around on the supply. If you choose to add say a 0.33 instead of the 0.1 all you're doing is giving yourself more coverage which you're welcome to do! There's a great EEV blog video about exactly this th-cam.com/video/BcJ6UdDx1vg/w-d-xo.html
Those ac-ac transformers are really hard to find in me experience, especially for European plug 230V. Where did you get yours? Also 300mA don’t sound like a lot. Definitely will be looking forward to your next power supply video
I just built a similar power supply, and I bought an AC-AC wall-wart transformer from a UK company named acadaptors-r-us, that specialises in them. It cost me about £20. It's a shame because the very next week, at a car boot sale, I found one from a Creative speaker system. I paid £1.50 for it and it's perfect for the job. Part number MAE180080BH4 just in case you are lucky enough to find one used.
No, they are not. Since everyone uses nowadays switched power supplies, such old transformers are easily found. Or you visit your local thrift shops and take as much as you want at home... Or you can also use Old AC-DC power bricks. You just have to rip out the inner electronics
Thanks, thrift store might be a good idea. I didn’t see the point of opening up power supply’s to rip out the inner electronics, it doesn’t really provide any safety advantage
I dont undertand how this will be very stable power supply. Zener drop voltage depends on current and current will not be stable here, because Vin in not constant.
@@TheAudioPhool Fair enough, I was trying for a name because saying “The AudioPhool is an engineering student…” seemed odd. But if you’d prefer using the channel name, I can do that too. (And yes, I am being serious. Thanks, I’ll post back when it’s published!)
I cannot believe the amount of work you put into your videos. The animations must take forever to make. And then you actually built an active load to properly show how it behaves instead of just connecting some resistive loads. Simply amazing and much appreciated!
The animations do take a good while to make. Partially because I'm not great at python 😅
Comments like this make it all worth it! Thanks so much for watching :)
I have a Master's degree in EE, and I learn something new from every video you make. Fantastic work, keep up the nice animations :D
High praise! Thanks so much :)
You got your degree from the circus clown lol? These things are basics.
@@sto2779 you must be the town idiot xD you don't learn every aspect of every field in school. I did mostly digital electronics, not analog. Also, who spends time replying to a year old comment 😅
@@sto2779 he bought the degree 😂
You deserve millions of subs, for such good beginner videos! :)
I am a retired electronics engineer and I always struggled to explain in simple terms to juniors, and you do it GREAT!!!!!!!!!!!!!!!!! :)
I appreciate and respect you for showing us the "how & why" that goes into the power supply design. I'd really, really like to see that "Proper lab bench power supply with all the bells & whistles" that you mention at 1:42 !
i know these animations are a lot of work but they're really elevating your explanations 🧑🍳💋
They are a lot of work but totally worth it! Really glad you're enjoying them :D
@@TheAudioPhool which software u r using for animation??
@@raghulravi6431 Manim!
3b1b.github.io/manim/
I've seen probably a dozen videos building the wall-wart Eurorack power supply. Yours is the first I've seen that gets it right - everyone else makes the assumption that you can draw some large fraction of the transformer's rated current as output current. Bravo for setting matters straight! Most of the clones of this design I've seen in other videos would have exploding caps or a smoking power transformer if operated even close to the claimed load.
With a linear regulator downstream, you can treat the load as a constant-current sink for a slightly simpler calculation. Linear regulators don't consume more than a tiny amount of current for housekeeping, so input current=output current is a decent approximation.
Also, you have to watch out for the capacitor's ripple current rating - the capacitive-doubler design is often pushing close to the limits for cheap caps. (Your two-cap design is much safer, good job!)
At some point I'll probably make my own video about this topic (and cite yours in hopes of sending a minuscule amount of traffic your way) and put up a Python script for designing such a beast. (I've got one right now, but it's "user friendly" in the sense of "if you're my friend I might let you use it"). I'd get a little more into what numbers to look for on the data sheet, I think.
