Excellent job, thanks! I notice at 20:26, your operating frequency is about 39kHz. At this frequency, nearly all current only flows very close to the surface of the conductor due to skin effect. The rule of thumb says it is a waste of copper to use wire much larger than the skin depth in radius. Skin depth at 39kHz is 0.33mm, suggesting #40 Litz wire, though something around #36 is almost as good and a lot cheaper. Cheaper still and not much worse in performance is finely-stranded conventional wire instead of Litz. Minimize the size of the solder joints, and do not add solder along the length of the wire where it is not needed for joints, and you will get less waste heat and losses and more power to the coil and work piece.
I used to work for Vulcan Forge in San Jose we used to forge 2 1/2 inch wrenches for Stanley and Proto in closed die forgings they forged at temperatures 2,300° f , and the induction heaters we used to use were also tuned to the coil, by connections made on the capacitor Bank, and every coil had deionized water going through it, and the billets were fed through it also had two rails liquid-cooled rails going down through the center of the coil, they were all made a stainless steel tubing. The forge had coils that can handle anywhere from a half inch material all the way up to 3 in Plus
Good job, I enjoyed the layout, the specs on components, and the very accurate description of the solitary drawback, and the immediate fix, using two cores, and 12 gauge wire. The "scope" shows the exact impact of fixing the core saturation issue. Quite a professional job, I have a 300 amp tig welder in my shop without running water, and found two water cooled motorcycle radiators, Honda, if I remember right, am running a "vibrator pump" circulating water when the welder is on, and it's worked a decade with no problems. I run ten or twenty percent radiator fluid, and have a glass "ball" quart canning jar, with a hole drilled in the bottom and near the top on the side for copper tube epoxied in, for a visual flow indicator and fluid quality view. I leave about half the jar air, for expansion, but haven't ever gotten the water more than warm. Thanks for a great video, good sound electronics, and sound engineering.
I doubt DIY coils are really worth it. Given that you need to probably buy excessive amount of 12AWG magnet wire as in whole spool, + the cores + the extra work, it's probably better to buy off the shelf inductors, for some people it might end up same as the parts cost, and you end up with superior inductors which will have lower losses. You will be spending 25-35 euro for the parts, meanwhile you can buy two high power inductors for around 15 each. Also if you end up buying different cores, you risk getting issues and having to buy different cores and wasting more time and wire.
Thanks for the video. Nice work! Just a note on why the temperature of the iron stops climbing. Resistance actually increases with heat but that is not the reason. At these power levels and frequencies the heating depends mostly on the magnetic properties of the workpiece. That is why a non magnetic workpiece does not heat as well. The magnetic workpiece actually becomes non magnetic when it reaches the curie point or curie temperature and does not react with the coil as strongly. For steel this is almost 1400 deg. F. You can mess around with a propane torch, a nail, and a magnet for a demonstration. Nickel and its alloys can do this at lower temperatures. Some magnetic coins are fun to play with (Canadian one dollar loonie).
Can't believe he said resistance decreases with temperature (never heard of superconductors buddy ?). At melting temp of iron its resistance has increased by 100%.
Eddy current losses versus hysteresis losses. Hysteresis losses are typically the bulk of your heating effectiveness with these frequencies and these metals. For metals that don't benefit from hysteresis losses (aka "nonmagnetic"), you generally need higher switching frequencies as well. Take all of this with a grain of salt and confirm yourself, but these are my experiences working in a place that designed these.
You must have triggered some Kens on another of your videos or else 79 people accidentally pressed the thumbs down button. Seriously, very nice board and very thoughtful design and follow-up to clean your sine. And I am grateful for the doubled toroid trick. When I lived in Houston there was a serious electronics surplus and salvage shop with buckets of fist sized toroids on down with maybe the largest being $10 to $15. I always swore to come back and raid their stock when I started prototyping induction furnaces but sadly I have moved too far away now and they don't do online sales. This was the type of place that used to have salvaged ships radars and teletype machines and copper vapor lasers and on and on. I could spend half a day drooling in the aisles.
And if you need anything just let me know I can just grab it for you when I go I only live out past Katy Mills so it's like a 45 min to 8 hours trip depending on how dumb i10 or West Park are being
around the 21 minute mark you state the internal resistance of the metal decreases with heat - it is actually increasing in resistance. This can be explained by faradays law of induction. because a constant Bflux is changing, a constant EMF is applied to the metal causing eddy currents. and P = V^2/R. which tells us as resistance increases, P decreases.
@@ichbineltyb3814 That's right. It's not the electrical resistance that's at issue here. He's heating a FERROUS metal. That means it's highly magnetic and the main heating effect is not from eddy currents but the continuous, rapid reversing of the magnetic field is 'jiggling' the magnetic domains of the metal, causing friction and hence heat. BUT as the metal heats up and passes the Curie point, the magnetic response suddenly drops and that strong heating effect ceases.
Worked for Sundstrand Aerospace in packaging Wild frequency input from a generator mounted to a jet engine gearbox and rectifies it and converts it to 400Hz 3 phase. One in production is a 30KW unit. So I was responsible for coming up with cooling and packaging. One suggestion is to attach that copper tubing to the circuit board for the straight buses for the capacitor bank. Attaching the inductor and semiconductors to the back side with the tubing will make it easier to use the liquid to cool the circuit. Just a thought.
