DIY bench power supply with adjustable current limit

Finishing my small linear bench power supply with an adjustable current limit. The voltage regulation uses LM317, the current limiter uses an OpAmp (MC33171) for better accuracy. Its output voltage can be adjusted from 1.25V to 5.5V and the current from 0 to 200mA. I build it mostly to test logic circuits and microcontrollers, but also circuits sensitive to interference. Using an iron mains transformer and a linear regulator, this PSU produces no switching noise. The transformer was BVEI3048013, later I replaced it using a bit more powerful BVEI3068016. Showing the process of designing it, building it, the full schematic and how does it work.
The previous episode:
• Building a linear benc...
The power supply on my website:
danyk.cz/miniz...
Please support my channel on Patreon:
/ diodegonewild
Instagram:
/ savage_danyk

[@jcxtra]

Thank you for sharing! I think I echo many people's thoughts in that you, bigclive and dave have taught me so much about electronics, I make my own things with atmel microcontrollers now... I mostly try and avoid mains just because it's a bit risky but you've taught me how to diagnose things and given me lots of useful knowledge. I think you're great, way better than any "formal" teacher I've had :)

[@DiodeGoneWild]

Thanks :)

[@benjaminzacharko8740]

I have to agree with this man. I've learned so much more about SMPS technology and the behaviors of MOSFET circuitry through your videos. A classrooms curriculum would pale in comparison.

[@marcinborkowicz2557]

Dave from EEVblog has a great (5-part, I believe?) video about his battery-operated benchlab power supply. It provides enourmous amount of the knowledge of good quality. Now, Danyk shows his own take and I'm so happy to see his project in created his own style, which provides another portion of knowledge. Good job and thank you!

[@marius9429]

Fr tnx to Danky and all the others I found my passion. One more year at highschool and I'll go straight to collage for electrical engineering. I can't thank those guys enough for their wonderful explanations with schematics and measurements. ❤

[@jcxtra]

@@marius9429 Good luck with your college in the future! :)

[@arduinoatolyem2121]

Its a 100% green power supply because you didnt even used acid to make the board

[@DiodeGoneWild]

It's also green because it's not built to end up in trash a year later ;)

[@Debraj1978]

21:45 = "Understanding the importance of fuse indcates your level of wisdom; on the other hand understanding the circuit indicates your level of knowledge and intelligence.": DiodeGoneWild

[@ZaneDaMagicPufferDragon]

Bloody Laborious lol 🤣 But Highly Educational & Entertaining Too!!! Thanks 😊 for sharing this with us and for making sure you cover everything you need to regardless of how long your video gets because it’s worth explaining it. I have learned so much from you and Big Clive and Julian Ilett and Fran Blanche and Dave at EEVBlog and so many more!!!!

[@goodwill7643]

17:21 - I heard that dust from fiberglass is not very healthy. Keep it in mind and take care.

[@DiodeGoneWild]

It's not easy to find something healthy these days...

[@goodwill7643]

@@DiodeGoneWild try to cut it with knife, then break it. But it is up to you, just wanted to warn you. :)

[@DiodeGoneWild]

Ok, maybe a knife next time ;) or a hand saw, it doesn't produce such a fine dust. Or maybe sheet metal scissors.

[@Broken_Yugo]

Silica dust, not quite asbestos bad, but bad enough to avoid.

[@jackofalltrades3727]

​@@DiodeGoneWild we don't want anything to happen to you , we need these videos to keep coming

[@rpbajb]

Love the botched circuit board. Hey, it works!

[@amkp40technology]

*18:57** 🤣🤣🤣👍🏻👏🏼👏🏼🙏🏻❤️❤️ Love You Sir ❤️.*

[@stephengray9207]

I am constantly amazed by your skill and expertise. I've been an electronics tech for 40+ years, and your skill is wonderful. It reminds me of me. when I worked with hundreds of others who did electronics for a living...and a couple of us (like me ) did it as a passion. If you want to be a superstar in electronics, then don't just follow this channel... support it on Patreon. even a dollar or a few a month is well worth it, Watch and digest everything, such wisdom and clarity. Even I've learned a lot... & I've been a tech for 40+ years!!!