My back of the envelope for a quick-n-dirty +/12V power supply started with a 14VAC, 40VA wall wart, designing for an output current of a third of an amp. Designed with Schottky rectifiers (SB2405 is what was on hand, if memory serves) on the input, but 1N4004's for the protection diodes.
Calculation showed 920 µF to maintain regulation, so went with two 680 µF on each rail (didn´t bother with that 1R resistor in between them, it doesn´t actually accomplish all that much). Calculation showed about a 37° (0.65 rad) conduction angle, a reasonably gentle ripple current of 565 mA per capacitor, and an input current of 1.36 Arms per rail (2.72 Arms total), just under the 40 VA rating of the transformer. All of these were for a 60 Hz power line because I live in Yankeeland.
Ludicrously inefficient, but functional. The wall wart gets kind of toasty. The capacitive doubler has a deservedly bad reputation.
Conclusion: Buy a switcher. Designing your own switching power supply is a PITA. I know. I've done it.
Also spend the $30 or so on an RCD-protected line cord because I can't figure out how to put a proper safety interlock on the rack and there will be musicians' fingers inside, wiring modules. Use a commercial IEC power entry module, and insulated spade connectors, and a safety cover on the terminal block, and all *should* be finger-safe, but I still don't trust musicians with mains voltage - or to remove the mains voltage when working on the rack. Plus, on gigs, there's too much chance that someone will spill a drink into the skiff.
what you teach our college professors can't explain at all with such a ease ,Thankyou so much for sharing your knowledge with all of us , love Respect from INDIA
So this is basically very similar to my Eurorack Powersupply I built a year ago. I'd like to share some optimizations I did, as I wanted to gather as much current from it as possible.
-I've started with an 11.5V halogen transformer 50VA which is not ideal at all, but due to optimisations, I can draw 500mA per rail continuously!
-The rectifier diodes are beefy Schottky diodes.
-I used no serial resistors, so there was no further voltage drop, but this means the transformer and diodes are more stressed.
-For the linearregulators, use LDO (low dropout regulators)! So you don't need as much headroom, 2V vs 50mV makes a BIG difference.
-Adjust the filtering capacitors to optimize between ripple voltage and inrush current.
-Use a large Heat sink. At full load, you have to dissipate quite some power in the regulators.
All solid tips. Great work! :D
Subscribed to you like a year ago and never got around to watching anything til now. This is fantastic!!! I'll be binging your videos this week, you are really great at using accessible language without leaving out critical info. People like you are great resources for folks like me, been into it for a couple years but don't have much education and havent been in school for a decade so it's not easy plowing through textbooks. Love the animations too, I really appreciate the work you put into this. Keep it up :)
Glad to have you along!!
I did this the other day for a extemely hard to find analogue video power supply!!!! I combined 3 wall warts, two 7v and one 5v. Connected the grounds together of 5v and 7v and the positive of the other 7v to the ground to make +7, -7v, and 5v. I then probed the connector and made an educated guess about which one connection was which and it worked!
Nice bodge job! 😅
Solid Job on that.
Looking forward to more videos
Some stats:
The spoken duration of the video is: 1263 seconds - 50.85463 of silence
4826 total words spoken in HMS: 00:20:12 of total time: 00:21:03 excluding 00:00:50 silence
238 de-silenced wpm
Thanks for watching :)
@@TheAudioPhool no worries. enjoy the summer.
@@twobob You too mate!
Great to see you after long time... please continue because your explanation is very unique
I'm planning to spend this summer making plenty more videos! Not going anywhere :)
Lovely...thanks
Instantly subscribed your channel because of the quality content that most mainstream youtubers dont provide... Hats off to you sir.. keep it up🎉👏👏
Seriously awesome. Especially how obvious it is that you’re enjoying yourself. Puts a smile on my face and new thoughts in my brain
Excellent video!! I have an associates degree in EE and remember understanding ripple voltage, but didn’t quite understand why it was occurring. You cleared it up for me in this video. The draw of the load. Thank you!!
