DIY 1 - 15V bench power supply

Building a small linear bench power supply with an adjustable output voltage (1.25 to 15V) and current limit (0 to 350mA). Choosing the transformer, explaining the schematic and how does it work, some tests.
This power supply on my website:
danyk.cz/miniz...
The 5.5V power supply:
• Building a linear benc...
Please support my channel on Patreon:
/ diodegonewild
Instagram:
/ savage_danyk

[@Langendimi555]

Perfect timing for a Friday evening. Better than Netflix 😅

[@catalinbadalan4463]

Anything is better than Netflix.. :)

[@imnotbeluga007]

Especially when it is free to watch. Either with a skippable ad or two, or none for modded YT client users.

[@adamjuszczak7437]

When i see his new video about making something it always makes my day

[@tigerelectronics5966]

Yesss!

[@ZaneDaMagicPufferDragon]

I love ❤your bloody awesome long videos because I always learn more about ACIENT Technology 🙃 I love when you scratch a circuit board literally from scratch, scratching away the copper and making tracks and vias and stuff like that it’s awesome 😎 to see!!! Please keep making bloody long videos and I’ll keep watching here and on Patreon indefinitely!!! 🥳🥳🥳

[@DiodeGoneWild]

Thanks :) I'm really glad that my unconventional videos and techniques are appreciated ;)

[@simonowen488]

@@DiodeGoneWild 10,000%!

[@vermillionreaper]

@@simonowen488 1,000,000,000%! actually

[@sklepa]

Love this build videos ❤ Long durations is an advantage not obstacle. Keep it up Danyk!

[@hullinstruments]

IT WOULD BE AWESOME.....if you made a video teaching the process of drawing and reading schematics. From simple to intermediate / advanced. You describe them better than anyone I've seen and would love a long-form video talking more about schematics. Specifically showing how to trace the current path on the circuit diagram. A lot of folks don't understand how the current travels through a circuit, and your diagrams would be the perfect opportunity to teach that particular aspect. Accompanied with drawing / reading diagrams

[@Scyth3934]

I agree!!!

[@deterdamel7380]

Your creations have always a very special style. It's not only your cat likes this, me too.

[@rich4239]

excellent....really appreciate the full explanation of component selection and heat sink selection

[@lolsypussy]

Your parts are from the soviet era. Amazing to see such contrast from what most similar content creators use nowadays 🙂

[@cube.in.]

Welcome to the Czech Republic :-D

[@moeinsp2027]

👍

[@yellowgreen5229]

This is the problem of the cold War against the workers. Capitalists are like the mafia, if you don't do capitalism they punish or murder you. DemocraticSocialismNOW

[@Mladjasmilic]

Communist era, not Soviet. Btw, word 'soviet' in Slavic languages means 'advisory' or 'parlamentary'.

[@richardlighthouse5328]

@@Mladjasmilic In English language, it is soviet.

[@BusyElectrons]

I also appreciate these longer build videos. Your explanations are very clear to me and I seem to always learn something by watching. Thank you!

[@d.k.9406]

Thanks! I like your bloody long videos most, because of more information and detailed explanations. I like all sorts of powersupplies too. I missed up today a discrete Linear powersupply, as COUNTERPART to the old switching single transistor phone charger. In fact i am interested in something from 24V AC input and 3 to 5 A output Have laying around some medium Transformers For Example my favourite Ringironcore with 6.67A output Good compromise in Power/weight/price/size

[@DiodeGoneWild]

Thanks ;) I'm really glad that somebody appreciates long videos over some 3 minute botches where the author doesn't even explain anything he's doing.

[@anonymoususer6448]

I really like the calculations of the heat sink ! As always on this channel: I learned something new again :-) Love it

[@ZaneDaMagicPufferDragon]

Me Too!!! 🎉🎉

[@jp040759]

I liked the way you tested the characteristics of the transformers. I never thought to do this kind of testing before choosing one for your application. .👍👍👍👍👍 I will definitely use the technique with the scope to test potentiometers in the future. Great idea.