You haven't seen hot until you've seen a 400 kW power supply on a 5' dia. chiller-cooled coil, heating a solid graphite billet surrounded by a few hundred pounds of powdered graphite. It was white hot for several hours and took all next day to cool. We were making the first woven carbon-carbon composite rocket nozzles and nose cones at SAI (before it was SAIC) for the Air Farce and NASA. Our technique was used to make the Galileo Jupiter probe's heat shield.
Brilliant. I will be trying to make 3 heaters with elongated pancake coils to achieve 9 kw heating on top and sides. Leaving the bottom open to get a workpiece in the contraption freely and prevent contamination of the coils by falling debris. Thanks again.
I am wondering if the metal would get hotter if the coils were wound in a tighter circle. I believe the metal needs to get at least yellow hot for proper forging or perhaps other uses. I am thinking a nice project would be for you to make a second coil to swap out with this one for times when you want to heat up something smaller and make it hotter. Not criticizing what you did in anyway. You are a great teacher and demonstrator on here. love your channel.
I used to work at a forging company these induction heaters and it would just get to brilliant yellow you'd almost need sunglasses to look at it and we used to forge at 2,300°, if the Hammerman get it too hot when the Closed Dies hit the metal it would explode and shoot sparks everywhere just like a flint on a lighter
Very interesting project and you are obviously very skilled in electronics. I was drawn to your video by the possibility of using an induction heater for heating some smaller projects in my black smithing work. Firing up my gas forge to do small work is not very efficient. While very informative I want to mention that the steel was no where near a forging temperature. The steel needs to be between 1200- 1300 degrees centigrade to be effectively forged. That is an incandescing yellow heat when viewed in shade. I am confident that you, given your expertise would be able to design an induction circuit to meet the temperature criteria. Again a great video and information thank you for sharing it.
Using coaxial cable braid is the best for reenforcing copper pcb tracks. Sweat it onto the tracks with a liberal amount of solder. Flexible and easy to work with.
The 12 ga magnet wire I obtained was STIFF. I had to clamp my toroid stack in a soft jawed vice while winding it, and it still looked really sloppy. Schematix made it look so easy.... Oh well, hopefully it will still get the job done.
I started my own induction heater project, so far you are way ahead of my progress but your project is looking exacly as I want my project to look, amazing, cant wait for the water cooling system, that is goint to be very interesting
A variety of coil sizes matched more closely to the size of the object being heated will greatly increase the temperature of the object Once worked for a company turning out induction heated rivets and such and our coils were not much more than .25" clearance or less around the part...also our coils had water circulating them for cooling
Johnny, if you are still there...? Could you please elaborate a bit more on what you mean by "coil sizes matched more closely to the size of the object" ? How much margin between the outer edge of an object (assuming it's closely centered in the coil) and the inner edge of the coil would you recommend? Thanks...
@@sempertard These devices operated at extremely high voltage and the coils were actually copper tubing, round or square, and shaped for the area of the product that needs hardening with water flowing through the tubes constantly. For instance, one item we did a lot of was a rivet slightly larger than a human thumb and the part would go from ice cold steel to bright red hot in a matter of seconds and then quickly removed and dropped into the quench tank edit: another critical component was the requirement to never ever let the part touch the coil and for that we had heat resistant cloth and a paper-like substance I would say maybe keep experimenting with coil sizes and if more or less turns in a coil makes a difference and if you want to have a smaller gap then make sure to have something non-conductive between the coil and the part being heated edit-edit: the gap shown at 21:40 is perfect and to make it even more perfect could be to make a dedicated crucible coil and turn the lower coils such that they maintain that same perfect gap all the way down, making your perfect crucible coil even more perfecter :-) Also if you can shed enough heat from the copper tube with water you might see higher temps on the crucible EDIT, EDIT, EDIT: just came back from your channel looking for a video about how the coil worked after hooking up to water and if you've melted anything in the crucible...so, did it work ??
The weak link with these circuits are the resonant capacitors. So many required to obtain the capacitive value and series current capability. Pity those old high power valve transmitter caps are not more common. Capacitors out of the high frequency TiG welders are made by Miller and can take a lot of current. What is really needed are water cooled capacitors. Must try making some.
If i understood the function of this device correctly, the resonance is used to create a high frequency AC to drive the work coil? One could use a microcontroller instead to run the semiconductors at the desired frequency like in an switchmode powersupply and get rid of the coils and capacitors alltogether right? Perhaps only one big cap to smoothen the draw on the supply.
I am a paramedic and have very little knowledge about scratch building any electronics. But I must give credit when due, you are very neat and tidy when you build sir. I wonder if the engineers who built the Apollo rocket systems had to be just as creative, neat, and tidy?
Alternatively you can fill the tube with water and freeze it I found it this works better than using sand. Have made heat exchanger coils for condensers for large distillation purposes
I was going to recommend just bending the whole work coil to match the trace shape and directly soldering components to it through the PCB, but your idea would be a lot easier...
