Building an Adjustable High Voltage Supply
Vložit
- čas přidán 4. 01. 2024
- I've wanted to build a high-voltage power supply for a long time, and in today's video, it finally happens. Using a ZVS driver, a flyback transformer, and a high-voltage cascade, I generate a voltage of 50 thousand volts. This video covers insulating the components with epoxy resin, a self-designed high-voltage connector, and integrating them into a custom enclosure to create a power supply capable of producing a variable voltage from 20 to 50 thousand volts.
Join my Patreon and support my projects! Your contribution means the world to me and helps bring my ideas to life. I truly appreciate your support! / advancedtinkering - Věda a technologie
Small tip: Watch out with black plastics, resins, insulations, etc. They are often colored black using carbon. Can bring some really nasty surprises.
Yes, in my first version I used only white PLA to avoid getting problems with carbon black as a pigment. But it seems like it makes basically no difference. The white PLA is also terrible for insulation.
The resin uses aluminum silicate as a filler and the data sheet states the dielectric strength.
But you are right! I should have mentioned that in the video.
weird they'd put carbon in insulating epoxy though
Hahahaha, I got zapped in the ear once using a bit of rubber air line to listen for where a Meech high voltage anti static thing was arcing.
@@fmdjFrom actual experimenting the carbon itself doesn't seem to make any difference at all. It seems to be the PARTICLE SIZE that causes problems. Using soot carbon makes no difference at all while ground up charcoal bricketts almost makes conductive epoxy!
@@glasslinger yes and I can confirm that epoxy mixed with pestle-and-mortar'd charcoal is kinda conductive, I've been able to stick copper wires to graphite sheets with that, and get a rather low resistance (I'm having some fun with batteries).
LV and HV, seem like a perfect power supply for tubes like x-ray tubes 😉
;)
@@AdvancedTinkeringIs 60kV gonna be enough? Most *ahem* vacuum tubes I know of like >70kV
@@apo_chromatic A bit on the low side, but dental vacuum tubes emit vacuum tube rays between 45 - 65 kV. Of course the big vacuum tube are usually in the range between 90 - 150 kV
@@diecksl yeah that’s true, I was probably a bit swayed by the absolutely tiny X ray tube I picked up recently that operates at 70 kV, it’s about as big as one of those small cans of soda
@@apo_chromaticdepends on energy of the x rays
You're a true master of tinkering. Using so many different skills to achieve what you want and on top making it aesthetically pleasing is amazing.
Thank you! I appreciate hearing that you like my projects!
next time i plug in china modules like legos and ppl are gonna call me einstein ffs
First off, really nice build!
My suggestion, make the output HV connector a bayonet lock quarter turn (similar to a mil-dtl-26482 or mil-dtl-38999 series I locking mechanism). There are also some plastic bayonet lock examples, such as Souriau UTP, ITT Cannon Trident, Amphenol Aquarius, etc. I feel like bayonet connectors are much more pleasant to use when you need to attach or detach a connector often, whilst screw-coupled connectors are better for semi-permanent applications where you would rarely or never disconnect the connector.
Moreover, if you wanted to, i believe you would be able to integrate a low-voltage safety interlock loop into the output connector at the opposite end of the HV terminals, so that HV power is only turned on when the low-voltage interlock loop is completed. All HV connectors I have used previously has had an HV interlock loop. Obviously it would be difficult to switch on/off the HV output, but the low voltage input to the voltage multiplier could easily be cut with a simple relay in series with the HV interlock.
Lastly, if you want really nice and professional looking front and back panels for your rack mounted equipment, you can use aluminium PCBs. They are extremely cheap from china compared to a custom machined panel, and they will make any cutout you want, and text/logos can easily be made with the silkscreen, solder resist, copper like you would for a normal PCB. Just epoxy two set of 1.5mm aluminium PCB's for a very sturdy 3mm front panel.
What makes design of these types of equipment so interesting are the choices that have to be made to work around space constraints, material properties, and real-world limitations of components.Seeing the failures are often more illuminating than the successes.
probably one of the neatest and safest diy high voltage projects I ever saw!! almost could be sold as a product
Very nice. Just a suggestion: I recommend, in addition to the switch with the protection, to insert a small circuit with a relay and two buttons to activate the high voltage. Such as start and stop systems for power tool motors. Maybe the stop button with a nice emergency mushroom.