[@fabiocattaneo9434]

Nice to see that somebody else uses this technic for building prototipe circuits pcb (... I believed to be the only one). I learned this technic a long time ago (ok,not so long... maybe 40 years:I was a fresly graduated young engineer) when I started to work for a telecommunication equipment company and I was sent to a tour in the prototipe building department and I saw some dental cutter (I am not joking ... "dental cutter here...for what????"). I was explained that the usage was for building prototipe pcs and they showed me how to cut traces into copper sheets. If you have enough skill (..not all have) is it possible do make nice works. Obviously do not pretend to build a controller with 40 or 80 pins chips or use smd chips, but for buildig prototipes with some 2.54 mm ics and insertion components it works nice and (this is a bonus) you can also use two layers copper sheets and have a large ground layer. Now I use exactly your technic at home for the circuits I buid at home for hobby, when it is possible to avoid to send a gerber file to China and whait for a "professional" pcb).Thanks for having unblocked the memory.

[@andymouse]

I never thought about visualizing a dirty pot that way, thanks !...May your salt flow freely, cheers.

[@AlexanderBukh]

genious, right? even if it is not enough to hear, you can osc plot it however fine you want

[@gordonwelcher9598]

This will be a nice power supply for the breadboard. An nice description of the design and operation, very educational. Linear power supplies seem simple but there are a lot of factors to consider. The diode voltage drops were not accounted for in the transformer calculations. Schottkey diodes are very nice to use but remember that some types have a bloody low reverse voltage breakdown. Your circuit descriptions are the best.

[@mikefromflorida8357]

The longer videos are always better. Keep them coming please.

[@piconano]

I love your long videos. I don't bother watching short tik tok style videos .

[@jms019]

I love his PCB construction technique and of course the finger remover.

[@stark_energy]

15:52 wow the PCB is artistic and archaic!

[@clemensruis]

That "slow blink" your cat did is how cats "smile" at us. Well, I guess in this case s/he smiled for the camera. 😀

[@CJWarlock]

Cool build. :) I appreciate the skill to build something fast and decent looking. So many years and I've never thought to put rubber seals to distance the potentiometer from the case. Ingenious solution! :) I've only used metal washers or a second nut. Cheers! :)

[@DiaconescuAlexandru2024]

It's always interesting to see your "peculiar" PCB manufacturing that ends up looking very nice. Although seeing an unsocketed IC hurt me a bit :(

[@houmer125]

Pan Kočičák s hláškami je skvělý :D

[@smoothjamie4046]

@11:40 to measure the junction temperature of the regulator use thr RthJL from the datasheet, and meaaure the temperature of one of the legs. The legs of the transistor are connected to the die. The final temp is the sum of the temp rise across the junction and the leg.

[@oddball119]

The way you created the pcb is really interesting. I guess it is easier than having to design one and then have it printed and shipped to you.

[@EJEuth]

WILD Video Proposal & C#5: Repurposing Unknown transformer. Posters/authors comment that nominal current or VA of an Unknown transformer (not those of yours still with some labeling) could be “guesstimated” by plotting AC Voltage vs Current The increase of AC consumed power is done by increasing loads (for instance: shower resistance, or salted water solution and metal plates as variable distance submerged electrodes. They claim that about 10% to 20% voltage drop (sometimes being called ‘transformer droop’ ?!?) is the limit, but I did not see this being done with temperature monitoring too. Others, estimate the VA by measuring the cross section of the central iron core, in cm2 and correlating to the maximum VA by some empirical formulas. So, there might be plenty of ideas to evolve in something that every country and user may face, sooner or later. Thanks again for sharing - always nice to see your channel!

[@OldSkoolF]

Love your videos and humor. Should do some videos on how the different windings in transformers work and effect each other. I have seen some with copper foil windings and don't understand how its different from thin wire. Thank you.