Interesting video. I was also thinking about a generator-based induction heater - getting into the 10-15kW range with a 20kVA second-hand genset. Potentially (in the future, after that) getting into higher power than that (Which is completely foreign to me, and will require a whole different set of considerations and learning curves) - but "medium DIY induction furnace" is definately one of the paths my creative endeavours seems to always realign to. I'd also be very interested in (and likely forced to) migrating to PF-corrected supply architectures as well. Subbed.
AWESOME I.H. and P.S. !!!!!!!!!!!!!!!!!!! I was going to comment on the capacitor diagonal connection, but you mentioned it as I was watching further. One thing that I have found done somewhat inadvertently in helical work-coils is in how many turns actually produce flux interaction with a work piece. If you get the opportunity, could you take the coil and connect a thin wire across the terminal leads near the coil, and then dip that 'closed loop' into a soap-bubble bath, then pull it out and just count how many 'closed' circular surfaces appear through the coil's center...these are the ones that contribute to heating / flux-coupling. I have found that it can often be less than it would seem by counting 'turns' by sight. Fractional loops at the ends can be very misleading. THANKS MUCH... Please let me know if / when you do it. I have seen cases when the true number of 'useful / legit' turns has been as many as two less than some large I.H.'s makers claim, even though it may look so. --dALe
Problem with the circuit is probably the Mosfets IRFP4668. They have a high gate capacity and tend to swing at the gate at the smallest cicuit induction in the gate path. Thus they dont switch fast and exact enough and build up a lot of heat quickly. Did you measure the gate signal with an oscilloscope? Would be interesting to see. They have to advantage of high voltage switching at high amperage, but they are hard to control in this self oscillating circuit.
@@papvendel7490 If you bought cheap at ebay, they were probably fake Mosfets - they burn easily. Maybe there was a problem with the gate signal, a ringing, and not properly switching. You have to measure the gate signal.
Instead of soldering those big wires to beef up the traces, I'd recommend the following on your next design: don't bus the caps. Each cap should have its own dedicated pair of traces going straight to the big terminals for the induction coil. I would make a roughly round PCB with the induction coil terminals in the center and the caps arranged in a circle around them. Also, I'd spread the cap traces over 4 layers so they can be really wide. Some heat sinking may be required at the terminals.
A good source of large inductors are old UPS's I have scrapped a number of APC 3000kva Rack mount UPS and they each have at least 6 large inductors, double the size of yours
Since you are using lead tin solder which is kinda regular solder. How about using a high quality solder that that has 6 percent silver. That would make the conductivity a lot better. What do you think fella ???
@@Schematix What else would you do to lower the electrical amperage of the unit Sir ??? Different or more capacitors, transistors. Heavier gauge wiring. WHAT Thanks
Another first class video. I will use this idea to make an instant water heater. Will pass a 3/4 inch galvanised steel pipe through the coil without touching the copper. Then push water through and will experiment with water flow . Please what is the frequency you are using. I thank you again for sharing. Firstclass project.
You have never welded have you? The way you soldered the board was how best to warp something when applying heat. Heat at one end and then move the heat slowly along it. You need to uniformly preheat what ever you can in an oven or by brushing with a flame or hot air gun. You should also solder the ends first then keep dividing the length by half for your next solder point.
Nice build not too bad how ever to help you along first All wire and components must be cleaned with 90%ipa (alcohol) One other thing when soldering you must cut the leeds so soldering covered it part of mil spec had to pass a 1 week course on it along with 2 year's of training was on a incoming nuke station USA wire wrapping was hard as well but have fun doing great job. Oh tools as well must be cleaned with ipa.
Thank you, this is fantastic... I am assuming that it heats up to 600-700 deg celcius? If one was to make the bus bar, would you solder it to the traces or just have it attached at correct points?
I'm not a julous type but I'm jealous of your tools and knowledge and how did you know it was the coil making the wave wrong and not something else these induction heaters look cool what happens when you only pass water thought it does it heat the water?
So, on the subject of coils: let's assume we want to use it for forging knives. Usually this means we start off from material that's relatively flat, having a more or less rectangular cross-section. Would it be helpful to have the coils conform to this shape rather than having them be round? Second question: since it's not a good idea to touch the coils while in operation, and since electricity can travel from the coil to your body through the tongs when the work piece touches the coil, how would you to about protecting the user from accidental contact? Would pottery enamel hold up, if we paint the coil in this? Will it adhere to the copper in the first place? Will it impact performance? Are there better options?
A couple of questions. 1-Is there a frequency that is chosen to heat steel? 2-I want to build this heater to temper knife blades. How could I modify the coil? Because it has a dimension of 300 mm x 3 mm thick.
I'm looking to build an induction cooktop heater. I don't like the ones in the market, rather than regulating the power to heater they seem to work on the on/off principle.
Hello, author of this channel, I have a question, if you add 4 more keys in sequence to the scheme, will it change the power in kilowatts to 5 kilowatts? If you have such a scheme or an idea for 5 kilowatts, please share it, thank you very much😊
Great stuff... Any idea how much all that would cost approximately and it's it scalable I.E. 10KW, 20KW, 30KW, etc. How about a more in depth explanation of the circuit say breaking it down into sub circuits like how is the Sine wave generated. Is there a PDF to accompany the build if one should try building it for themselves.
