New Induction Heater circuit with no center tap.

[tcmtech]

IR2110 (500 volt) or IR2113 (600 volt) from digikey for ~$3 or ~$4.50

 

[tcmtech]

Yes, that is true for many parts. A non-related area but, when I had a Ford Granada, the track rod ends were, say, $10UK each, but an identical track rod end for a Ford Escort (much smaller automobile) was £2.50UK.

Been a normal event in my life since my teens.
Lincoln town car with a 302 V8 starter, $200. 74 Ford F100 pickup with same 302 V8 starter, $50. Different part number and boxes on the auto parts supply store shelf but identical units right down to the last digit stamped on both staters inside the boxes.
And that most often followed by some ignorant parts jockey still insisting that the $50 one wont fit the lincoln engine for reasons he can't begin to explain given the overwhelming evidence that they are in fact the same units right in front of him.

 

[spec]

I use to run into that all the time at work name brand parts were unreasonable expensive but common identical parts were cheap, some companies were more expensive than others. Honeywell was the worse a $125 micro switch was only $1 from the Micro switch manufacturer, if you know what the replacement part is and where to buy it, it saves a lot of money. Now days companies call those, proprietary parts, I call it greedy deceptions.

Ball bearing races are the same. At the garage where I worked evenings and weekends, we had an SKF bearing catalogue and just read the bearing reference off, and ordered direct from SKF. The cost difference was astronomical. Mind you, the garage still charged the customer the auto manufacturer's price.

What IGBT driver do I need? Where do I buy it? How much does it cost?

See post #31. Texas instruments UCC21520, 2 of required, around $6.00US each from DigiKey, Mouser etc.
https://www.digikey.com/product-detail/en/texas-instruments/UCC21520DW/296-44611-5-ND/6140192
https://www.mouser.co.uk/ProductDetail/Texas-Instruments/UCC21520DW/?qs=sGAEpiMZZMv3em63uDEI/FbxacumN%2bnAg5al5y18oH4=
https://www.ti.com/lit/an/slua778a/slua778a.pdf


spec

(Our posts crossed TCM)

 

[tcmtech]

(Our posts crossed TCM)

Not too worried about it. I figure about 10% of my posts ever get read all the way through anyways and of those maybe 10% of what they did read gets remembered correctly if at all.

 

[gary350]

I have never worked with these parks before, I have no clue what I am doing. This is the circuit in the pdf file it has minimum parts for no part values on the capacitors and diode? It has no dead zone. Do I still need a capacitor bank on the load coil? How does the Hz frequency get adjusted or set?

I am thinking I can connect 120 vac house voltage to a bridge rectifier and filter to the 600V and ground terminal start out with 170VDC until i get this thing working. Do i need current limiting to 14 amps to keep it from blowing the 15 amp house breaker?

What about the capacitor between the 2 IBGTs?

Looks like i need a 15vdc input power supply to the IR2113 data says 2 amps. I read somewhere the IGBTs need 9 amps.?

The other driver UCC21520 says it has a dead zone and it requires more parts with no part values on the drawing.?

**broken link removed**

 

[tcmtech]

The two outputs of your TL494 IC go to the Lin and Hin terminals of the driver IC and the driver IC takes those signals and conditions them to drive the gates of the switching devices.

As for the 9 map ratings on the IGBT's that the likely instantaneous peak pulse current the gates could theoretically draw if you had a driver circuit that had crazy high rise times. Realistically the 2 map rating of the driver IC outputs is more than enough to run even the largest IGBT units at many tens of KHz hard switching frequencies without issue.

As for your circuit you still have it drawn wrong. I could explain it but really you have to do some of this stuff yourself. Google has way more patience and educational merit than I can ever supply.

 

[spec]

I have never worked with these parks before, I have no clue what I am doing.

The other driver UCC21520 says it has a dead zone and it requires more parts with no part values on the drawing.?

The UCC21520 is no more difficult to use and understand than standard gate drivers. If you do not want dead-zone, simply do not use it.

The UCC1250 requires no extra components, to speak of. The gate drive values can be extracted from the MOSFET/IGBT data sheet or from the other driver data sheets.

Quite simply the UCC21520 is the best performing driver on the market at the moment. It has the advantage of high speed, high current drive capability, and most of all, complete isolation between the calm of the low level control electronics and the hell and fury at the gates of the high power, high voltage, output transistors. And later on, you will be able to get the benifits of dead zone to hopefully improve the performance of your circuit.

Remember if you need a particular circuit or a value sorted, just ask.

spec

 

[spec]

Here you are Gary from the UC21520 data sheet:

You just need two of these circuits and connect the untapped primary of your induction heater between the two outputs.

We can suggest values and how to drive the completed bridge- that is not the difficult part.

The difficult part is getting the gate drives right and getting the high-power transistors (MOSFET or IGBT) happy driving the induction heater transformer.

The mains power supply will also beed to be sorted too.

spec

 

[gary350]

Here you are Gary from the UC21520 data sheet:

View attachment 102172​

You just need two of these circuits and connect the untapped primary of your induction heater between the two outputs.