The relay practically de-energizes when the power goes out and there is no risk of leaving it on in the event of a blackout or similar problem.
I'm so glad you shared the copy template trick. I forgot how useful that is.
As soon as I saw the title I thought: "Advanced Thinkering", I knew it was your video before I even saw the name of the channel. Thank you so much for producing exactly the content I spent years looking for in this platform!
Thank you so much! I appreciate it! I'm struggling a bit making thumbnails that make someone recognize my channel. So I'm glad to hear, that you did!
My tip for PLA. Don't use glue get some DCM (Dichloromethane)!
It welds the parts to each other excellently by melting the top surface of the PLA and it holds in seconds (full strength takes a little longer but DCM is very volatile).
You can also weld acrylic parts basically optically transparent with it.
I just feel like it's a good opportunity to mention some (not a complete list) of the dangers of DCM. It's a likely carcinogen and additionally acutely toxic when used in a closed space so either do it indoors in a fume hood or go outdoors for it. Additionally I like to use a syringe to distribute the DCM for welding but make sure there is no point on the needle, blunt needles only, severe tissue damage is known from accidental DCM injection into a finger to the point of needing emergency surgery and barely keeping the finger. Also it's a quite mild but still green house gas so ideally make sure any waste and especially the main bottle are stored correctly and in bottles where it won't just escape, it's a pretty small molecule and will diffuse through most regular plastics.
Men will do literally anything to install THAT switch.
Good job, doesn't look to unsafe from a electrical point of view. You are probably the first non-electronics "DIY" CZcamsr that uses the correct wire end terminals. Good job. Do you have a link to the sortiment of those gold plated banana jacks, that looks very useful. Thank you :-)
Thank you! That's comforting to know.
Since CZcams does not seem to like links in comments, here is the ASIN number: B083FNTK5R
If you search for it on Amazon you should find it (if it's available in your country). Otherwise you should be able to find it by searching for "banana plug connector male female" or something similar.
@@AdvancedTinkering Thank you! Glad to hear that :-)
And yet still, somewhere in Germany a member of the VDE is seething with rage, writing a long winded ranty comment on why what AT just did is the most illegal and immoral thing to have ever been done with electricity. Because that's all VDE members ever do 😀
That's a lot of work. Great well thought out design.
Amazing work again! Love your own solutions, the precisement, etc. Keep on!
Your channel always get's me excited for a new video! The "shhhh, nobody needs to know" comment to your vacuum had me rolling on the floor laughing hahaha
Awesome project! I made pretty much the same thing a couple of years ago but I used PWM to drive the ZVS and high voltage banana connectors that are made for HV laser supplies. Pretty much what you already made but without the resin hassle. Also I would recommend to use a pneumatic switch from a washing machine along with rubber tube and a squeeze bulb as your HV enable switch. This way you will be always isolated from the device if something goes wrong.
I would use this option for the sheer coolness of the thing.
For nerd points, i would probably use Lego pneumatics as pump.
Thinking about it, you could have lego Pneumatics actuator that actuates the switch.
Nice! Looks like a solid build. Hope it serves you well!
Thank you!
that is some scary shit.
Please stay safe, I want to see new videos by you, having to re-watch only old ones would be sad.
I get more & more fascinated with Plasma each year. It's just something i always enjoy seeing any sort of science experiment being done with it or learning anything about it's behavior. Great build. I've done electrical work for remodeling jobs but this type of stuff seems like such a challenge to consider all the variables.
another well designed project! excellent ! I would have made it a negative output for powering a fusor! but this is pretty good for Xrays! Awesome!
I think you did a great job of that! I can't wait to see what uses you put it to.
Thank you! Yes, I'm very excited for the project in which I'm using this power supply.
Happy new year! Good job on your power supply too!
Thanks! Happy new year to you too!
i built something similar, got 20cm arcs and have never dared to turn it on again. the arcs were so loud it hurt. This was a beautiful build. i need to check this channel out
I am impressed 👍👍👍
Can't wait for them ZVS's to blow when they hit zero, they always do. Great video.
Not bad. The big-ass 1G resistors are a nice touch.
Beautiful build mate, nicely done! Yeah it didn't quite make it to 60kV (but 50-55kv is super close enough) but its 100000x more clean than what I would have had the patience to build.
Pretty damn useful mate!
@14:01 RIP Headphone users. Amazing video, thanks!