[@odieadog4086]

If I understand correctly, in the current regulation mode, this supply always has about 5mA rising current slope, when the output voltage goes from the set voltage down to complete short circuit. This comes from the fact that we are regulating the current of the voltage regulator together with the output current. So when the circuit goes into current regulation, the voltage regulator looses the abilty to regulate output voltage so the current into adjust potentiometer is not a constant 5mA any more, but slowly dropping towards 0mA as the output voltage is nearing full short circuit. So the 5mA that usually flow into voltage adjust potentiometer is eventually flowing into the shorted output. Since the slope is always 5mA, we notice this more, when setting low currents, say 10mA, than when when we set higher currents, say 100mA. Which is, of course, totally acceptable for current limiting, but one should know, it's not meant to be a good current source.

[@wrrr10g53]

Excellent video as always. I really like when you use heritage Tesla parts from the Soviet days, awesome. Thanks Dan!

[@mariushmedias]

Use a transformer with two secondary windings so that you can switch between them to have lower voltage drop (why 15v in - 3.3v out on regulator when you can have 8v in - 3.3v out). Bonus, you can parallel the secondary windings to get double current on secondary, but lower output voltage range. Often as little as 10-20mA is enough to put some load on transformer, or you could add one or two 1n400x diodes to drop around 0.7v per diode if the maximum voltage is too close to regulator maximum. Up to 11 minutes in video, no mention about capacitor after rectifier to smooth out ... a reasonably value is around 2200uF per A of current... or approximate with formula C = current / [ 2 x ac frequency x (Vdc peak - Vdc min desired) ]

[@zsomborgameplayoffical]

Good idea

[@Broken_Yugo]

Not sure a tap changer is really in the spirit of a power supply using a 7812 as a reference.

[@EJEuth]

C#2: Organized Repurposing of components - Nice… Knowing how thing were/are done in Eastern Europe = nice too! C#3: Comments of potentiometer running under constant current and power dissipation. Good highlight, rarely seen in other videos.

[@michaelfisher9671]

Never too long!

[@Antek1234l]

Very simple and nice schematic! Well done!

[@user.A9]

Missed you! Welcome back!

[@vanhetgoor]

The heat-sink INSIDE the box, that is a new concept.

[@akhilesh7313]

Thank you for the video, learning a lot....from you, please requesting you to upload more frequently 😀

[@moeinsp2027]

Nicee there's a long video from my favourite content creator engineer 🤙much respect 👍

[@HugforYou]

Another great project! I love this channel!

[@eeezzz9800]

Amazing Video! Good Bench Power Supply!

[@sabaubogdan6997]

Wow the jlc-pcb diy pcb kit looks rough but im sure they will come through. I use this pcb technique quite often after i got sick of perf-board and didn't want to order pcb-s from china or to use some kind of dangerous chemical corrosion. I actually learned about it from you. Then i found some other ones on KZbin. The Dremel is too rough for me and triggers a bit of ocd. I use a scoring knife and ruler for a nicer result. But if it works it works.

[@plainedgedsaw1694]

You can find thermal resistance in the datasheet, for LM317T (from ST) it's 3 °C/W for TO220, funnily enough, it's 4 for the thicc TO3. So in your case, die is about 18°C hotter than the heatsink, unless the thin tab has bent under the pressure from the screw and flopped it's bottom half away from the heatsink.

[@robertneill3057]

The power dissipation is 5W higher for the TO3 package (20W) when compared to the TO220 package (15W).

[@d.t.4523]

Keep working. Good luck! 👍

[@hernancoronel]

Great video Dani! What does your wife say about the mess with the dremel? LOL! Thank you for the video!

[@zx8401ztv]

Nice little psu, i was shocked seeing you make a small linear supply. You have made massive switchmode beasts that could power a jumbo jet easy, long mains cable though :).

[@DiodeGoneWild]

Sometimes I like to keep it simple ;). To test low power loads, linear power supply is fine. The short cable with crocodiles was just for test, then it got a 3m long cable ;).

[@EJEuth]

C#4: Transformer selection among existing used ones = nice to see highlights of protection, fuse or PTC. I found this somehow detailed, but still brief - paradoxically, I know. Detailed when you compared amongst 3 options, interesting to know the short circuit current, but I felt missing the voltage under a given load, say 12V 5W and 10W lamps (car lamps would work here), under AC voltage measurements.

