Home Made 30 kV ZVS High Voltage Power Supply and Tear-down

[@acompletelynormalhuman6392]

0:25 at least he's honest... also I love that shirt

[@kVital_2023]

Nice! I had built the same HV PS some time ago. The only remark is that the ZVS driver is designed for an induction heater, where is hard/impossible to make two driving coils, so it has 2 inductors on the PCB. There is another type of ZVS driver with only one inductor and which is more efficient to drive an HV transformer (but it requires 2 coils on primary). You can also modify yours (the schematics are very similar). This will produce less heat of transistors.

[@LesLaboratory]

Well spotted! I used to build my own, running at insane powers, but these one happened to be dirt cheap (cheaper than I can make them, so just snapped them up) They might not be that efficient, but they don't break a sweat.

[@arnaud7671]

7:30 I see that moment where you almost forgot the capacitor could still hold a charge ^^

[@LesLaboratory]

LOL, yeah, no touchy!

[@sd17nano]

excellent video. thanks a lot!

[@LesLaboratory]

Thanks!

[@andymouse]

Smashing tutorial !...cheers.

[@LesLaboratory]

You are welcome!

[@sahinbozkurt886]

Thanks for this video. This was very beneficial for my project. But, can you share the components list with us?

[@dalenassar9152]

At about 1:15 when you show the inside of the HV power supply, The HV ground looks awfully close to the HV out. How do you prevent arcing?? THANKS FOR ALL OF THESE OUTSTAANDING VIDEOS!!!

[@davidosterberg]

Nice. I see you have 300 subscribers. Keep it up and you will have 100k subscribers in a year.

[@LesLaboratory]

Thanks! Now that would really be something!

[@GaryMcKinnonUFO]

I'm having trouble sourcing AC flybacks unless i order from China and wait 3 weeks, did you source yours in the UK and do you have a link ?

[@LesLaboratory]

It came out of old equipment, a radar set if I recall! It is however entirely possible to wind your own secondaries for these. I built a few over the years or x-ray stuff I was building.

[@GaryMcKinnonUFO]

Thanks Les. @@LesLaboratory

[@en2oh]

I just found your channel. Great resource. Thanks! In terms of controlling the output voltage, will the ZVS module tolerate 0-? Volt to make a practical variable voltage power supply? Thanks, Doug

[@LesLaboratory]

Cheers! Normally you can run them up to 36v or so (some can be pushed as hard as 48v) There it a trap though, you cant start them at 0v, or even soft start them for that matter. They need the turn-on inrush current to start oscillating (normally around 12v)

[@gunrunner110]

@@LesLaboratory is there a possibility to modify a ZVS driver to start them at 0 Volts? Or what would be your solution to get a fully variable HV power supply? And one question about the ground cable: does it need to be rated for HV too? Because I see a lot of HV supplies where only the output cable is a HV cable . But won't there be running HV through the ground cable too? Sorry for my crude English and thank you for your help!

[@nicolaskrause7966]

How much current can the supply produce at maximum voltage? Very much enjoying your videos! Just discovered them via Twitter!

[@LesLaboratory]

Awesome! The design I am using is current limited to about 8mA, but ZVS drivers are capable of pushing hundreds of watts through a line output transformer. I'm not sure how long a standard monitor transformer would stand up to the punishment, but they are cheap, so give it a go! You can buy CO2 laser transformers pretty cheaply as well, and I am willing to bet you could drive one with a ZVS successfully.

[@chouaibchouaib4880]

Does she gave a regulated and stabilized a DC voltage and current?

[@johnnycash4034]

Where did you buy the high voltage insulator? Looks cool

[@LesLaboratory]

Just off of eBay. They show up from time to time.

[@florinpetrache7834]

i dont have time to find spheres because my problem but needles are ok i believe. MY CONCERN IS NOT TO PIERCE MY CAPACITIVE DOORKNOOB(40kv 1nf with 20% tolerance).my source is a high voltage generator with a voltage multiplier that it said to have 40kv but in reality it has about 20 kv because the arc jumps at 20mm.Of course i put a resistor in series(about 3Mohm 3W)for protect the source.I think you can power and you your lasers with these modules( period max 5 sec)to not heating up(especially the small one).The 1000kv ( more than 20kv at 7,4v) 3.6v -7.4v input and the 400kv (20kv) 3.6v-6v input.for power source i use two 3,7v acumulators.