I find some teachers assume things are obvious which aren't as obvious as they think it is! I'm sure I'm guilty of the same sometimes :)
Thank you, I learned some new things! A note about the 2 volts that you specified for headroom. Two volts is the minimum to prevent drop out. Three volts is the minimum required to allow the regulator to achieve its minimum ripple factor. Read the fine print in the data sheet and application notes. Also, read the testing perimeters that the manufacture uses to achieve the best ripple results, for marketing bragging rights. You will find that they use a minimum of 3 volts of extra headroom. HTH. Cheers!
THE BEST video on power supplies I've ever seen. I will say... you mentioned that this video is meant for a total beginner. I am a total beginner, and more than 50% of this was all WAY over my head. BUT... I have kind of learned to expect this. I realize that "beginner" can be very relative to someone who already knows electronics. Maybe you meant... "Not new to electronics, but... New to power mgmt." or... "...new to audio power..." Either way... Brilliant video. I'm going to support you on Patreon!!!
Wow! Thanks!!
Yeah beginner friendly is always a relative term I guess. Feel free to shoot me a message on Patreon if you need any help on anything specific!
My advice as a beginner is that you will rarely understand everything first time. You have to gradually build up your knowledge through repeated attempts at understanding various things and slowly more.and more of it will make sense. At least that's how I found it!
@@TheAudioPhool Yes. I agree. Being comfortable enough with the idea of not understanding everything, but not giving up... is a tool that I've used all my life for all my endeavors. I work in computers, and you'd better believe, I have daily experiences where I am talking with someone about something that I barely grasp myself. I will definitely reach out to you on Patreon if (when) I come to something that is specific. I have also had to learn, however, that there is an art to asking good questions, and even though I WANT to send you 10 questions per video... I realize that my $7/month or so USD doesn't exactly translate into you becoming my own personal tutor. :-) Ha ha. But... I do think that maybe a "One question per month" limit is a good place for me to shoot for. Ha ha ha... I am just incredibly grateful that you are making these videos. They are basically EXACTLY what I've always wanted to see on TH-cam.
One of the best resources on the subject that I've come across- thanks for this excellent video!
This is a concise and intuitive lesson in every aspect of building a power supply. Amazing visual explanation of the diode's and capacitor's mysterious actions in this circuit.
Great work!💯💯
Your channel is soo good I hope you can get back to posting some new videos soon, I believe you are busy but seeing that "proper" power supply video one day would be amazing, cheers! :)
It's worth noting that it is _disadvantageous_ to use larger filter capacitors than is necessary. This is because the power dissipated in the linear regulators is the product of the average voltage across the regulator and the current drawn. By using capacitors that are only a little larger than the minimum needed (allowing for tolerances, of course), you maximise the ripple voltage, which consequently minimises the average voltage across the regulator and hence minimises the power dissipated in it. It also minimises the inrush current on startup and maximises the conduction angle, so this is a good example of not over-engineering a design.
Something that may not be obvious for a beginner: A linear regulator produces heat according to the excess voltage that it has to "burn". E.g. with a capacitor voltage of 19V and a output voltage of 15V, running at 400mA, we will burn (19V-15V)*0.4A = 1.6W of heat. Your regulator will need a small heatsink attached to it to be able to dissipate that heat.
I do show my circuit with a heatsink but it was probably a bit of an oversight not mentioning it explicitly in the video. Thanks for pointing it out!
I have never seen a diode arrangement like that for a dual voltage output linear regulated supply. I always have used full wave bridge rectification if the transformer has a center tap. However, the more popular general power transformer output winding design for dual voltage output is to have two separate equal voltage output windings. Some transformers are wound so you can actually have two separate full wave bridges one dedicated to the positive supply and one for the negative supply. This has the advantage of spreading the heat out at the cost of two smaller bridges versus one large bridge. With the two "floating" supplies approach you can choose where to locate the single point star ground relative to the loads you are driving, like an audio power amplifier. This will prevent mixing of different rail currents through a long common ground impedance.
i wish these animations were more coomon, it helps understanding the circuit like A LOT! I whis i had these in my study materials.
I know I'm preaching to the choir here, but damn, these animations are so good!
Superb explanations and animations. Great work.