[@shafiahmad6795]

Welcome back 🔙

[@W1RMD]

It's funny, I was just thinking I haven't seen one of your videos in a while. A current/ voltage limiting supply is a MUST for every tech. Unless you NEVER make a mistake. I'm equally interested in you really cool VFD temperature display. Do you have a video on that? Thank you for sharing!

[@DiodeGoneWild]

The thermometer is this: kzbin.info/www/bejne/j5ucZoCumd2Undk sadly, the display is just green LED (525nm). If it was a VFD, it wouldn't run for months on a battery ;).

[@nubie1100]

1:57 ... my favourite 🤣🤣🤣too high to beee practicauuuulllllll

[@cda32]

I love seeing projects that don't need people to first build a reflow capable electronics lab and design PCB gerbers to acceptable fab standards.

[@rkan2]

A lost Art!

[@amkp40technology]

*Very interesting explanations 😀😀😀👍🏻👍🏻👏🏼👏🏼🙏🏻❤️❤️🔥🔥👏🏼👏🏼*

[@SINHRO-FAZA]

Было бы интересно взглянуть через осциллограф на шумы, которые будут на выходе этого блока питания, при разной нагрузке.

[@labiadh_chokri]

You can add a digital or analog modul to show the voltage and current value.

[@danielraducu9073]

you are one big piece of a beautiful man, thank you for your care and great interest in helping others, i love the fact that your videos are in great detail and i will definetly build this summer this power supply, i want to use it for recharging rechargeable batteries that can take only a current limit and no more to increase their life span, personally i am kindda of a radio guy, my biggest dream is to learn to DESIGN and BUILD a superheterodyne, AM modulation for a remote control 27Mhz, i was not able to find this type of knowledge on the internet yet, i want to be able to compute the filters, why this number of turns, and why this AWG, I want to understand why the IF filter is sort of a small transformer and not just sort of an RFfilter(1 coil + 1 cap), i want to understand the purpose of the tap in some IF filter, i surfed the internet left and right and all the superheterodyne videos and websites speak theory, freq component theory, what it does but nothing about DESIGNING one from scratch, sometimes it feels like there are some secrets that have been under the lock and key, i ve almost checked most of the roumanian literature on radios/ radio schematics (1950-1989) and some western radio books but i still was not able to find what I seek. I just seen your channel early this morning and saw your repairing old radios and i thought for a second that you might be the guy to help me with this, and show to your community too this type of challenge. Please , please , please make videos on this topic, or radio projects, building too not just repairing( repairing someth needs more knowledge than learning how to make one from scratch i think...)

[@EJEuth]

Very interesting sharing, thanks! Few comments, piecewise: Hint - solution is in the name DIODEgoneWild C#1: Headroom estimation is done at short circuit and No Load voltages, BUT… #1A This is not true, as in low CC values, voltage drop is consumed by the MOSFET, not the linear regulator, that would still be Ok. #1B At Full current, (mosfet not limiting current)m Vin-Vout might be only for fully charged capacitor, then, showing output. #1C Otherwise, capacitor voltage will always be lower, due to transformer loading. Mitigation for risk at 1B: 2~5R power resistor at input of LM, and/or 2~4Diodes (used wildly) in series, at Vin of LM 😊

[@labiadh_chokri]

Nice video, how you measured the short-circuit output current without blowing the internal fuse of the transformer.

[@German_byte]

In Australia we call the steel sponge a “curly girl”. 😂

[@German_byte]

Thanks

[@DiodeGoneWild]

Thank you ;)

[@snakezdewiggle6084]

Great project / video. Thank you. ;)

[@odieadog4086]

Thank you for a nice video reminding me not to overcomplicate every design I try to make. However, did you test the stability of voltage or current regulation with different capacitive loads? Does it ring excessively or oscillate with any? Also opamp driving a mos-fet with a huge input capacitance usually requires some frequ. compensation in the feed back loop - were you lucky that it did not tend to ring a lot, or is it something about the particular configuration that makes it stable?

