Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

10 Kilowatt Induction Heater

Status
Not open for further replies.

MOSFET KILLER

New Member
Hello Everyone,
I have started this thread to post schematics, pictures, block diagrams and questions regarding my 10,000 watt induction heater project. I will outline most of the major components of the circuit and describe their function.

So, lets start with the power supply, I have a 240VAC, 60 Ampere line (14.4kVA). This is then passed through soft start resistors to charge the filter capacitors and prevent a massive current surge from tripping fail safes. Since a bridge rectifier distorts the current waveform I have included an active power factor correction circuit. I have chose to limit the output power by varying the DC inverter supply voltage. To do this I have placed a buck regulator on the ~400VDC output of my PFC pre-regulator. The buck regulator will give me control of the output power from 0 to 10,000 watts.

On to the frequency control circuitry, the two inputs are from a current transformer and the high side oscillator. They are compared by a type II phase detector and filtered by an RC network. As you can see it is essentially a phase locked loop that keeps the oscillator (a TL494) on the resonant frequency of the series resonant tank which can vary with loading. I also have control over the phase angle to keep the power factor low (as close to 1 as possible).

The IGBTs in my half bridge inverter are driven by a pair of identical IGBT drivers that consist of optically coupled TC4422 IGBT driver ICs, undervoltage lockout, and antisaturation detection circuits. Both the high and low side IGBT drivers are coupled to the oscillator with optocouplers for equal timing. The antisaturation detection circuits compare the IGBT collector voltage to a reference voltage and will give a high output if the collector and gate are both high at the same time.

My current detector provides feedback to the buck regulator to limit output power if the current becomes too high. This circuit will also trip an overcurrent latch if the current is high enough to indicate a fault.

I have also included safety interlocks since I am building a commercial model induction heater. They cover things such as cooling water and component temperatures.

The IGBT drivers drive a half bridge of IGBTs that turn the ~400VDC output of the power supply into high frequency AC (about 20 kHz). The inverter output supplies the primary of a ferrite transformer with a 10:1 turns ratio. The secondary is stepped down in voltage and isolated from the mains. The secondary supplies AC to a series resonant LC circuit, I opted to use series resonance after discussions with **broken link removed**. He told me that the capacitor in a parallel resonant circuit will periodically short circuit the inverter during switching, but in series switching is better.

I have covered most of the important components of this induction heater, questions, comments, suggestions are welcome.

Thank You,
Alex Sears
 

Attachments

  • 10,000 Induction Heater Block Diagram.jpg
    10,000 Induction Heater Block Diagram.jpg
    54.2 KB · Views: 12,073
Last edited:
hmmm there are two details here....

1) what you are asking is well beyond most posters here.

2) what you are doing does not appear to be building but designing a COMMERCIAL unit for SALE, and people take offense at being used for other people's gain. The typical response in the forum is "go hire a contract design firm"

while the exception to this is the occasional hobbyist that is trying to follow his dream, 10KW is beyond what a hobbyist would need for his hobby. I could imagine one looking for an induction heater to build his own sand casting setup, but again not 10KW. That is the point an electric stove circuit is at and fully half what many houses have available total before blowing the main breaker. No one here is dumb enough to believe that you are going to hire an electrician to add a circuit to your house or pull the plug on your kitchen stove to power up your "toy".

Dan
 
Last edited:
Well, right now I am in the process of designing the power supply, I was wondering how much filter capacitance would be required. The supply is 240V RMS which is then passed through my PFC pre-regulator that outputs ~400VDC the inverter will draw 50A RMS. The first filter capacitor will have to be the biggest because it has to smooth out 120Hz, the one after the buck regulator will be smaller since it operates at >50kHz. I will also place some smaller polypropylene capacitors at the output of the buck regulator for high frequency filtering.
 
Last edited:
The amount of capacitance required depends upon the amount of ripple voltage you can tolerate.

Would it be possible to operate the unit from 3-phase power? That greatly reduces the rectifier output ripple and also the amount of filter capacitance required for a given ripple.
 
I would if I could, it's too bad that I can't get 3ph at my house in the suburbs. I will be using the inverter current feedback to keep the frequency on resonance so I don't think that my control circuit can handle much ripple. On the other hand I may be able to filter it out with a high pass filter network and be able to use less filter capacitance. I will need to use high voltage capacitors due to the ~400VDC output from the PFC pre-regulator.

