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.

1GBT Induction Heater Circuit Design help needed.

Status
Not open for further replies.

gary350

Well-Known Member
Does the 1GBT operate the same as a Mosfet?

I took an existing mosfet circuit and put the 1GBT in the circuit, will this work?

Before I go to the trouble of doing the math I need to make sure this circuit will work. I am going to use water cooled 1/2" copper tubing as wire. The circuit will be self contained in a case with no power supply transformer. My 225 amp Lincoln stick welder will be the power supply transformer. Pump, radiator and cooling fans will keep the induction heater circuit cool. I have 600 amp 600 volt diodes to convert welder AC to DC. I also have 800A 600V diodes but dont think they are needed.

When I need to use the induction heater I will clamp the welder cables to the terminals on the back side of the induction heater then set the amp load on the welder with the selector switch.

I am not sure the 1N5819 diodes are correct for this 1GBT circuit?

I also need to determine the corrent resistor value for the gate circuit?

I need to calculate the correct choke uh and experement building a large choke coil that will work?

I may need to put the power supply capacitors and rectifiers in a seperate case with cooling the capacitors will take up a lot of space.

I want this circuit to be bullet proof. All the parts need to be over kill and cooled. Induction coils will attaching to the front of the unit with brass and copper tubing fittings. I have some high power water cooled resistors 1 ohm and 10 ohms I'm not sure if I can use these. The cooling system will be filled with transformer oil.

What is gate current on this 1BGT ?


**broken link removed**
 
Last edited:
To fully understand what you are asking, you must define objectives.
Source power (V,A,W)
Load power (f, W)
Load type? iron or Fe + Al

1N5819 is only rated for 1A . Why these?
 
To fully understand what you are asking, you must define objectives.
Source power (V,A,W)
Load power (f, W)
Load type? iron or Fe + Al

1N5819 is only rated for 1A . Why these?

Because 1N5819 are already in this mickey mouse circuit. 1GBT replace the Mosfets, already said that above. Diodes probably need to be replaced or 20 diodes put in parallel for 20 amps. I dont know gate amps yet.

My welder is 65 volts AC. Amps will probably never been 200 but never know until I test it to see if it will do what needs to be done that is what the selector does on the welder. I need to heat 2"and 5" diameter pipe red hot. Assume max load is 65 volts, 100 amps = 92 VDC at 100 amps = 9200 watts. 200 amps would te 18KW and 225 amps will be almost 21KW.

I finally found a data sheet that is not trash advertisements. Max gate voltate 20. Threshold 4.5/6/7.5. If I am reading this correctly cutoff current is only 2ma?

The circuit needs to be changed. The gates need their own power supply voltage.
 
Last edited:
I get the feeling the OP has a hands on mechanical background and isn't much into specifications. Make it big, make it strong, if anything breaks , make it bigger next time.
 
Do you mean IGBT (not 1GBT)? IGBT stands for Insulated Gate Bipolar Transistor.

To answer your question (well, not really, but partially), an IGBT is effectively a MOSFET switching a bipolar transistor. When looking from the gate it resembles a MOSFET, but when looking from the collector/emitter, it looks like a BJT.

IGBT_equivalent_circuit_EN.svg.png
 
I notice your circuit is asymmetrical. Is that intentional?
 
Do you mean IGBT (not 1GBT)? IGBT stands for Insulated Gate Bipolar Transistor.

To answer your question (well, not really, but partially), an IGBT is effectively a MOSFET switching a bipolar transistor. When looking from the gate it resembles a MOSFET, but when looking from the collector/emitter, it looks like a BJT.

View attachment 92630

$_12.JPG


**broken link removed**
 
The data sheet says gate is 7.5v 2ma. Does this mean the internal resistance of the 1GBT limits current to 2ma or does the gate resistor do that? I'm not sure I know how to calculate the correct value of the gate resistor.
 
The gate charge is 3000nC or 3 uC. The switching overheads are about 450ns. The suggested Gate resistor appears to be 2.1 ohms. I generally use 4.7 Ω or more to limit ringing also a ferrite bead close to the gate will help.
If your gate drive can deliver say 3A then you can charge the gate in about 3uS ( as 1 C=1A*1S). Add in switching overheads and you can see you will switch on or off in about 4uS. Increase the gate current for faster charging. To achieve higher gate currents with the gate resistor you should drive the Vge to perhaps 18V for cleaner switching.