Glad you're enjoying it! The animations are a fair bit of work but I think have really taken the videos to a higher level. People seem to be enjoying them!
Thank you for making this so easy to understand. You would make a great teacher. Looking forward to the next video! BTW cool shirt
You're welcome! Thanks for leaving a comment :D
Wow, what a truly educational video. Learnt so much that others just dont teach. Perhaps we should rename you "Carlsberg". If Carlsberg did Electronics videos.......
Dude you are amazing.Keep up the good work ❤️
Thanks! I will 😁
Brilliant job, and amazing explanation.
This is a monumentally amazing video.
There are variable 78 and 79 series regulators. They have a GU(1C) suffix in a 4 terminal power package. The output voltage specifications between the 2 polarity types do not match however with the 79GU regualators offering a slightly greater voltage range at the low end. The 78GU1C is often used in dc servo motor control for a constant speed. Come across them in audio applications in tape decks etc.
They have been somewhat superceeded by the LM317 and LM337.
Amazing video, great clarify, leant a lot; thanks a million !. Regards from India.
Wow ! I will come back for more.
That looks like a Hinton instruments PSU in the thumbnail. Would love to get one of those in my new case. Great vid btw.
Sir awesome video explained in a simple way , when is your next video on same topic i.e. how to design power supply in detail , explaining the working of every components , why it is used. Waiting for your next video . Thanks
This was an amazing video! Really loved the effort put into the animation! Great job!
It was really cool to see the oscilloscope traces for this. I've seen the diagrams with ripple current etc. loads of times, but somehow it makes it a lot more real to see the traces as you adjusted the load in real time. Very nice to see how in practice it exactly matched the theoretical behaviour!
Seeing is believing!!
You have done a super job. Ask utube to provide a soft button so viewers can make a donation without said donation becoming monthly. Some channels have this set up to buy the creator a beer, coffee etc.. Some viewers can not make regular Patreon payments but, can make occasional expressions $ to reward your work.
Outstanding tutorial!
when the world needed him most, he vanished.
Watched this video. Instant subscribe.
Phenomenal explaination.👍
I would describe an diode in forward bias as a 0.7 V voltage source, not a short. A short means that both terminals have the same voltage. What's important is that current is not limited (like in an voltage source).
Please, make a lab power supply design. It could even be a course like some Phil's Lab videos on KiCAD with more detail on paid course or for Patreon.
I have seen few channels playing bringing up the idea, but it's never done or left not finished.
There is plenty of videos with simple power supplies - from just chopping of leads of a computer ATX power supply and attaching banana plugs one way or another, maybe adding some Chinese voltmeter ir DC-DC converter module.
EEVblog Dave started a design years ago, but it tracked different way to a pocket supply, that got wery advanced but wasn't finished.
It is nowhere else to be found - how to make custom LCDs and heatsinks. Interesting.
But it would be more practical to understand the necessary principles, what to be cautious about and build a real bench lab power supply. With some bells and whistles added.
But I believe I'm not the only one who has done a fair share of simple L78xx/L79xx and L317 circuits, but would like to make a proper lab/bench power supply but lack the skills.
Chinese modules are useful, but you don't really learn much by wiring them together and they have significant EMC and ripple drawbacks.
Thank you for this video, circuits shown are simple and known, but you shared a good amount of knowledge and details to be aware of, many others skip over or probably don't know themselves.
I think microreps and Signal path had idea of a lab power supply, but it hasn't came to life.
1 subscribe, 1 like for your effort sir, thank you for making this amazing content
This is a fantastic video, so thanks for that. But I've got a question that occurred to me that I'm hoping you can offer an opinion on:
Is there anything to be gained as far as maximum current output by using a transformer that outputs a voltage higher than than the regulated output, e.g. using a 15vac xformer to produce +/-12v? Using the same assumption that you did for 12V, i.e. minimum ripple V = 12+2+.5+1=15.5V, and assuming the max ripple V for a 15V xformer to be 20.5V gives an allowable ripple = 5V. Assuming a capacitance of 2000uF and rearranging the equation @ 12:35 to solve for Idc gives 600ma. (I used f=60Hz for US power...)