Unlike what Ubergeek63 thinks, I am not designing a unit for sale. Selling a commercial unit is way beyond my capibilities, who is going to trust a 15 year old. Now, do not start to lecture me about my age because I have been learning about electronics for >10 years! I am simply designing an induction furnace for my own projects.
 
In the larger inverter based welders and plasma cutters that are single phase compatible the most common capacitance to current standard falls between 200 uf and 600 uf per amp.

So if your power supply draws around 50 amps at 240 volts AC a capacitor bank in the 450 volt 10,000 to 30,000 uf range may be sufficient.

Emphasis on the BANK setup. Preferably around 6 to 12 smaller capacitors in parallel is most common and gives the greatest service life.
 
I would if I could, it's too bad that I can't get 3ph at my house in the suburbs. I will be using the inverter current feedback to keep the frequency on resonance so I don't think that my control circuit can handle much ripple. On the other hand I may be able to filter it out with a high pass filter network and be able to use less filter capacitance. I will need to use high voltage capacitors due to the ~400VDC output from the PFC pre-regulator.

Unlike what Ubergeek63 thinks, I am not designing a unit for sale. Selling a commercial unit is way beyond my capibilities, who is going to trust a 15 year old. Now, do not start to lecture me about my age because I have been learning about electronics for >10 years! I am simply designing an induction furnace for my own projects.

Actually your undergrad mentor has no business advising you in this nor does anyone have any business advising you to do anything but give up. The fact is it is far beyond you and more than most degreed engineers that do not have a power electronics specialization would dare attempt.
 
15 years old? Stop the project immediately, you'll kill yourself or at the very least vaporize large amount of metal and cause property damage if you chose so much as a single wrong component at those power levels.
In order to even draw that much power you would have to tap DIRECTLY into the breaker box at your home, which is illegal.

I'm 32 and consider myself relatively competent with electronics with several years of casual sometimes serious study, and I wouldn't even attempt to do something of that scale.
 
I am not just going to stick a bunch of components together half assed and plug it in to the 240V. I will test each individual part at low power before connecting everything together and testing at low power before even thinking about plugging it in to mains.
 
Then start your design off with the intent to build a 1 10 or 100 watt induction heater and scale up from there AFTER you have the lower wattage portions complete first.
 
I have already constructed a 1kw heater about a week ago. I had about 1000 watts RMS of inverter output 15 amps peak, 10 amps RMS, 106 volts RMS, PF >.9 all in all, pretty efficent. The major design flaw was that the transformer was only rated for 2 amps RMS max, I only ran the heater for about 30 second pulses and allowed the transformer about half an hour to cool between uses. I did not operate it from the mains because the setup was not really well insulated on my table and I did not want to get electrocuted. I have been able to heat a hardened steel screwdriver bit for an electric drill to bright orange in that time though. I have experience with switching electronics but I appreciate you concern.
 
I have finished the buck regulator controller schematic. I used an MC34060A PWM controller, it covers 0% - 100% so I have full control over the output power. I have optocoupled the gate driver section to the PWM controller because I will use a floating supply for the gate driver for the same reason one would be used for a high side gate driver. I have not labelled the RC timing components on the PWM section because I want to determine the best operating frequency once I choose some other dependent components; inductor, transistor, etc.
 

Attachments

  • Buck Regulator Controller.jpg
    Buck Regulator Controller.jpg
    40.2 KB · Views: 4,810
The operating frequency is also dependent on the actual target. Higher generally being better.
 
Yes, also I dont want a massive expensive inductor but I want to be able to use IGBTs.
 
arc flash hazards

Young Man,
If you want to live to be a middle aged man then you don't want to build this project. It appears you have no training in the hazards associated with power electronics. Do you understand what an arc flash is, and what it will do to you? At those voltage, current ratings a short circuit will cause the metal in the device to vaporize within .1 seconds, it will reach approximately 30,000 degrees. The fire ball will expand as much as 6 feet from the source. You will be engulfed, and it will burn you to the bone. The shock wave can collapse your lungs. inhaling the vapor will totally destroy your lungs. If you survived which is doubtful you would be disabled for the remainder of your life. As a side note you would probably burn down your fathers house. This type of project only belongs in a concrete building with proper shielding, and safe guards. Not to mention the required by federal law P.P.E. that you would be required to ware when working on live power electronics. My best advice to you is find another project. Go to school get an engineering degree, learn about the hazards of what you want to do. Then pursue it from a point of knowledge, and safety.
 