Hi, your video was excellent, can you please make a video on how to control a induction cooktop through a PID controller as hundreds of thousands of people use induction cooktop for brewing and it will be very helpful for all of us
That shouldn't be that hard to do. There are PID temperature controllers for industrial use that come with or are designed for standard Pt100 temperature probes and that have a simple open/close contact that can be programmed for a certain temperature. You can use that contact to switch a relay on and off, and wire the cable that powers your cooktop through that relay. If you google "pid pt100" you should be able to find one that fits your purpose, and for the realy you only need to watch out for that the contacts are rated for the voltage and amperage of your cooktop. You might also need a 24v dc powersupply both for the PID and the relay, as industrial equipment is usually powered by 24v DC (but you can get PID's and relays that run off the mains too).
Thanks buddy for the learnings 👍🏻 i have been using a cheap inverter arc welder for a power supply . It would be good to see your adaptation of using one to power your zvs .
Just wondering if you had a target resonance for the output? We use our induction heaters for copper brazing, but I can't recall the exact kHz value that works best for non ferrous metals. Would be good to have a small mobile unit like this for the smaller connections of leads in the series and parallel circuits. Great build! 👍
For my application I was targeting what would be referred to as 'medium-frequency' the end result was around 39~40kHz. For ferrous metal, my understanding is mid~low frequency yields quicker heat time due to the 'skin effect' of higher frequencies
@@Schematix Does this include the graphite cruicible? I have not found any info on what frequency they heat best. Ive got a 200ml size and made a 135mm diameter coil for it. 11 turns, about 5 meters, I will need to shorten it after some testing. Thanks for the video.
I have a distrust for these self starting "ZVS driver"s, they are picky about supply rail rise time, and also the frequency varies a lot with load (could be a good thing as it tracks for resonance). The problem is, unlike lower powered versions, once out of sync it could go boom. I'd love to see one implemented using a soft switching phase shifted full bridge! Maybe with big ass GaN dies even!
if you could drive a transformer core at them frequencies couldnt you make the core from an old microwave transformer? or even use the ballast coil from a treadmill? if it has to be fereite (iron/ceramic) you could make one using. 1. flyback transformer core from old tvs (they usually are split). 2. tv horizontal deflection yoke cores they are usually split and can be hand wound. all fereite cores are made of iron and ceramic and are fragile and can break very easy. the iron makes the magnetics work and the ceramic gives the shape and maybe allows the cores to be much more effiecent operation.
Excellent job, thanks! I notice at 20:26, your operating frequency is about 39kHz. At this frequency, nearly all current only flows very close to the surface of the conductor due to skin effect. The rule of thumb says it is a waste of copper to use wire much larger than the skin depth in radius. Skin depth at 39kHz is 0.33mm, suggesting #40 Litz wire, though something around #36 is almost as good and a lot cheaper. Cheaper still and not much worse in performance is finely-stranded conventional wire instead of Litz. Minimize the size of the solder joints, and do not add solder along the length of the wire where it is not needed for joints, and you will get less waste heat and losses and more power to the coil and work piece.
I used to work for Vulcan Forge in San Jose we used to forge 2 1/2 inch wrenches for Stanley and Proto in closed die forgings they forged at temperatures 2,300° f , and the induction heaters we used to use were also tuned to the coil, by connections made on the capacitor Bank, and every coil had deionized water going through it, and the billets were fed through it also had two rails liquid-cooled rails going down through the center of the coil, they were all made a stainless steel tubing. The forge had coils that can handle anywhere from a half inch material all the way up to 3 in Plus
What were the advantages of tuning it?
Did they tell you what happens if you don't tune it?
Good job, I enjoyed the layout, the specs on components, and the very accurate description of the solitary drawback, and the immediate fix, using two cores, and 12 gauge wire. The "scope" shows the exact impact of fixing the core saturation issue. Quite a professional job, I have a 300 amp tig welder in my shop without running water, and found two water cooled motorcycle radiators, Honda, if I remember right, am running a "vibrator pump" circulating water when the welder is on, and it's worked a decade with no problems. I run ten or twenty percent radiator fluid, and have a glass "ball" quart canning jar, with a hole drilled in the bottom and near the top on the side for copper tube epoxied in, for a visual flow indicator and fluid quality view. I leave about half the jar air, for expansion, but haven't ever gotten the water more than warm. Thanks for a great video, good sound electronics, and sound engineering.
I doubt DIY coils are really worth it.
Given that you need to probably buy excessive amount of 12AWG magnet wire as in whole spool, + the cores + the extra work, it's probably better to buy off the shelf inductors, for some people it might end up same as the parts cost, and you end up with superior inductors which will have lower losses.
You will be spending 25-35 euro for the parts, meanwhile you can buy two high power inductors for around 15 each.
Also if you end up buying different cores, you risk getting issues and having to buy different cores and wasting more time and wire.
Thanks for the video. Nice work! Just a note on why the temperature of the iron stops climbing. Resistance actually increases with heat but that is not the reason. At these power levels and frequencies the heating depends mostly on the magnetic properties of the workpiece. That is why a non magnetic workpiece does not heat as well. The magnetic workpiece actually becomes non magnetic when it reaches the curie point or curie temperature and does not react with the coil as strongly. For steel this is almost 1400 deg. F. You can mess around with a propane torch, a nail, and a magnet for a demonstration. Nickel and its alloys can do this at lower temperatures. Some magnetic coins are fun to play with (Canadian one dollar loonie).