We can suggest values and how to drive the completed bridge- that is not the difficult part.

The difficult part is getting the gate drives right and getting the high-power transistors (MOSFET or IGBT) happy driving the induction heater transformer.

The mains power supply will also beed to be sorted too.

spec

I understand what Dead Zone is that will be a much better circuit but never worked with this.

I have no problem with extra parts, I just don't know how to find the values of capacitors, resistors, diodes in that circuit.

What is the rectangle box to the left where it says, VCC, PMW, uC, Analog or Digital?

I don't know the abbreviation, VCC, VDD, VSS, SW?

I think HV DC link is probably the power supply?

Where does the load coil connect and does it have a capacitor bank like the mosfet circuit?

What is the symbol to the left of HV DC Link it looks like an Antenna symbol?

 

[tcmtech]

Quite simply the UCC21520 is the best performing driver on the market at the moment.

I will have to look into those now. The old IR2113 has been my 'go to' H/L driver IC for years.

 

[tcmtech]

I have no problem with extra parts, I just don't know how to find the values of capacitors, resistors, diodes in that circuit.

What is the rectangle box to the left where it says, VCC, PMW, uC, Analog or Digital?

I don't know the abbreviation, VCC, VDD, VSS, SW?

I think HV DC link is probably the power supply?

Where does the load coil connect and does it have a capacitor bank like the mosfet circuit?

What is the symbol to the left of HV DC Link it looks like an Antenna symbol?

Application notes on how to hook up your circuit.
https://www.ti.com/lit/an/slua778a/slua778a.pdf

Complete driver circuit board ready to go. ~$30.
https://www.mouser.com/ProductDetail/Texas-Instruments/UCC21520EVM-286/?qs=SMCV84mnLhlC5iOQdcy6tw==

 

[spec]

I understand what Dead Zone is that will be a much better circuit but never worked with this.

It is dead easy and just amounts to a resistor or voltage value- we can sort that

I have no problem with extra parts.

Great

I just don't know how to find the values of capacitors, resistors, diodes in that circuit.

We can sort that- don't worry it is no big deal

What is the rectangle box to the left where it says, VCC, PMW, uC, Analog or Digital?

It is just that- a box showing lots of options. It is a general way to shoe how the various signals could be controlled. In your application you would have just one in[put from an oscillator and that would be it. No sweat. It may look complex if you are not familiar with this part but there is nothing complicated really. As I say, we can show you all the connections you will need.

I don't know the abbreviation, VCC, VDD, VSS, SW?

This is dead simple: VCC is the positive supply line for the logic on the chip. The ground symbol is the negative line of the same supply.
VDD is the positive boot strap supply that is generated to drive the top transistor gate with enough voltage to turn it on.
VSS is just the bridge 0V line (earth or chassis).
SW stands for switch and is just sloppy labeling. It is the output from that half of the bridge (where one end of your coil would connect).

I think HV DC link is probably the power supply?

That is correct. It is just sloppy labeling again.

Where does the load coil connect and does it have a capacitor bank like the mosfet circuit?

One end of the coil connects to the point marked SW on the left half bridge and the other end of the coil connects to the same point point on the right half of the bridge. No capacitors are involved.

What is the symbol to the left of HV DC Link it looks like an Antenna symbol?

It is absolutely nothing- just sloppy drawing. You can rub it out if you want.

Honestly, although this chip and the signals may seem complicated, they are not. In fact, the way you will be using it will not be much different to other drivers you have used already

spec

 

[spec]

I will have to look into those now. The old IR2113 has been my 'go to' H/L driver IC for years.

The IR2113 is a damn good chip and, as you say, has been the go-to driver chip for a long time.

But the UCC21520 is the latest thing. It can source 4A and sink 6A but, the fact that you can completely isolate the logic circuits, the low side drive, and the high side drive, is the deal maker. It also means that there is practically no limit to the bridge positive supply line voltage. They are fast too, with a small timing skew.

When these chips first came out they were a bit pricey but, now they have come down the $6US level, one of, they are much more attractive, especially as the price is still dropping. I have some with milk for breakfast every morning.

spec

 

[tcmtech]

That full isolation and ability to run as a dual fully isolated high side driver is what got my interest that and the less than $7 price tag.

The stuff I work with doesn't need the high speed though. I can't say I have ever had an application where I needed to run more than a few tens of KHz ever came up.

 

[spec]

That full isolation and ability to run as a dual fully isolated high side driver is what got my interest that and the less than $7 price tag.

The stuff I work with doesn't need the high speed though. I can't say I have ever had an application where I needed to run more than a few tens of KHz ever came up.

Yeah the isolation is good- it gives a great deal of flexibility in your design.

It's not the switching frequency that is the big deal to me. It is the how fast you can turn the MOSFET/IGBT gates on and off, and also the timing skew, which if too big, means you have to have more dead time in a bridge or a current limiting configuration, an inductor in the power line for example.