[@richardhalliday6469]

Marvellous - the not printed circuit board looks like it was drawn by David Hockney - perfect - a sturdy and repairable PSU.

[@fredflintstone1]

great video definately "Not Dodgy" keep them coming ;-)

[@jimadams7765]

That was an excellent tutorial mini-series. I learnt a lot from it. Thank you.

[@DumahBrazorf]

Loved how he dipped the transistor in the heat transfer compound like a potato chip in the mayo, rotflmao!!!

[@gacherumburu9958]

Informative and entertaining. Thanks 👍

[@danielmantione]

PCBs a la Danyk! ;) No CAD tools, no gerber files, not even etching by hand. Real men mill their PCB!

[@pe1dnn]

I can see why this is not sponsored by a PCB manufacturer... Great job.

[@Alchemetica]

Love the post apocalyptic circuit board design.

[@crossan77]

I really hope you were wearing a mask when you cut that board!

[@DiodeGoneWild]

Hell no, we've burnt all masks several months ago...

[@t1d100]

Excellent. Thank you. I learned some new things.

[@unclepecos5426]

Oh, I like the way you build the PCB 😃 I also noticed that I laugh a lot while watching your videos. They are entertaining and educational at the same time. Thanks for your great effort.

[@TheInventor202]

Thank you for the video

[@omsingharjit]

You could made using smps or can Build by modifying 12V 10 Amp smps

[@edgeeffect]

I wish I'd thought of your pad of wet towel to keep the shaft cool whilst cutting it.... I'd have missed burning my fingers so many times. :(

[@Speeder84XL]

Really nice! Very interesting and well explained about having high voltage drop on MOSFETs and why "safe operating area" differs from the numbers derived from maximum power dissipation (I didn't know about that). I have mostly used MOSFETs for switching (unless in very low current applications), so I havn't really run in to the problem. But it's still good to know if I some time get the need to operate a MOSFET in linear mode at relatively high current (for example to make a simple design of something where efficiency really doesn't matter - like a variable load for testing power supplies).

[@liam3284]

using them as switches, you run into trouble when they de-saturate.

[@Speeder84XL]

@@liam3284 Yes, but if desaturation during switching happens, it's usually caused by a fault or by the gates getting too low voltage (usually caused by the whole circuit getting too low input voltage). If that happens, it means the rated "maximum power dissipation" of the transistors is often exceeded as well (and may lead to failure anyway). But, this fact that they may handle less if the voltage drop is large, can sometimes be of interest for a switching circuit as well. When for example evaluating the need for an "under voltage lockout". High power circuits pretty much always need it, since the power dissipation will easely reach destructive levels if the transistors go into linear mode. But if it's a low to medium power one, the transistors can sometimes be allowed to go into linear mode if the input voltage get too low - since the power dissipation will stay below damaging levels anyway. Which allow for a simpler circuit with less components. But it can be good to know some more margins may be needed, depending on how large voltage drop they may be subjected to during that condition.

[@WagTsX]

interesting, I wonder how much is the time between CV to CC like when you set to like 10mA - 5.5V and then power a red LED directly. If I do so with my Riden's bench PSUs, the LED sometimes burn, sometimes not.

[@igrewold]

Red color LED has a voltage drop of 1.8 (or ranges 1.7 to 2.2 ) volt

[@jankomuzykant1844]

I think it mainly depends of output capacitor which slows down CC timing

[@DiodeGoneWild]

There's always a high current pulse when you connect an LED to an already running bench power supply. The output capacitor discharges into the LED and also the current limit has a delay (the sensing circuitry, the power transistor...). The safe way is probably to connect the LED and then power the power supply, or to use some series resistor, despite the current is then limited by the psu.

[@Mark1024MAK]

Actually the safest way is to short out the output terminals, set the output current limit, connect the LED in parallel with the still shorted output terminals, then disconnect the short circuit.