Can't believe he said resistance decreases with temperature (never heard of superconductors buddy ?). At melting temp of iron its resistance has increased by 100%.
Eddy current losses versus hysteresis losses. Hysteresis losses are typically the bulk of your heating effectiveness with these frequencies and these metals. For metals that don't benefit from hysteresis losses (aka "nonmagnetic"), you generally need higher switching frequencies as well. Take all of this with a grain of salt and confirm yourself, but these are my experiences working in a place that designed these.
You must have triggered some Kens on another of your videos or else 79 people accidentally pressed the thumbs down button. Seriously, very nice board and very thoughtful design and follow-up to clean your sine. And I am grateful for the doubled toroid trick. When I lived in Houston there was a serious electronics surplus and salvage shop with buckets of fist sized toroids on down with maybe the largest being $10 to $15. I always swore to come back and raid their stock when I started prototyping induction furnaces but sadly I have moved too far away now and they don't do online sales. This was the type of place that used to have salvaged ships radars and teletype machines and copper vapor lasers and on and on. I could spend half a day drooling in the aisles.
I love EPO too moved to out of town but still get over their when I can
And if you need anything just let me know I can just grab it for you when I go I only live out past Katy Mills so it's like a 45 min to 8 hours trip depending on how dumb i10 or West Park are being
Loved this and am going to do it. On a side note, loved the fly checking out your work. It always adds to the video to have an assistant
Awesome! Thank you!
around the 21 minute mark you state the internal resistance of the metal decreases with heat - it is actually increasing in resistance.
This can be explained by faradays law of induction. because a constant Bflux is changing, a constant EMF is applied to the metal causing eddy currents. and P = V^2/R. which tells us as resistance increases, P decreases.
AFAIK it has to do with curie temperature
@@ichbineltyb3814 That's right. It's not the electrical resistance that's at issue here. He's heating a FERROUS metal. That means it's highly magnetic and the main heating effect is not from eddy currents but the continuous, rapid reversing of the magnetic field is 'jiggling' the magnetic domains of the metal, causing friction and hence heat. BUT as the metal heats up and passes the Curie point, the magnetic response suddenly drops and that strong heating effect ceases.
Correct i think he just got mixxed up
I'm sure it just a slip of the tongue.
I'm pretty sure he just misspoke.
Worked for Sundstrand Aerospace in packaging Wild frequency input from a generator mounted to a jet engine gearbox and rectifies it and converts it to 400Hz 3 phase. One in production is a 30KW unit. So I was responsible for coming up with cooling and packaging. One suggestion is to attach that copper tubing to the circuit board for the straight buses for the capacitor bank. Attaching the inductor and semiconductors to the back side with the tubing will make it easier to use the liquid to cool the circuit. Just a thought.
Thank you, for circuit, and board, files. You do an excellent, and thorough work, in your presentations. Thank You again.
You haven't seen hot until you've seen a 400 kW power supply on a 5' dia. chiller-cooled coil, heating a solid graphite billet surrounded by a few hundred pounds of powdered graphite. It was white hot for several hours and took all next day to cool. We were making the first woven carbon-carbon composite rocket nozzles and nose cones at SAI (before it was SAIC) for the Air Farce and NASA. Our technique was used to make the Galileo Jupiter probe's heat shield.
Actually, I've seen hot. In fact, I see hot every day. So have you, and everyone else you or I know. It hangs in the sky. 😉
This sounds awesome. Would love to know more about how this was done.
Please talk to TH-camr smarter every day.
Great to hear you say 'Soldering' as opposed to 'Soddering' ! Cheers mate; keep up the great work...
Nothing wrong with regional dialects having different pronunciations.
You do good work. People could learn a lot about doing things well regardless of the projects themselves. Thank you for the effort.
for larger inductor rings check out ham radio supply sources, large rings are used with lower frequency "HF' antennas.
funniest ad I've seen in a long while! You rock.
That board layout is absolutely phallic
Brilliant. I will be trying to make 3 heaters with elongated pancake coils to achieve 9 kw heating on top and sides. Leaving the bottom open to get a workpiece in the contraption freely and prevent contamination of the coils by falling debris. Thanks again.
Much better with the music by not making it too loud. Great video by the way!
LOL - when you pulled out the 2.0 model I audibly cheered I was so happy to see that! This is great, thank you for this video.
another tip - if you own one, pre-heat the pcb and copper wire with a heatgun which helps out when soldering with a undersized iron
I am wondering if the metal would get hotter if the coils were wound in a tighter circle. I believe the metal needs to get at least yellow hot for proper forging or perhaps other uses. I am thinking a nice project would be for you to make a second coil to swap out with this one for times when you want to heat up something smaller and make it hotter. Not criticizing what you did in anyway. You are a great teacher and demonstrator on here. love your channel.
He has already said for best performance to match coil with work piece. This coil was designed to match his crucible for melting metal.
@@mevk1 That crucible should be able to melt some serious aluminum.