Is this a reasonable assumption? Or is there something really basic that I'm overlooking? Google seems to be pretty useless when it comes to the specifics of charging and discharging of filter caps in linear power supplies...
Brilliant!
Quickest comment ever!
If you add too much capacitance you can blow your input transformer. It's a balancing act.
spot on mate. Most people assume more capacitance = less ripple = better.
Great video. Thanks
Thanks for watching
Great video thanks: I have a 220/24vAC transf which I'd like to use for +/-15vdc- can we take same design and just modify resistors/capacitor values?
Go into the maths and theory! There are too many simple and dumbed down channels that shy away from showing the actual physics and engineering --- don't fall into the same trap!
people dont watch your videos if you go too deep unfortunately
I enjoyed that! Thanks!
You're extremely welcome :)
Thanks professor !
Great series 👏
Had to sub again. Satan needs to keep his Meat Gazers off my buttons, and let me control the flow of information I receive.
On a plus note. I like the illustrations you’ve added to your explanations.
God Bless.
So weird!
Thanks for resubbing anyway :)
Thank you very much . I have a question please. From your experience, what are the factors that damage the double rail operational amplifier IC? And I have another question. For example, I forgot to provide it with the supply voltages, and at the same time I provided it with voltage at the two inputs. Will it be damaged?
Where did you buy your ac/ac adaptor from? I can't seem to find any! (im also from the uk)
Thank you
what about smps? will there be a video covering such topic?
I was actually hoping that this was about designing actual transformers, since there seems to be a lot of money to be saved for many electronics projects if you wind your own transformers. Also, single voltage supplies are very very common these days. I've got a ton of 5V, 7.2V, 9V and 12V powersupplies lying around, just from all of my electronics gadgets. It would be nice to power my audio electronics projects with those, but these projects usually require a positive and negative rail, so what is the optimal way of creating a split dual supply from a single supply? A voltage regulator seems to be the easiest solution, but not exactly the most efficient solution. Would be nice with something more efficient, but still very stable. Buck converters perhaps?
What kind of power supply is this? Linear power supply? The content is excellent, please make a video on how to make a SMPS (Switch Mode Power Supply).
Thank you, wish I'd seen this years ago. I've built a couple of these--more crudely than you described, though they seem to work just fine. More recently I used a charge pump (icl7662) and it seems to sound fantastic also (used with preamps for a guitar and a spring reverb driver). Is there any reason one might want to choose between methods for split rail?
Glad you're enjoying it!
Mostly it's whatever works for you. Any project will have a certain input voltage, certain current draw and requirements for how "noisy" the supply can be and that dictates what topology you use for your supply. Definitely an interesting video in there somewhere comparing different supply designs!
So nice
Thanks for watching!
Question. If i don't have a power transitor, can i use a power mosfet? How about doing a Darlington circuit with a lower power transistor?
would it be simple enough to do this with a usb-c connection instead of a wall wart?
I think in the schematics you forgot to add the negative voltage regulator LM7915
i love your videos and am slowley begining to actually understand themm.... i am beggining to repair ... audio amplifiers ...you know how sometimes you need these symetric powers uspllies ...and you dont havea center tap transformer so iwas looking for an alternative circuit givin only a normal transformer ....thisis very usfual and the math is cool to ...but can they handle a serious current draw? because its happend to me that i i overheat the voltage regulator when in a audio circuit
This one can on paper handle up to an amp depending on what the load looks like. If your voltage regulators are overheading then you just need to put a bigger heatsink on them. If you look in the datasheet for the regulator there should besome calculations to show you how much heat you need to dissipate for every watt of power dropped across the regulator (assuming you're using a linear regulator)
I want to know how to build a +/- 12V PSU powered by a lithium battery... i.e. battery powered synths... But I've not found any good videos on the subject. I know that it involves a floating ground but, I'm certain there are lots of little gotcha's involved you need to plan for and am just not sure how to proceed.