That is assuming that I take ZERO safety precautions, I am going to have short circuit protection and several redundant protection circuits that will prevent such a greatly exaggerated event from occuring. If you would look at some of my schematics you would see that I know what I am doing and I have taken appropriate safety precautions such as overcurrent protection curcuits that can shut down the unit in microseconds. I have also been learning about and experimenting with power electronics for about 2 years now, now I know this does not qualify me for anything but it should allow me to persue this safely.
 
I say go for it! When I was 15 I was building Tesla coils powering them off of 15 KV sign transformers plus working with and rewiring 240 volt mains powered equipment all the time. You are the best source of knowledge for what you are comfortable working with.

As far as a 6 foot arc flask from a 240 volt line, Huh? :confused:
The last 6 foot arc flash I saw generated by a man made power source took a 4160 volt AC 500 amp service going dead short in a 3000 HP motor controller at a coal mine about 16 years ago when I was a student co-op worker.

10 Kw is about what a standard 250 amp stick welder draws from a 50 amp line at full output or what your average electric oven draws at full power. For what its worth a the open circuit voltage on most stick welders is around 80 - 100+ volts but the inductive spikes off of the leads during welding can go to several hundred volts.

I dont know where some of these people live but where I am from every farm and most garages have at least one or more 240 volt sockets and most are on a 30, 50, 60, or even 100 amp fuse or breaker. I have four independent 100 amp sockets in my shop, one outside my shop and one off my main power panel.
There is nothing illegal about using a 240 volt power source. Its basically just two 120 volt lines 180 degrees out of phase from each other and uses a pair of higher amp protection devices.

Sure there is the potential to kill or severely injure yourself but from what I see you working with already I would say your probably being far safer and realistic about it than most others on this forum. Where I come from at age 15 we are handling high powered hunting rifles, heavy farm equipment, vehicles on roads, compound bows with wicked arrows, motorcycles, four wheelers, and obviously still living to be older and wiser from it.

You most likely all ready know the basics of safety and are aware of this but but just in case wear rubber soled foot wear and put your live circuits on dry wood surfaces if anything is exposed and try your best to keep one hand in your pocket or at least firmly grabbing something non conductive. By doing so you are greatly reducing the chance of any large current from passing through your heart or body. You may pass a heavy current through your hand and burn yourself but you will live at least! :)

Have fun and never be afraid learn something that most others are to afraid to learn. You will be a better, smarter, and wiser person for it I promise! :)
 
That is assuming that I take ZERO safety precautions, I am going to have short circuit protection and several redundant protection circuits that will prevent such a greatly exaggerated event from occuring. If you would look at some of my schematics you would see that I know what I am doing and I have taken appropriate safety precautions such as overcurrent protection curcuits that can shut down the unit in microseconds. I have also been learning about and experimenting with power electronics for about 2 years now, now I know this does not qualify me for anything but it should allow me to persue this safely.

If you have this stuff sitting on a table, and you have no shielding or PPE no matter what you think your design is capable of you are taking ZERO safety precautions. What I described to you is not exaggerated. I saw it happen to a guy three years ago. and he had 25 years experience as a licensed professional. If you are only 15 there is no way you know enough to do this right. I read the part where you have an underrated transformer and had to give it several hours to cool between uses. Every time you overheat it that way you are slowly breaking down it's dialectic material. At some point it will fail and start an arc flash event. I fear for you and your family's safety I wonder if your parents know what you are doing in their house. You need to do a reality check. I am pretty sure this will not change your mind. Go find a professional power electronics engineer and show him what you are doing and let him read this. I have no doubt he will verify what I have told you
 
tcmtech, you live on a farm, we're talking city folk residential here =) Place around here only have 100 or 200amp SERVICE not breakers.
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top