I used to work at a forging company these induction heaters and it would just get to brilliant yellow you'd almost need sunglasses to look at it and we used to forge at 2,300°, if the Hammerman get it too hot when the Closed Dies hit the metal it would explode and shoot sparks everywhere just like a flint on a lighter
Very interesting project and you are obviously very skilled in electronics. I was drawn to your video by the possibility of using an induction heater for heating some smaller projects in my black smithing work. Firing up my gas forge to do small work is not very efficient. While very informative I want to mention that the steel was no where near a forging temperature. The steel needs to be between 1200- 1300 degrees centigrade to be effectively forged. That is an incandescing yellow heat when viewed in shade. I am confident that you, given your expertise would be able to design an induction circuit to meet the temperature criteria. Again a great video and information thank you for sharing it.
Using coaxial cable braid is the best for reenforcing copper pcb tracks. Sweat it onto the tracks with a liberal amount of solder. Flexible and easy to work with.
The 12 ga magnet wire I obtained was STIFF. I had to clamp my toroid stack in a soft jawed vice while winding it, and it still looked really sloppy. Schematix made it look so easy.... Oh well, hopefully it will still get the job done.
I started my own induction heater project, so far you are way ahead of my progress but your project is looking exacly as I want my project to look, amazing, cant wait for the water cooling system, that is goint to be very interesting
You can do it!
I am Nasuh Jaber
A variety of coil sizes matched more closely to the size of the object being heated will greatly increase the temperature of the object
Once worked for a company turning out induction heated rivets and such and our coils were not much more than .25" clearance or less around the part...also our coils had water circulating them for cooling
Johnny, if you are still there...? Could you please elaborate a bit more on what you mean by "coil sizes matched more closely to the size of the object" ? How much margin between the outer edge of an object (assuming it's closely centered in the coil) and the inner edge of the coil would you recommend? Thanks...
@@sempertard
These devices operated at extremely high voltage and the coils were actually copper tubing, round or square, and shaped for the area of the product that needs hardening with water flowing through the tubes constantly. For instance, one item we did a lot of was a rivet slightly larger than a human thumb and the part would go from ice cold steel to bright red hot in a matter of seconds and then quickly removed and dropped into the quench tank
edit: another critical component was the requirement to never ever let the part touch the coil and for that we had heat resistant cloth and a paper-like substance
I would say maybe keep experimenting with coil sizes and if more or less turns in a coil makes a difference and if you want to have a smaller gap then make sure to have something non-conductive between the coil and the part being heated
edit-edit: the gap shown at 21:40 is perfect and to make it even more perfect could be to make a dedicated crucible coil and turn the lower coils such that they maintain that same perfect gap all the way down, making your perfect crucible coil even more perfecter :-)
Also if you can shed enough heat from the copper tube with water you might see higher temps on the crucible
EDIT, EDIT, EDIT: just came back from your channel looking for a video about how the coil worked after hooking up to water and if you've melted anything in the crucible...so, did it work ??
Finally an Induction heater build in a language that I can understand
The weak link with these circuits are the resonant capacitors. So many required to obtain the capacitive value and series current capability. Pity those old high power valve transmitter caps are not more common. Capacitors out of the high frequency TiG welders are made by Miller and can take a lot of current. What is really needed are water cooled capacitors. Must try making some.
If i understood the function of this device correctly, the resonance is used to create a high frequency AC to drive the work coil? One could use a microcontroller instead to run the semiconductors at the desired frequency like in an switchmode powersupply and get rid of the coils and capacitors alltogether right? Perhaps only one big cap to smoothen the draw on the supply.
I'm afraid I know nothing about electrical things, but I sure would like to have something like that to harden items I machine. I'll be watching!
I am a paramedic and have very little knowledge about scratch building any electronics. But I must give credit when due, you are very neat and tidy when you build sir. I wonder if the engineers who built the Apollo rocket systems had to be just as creative, neat, and tidy?
Alternatively you can fill the tube with water and freeze it I found it this works better than using sand. Have made heat exchanger coils for condensers for large distillation purposes
if you use copper brake line to reinforce the traces. you could also use the reinforcemet for watercooling the PCB. possibly.
I was going to recommend just bending the whole work coil to match the trace shape and directly soldering components to it through the PCB, but your idea would be a lot easier...
@Mikron this video is perfect and you should aim to have a build just like this.
brother you are very good person
Excellent video rarely found on this subject
I made a similar induction heater and if you have them, use copper nuts and bolts to cut down on resistance at the joints
Interesting video. I was also thinking about a generator-based induction heater - getting into the 10-15kW range with a 20kVA second-hand genset. Potentially (in the future, after that) getting into higher power than that (Which is completely foreign to me, and will require a whole different set of considerations and learning curves) - but "medium DIY induction furnace" is definately one of the paths my creative endeavours seems to always realign to. I'd also be very interested in (and likely forced to) migrating to PF-corrected supply architectures as well. Subbed.
I like those clip-on TO-220 heat sinks.
Very impressive. Thanks for posting.
I would really like to see a video of you using the crucible! What a fantastic idea!!!
Can you tell me how people melt metal using induction heating ??? Look forward to see more of your creation.
Really cool project. I cant help but see that it resembles the Neo Armstrong Cyclone Jet Armstrong Cannon
Great job.😃i too try to find ways to lower costs for various electronic parts. things get very expensive.
AWESOME I.H. and P.S. !!!!!!!!!!!!!!!!!!!
I was going to comment on the capacitor diagonal connection, but you mentioned it as I was watching further.