Likewise, I was not able to find the follow up video you mentioned.
I'm a bit late to this party with this one, but was wondering about the capacitors on each side of the 7815/7915 regulators. Your schematic shows all 0.1uF, but all the 7815 data sheets I have seen show 0.33uF and 0.1uF for the 7815 and 2.2uF and 1uF for the 7915. Clearly this has not caused any problems for you, but why would the chip manufacturers specify four different values of capacitor if one is enough? Does it make that much difference?
Great question! It's basically to do with frequency response. If you have calculated that you need a certain amount of capacitance, you can choose to implement that however you like. If you use different sized capacitors then you are effectively covering different areas of the frequency spectrum because they have differing reactances. Spreading it out is a 'safer' option. Since we are mostly filtering low frequency information (50Hz into DC) the HF end of the spectrum doesn't really need much coverage in this instance, which is what the smaller values cover. You could probably omit those smaller capacitors entirely since we have SO MUCH capacitance before and after it, but i always prefer to have a little bit of high freqeuency coverage so that we don't get little blips in the output from the diodes switching on & off and any noise that mightbe floating around on the supply. If you choose to add say a 0.33 instead of the 0.1 all you're doing is giving yourself more coverage which you're welcome to do!
There's a great EEV blog video about exactly this
th-cam.com/video/BcJ6UdDx1vg/w-d-xo.html
Have seen 470nF capacitors used in RS Electronics datasheets.
Those ac-ac transformers are really hard to find in me experience, especially for European plug 230V. Where did you get yours? Also 300mA don’t sound like a lot.
Definitely will be looking forward to your next power supply video
I just built a similar power supply, and I bought an AC-AC wall-wart transformer from a UK company named acadaptors-r-us, that specialises in them. It cost me about £20. It's a shame because the very next week, at a car boot sale, I found one from a Creative speaker system. I paid £1.50 for it and it's perfect for the job. Part number MAE180080BH4 just in case you are lucky enough to find one used.
No, they are not. Since everyone uses nowadays switched power supplies, such old transformers are easily found. Or you visit your local thrift shops and take as much as you want at home...
Or you can also use Old AC-DC power bricks.
You just have to rip out the inner electronics
I got mine from a place that sells them specifically for the UK. I will have a look around to see if I can find some in Europe and get back to you :)
Thanks, thrift store might be a good idea. I didn’t see the point of opening up power supply’s to rip out the inner electronics, it doesn’t really provide any safety advantage
@@woosix7735 no I wouldn't recommend doing that for this project.
Merci.
Merci to you too! Thanks for watching :)
All that is missing is the follow up video of this project.
Hi hope you are doing well and safe..it's been a long time
I wish I could like more than once1
You'll have to just like all my other videos instead!
what voltages does a 7915 put out?
Negative 15V DC.
The capacitor is it polarized cap type?
The large ones are polarised electrolytic type yes.
8:38 22 VA means 22 Watt right? So 22 VA Does it mean multiplying input AC 220 voltage by the input current?
Why choosing frequency to be 50hz instead of 100hz because after rectification the AC frequency double
Because for the capacitor each side acts like half wave rectification.
@@TheAudioPhool alright thanks
mains?
класс))) спасибо!!!
I dont undertand how this will be very stable power supply. Zener drop voltage depends on current and current will not be stable here, because Vin in not constant.
This breadboard got realy hot between 15 and 20? (just for the algorh...)
Are you ok? You haven’t made a video in a very long time.
What's this guy's name? I need to cite him in a thesis.
If you're being serious i'm pretty sure you can just cite the channel name. Also i want to read your thesis!
@@TheAudioPhool Fair enough, I was trying for a name because saying “The AudioPhool is an engineering student…” seemed odd. But if you’d prefer using the channel name, I can do that too. (And yes, I am being serious. Thanks, I’ll post back when it’s published!)
@@questionman5 I'm a bit of a private person online. Sorry for being a spoilsport!
10:18 poor power supply getting spanked over and over again…
@electroboomb
Got you over the $200/mo Milestone! 🦾🤖