[@DiodeGoneWild]

It seems to be stable as it is, I didn't have to do any effort specifically to make it stable. It might be because the opamp isn't very fast.

[@birtanerkanli8693]

Thanks!

[@worroSfOretsevraH]

Are you going to review/test Lidl "intelligent" car chargers? Would be interesting.

[@uwepolifka4583]

The voltage regulator potentiometer needs its contact to ground to work. I think when the mosfet becomes active it will cut off the ground from the voltage regulator.

[@DiodeGoneWild]

it cuts off the whole regulator, so it's not a problem.

[@Killerspieler0815]

Add a MOV etc. to the primary side of the transformer , as heatsink maybe use a 20 years old all Aluminum CPU-cooler (like for 60 Watts Athlon-XP Socket-A ) with a fan ... I wounder how hot this heatsinked sillicon is in summer (my room gets maximum 43°C ambient) ...

[@t1d100]

Excellent.

[@goggymagogy8911]

@DiodeGoneWild You forgot feedback resistor in opamp, to limit voltage amplification gain. Also, you forgot tiny capacitor there, to avoid unintentional self-generation.

[@DiodeGoneWild]

The capacitor in a feedback would make sense, to limit the gain only in a short run, but to keep it infinite in a long run. Or a resistor capacitor series combination as a negative feedback. I actually thought it will be necessary to add it, but it turned out to be stable without it.

[@goggymagogy8911]

@@DiodeGoneWild In a tons of power supply schematics it present, that's why I think it is necessary here too. May be You have "play" with different loads or "long wires connections" to obtain mode where it starts self generate.

[@Its-07]

Amazing 😅

[@ruimvp]

By my experience I am not convinced with the current sensor. It must be a super hyper special opamp to deal precisely with that threshold and voltage difference. But my days of electronic artisan had passed. My last diy project was in 2018 when I built a circuit that commutes automatically the input of home devices between the solar inverter and the network in function of the battery voltage. Above 13,5 V solar and under 11,5 V network.

[@plainedgedsaw1694]

Most linear psu's made in 60s and 70s used transistor Vbe as current sensor, 723 does that too.

[@ВіталійГрабчук-р5я]

"the board came out nicely" - man it looks medieval :)

[@Katanium]

Very cool!

[@aicisha]

11:28 you can calculate the junction temperature if you know power dissipated, the temperature of the heatsink and Rth j-c from datasheet of the regulator. For example: Tj = 80 + (3K/W) * 6W = 80 + 18 = 98C, right?

[@DiodeGoneWild]

Of course, but there's also the case to sink thermal resistance, it's much harder to estimate. And the insulation pad.

[@aicisha]

@@DiodeGoneWild you are right, forgot about Rth case-heatsink ☹️

[@user-dw6fj1py1o]

Cool!👍

[@Fajardo3]

Sir can you please make a review of battery tester namely topdon etc... thanks 🙏

[@fargodandy7827]

Build switching one without noise!!

[@4L1LAT]

I have question. What is operating frequency of this power supply . And can I use it to radio Sistems?

[@joejane9977]

the cat has the best input

[@lmwlmw4468]

Nice.

[@alexloktionoff6833]

Nice! :D

[@zsomborgameplayoffical]

I like this content!

[@AdamV20]

I built power supply based on lm350, and when it's loaded to much/overheating, it cycles (oscilates)

[@janzahradnik8089]

Have you got some nixies IN1- or IN-12A?

[@diyelectronics5375]

Can you make a video nixíe tube clock step by step i have a nixíe tube N12 and i don't how to build in.

[@AugustineAriola]

Where did you learn electronics ? From what school ?

[@অৰি]

Where do you get those project boxes from??

[@tagalogelectronics6615]

😎

[@c.schandra2134]

Can you build smps based laboratory power supply. I have 12v power supply 60a but don't know how to make lab bench power supply with it.....

[@ElectronPcRepair]

How can I upgrade this or my 15 vDc power supply to 30v 5 Amps?