[@fluffyblue4006]

Nice video, good explanations. I didn't know about that mosfet hotspot problem, yet. Thanks. When building the bigger one, will you also be using relays for selecting appropriate secondary taps of the transformer, in order to keep the dissipation down in the regulator? With that, you could create a 0 to 24V 3A power supply without the need for a big heatsink. Please do, because else it would just be an upscaled version of what you did in this video.

[@FerrybigGaming]

You could do this with an potentiometer for the voltage where you have access to the shaft. Just 3d print an inner wheel, which has a thicker section that activates a micro switch, which switches to the higher voltage tap. This would not work for the current limiter dissipation

[@CamelCasee]

18:19 that explains why I've never been on a date.

[@16BITMEME]

😱😍😍😍😍 Thanks for awesome Video DANYk please make more diy video like this.... I love your Long Videos ..... Nice explain

This makes me laugh.😆😂

[@martinda7446]

I really enjoy drawing my circuits out by hand and etching them, they can have all sorts of added bits for fun. Cats included. Edit, I've been doing it that way (for me) since 1978.

[@Katanium]

Amazing content I love watching your videos they cure my stupidity! 😊

[@phenomenon9114]

Thanks for sharing with us ❤️ Can l put a power transistor on the out put of the lm317 to a make it more powerfull.

[@DiodeGoneWild]

LM317 datasheet shows couple ways of how to increase its current capability using an external transistor. At least some versions of the datasheet. But for more current, you might also have a double opamp and just one transistor, that both limits the current and regulates the voltage.

[@Broken_Yugo]

Very common trick, plenty of examples in linear regulator datasheets and around the web.

[@phenomenon9114]

@@DiodeGoneWild thankyou sir for reply on my comment u made my day

[@poptartmcjelly7054]

You could use a double op-amp to control the MOSFET to adjust both current and voltage. This way you would no longer need the lm317.

[@DiodeGoneWild]

Of course :) but I decided to use LM317 because it's ready made, simple and almost guaranteed to work well. In very small power supply, this makes sense. Bigger linear power supplies of course would use just a single power component, not two.

[@plainedgedsaw1694]

Or make positive side current sensor and make it pull the adj of 317 to -1.25.

[@SaltyPuglord]

One thing you get "for free" with most LM317s, is over-temperature protection. It's a nice "free" gift that would otherwise require designing and debugging some additional circuitry. Using an opamp will get you more precision on your output voltage, but using a 317 with overtemp protection will get you more safety overall. In a power supply, I'd rather go safer. (Most power supply rails are only supposed to be accurate to ~3% anyway.)

[@rilosvideos877]

It seems you really think about everything that could happen to the circuit to make it waterproof! Great knowledge and teaching as always! Could you please do such a diy power supply as a switch-mode PSU with much higher power, lets say 0 - 50V / 0...10 A? It could also be a modification of a computer- or server-PSU, making the voltage adjustable?

[@jg6780]

Instead of using dremel to remove copper you can cut some pcb patches using metal shears, and then glue them with super-glue to pcb plane. It is fast and clean.

[@Uraim]

does it work with higher voltages? like about 20-25volts dc on bridge rectifier? The mostly needed thing is the current limit

[@windward2818]

Maybe you can comment on the wild mains voltage specs for eastern Europe that I have to meet in universal power supply design. The specification I see often for universal power supply design gives a high line mains voltage of 265VAC @ 50Hz. Very similar to the UK high line. P-Channel MOSFETs have come along way in terms of Rds on. For example the infineon IPB110P06LM is 16milliOhm @ -4.5 Vgs, -91Amp (The SOA looks OK).

Amazing, truly amazing! Thank you very much.

