A cheap 120W Chinese induction heater module uses a simple resonant circuit for workshop heating tasks like loosening stuck screws and hardening small metal parts.
Its design uses two MOSFETs, two fast diodes, two inductors, and an external coil plus high-capacitance resonant elements.
On 12V DC, it drew 1.8A idle and peaked around 12A, reaching nearly 150W.
It heated a 6.5mm screwdriver red-hot in 2 minutes, an eight screw in 3 minutes, and a bolt with nuts in 4 minutes.
A universal PSU became unstable from interference, and the 120W rating remains limiting for larger or awkward workpieces.
Well, he cooperates (or rather learn) from a Polish inventor who can do wonders with power supplies. He also does heaters, but mainly power supplies for electron guns and magnetrons (To see a 100kV regulated power supply was something, especially since it wasn't that big).
I hope that I will be able to praise my toy by the end of the year, whether it is a success when I heat something up, or a show of what it looks like when something goes wrong
@_lazor_ if you have a ready construction, be sure to present it in DIY. This 100kV power supply had some huge resistive divider in the feedback loop or some kind of black magic?
@robokop yes, I used a transformer soldering iron, but there the tip attracted the cut ends of the resistors on each side, and the electromagnetic field did not heat up the nearby sections to redness
Yes, a specially made voltage divider, even showed me samples, which he ordered with such a divider directly on the PCB, where on the one hand the resistance and on the other hand the path led as the filter capacity. Such tensions are magic.
This is my idea, because I have enough topics with power supplies in the half bridge topology where everything is done without thinking, and the topic is really extensive and writing everything from scratch is tiring, so it will excel with a calculation and an example of a PCB in a circuit maker . Progress can even be watched: https://circuitmaker.com/Projects/Details/Marcin-Szajner/inductive-heater-STM32F334
However, the work is progressing slowly.
Imagine a transformer with tens or even hundreds of amps at the tip and with a frequency not 50Hz but 50kHz and more.
@_lazor_ if you have a ready construction, be sure to present it in DIY. This 100kV power supply had some huge resistive divider in the feedback loop or some kind of black magic?
Such power supplies are not unusual - they are commonly used for insulation testing (old electricians and power engineers remember "kenetrons"). The voltage was also regulated there - because connecting full voltage immediately triggered the short-circuit protection system.
TechEkspert wrote:
@robokop yes, I used a transformer soldering iron, but there the tip attracted the cut ends of the resistors on each side, and the electromagnetic field did not heat up the nearby sections to redness
@_lazor_ if you have a ready construction, be sure to present it in DIY. This 100kV power supply had some huge resistive divider in the feedback loop or some kind of black magic?
Such power supplies are not unusual - they are commonly used for insulation testing (old electricians and power engineers remember "kenetrons"). The voltage was also regulated there - because connecting full voltage immediately triggered the short-circuit protection system.
But for 10kW? And I will add that it was a power supply for an electron gun, so at this 100kV potential there was another power supply to generate voltage on the Wehnelt cylinder.
As for the directionality, it was more about the fact that on one side there was a magnetic armature, so made that it did not melt and only one side of the coil caused heating.
But for 10kW? And I will add that it was a power supply for an electron gun, so at this 100kV potential there was another power supply to generate voltage on the Wehnelt cylinder.
Power supply with a voltage of 100kV with a power of 10kW ??? Some invention to knock down planes?
I don't know what it went to, and if I did, I probably wouldn't say it anyway. However, there are devices such as clystrons where such a power supply can work.
But the field comes out only on one side, thanks to a piece of jumper made of an undefined material, you can also see a lot of Kapton tape on the wires leading to the coil.
In general, magnetic circuits are probably not my forte, just like advanced inductance details are rather a mystery to me, although about the basics of inductance I once managed to do material: Some applications of inductance :)
Anyway, I think that making material about something, making a project, etc. is the best way to learn something new.
Hello. A bit old topic, but I will add my observations. I bought this toy some time ago, straight from a Chinese on ebay. It is admittedly low power, but quite a nice and useful device. In my DIY workshop, she has helped me many times to even remove or attach metal axles in plastic wheels of car models. Of course, I also started experimenting with the converter, which ended up puncturing the mosfets and giving her a vacation in a drawer. Recently, I found this post and, like the author of the article, replaced the transistors. There were no dedicated (in my case, the casings of the pierced mosfets were not damaged and the inscriptions could be read) or those suggested by the author of the post. So I soldered BUZ 11 and the heater came back to life. The transistors are big as for this design, but like the author, I have the impression that the converter works better than with the original ones: the PCB does not heat up, the induction heats up the elements faster. The article helped me a lot. I greet the author.
✨ The discussion revolves around the utility of a 120W induction heater, particularly its effectiveness in workshop applications such as unscrewing stuck screws and hardening small metals. Users express skepticism about the heater's low power, noting that while it can achieve high temperatures, it may not be suitable for larger tasks or metals like steel. Some participants suggest alternatives with higher power ratings, while others share experiences of modifying the device for better performance. The conversation also touches on the importance of the heating method, the materials being heated, and the potential for radio interference. Overall, the consensus leans towards the device being more of a novelty than a practical workshop tool, although it can be useful for specific small-scale applications.
What maximum temperature can a 120 W ZVS heater reach?
Steel can be brought to visible red heat (~900 °C), limited by the Curie point (~770 °C) where heating slows [Wikipedia; Elektrode, vodiczka, #17240393]. Small samples reach that in 2–3 minutes; larger masses plateau lower because of heat loss [Elektroda, bestler, post #17239210]
Does the metal type change achievable temperature?
Yes. Ferromagnetic steels heat fastest until the Curie point; non-magnetic metals rely solely on eddy-current losses. Aluminium hardly heats unless placed in a crucible, while copper needs higher power density [Elektroda, CMS, post #17242838]
Is temperature adjustable on these modules?
No built-in control exists; output follows input voltage and load coupling [Elektroda, domel1, post #17239324] Add external PWM or a DC bench supply with current limit for basic regulation.
Users report noticeable noise; automatic PSU relays chattered and FM reception degraded within 1 m [Elektroda, bestler, #17239210; TechEkspert, #17241205]. Shielded leads and ferrite beads help.
Do FM radios and phones still work nearby?
FM sensitivity drops, but a mobile phone kept service at 2 m in informal tests [Elektroda, bestler, post #17241505] Results vary with coil size and housing.
Can the coil be shortened to boost efficiency?
Yes—fewer turns reduce inductance, raising resonance. Match it with the capacitor bank to stay in the 100–200 kHz sweet spot for steel [Elektroda, bestler, post #17241256]
May I use a UPS transformer instead of the twin chokes?
No. The two toroidal chokes must not be magnetically coupled; a transformer would saturate and kill ZVS action [Elektroda, lazor, post #17251861]
What power supply works best?
Server SMPS bricks (12 V / 60–100 A) cost little and handle surge current better than variable bench PSUs that may trip [Elektroda, robokop, post #17239849]
Fit low-Rds(on) TO-251/TO-220 MOSFETs rated ≥50 A, 40 V.
Add a small heat-sink and test at 5 A before full power.
Any safety tips for workshop use?
Keep coil insulated; hot metal can spit. Use eye protection, ventilate fumes, and avoid placing bank cards or watches within 20 cm due to strong magnetic fields.