Nice tidy build
Thanks!
Awesome build,
I would suggest considering the construction of a custom half-bridge inverter. It can be simplified by utilizing a self-oscillating chip such as the IR21531. This approach allows for greater control over the frequency, facilitating improved impedance matching between the transformer and multiplier.
RG8 coax, with a solid PE dielectric, works pretty well for HV cable up to 60 - 80 kV. If you drill out the middle of a UHF connector, you can make an easy, reliable and safe (relatively speaking) HV connector. That's how it's done on Glassman and Spellman HV power supplies, including some that go to ~80 kV. Cheap and effective.
I don't care for potting the multiplier modules. The usual failure for these kinds of supplies is the diodes and (occasionally) the caps in the multiplier, and it WILL blow sooner or later. Glassman does all of it in air. For some of the supplies I've built where I needed to make it more compact, filling it with mineral oil works great, and you still stand a chance at repairing it when it inevitably fails.
Seems like it'd be a good idea to also add some sort of lock toggle to the HV connector
Hmm, I wonder what a vacuum chamber channel would want with a high voltage power supply :p
Also, its possible the fact the resistors are wire wound has more of an impact on the current than their resistance, they're probably acting like tiny jnductors
If you're really keen for 60kv, you can just freehand solder a voltage multiplier, or add a 4th stage to the same one as you used
Ahh, I can see, you're a real man of culture! Labeling your label maker as a label maker is such a nice thing to do! 😂
8:52 Damn that's an expensive vacuum chamber😂😂
Hello! Great work! I maded myself a few high voltage power supplies and one thing that I have discovered is that the oscillating high current of the zvs driver and the metal case are not made for each other. Part of the power that you are putting into the driver doesn't convert to high voltage but is induced in the metal case and lost as heat. Even though you have a nice output I think it would have been better if you have used a plastic or wooden box.
maybe add a microswitch and relay to not energize the ZVS unless the HV connector is screwed in....
Great idea! Maybe a magnetic switch would work for that.
@@AdvancedTinkering yea a reed switch could also work
+points for the comment "for... reasons" 😂
Very nicely done. Still deadly, though...
This bloke is impressive.
ah yes, the 50kV bannana plug, better dont get them mixed up :P (oh also, youre amazing at 3D design)
Very nice design. Still have to make mine at some point. But it will be a lot crappier xD
Great video and an absolutely beautiful design! Watching this video also answers some of the questions I had for a power supply I was trying to design as well.
I'm looking to design a variable supply (0-15kV, 0-5 amp, or as close as I can get) for tube experiments. Not x-ray or anything like that, but I had some ideas involving magnetrons. Any thoughts on where to start? Maybe A design similar to yours could work?
Wax makes a good temporary potting medium if you wanted to test it potted but then remove it. Back in the day wax and wax paper was all they had!
I recall a teardown of an x-ray machine from Mikeselectricstuff where a PL-259/SO-239 Rf connector shell was used with polyethylene dielectric RG-8 coaxial cable as the HV cable. The center pin was a 4mm banana plug on the end of the cable with the insulated core sticking out 150 mm ahead of the backshell. The insulated core of the coax was pushed into a plastic sleeve and mated with a 4mm in the back. I think it was also insulated with silicone grease or something.
Wau: Das nennt sich Gruendlichkeit.
Hi amazing video it is a super cool system are you doing well
maybe test the voltage reading with lower voltages. (like avoltage that you can actually measure with another meter)
Because the resistor might be off, and it will be more off due to slight conductivity of your epoxy resin.
Nice Video :)
When you create a spark, how low the output voltage dives? I wonder if the 10W 10Kohm resistors are not handling too much current, if not they will open.
Very nice project.
I recognize that driver and multiplier PCB everytime I see em. Highvoltageshop stuff is just a great "supplier"for the obscure world of all things HV. Reminds me to get some parts for my large multiplier.
What resin did you use? That stuff would work wonders in my planned 120kV cascade, regular epoxy just can't hold back anything over 40kV.
Yes, it's great that there are shops selling niche components. The resin I used is from Demak. The clear resin is SEPOX225. The black resin is XE661 HV. Unfortunately, it seems that it's not readily available for purchase and one has to inquire with the manufacturer. I got it from a friend who is a polymer chemist after telling him that I couldn't find reliable breakdown voltage data for commonly available resins. If you come across a good alternative that's easily accessible, please let me know.