I wounder what the next thing will be. Seems like something new get introduced every month. It is a job to take it all in.

spec

 

[spec]

POST ISSUE 03 of 2016_11_14

Gary, it seems to me that there are two good approaches for your induction heater:
(1) Full bridge: four MOSFET/IGBT
(2) Half bridge with +- supplies: two MOSFET/IGBT

I think (2) has a lot of benefits so, for your consideration, I knocked out a quick block diagram to illustrating the approach using a single UCC21520 chip and two high power MOSFET/IGBT:

​

I have been a bit lavish with the 15V power supplies, and while two are not absolutely essential, they do give complete isolation between the logic and bridge. They also improve performance and make the operation of the circuit easier to understand. And PSUs are dirt cheap on the net anyway.

spec

ERRATA
(1) Change fo to 200KHz
(2) Change power supplies to 12V @ 1A

 

[gary350]

Gary, it seems to me that there are two good approaches for your induction heater:
(1) Full bridge (four transistors (MOSFET/IGBT)
(2) Half bridge (two MOSFET/IGBT) with +- supplies.

I think (2) has a lot of benefits so, for your consideration, I knocked out a quick block diagram to illustrating the approach using a single UCC21520 chip and two high power transistors:

**broken link removed**
​

I have been a bit lavish with the 15V power supplies, and while two are not absolutely essential, they do give complete isolation between the logic and bridge. They also improve performance and make the operation of the circuit easier to understand. And PSUs are dirt cheap on the net anyway.

spec

I have 1200W MOT wired for 21V already but no CT. I have a 1400W MOT I can make into 22V with CT. I was hoping to run the 1200V IGBTs on 120 vac with bridge rectifier and filter = 170 VDC on a 15 amp breaker. Low voltage is probably safer than to have a work coil running at 170V that might zap the heck out of someone.

Looks like 1 of the diodes is backwards.

Where do i get a 100KHz OSC .

What is divide by 2 ?

How many amps are the 15v PS ?

What are part values on diodes and capacitors?

Where does -22v go ?

What connects to the other 2 terminals of Q1 and Q2 ?

**broken link removed**

 

[tcmtech]

Where do i get a 100KHz OSC .

What is divide by 2 ?

How many amps are the 15v PS ?

I would just put your TL494 IC in that spot and use it to generate the variable frequency and PWM for everything.

As for the power supply I have never ran anything but one 12 - 15 volt 1 amp unit for such control circuits on anything I have ever built. Although in your case being the driver IC's Spec recommends are fully isolated output devices I would run a separate power supply for that side of the system.

As for current limiting if you want to run your circuit on full 170 VDC power putting your coil tank circuit capacitors in series with your heating coil will work to limit your overall current draw through simple capacitive impedance as I alluded to and showed in my reference to the high powered H bridge systems used in the big welding power supplies.

 

[spec]

I have 1200W MOT wired for 21V already but no CT. I have a 1400W MOT I can make into 22V with CT.

Great. I only put in +- 22V supplies because that was on one of the schematics that you previously posted. The bridge plus and minus supply voltages can be whatever you like, but they must be the same magnitude and must match the V/I characteristics of your coil for maximum heating and efficiency.

I was hoping to run the 1200V IGBTs on 120 vac with bridge rectifier and filter = 170 VDC on a 15 amp breaker. Low voltage is probably safer than to have a work coil running at 170V that might zap the heck out of someone.

From a safety point of view, you are only allowed to have a maximum of 30V exposed to human contact. Any higher voltage and you would need to insulate the coil, which would be difficult, but not impossible. I would suggest going for 24V, nominal (30V max), as this is an industry standard voltage. It will also fit in with a possible future approach of using lead acid batteries in parallel with the bridge +-22V mains supply to enhance the induction heater supply short-term current capability.

Looks like 1 of the diodes is backwards.

Well spotted- please collect your prize from the mods. (sketch corrected by the way)

Where do i get a 100KHz OSC .

What is divide by 2 ?

Easy peasy- I will post a schematic tomorrow. Its getting late here. The divide by two just ensures that you get an exact 1:1 mark-to-space ratio of the input waveform feeding the bridge driver. By the way the oscillator should be 200KHz to account for the divide by two stage.

How many amps are the 15v PS?

Change PSUs to 12V at 1A (see ERRATA) Around 10 bucks each supply on ebay.

What are part values on diodes and capacitors?

That will have to be analyzed and specified, but no big deal

Where does -22v go ?

-22V is the bridge negative supply rail.

What connects to the other 2 terminals of Q1 and Q2?

That will be calculated when we have details of the MOSFET/IGBT that you intend to use, but almost certainly a resistor with a diode in series with a resistor in parallel with the resistor.

As I said before, don't worry! There is a standard procedure for establishing the design and all the values. Then you will be able to build a good test-bed for your induction heating investigations.

I don't like to be a nag, but we will also have to work out an optimum physical layout to get the best performance, but from what I have seen of your work you are pretty good at the physical side.

spec

 

[spec]

Hi Again Gary,

This is all there is to the oscillator and divide by two. The chips are hellish expensive though- about 30 cents each.

spec

​

ERRATA
(1) Change 15V SUPPLY LINE to 12V SUPPLY LINE.