**broken link removed**

PS: Use a gate drive that can source AND sink. Just sourcing and relying on a resistor to sink will slow your switching and leave the collector-emitter in an active region and cause excessive heating.
 
The data sheet says gate is 7.5v 2ma. Does this mean the internal resistance of the 1GBT limits current to 2ma or does the gate resistor do that? I'm not sure I know how to calculate the correct value of the gate resistor.

The 7.5V spec is for the maximum emitter-gate turn on threshold. That means that it might not turn on at all with a voltage less than that. I would suggest a minimum voltage of 15.

But you're going to have a big problem with the passive gate drive circuit you are using. The gate capacitance, Ciss, for that IGBT is 120nF, which is huge, and will run Very Slow with a passive gate driver.

If you want to use IGBT's, you really need to use a real PWM controller and some high current gate drivers.
 
The 7.5V spec is for the maximum emitter-gate turn on threshold. That means that it might not turn on at all with a voltage less than that. I would suggest a minimum voltage of 15.

But you're going to have a big problem with the passive gate drive circuit you are using. The gate capacitance, Ciss, for that IGBT is 120nF, which is huge, and will run Very Slow with a passive gate driver.

If you want to use IGBT's, you really need to use a real PWM controller and some high current gate drivers.

How do I do a PWM controller?
 
The UCC37322 is not in any of my books. Datasheet online it not telling me what I need to know. I operates on 15vdc, one IN and 2 OUTs. But what sets the frequency of this device? All the circuit drawings show both outputs 7 and 8 connected in parallel, how does that work?

It seems to me the outputs need to be 180 degrees apart, one to trigger #1 IGBT and one to trigger #2 IGBT at a pre determines Hz of about 100KHz.
 
You use 'books' for cross reference?
Try the internet.
https://www.newark.com/on-semicondu...3&langId=-1&storeId=10194&gs=true&st=UCC37322

EDIT: these are gate drivers, which are fed by the PWM....read the whole guide dude.

EDIT2: The Gate driver runs a push pull source and sink.....so the push and pull pins are tied together to provide a higher current path. You'd need a separate UC37322 for each IGBT.

Please note that there are several pages to this guide...this one:
https://www.instructables.com/id/30-kVA-Induction-Heater/step3/Principle-of-Operation/
discusses the operation principle.
 
Last edited:
It does not seem right the circuit is forced to run at a different Hz than the LC resonance frequency. Data sheet tells me nothing about how to set the Hz of the TL494 or how to determine the value of the other parts. Data sheet tells me nothing about how to determine part values for the UCC37322 either. I dont like to copy someone elses circuit I would rather LEARN how it works.

002_zpstsiwrkta.jpg
 
Last edited:
So what exactly is wrong with using the correct driver IC's and other related circuitry to properly control your IGBT's the right way?
 
It does not seem right the circuit is forced to run at a different Hz than the LC resonance frequency. Data sheet tells me nothing about how to set the Hz of the TL494 or how to determine the value of the other parts. Data sheet tells me nothing about how to determine part values for the UCC37322 either. I dont like to copy someone elses circuit I would rather LEARN how it works.

Learning is admirable. Learning from an example is very efficient.
You approach to take an example circuit and then just change the parts because you think it could work requires much experience and understanding of how and why the components work. Your questions show a lack of such knowledge.
Thus, make something that works, learn from it , then perhaps make your own with your own alterations. Tweaking high power and high heat designs without understanding will get expensive and perhaps dangerous fast.
Perhaps you should make a very small unit to test your own ideas. At least it won't explode and damage things.

The forum is here to help , but we can't teach the discipline. That's your effort.

EDIT: I don't know what datasheet you're looking at but the specifications given DO inform on what conditions the part will work under. Then you use various methods to create those conditions as per your needs. If the TL494 spec says it can oscillate between 1Khz and 300Khz based on the given relaxation formula, how is it you cannot calculate the R & C that u need?
 
So what exactly is wrong with using the correct driver IC's and other related circuitry to properly control your IGBT's the right way?

I would like to read the Data Sheet and learn how these devices work. I have no clue what the 20K or the 50K variable resistors do? I looked at several data sheets and never did learn how to set the Hz of the TL494. One data sheet says pins 8 and 11 are the outputs but a different data sheet says pins 9 and 10 are the outputs. Frustrated.
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top