One thing that I have found done somewhat inadvertently in helical work-coils is in how many turns actually produce flux interaction with a work piece.
If you get the opportunity, could you take the coil and connect a thin wire across the terminal leads near the coil, and then dip that 'closed loop' into a soap-bubble bath, then pull it out and just count how many 'closed' circular surfaces appear through the coil's center...these are the ones that contribute to heating / flux-coupling. I have found that it can often be less than it would seem by counting 'turns' by sight. Fractional loops at the ends can be very misleading.
THANKS MUCH...
Please let me know if / when you do it. I have seen cases when the true number of 'useful / legit' turns has been as many as two less than some large I.H.'s makers claim, even though it may look so.
--dALe
2.5mm is single strand solid here in the UK. I’ve made psu busbars from hammering small bore copper pipe flat.
multi stranded wire is more rf efficient ie less loss
Nice tankless water heater!!!
Problem with the circuit is probably the Mosfets IRFP4668. They have a high gate capacity and tend to swing at the gate at the smallest cicuit induction in the gate path. Thus they dont switch fast and exact enough and build up a lot of heat quickly. Did you measure the gate signal with an oscilloscope? Would be interesting to see. They have to advantage of high voltage switching at high amperage, but they are hard to control in this self oscillating circuit.
@@papvendel7490 If you bought cheap at ebay, they were probably fake Mosfets - they burn easily. Maybe there was a problem with the gate signal, a ringing, and not properly switching. You have to measure the gate signal.
You are a very talented person. It was also very entertaining to watch you build it.
Great heater/melter. Can't wait to see where this one takes you.
Thanks for the free Circuit Schematic. 😊😊
PC water cooling parts would be perfect for this project. It's 1/4 inch standard, comes in many sizes and uses standard PC fan sizes.
Omg 😲this is amazing! 👏 I wish I had the tools and money to do the projects you do.
yeah man you gotta get that steel a bright orange
Nice project!! I want to see you melting cast iron and bronze using this induction heater
if you hold somthing with a metal bezel that heats up , it seems like nice heat controll on object
Instead of soldering those big wires to beef up the traces, I'd recommend the following on your next design: don't bus the caps. Each cap should have its own dedicated pair of traces going straight to the big terminals for the induction coil. I would make a roughly round PCB with the induction coil terminals in the center and the caps arranged in a circle around them. Also, I'd spread the cap traces over 4 layers so they can be really wide. Some heat sinking may be required at the terminals.
NICE INSIGHT
you could also manufacture your own toroid rings capable of handling higher inductance loads
Amazing build and video
Fantastic work. Please scale this design up to 5-10 KW.
I think he bought a pair of them. If everyone requests maybe he will parallel them -- then display 6KW of glorious induction powerage.
A good source of large inductors are old UPS's I have scrapped a number of APC 3000kva Rack mount UPS and they each have at least 6 large inductors, double the size of yours
Thanks for this video, very helpful.
Great video. Try hooking up a high voltage transformer to the output next 😼
Yeah I rekon!
Since you are using lead tin solder which is kinda regular solder. How about using a high quality solder that that has 6 percent silver. That would make the conductivity a lot better. What do you think fella ???
Silver solder would be a better option than lead based solder for this application (assuming you can afford it ;)
@@Schematix What else would you do to lower the electrical amperage of the unit Sir ??? Different or more capacitors, transistors. Heavier gauge wiring. WHAT Thanks
Thanks very much for not using music background...
Where you add copper to tracks could the actual tracks be increased to say an inch wide. There seems to be plenty of board to use.
Very good
Another first class video. I will use this idea to make an instant water heater. Will pass a 3/4 inch galvanised steel pipe through the coil without touching the copper. Then push water through and will experiment with water flow . Please what is the frequency you are using. I thank you again for sharing. Firstclass project.
Please also if you can tell us the copper coil inductance.
You can get cheap contact thermometers from ali, then you can read from specific points
You have never welded have you? The way you soldered the board was how best to warp something when applying heat.
Heat at one end and then move the heat slowly along it. You need to uniformly preheat what ever you can in an oven or by brushing with a flame or hot air gun. You should also solder the ends first then keep dividing the length by half for your next solder point.
Some nice ingenuity employed here
Nice build not too bad how ever to help you along first All wire and components must be cleaned with 90%ipa (alcohol) One other thing when soldering you must cut the leeds so soldering covered it part of mil spec had to pass a 1 week course on it along with 2 year's of training was on a incoming nuke station USA wire wrapping was hard as well but have fun doing great job. Oh tools as well must be cleaned with ipa.
Ce type doit être alcoolique. Autrement il aurait précisé pourquoi ce lavage. Pas fort.
Jlcpcb isn't paying you enough. Loved the ad
Thank you, this is fantastic... I am assuming that it heats up to 600-700 deg celcius?
If one was to make the bus bar, would you solder it to the traces or just have it attached at correct points?
I'm not a julous type but I'm jealous of your tools and knowledge and how did you know it was the coil making the wave wrong and not something else these induction heaters look cool what happens when you only pass water thought it does it heat the water?
So, on the subject of coils: let's assume we want to use it for forging knives. Usually this means we start off from material that's relatively flat, having a more or less rectangular cross-section. Would it be helpful to have the coils conform to this shape rather than having them be round?