Thanks for posting that shop name. I also am hunting for a better insulating epoxy. The best i have tried yet is actually my wifes art stuff - crystal clear something its called, long set time, and almost no bubbles at all. It says its actually waterbased, go figure ha ha. Im off to chech that hv shop, thx again
@@AdvancedTinkeringbeautiful build btw, and excellent presentation/vid.
@@jonlitchfield8888 Currently have a 75kV peak multiplier cast in polyester resin and one in polyurethane. Lets see which resin holds up more.
After 7 seconds it went from infortainment to entertainment😅
Cool connector you got there, potting and all, I've seen ptfe and pvc sleeves used in a similar manner.
As EE the rest of it can look a bit sketchy - moving the feedback pot on a long wires across a case where high frequency and voltages are present - thats just asking for problems, to me all those zvs drivers look sketchy AF, but thats just me being nitpicky.
Regarding earthing - safest and most reliable would be riveted and/or crimped connection - everything bolted needs to be checked and retightened as needed.
You are right about the drivers! HEY! The mickey mouse trade mark is expired I hear! The drivers look Mickey Mouse! :)
Hi. Congrats. Can you explain what you did to generate AC from the high voltage DC that you got from the flyback transformer? Thanks
Hi! It's an AC flyback transformer. So the output can be used for the voltage multiplier as is.
Will you be sharing the modules that you have used? anbd P/M of the rest of the elements.
I haven't seen you using a pilot hole drillbit for drilling holes for the powersupply.
Metal drillbits are cheap to get ;)
Klingt als würdest du deinem Ziel des Plasma Sputtering einen Schritt näher sein. Sehr cool!
Das Netzgerät ist zwar nicht für das Sputtern gedacht, aber an dem Projekt wird auch gerade gearbeitet. Danke!
Spellman is reusing type N RF connector for the captive nut on their HV PSUs.
Is not the same. Are HV voltage rated.
@@fabiobosco7545I I said connector shells.
Einfach wunderbar! Works and looks beautiful. If you would want to measure the current, a shunt resistor can be used, then just measure the voltage across it and I = V/R. Have you thought about using this supply to drive a fusor? ;)
Thanks! At some point I would like to build a fusor. But thats still a long way to go.
Mh.... yes a high voltage shunt resistor should work. I will take a look into it.
Did you post the printed parts somewhere?
I don't think you have mentioned it, but you can't use a regular flyback transformer from an old CRT TV. These have a rectifing diode in series with the output coil which provides more or less DC, which doesn't work with voltage multipliers. You need an AC flyback which are somewhat harder to get.
Awesome. But, most voltage multipliers that I know work with a spark gap. Setting it in resin is awesome but they work better with having multiple spark gaps. Anyway it works for you. I have built my own and can get much higher outputs.
I think Marx generators use spark gaps.
@AdvancedTinkering I think you're right, he's using a Marx generator, while you're using Greinacher multiplier.
@@xxportalxx.Cockroft-Walton multiplier. There are constant fights between Europeans and Americans over who originated something. Cockroft and Walton were the actual inventors of the circuit.
Ahahahaha "Shhhhhh... nobody needs to know.." 😂😊
Wenn nur 1/3 aller Geräte so gut gebaut wären wie deine wäre unsere Welt besser drann!
Vor allem im Bezug auf die Nachhaltigkeit!
So viele Dinge sind wegen den letzten paar Cent Ersparnis unnötig schlecht konstruiert... Was dazu führt das der Kram nach relativ kurzer Zeit kaputt geht, und das oftmals katastrophal.
Ich schweife ab... Geile Projekte die du da machst! 👍
Danke! Freut mich, dass dir die Projekte gefallen!
Please let us know where we can find that exact 30A 800W step down converter, the serial volt switching unit and the 24V 20A powersupply, thank you!
Great idea with scanning and printing power supply; i always used ruler and/or caliper, wasting a lot of time.
I hope the distance between your caps sufficient. Having them back-to-back like that has cost me a good handful of ceramic caps.
I haven't had any problems (at least for now). I think the few millimeters of epoxy between them should be good enough.
But if something fails in the future, I will mention it in a pinned comment.
I think that your design needs short/overcurrent protection. You could add overcurrent protection very easily: just use comparator with hysteresis and connect it to the resistor shunt, then in case of fault restart PSU.