[@LesLaboratory]

You really do need the time to make at least reasonable measurements. It is worth it in the end!

[@arktik75]

Hi Les!What is the voltage and the current wich you feed the zvs?I just burned my hv power suply and I want to build one based on your design.I have an 1000w zvs.Many thanks!

[@LesLaboratory]

I normally feed it off of a variac upto about 20v but a low current may be a couple of amps or so.

[@messageobliquespe100]

I had that calculator - what a bore ache having to push the buttons down - Casio from then on some 40yrs ago!

[@LesLaboratory]

But it is cool but!

[@florinpetrache7834]

for my mini tea nitrogen laser ,it will work with a crt flyback with ballast resistor or should i built an multiplier?it will work a multiplier with 4 diodes or six 15kv, 550mA ,and 20kv 1nf ceramic capacitors with a balast resistor 3,3Mohm?and which is the voltage of your flyback?

[@LesLaboratory]

it depends, if it is a DC flyback, a multiplier is not needed, nor will it make any difference. My flyback is AC and puts out about 10kV hence the multiplier.

[@ひるまんせちよ]

Stupid question but if you varied the input voltage of the zvs driver, would it increase/decrease the voltage of the flyback?

[@LesLaboratory]

Yes, in fact when I drive the home made Nitrogen Laser with it, I hook it up to a home made variable power supply.

[@ひるまんせちよ]

@@LesLaboratory you hook the zvs to the variable power supply?

[@LesLaboratory]

@@ひるまんせちよ yes, see this video for an example: kzbin.info/www/bejne/pnTRiZmqqq9skNE

[@ひるまんせちよ]

@@LesLaboratory just checked it out 👍 This will be helpful for my current project, thank you!

[@LesLaboratory]

@@ひるまんせちよ Cool! Just one gotcha though, you cannot start a ZVS from 0v. They need a kick at turn on to start oscillating. I generally start at around 10+ volts or so.

[@florinpetrache7834]

I made a multiplier with 10 capacitors. The diodes are fast (restoration time 150ns) but will be able to handle the high frequency of the ZVS? What do you say?I'm afraid of burning the diodes because of the high frequency.

[@LesLaboratory]

Assuming that the transformer resonates at about 20kHz, the period will be 50uS, so 150ns recovery time is more than adequate. Use a ballast resistor on the output of the multiplier, else a discharge may exceed the current rating of the diodes.

[@GliTaNio]

Which side is positive and negative?

[@LesLaboratory]

Positive is at the Insulated output.

[@aaamet]

Great video! I'm planning to build something like this for a project. Do I have to use a flyback transformer? Since ZVS driver works at the resonance frequency, would it be okay to use a transformer out of a microwave oven?

[@LesLaboratory]

Thanks! The LC in the ZVS driver behave as a resonant tank circuit, and thee normally resonate at round 20kHz or so. The heavy Iron core of a microwave oven transformer it much too lossy at these kinds of frequencies. Flyback Transformers can be had very cheaply off of sites like eBay and make for a compact, high voltage supplies that are much safer than Microwave Oven transformers!

[@amirhakimi5453]

Hey, I got a bunch of question. What is the output of your ZVS driver? Is it AC or DC? Where did you get an AC flyback transformer as I have a hard time looking up for one? And did you use AC for your initial input into your ZVS?

[@LesLaboratory]

The output from a ZVS is AC. The input is DC. It is possible to wind your own flyback if you can't get hold of one.

[@florinpetrache7834]

my doorknob capacitor have 40kv 1000pf.For peaker capacitor is good to have aprox 500pf ,about half of my doorknob capacitor?Which is the capacity i need?

[@gastjjs]

Hi, can you explain the formula for getting the right output at the voltmeter?? thank you

[@LesLaboratory]

Better than that, there is a good tutorial on it here: www.eeeguide.com/series-resistance-microammeter/

[@medeirossergio9924]

Nice HVPS!!!! Could you tel me how you use a digital panel voltage? Thanks!!!

[@LesLaboratory]

Same as with the Analog meter, a suitable high voltage divider, plus the meter. The catch is though that unless the digital meter is sufficiently well designed, you will run into problems. I have had endless issues with inexpensive digital panel meters. Analog just works! Anyway, see here for a good resource: www.repairfaq.org/sam/hvprobe.htm

[@florinpetrache7834]

your diodes from multiplier are 15v high frequency diodes or regular diodes?