Second question: since it's not a good idea to touch the coils while in operation, and since electricity can travel from the coil to your body through the tongs when the work piece touches the coil, how would you to about protecting the user from accidental contact? Would pottery enamel hold up, if we paint the coil in this? Will it adhere to the copper in the first place? Will it impact performance? Are there better options?
A couple of questions.
1-Is there a frequency that is chosen to heat steel?
2-I want to build this heater to temper knife blades. How could I modify the coil? Because it has a dimension of 300 mm x 3 mm thick.
I see, thank you, I will use your inductor oversaturation tips also the frequency, as reference to make similar inductor
Thanks for the great video. It was amazing. well done.
Welldone cant wait for the foundry video
I'm looking to build an induction cooktop heater. I don't like the ones in the market, rather than regulating the power to heater they seem to work on the on/off principle.
You deserve more subscribers mate thanks for the content. 👍🏻
What was the temperature it got up to when making the metal red? What is the highest temperature it can do?
i assume if you want to melt metal you have to switch it using a controller and avoid auto oscillation, you will have to use IGBTs
Hello, author of this channel, I have a question, if you add 4 more keys in sequence to the scheme, will it change the power in kilowatts to 5 kilowatts? If you have such a scheme or an idea for 5 kilowatts, please share it, thank you very much😊
Great videos !
Hello sir you did a great work.please can you buy welding machine schematics from 120amp to 200amps
Given the equilibrium you mention at #21:05, how do induction furnaces melt iron? Building an induction furnace is what I'd like to do.
beautiful
Now I got a Norwegian Made CAN-Bus controller for this Huawei R4850G2 Powersuply reddy for you :D
Great stuff... Any idea how much all that would cost approximately and it's it scalable I.E. 10KW, 20KW, 30KW, etc.
How about a more in depth explanation of the circuit say breaking it down into sub circuits like how is the Sine wave generated.
Is there a PDF to accompany the build if one should try building it for themselves.
Hi, your video was excellent, can you please make a video on how to control a induction cooktop through a PID controller as hundreds of thousands of people use induction cooktop for brewing and it will be very helpful for all of us
That shouldn't be that hard to do. There are PID temperature controllers for industrial use that come with or are designed for standard Pt100 temperature probes and that have a simple open/close contact that can be programmed for a certain temperature. You can use that contact to switch a relay on and off, and wire the cable that powers your cooktop through that relay. If you google "pid pt100" you should be able to find one that fits your purpose, and for the realy you only need to watch out for that the contacts are rated for the voltage and amperage of your cooktop. You might also need a 24v dc powersupply both for the PID and the relay, as industrial equipment is usually powered by 24v DC (but you can get PID's and relays that run off the mains too).
Thanks buddy for the learnings 👍🏻 i have been using a cheap inverter arc welder for a power supply . It would be good to see your adaptation of using one to power your zvs .
really really nice! are there any bits inside that are leathal? I dont really want to get hurt or worse. Superb video thanks!
So iron and steel can be heated with an induction heater. What other metal's can be excited ? Can sulfate forms of various metal be induced with heat?
Just wondering if you had a target resonance for the output? We use our induction heaters for copper brazing, but I can't recall the exact kHz value that works best for non ferrous metals. Would be good to have a small mobile unit like this for the smaller connections of leads in the series and parallel circuits. Great build! 👍
For my application I was targeting what would be referred to as 'medium-frequency' the end result was around 39~40kHz. For ferrous metal, my understanding is mid~low frequency yields quicker heat time due to the 'skin effect' of higher frequencies
@@Schematix Thanks! Yes I assumed you must of had a specific inductance level to feed the resonance tank based upon your toroid windings. Nice work.
@@Schematix Does this include the graphite cruicible? I have not found any info on what frequency they heat best. Ive got a 200ml size and made a 135mm diameter coil for it. 11 turns, about 5 meters, I will need to shorten it after some testing.
Thanks for the video.
Esde projeto pra usar em brasagem de tubos de cobre seria importante pois e portátil me favoreceria muito
I use desoldering wire for reinforcement.
I have a distrust for these self starting "ZVS driver"s, they are picky about supply rail rise time, and also the frequency varies a lot with load (could be a good thing as it tracks for resonance). The problem is, unlike lower powered versions, once out of sync it could go boom. I'd love to see one implemented using a soft switching phase shifted full bridge! Maybe with big ass GaN dies even!
Yup it will burn instantly.
if you could drive a transformer core at them frequencies couldnt you make the core from an old microwave transformer?
or even use the ballast coil from a treadmill?
if it has to be fereite (iron/ceramic) you could make one using.
1. flyback transformer core from old tvs (they usually are split).
2. tv horizontal deflection yoke cores they are usually split and can be hand wound.
all fereite cores are made of iron and ceramic and are fragile and can break very easy.
the iron makes the magnetics work and the ceramic gives the shape and maybe allows the cores to be much more effiecent operation.
Schematix wi do r2 r3 r4 r1 are in the scheme 1k value and 150 ohm and in your video ar 47ohm and 470ohm thx for the answer
with that you can also melt metal in a vacuumed controled area.
Can you do a smaller build of this for those of us who uses an induction heater in place of a butane torch for our vaporizers for weed?