[@ElectronPCFix]

How can I add short circuit protection using relay

[@volleswerkfullorganpower249]

where have you been? 🧐

[@iceberg789]

but, is this ok to put current limiting before voltage regulator ? if less input voltage is fed, will not lm317 misbehave , and may be, may even get a short service span ?

[@plainedgedsaw1694]

There's application note in (TI or natsemi i think) datasheet that uses second 317 in current reg mode before the voltage regulator. And they called it a "laboratory power supply).

[@iceberg789]

@@plainedgedsaw1694 but there, a fixed -10v applied for adjustment biasing. so that ic never runs out of voltage difference. here, where you are controlling current, voltage difference will come very low in adjustment pins during current limiting conditions.

[@plainedgedsaw1694]

@@iceberg789 the -10V is clamped to -1.25V. That's where the bottom of the divider goes. That's also the voltage that will be present between any part of the regulator when current limit kicks in... And is further divided by voltage setting divider. Say if you have it set for 0-10V variable, it is set at 5V, voltage going into the adj pin will probably be around half of the 1.25V.

[@iceberg789]

@@plainedgedsaw1694 here in this video ? i thought mosfet may take it near to 0. otherwise it ill not be very effective current control.

[@aicisha]

12:25 you do not have a cheap bench vise? :(

[@DiodeGoneWild]

I have a vise in the garage, but I work in my living room.

[@aicisha]

@@DiodeGoneWild watch your fingers! 😵

[@nigruskaja]

What can I use instead of FQP33N10? IRF540 is good? Nowhere can buy this mosfet, alli is 90% fake, and long delivery time.

[@DarrenDignam]

Upvote for cat

[@miftosaurus]

... termometrul ăla cu mercur de la final face toţi banii!!! :))))))))))))))))))))))))))))))))))))

[@objection_your_honor]

If you lookup Russian MacGyver, you'll see a picture of Danyk!

[@jendak7921]

What? :-O He is from the Czech Republic.

[@objection_your_honor]

@@jendak7921 Czechoslovakia was a satellite state of the Soviet Union.

[@yousouf4047]

Can you make a high current one with 2n3055 transistor

[@brucepickess8097]

Ahhhhhh yes, the mother of all power transistors from the past. Yes and darlington a pair.👍😏

[@kalashnikov_47z]

Amazing video I like this type video 😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍😍

[@wowyummyyy]

L I K E 👍 👍 👍 👍 👍💯💥 💯 💥💯🤩 🤩 🤩🤩

[@diyelectronics5375]

Use LM338T much better i use that they are over temp protection build see in datasheet i dont like use LM317 it is doggy for me. in my opinion

[@tiborbogi7457]

I think DGW just recycling what he have in a drawer or in the box in the shed.

[@imnotbeluga007]

LM338 is also a voltage regulator, but with a max current flow of 5A. But he only requires 500mA or 0,5A max. So, it doesn't matter, and the LM317/338 have overheat protection.

[@keithking1985]

👍🇮🇪💚🙏🏻

[@Metalloys]

Hello. Just a small comment regarding no load power consumption : this is probably just the magnetization current of the core. You probably just measured the current without the phase angle (you should find it very close to 90 degrees). This is not real active power drawn.

[@SAHILKHAN-lu8oq]

Then you don't know how he measured that. He used some four quadrant analog multiplier chip (AD633), which outputs proportional mV according to real power. KZbin doesn't allow pasting links in comments, otherwise I could show you the page where he talked about it.

[@DiodeGoneWild]

The 1.1W no load power is a real (active) power. About 1W is the transformer, probably mostly iron core eddy current losses and hysteresis losses, and also resistive losses in the primary due to the magnetizing current.

[@Basement-Science]

Only big high-power transformers have close to 90° phase angle at no load. Any small transformers like this may even be mostly resistive because the primary winding is just so long and thin (thousands of turns).

[@javasyndicate713]

the cat almost 10 yrs old, though 😀

[@DiodeGoneWild]

No, she's just 7 years ;)