[@LesLaboratory]

High frequency diodes. Mine are each rated 30kV, but 15 kV will probably do the job.

[@liajaj8157]

quick question, can you build a 15kv DC power supply that can deliver negative voltage with NO AMPS???

[@LesLaboratory]

You coudl easily make the supply negative, bu reversing the polarity of the diodes in the Cockroft-Walton multipler. I am not sure what you mean by no amps though? I'm sure if you wanted almost zero current, you could use a multi Giga ohm resistor on the output.

[@kevinscollan8293]

Where dya get one of those flybacks? all the ones I've come across have diodes inside to rectify or a double cicuit potted inside as well.😎💕🖖

[@LesLaboratory]

I got mine out of an old Black and white monitor. You could try eBay, but it is possible to wind your own secondary on the core from a color monitor.

[@Convolutedtubules]

Neat build! What is the simplest/best way to power a voltage multiplier that requires 2.5Kv at 50KHz? Can a ZVS induction heater be used, if so then how?

[@LesLaboratory]

The ZVS driver I use in the video is designed for induction heaters. It would simply be a case of acquiring a suitable transformer or winding your own. You can use the ferrite cores out of old monitor transformers.

[@Convolutedtubules]

@@LesLaboratory I require a high frequency of 50Khz.

[@florinpetrache7834]

so ,a spark which jumps at 20mm lenght is 20kv?i dont have a high voltage probe to measure the voltage of my power supply.my power supply forms an arc about 18mm, so is 18kv.

[@LesLaboratory]

Yes, more or less. There is a document here that may help you: www.kronjaeger.com/hv/hv/msr/spk/index.html

[@florinpetrache7834]

how many turns you have to the primary? 20 turns is good?

[@LesLaboratory]

Yep, about 20 turns. It depends on the flyback.

[@Telectronics]

Even If it´s true that electronic gets distorted when high voltage is near I saw high voltage power supplies with digital meters so it must work some how. I could even get analog meters jumping like crazy so they are not immune too. I did not try hard enough myself but maybe you could try shielding, filtering and grounding(earthing) in one or more combined way.

[@LesLaboratory]

Sure with shielding you can get digital meters to behave. unfortunately I only had a plastic case for this project. Analog meters are however quite robust, and some are design for direct reading of EHT, like electrostatic voltmeters.

[@Telectronics]

Yes they are more forgiving but If you use a marx generator on the same ground and you don´t earth anything they say I am out now xD If not banging with a marx on the same ground they also were good enough for me :) I guess there are oscillations when the marx strikes because sometimes I have even flash overs of my 40kv resistors which are plenty for 25-30Kv with a flyback transformer. Still I don´t think a mains earth is made for a marx too so I leave the ground floating and power it from a battery.

[@nicolaskrause7966]

Just re-watching this great video again, from some other research online there's no easy way to make a flyback based circuit such as this a negative high voltage power supply is there?

[@nicolaskrause7966]

ie: It's not possible to reverse the diodes in the CW multiplier to obtain a negative output is it?

[@LesLaboratory]

It is absolutely possible! If you want real high voltage with reduce risk of the multiplier flashing over, you can drive two flybacks and build one positive and one negative too.

[@florinpetrache7834]

i am the guy with last comment with the helium neon and nitrogen laser.what is the power of your high voltage resistor(20 Megaohms )?

[@LesLaboratory]

3 x 20meg in paralell

[@drrrrockzo]

I was literally just looking at my power supply thinking that I need to enclose it. Quick question about the ammeter, if I'm thinking right 1Gigaohm makes it read out in KV/uA and the meter's internal resistance is so low by comparison that it can be ignored, correct?

[@LesLaboratory]

Almost, the meters resistance is about 2.9k, so I end up with quite an odd division. I forget where the doc is that I used for my math. IIRC I had to measure the voltage required for FSD then work from there There is a document on measuring high voltages here : hotstreamer.deanostoybox.com/TeslaCoils/OtherPapers/JR-Lucas/HV_Chap6.pdf

[@drrrrockzo]

@@LesLaboratory Thanks Les, that link has some really useful stuff in it.