[@RanaRao_Chandrachur]

Next project *simplest but yet powerful smps using NO FANCY CHIP* 🙏🙏 SIR, please cover this topic with all calculations in a *BLOODY LONG* video. 🙏🙏

[@128Wojtek]

Thank you so much for this video! Thanks to your film I've finally built a power supply with adjustable current limiter (of course with variable voltage) 🤠

[@omsingharjit]

Plz make device for OCD Peoples

[@nevillewaiting485]

The data has to be shown "scientifically" or log 10 or it wouldnt fit on the scale mate, as I'm sure you know🙂 the sheet would need to be 100 times bigger, so as any usefull information could be displayed. Also, as I'm sure you also know, it's a very common scale, you ask your cat, she will explain things to you 🙂Kind regards, Neville

[@markonikolovsk22]

Diode: I like how the layout of the bridge rectifier came out. Meanwhile... Me: that looks like the symbol that shall not be spoken!😆

[@UnulOarecare]

A quick question please - I replaced the connection between the motherboard and the power supply - the new connector has a white wire and the power supply doesn't - any idea why is missing?

[@ohmware2020]

hi, i have all the same parts except for the transformer, which i have 28-0-28, can i use it? will the design alter if i use different transformer?

[@d.k.9406]

Danke! Thanks again for more than just one way of adjusting and on top explaining the way how they regulate!

[@Gengh13]

Functional but it would definitely go in the dodgy category.

[@sanjikaneki6226]

Why not use a pmos on the high side and tge sense resistor also there? This way your ground would be cleaner . Also why do you always use old components? Yes for mosfets you said why but I don't understand why you do it for op amps and ldo s too.

[@omsingharjit]

Once you promised to convert your 24 v 10 amp into 30 volt adjustable powersupply i wish you make it soon

[@EasyOne]

The hardest way to make pcb

[@michaelseitz8938]

Beautiful 😊

[@tony42hajdu]

excelent asmr video! love your sense of humor

[@NaoPb]

Ah,, just what I was looking for. Thanks!

[@sanjaymogiya6722]

Your way of speaking English is amazing 😍

[@MC-01]

I was a bit bothered on the cut out from the bridge rectifier came out, when you take a close look at it you'll see it

[@Phil8sheo]

LMAO, I was about to give up trying to figure out wtf you were talking about. I promise you, no brown shirt sympathizer running this channel, just a coincidence.

[@lorenzorentniop717]

You are the best!💪

[@Yorumcu63]

You are excellent in electronics. But I was shocked to see that you made pcb paths with a drill.😯 A PCB pen and some hot Fe2Cl3 were enough No doubt you know that. But this is speed result😀

[@technixbul]

Try proteus isis for schematics and PCBs, you can even virtually test whole schematics and MCU programs on it, its the perfect program

[@Paxmax]

I rate this video extremely Undodgy with a nice amount of bodge that warms the builders soul... nice anti-OCD... test of character rite there, hehehehehe.

[@martinda7446]

That was wonderful. They just keep getting better. (The videos) (and the animals).

[@janzahradnik8089]

Can you make a video where you make a switching power supply in to 10A and power 100W? And calculating a pulse transforer?

[@Mark1024MAK]

The ST L200CV is/was a nice voltage regulator with built-in adjustable current limiting. But it was always more expensive than a LM317T and apparently ST have discontinued the L200CV now ☹️

[@omsingharjit]

14:20 this is going to be what...? Asmr

[@iceberg789]

neat.

[@bdf2718]

You put on too much heatsink compound. It has a thermal resistance, but it's a lower thermal resistance than air. It's meant to fill in microscopic pits and scratches in the metal surfaces where otherwise you would have air. Spread it on thickly and it prevents metal to metal contact.

[@FulvioGa]

We cannot know how much safe this power supply is, because to know that we should destroy the transformer.

[@hernancoronel]

Awesome Daniel! Thank you for sharing so much knowledge again!

[@jp040759]

EXCELLENT. Glad you are not OCD. THANX for the great effort!!!!!!!!!!!!!!!!!!!!!!!!!!!

[@MissNorington]

Finally, a DIY linear power supply on KZbin! Yay! I have waited for years!