I built an adjustable 60 kVdc supply (2 mA) 15 years ago, and have never used it to produce even one arc. I may be alone in the world.
and how much current did you deliver at 50 kV? Do you have some ripple values?
5:15 by far not an electronics expert either, but the primary of the flyback transformer being so close to the other transformer worries me slightly, I dunno if it should or not...
Anyways, much more cool than my repurposed ATX power supply ;)
He labeled the labeling machine. :)
Can you tell me where you get this fly-back transformer ? I was unable to find it by part number. It does look decent for my project.
It's from a shop called "highvoltageshop". I know, very creative name.
But I don't know if they ship to your location.
There are also a lot of flyback transformers on ebay that would work fine.
3:37 ground distribution point now that's what we call "ZHE GERMAN ENGINEERING"
+Advanced Tinkering can you please tell what temperature did you use to make cesium,did you preheat the furnace, and did you slowly heat the cesium chloride or directly set it to the temperature required.And can magnesium be used instead of lithium.
I think with this much of insulation you can easily go up-to 100kV.(Talking about Voltage multiplier)
I actually thought about building a second voltage multiplier that I can install depending on the voltage I need.
Just to validate to be sure as i missed it somehow but the multiplier is directly connected to the HV output of the flyback?, even with the DC rectified output ??
No, the flyback transformer is not rectified. It has a DC output.
I would really love to know what type of 80kv cable you are using (looks shielded), i am looking for a somewhat flexible cable in the same voltage range and only can find extremely hard to bend cable
It is from "highvoltageshop". It actually is very stiff. But I think they also sell silicone insulated HV cable rated for 100 kV. It should be more flexible.
You could add another stage of multiplier into the output wand itself but then your meter wouldn't correlate anymore.
That wouldn't be possible since the output is DC.
Is there a reason you went for half wave multiplier instead of a full wave? I've just been searching the youtubes for examples of full wave but no one ever does it.
I dont know why, but i want one
Beware that potenciometers wear up over time, sometimes leading to random peaks when rotated, like the popcorn sound in old radios when adjusting volume.
On a second tought, since you have the safety resistors to not lower 12V then the wear up shouldn't be a problem
Thanks for the advice!
I like it without the resistors. It will make a nice stun gun for protection if made portable
I know you needed voltage, but did you measure currents it can produce? I want to convince my colleague to redesign capacitor battery charger for our experimental setup to be something like this. Right now our charger is, as we call it, made from "shit and sticks", as it was "temporally" solution 14 years ago and nobody bothers to update it.
Awesome build! Looks like a fun way to die. 😂
That's great, but for example, will it kill us to get close to this? Or if such a project is going to happen, how will we know what will kill us or not?
Can you explain why the arcs never go in straight lines?
I can't say for sure. But I suspect it's the same effect as in the jacobs ladder. The ionized hot air rises and thus the arc rises with it (since it is most conductive).
But I feel like it only partially explains the behavior of the arcs.
... and what about a variable current limit like the Heinzingers have ...
The biggest problems I suppose with projects like this is that you have to perform numerous tests to actually deem it 'safe'. An uncountable number of tests for a full-on supply if you actually want to sell it... Considering the effort and parts, I would deem this project around $1k-$1.5k as a starting value if you build it yourself. (Including man hours)
This cannot be sold. The liability would make insurance impossible to acquire!
what happen if you connect low voltage devices on this converter？
either a x-ray power supply or a scanning electron microscope supply
Is a AC-DC or AC-AC flyback transformer that you used?
It is an AC-AC flyback transformer. If it has a DC output it will not work with the voltage multiplier.
@@AdvancedTinkering where did you purchase it? Do you have a link where i can get It?
Eh.. toss that resistor pack in something like a load tester and get it off that power supply.. unnecessary.. may wanna put a safety pull on it tho.. connected to a wrist strap so in case you accidentally find out what electricity tastes like you will be able to remember it..
I have had this idea: What would happen if you connected a pipe to a compressed air gun and put some gasoline in the pipe. Then connect the other end to to a plastic bottle. Also put another hole with a spark plug in the bottle. Would it explode? Because high pressure, vaporized fuel, air and ignition is there... Just an idea
Thank you for the email regarding
Be careful with black materials when working with HV. Black pigment is many times carbon, which conducts electricity.
It looks like you forgot to print a label for your label printer.
Necesito el diagrama por favor
name epoxi ?