[@HendraGunawan-tz4fh]

Mr. I have welding machine ,and the machine need power electric 220 volt 3300 Watt .my question is ,can this device able to make my machine work.thank you for your video

[@باترینکاس]

Ok

[@PosthumanKindergarten]

a.... " "small" " .... high voltage capacitor, sure :D if i had one of those i'd keep it exposed on a pedestal

[@florinpetrache7834]

so the formula U[kV]=30*d[cm] is true for sperical electrodes with a spacing much less than their diameter,not for needle like i have?what i see in graph is 2cm is aprox 20kv, and in tabel 20kv is 1,75cm for needle.Someone say i have 3kv per 1 mm distance beetwen the electrodes examples: → 10 mm * 3 kv/mm = 30 kv → 20 mm * 3 kv/mm = 60 kv i believe this examples is for spherical electrodes not for needles.

[@LesLaboratory]

You are better off using spheres. They are easily obtained (ball bearings, dome-nuts etc). Needle gaps are notoriously unpredictable as a measuring tool, because of the Corona discharge they generate.

[@dalenassar9152]

OK I looked aagain and I see that GND is NOT close to +HV and that the close spacing is at the same potential...sorry.

[@majidjanabi4976]

Can you build ac variable high voltage up to 100kv with variable frequency up to 30k Hz. The output power around 100 W? If you interested, please send me a qoute. Thanks

[@florinpetrache7834]

yes ,but how many watts each?

[@LesLaboratory]

The ones I used were 4Watts each. They do get quite hot in use, but the magic smoke stays in, so it's all good!

[@mohinderkaur6671]

shrouded hv terminal so it does not stick out for safety

[@brentonrowland8613]

Excellent nitrogen laser videos, thanks for the inspiration. I have not found any AC flyback transformers anywhere online. Do you have any suggestions for a source or links to an alternative design I could use? Would an ignition coil such as the ENA coil for Subaru Impreza shown in Plasma Channel's "The 50,000V Subaru Outback Powersource" video be a replacement?

[@LesLaboratory]

Thanks! If you look on eBay for "Vintage Flyback" you will find loads of them. Usually the windings in these are either pancake shaped, or visible, or they lack the black plastic casings of 'modern' ones. Most more modern (80's, 90's) Black and white TV flybacks are AC, but any modern-ish color monitor flybacks are DC (they have Diode split secondary's, that forms an integral voltage multiplier) It is possible to wind your own secondary if you are so inclined. I have done this and it is not as many turns as you might think (about 600 - 800), but you do have to insulate each layer and do a decent job. You can get new modern CO2 laser flyback transformers as well, which may be AC, however, I have not played with these.

[@brentonrowland8613]

@@LesLaboratory Thanks for the rapid reply. Unfortunately getting them to Australia makes them a bit expensive. I will keep hunting. I would like some details of your linear variac lab power supply you use. Have you done a video on that or have a schematic?

[@LesLaboratory]

@@brentonrowland8613 There is no diagram or video, but it was dead straightforward. Simply a small variac feeding a 24v transformer (for isolation!) that is then rectified with a bridge rectifier and a smoothing cap. It was a junk box build apart from the case. On a side note, you can not bring up a ZVS driver from 0v! The ZVS needs the initial 'kick' from power on to get it to oscillate. I normally start at about 10v (the minimum) then work up from there.

[@martin_flow_uy]

Hello, your channel is very interesting. I was wondering if there is any possibility that I send you the money and you send me the two 500 Mohm resistors? Otherwise, would you please be so kind as to send me the link of the store where they sell it? Thanks a lot.

[@mohammadamini9781]

Thanks a lot for video. It's great. How can I make this power supply with 60 kV output with the same frequency? Can you please help me and share the circuit schematic?

[@LesLaboratory]

You could add more stages to the CW multiplier, but you would have to insulate it well (for example submerge in oil). ZVS drivers have been around forever, and there are plenty of schematic on the net. If there is an appetite for it, I may well do a video in them in the future, but right now, I am in the middle of a CO2 Laser video... I am curious, what is your intended application?

[@mohammadamini9781]

@@LesLaboratory Thank you very much for your response. You are really generous that you want to do a video for it. That's so useful for me. Can you please suggest an applicable schematic to me for 80 kV output? I want it for ozone generation, but with 80 kV output.