[@chaitanyabhave5255]

Bloody ling but still very useful full of fun than typical Time laps videos like other

[@ucmetoo1]

at about 24:00 you can add several diodes in series to the base of the BC327 transistor to turn the LED off. alternately you can add a low voltage zener instead.

[@Pirelli.]

Another great video!! Thank you Dany!! God bless you! 😊

[@liam3284]

Next step, a low voltage PFC circuit after the bridge rectifier.

[@artursmihelsons415]

Great video, as always! This building method I call "Quick And Dirty".. 😂 I'm used it by my self, even in repairs, when replacing original transformer with PCB soldered one and created PCB by available place..

[@omsingharjit]

21:40 but i have ocd very high 😒

[@miftosaurus]

... waiting patiently for 24V/2A version. :D

[@cups3]

God do I wish you had been one of my instructors. You are excellent on teaching the practical real life way things work. I left Tech school with two years of a lot of theory , only to find out in my first few years, "It isn't necessarily so". PS: Circuit board is excellent ! 👍

[@TheSalemSaberhagen]

Všechno je perfektní, ale za tu DPS bych se styděl.

[@xntumrfo9ivrnwf]

How do you create the tracks for the 8 pin IC? Surely they're too small for a Dremel?

[@zaraak323i]

That's hardly an Usagi Electric board, but it works!

[@e.a297]

Please make a video about synchronous power supplies

[@andymouse]

So much useful info wonderfully presented, can't thank you enough....cheers.

[@therealjammit]

I had an idea: What if you use a current transformer on the the AC current right before the diode bridge to sense the current?

[@JohnZmith]

Grinded Circuit Board ! / GCB

[@albin6382]

Why not just rectify mains and then regulate from that?

[@imnotbeluga007]

Well, good luck regulating 325V (peak of 230V RMS sinewave) down to low voltage values. At 1A, 12V output you would be dissipating 313W of power as heat, which is impossible to dissipate in such a component package. Also, the regulator would go short-circuit at just over 40V, and this is over 300V.. Switch mode power supplies use a different technique, with rated components for it. What he built is a linear power supply.

[@VIPINSAINI_20]

25:50 15V 0.5A is good to see But I need a 30V 5A Dual Rails version 😑

[@hall117]

Muito top... eu vi um circuito com controle de corrente para LM317T com apenas 7 componentes (1 LM317T, 1 BD135, 2 potenciômetros, 3 resistores) ,,, no Proteus ele funciona até 1A tranquilo... seria muito top analisa-lo ... como não posso deixar links vou colocar somente o nome do vídeo e do canal... ( controle de corrente para lm317 lm350 lm338t Elizeu Silva ) Valeu...

[@crono331]

As my old boss used to say, is not elegant but it works

[@Delali]

This is really cool. Thanks so much

[@sforman2622]

Je škoda, že tomu tak rozumíš, ale provedení desky je příšerné. Hrozný bastl. Já mám opačný problém. Desky a osazení jak víno, ale takto do hloubky tomu neholduji, že bych i navrhnul něco :-D

[@imnotbeluga007]

I broušák někdy stačí, a netřeba žádnou chemii..

[@L2.Lagrange]

Really awesome video

[@englishrupe01]

Thank you so much!

[@gkolko]

congrats bro. Welldone.

[@lupoal4113]

amazing!!!

[@mohammadrezamim272]

I'm a little confused, Or I missed the point you explained this. Why you placed current limiter before lm317? In that point it doesn't measure actual current going into load but measures load current + lm317 consumption. Or I'm wrong?

[@argcargv]

If you did it the other way around you would loose voltage regulation performance. That is voltage regulator usually has a fairly consistent and small bias current, so the issue is much smaller for current regulation.

[@Mark1024MAK]

Did you watch his previous video? He explained it in that video.

[@bandit4550]

looks good

[@localgamingmaxyt7929]

Your cat is very very cute ❤️❤️

[@msansjr]

Double video, doubly bloody awesome!

[@keithking1985]

👍🇮🇪

[@DiaconescuAlexandru2024]

06:48 I feel very called out 😅

[@igrewold]

Cool man Thanks