IRFP250 MOSFET circuit design help.

[gary350]

The 20 awg should handle the current. The 24 is a little small.

How large do the Heat Sinks need to be on the MOSFET?

I wish I had a way to set each MOSFET to run at 200kHz using a 555 timer will be guess work?

There are several online inductor calculators each one gives me a different answer but some are in the same ball park. One calculator says #20 wire it good for 180 watts. I remembered from my experence building coils 40 years ago the more compact the coil is the high the uh value. I have a box full of cardboard coil forms so I will built this.

Inductance 0.05 mH
DC Resistance 0.11 Ohms
Wire Gauge 20 AWG
Wire Diameter 32 mils (1 mil = .001 in)
Coil Length 0.5 in
Coil Inner Diameter 0.63 in
Coil Outer Diameter 0.88 in
Average Turn Diameter 0.71 in
Wire Length 10.81 feet
Copper Weight 0.03 pounds
Turns 58
Levels 3.71
Turns/Level 15.63

 

[ronv]

How large do the Heat Sinks need to be on the MOSFET?

Each fet will dissapate about 1.2 watts (4 amps across .075 ohms) so only a small heatsink will be needed. One of the little ones like you can find at radio shack should do.

I wish I had a way to set each MOSFET to run at 200kHz using a 555 timer will be guess work?

What test equipment do you have available? Can you measure AC current? It will need to be adjusted to compensate for the capacitors and inductor anyway. The target frequency is about 178 Khz because of the larger capacitors across the work coil. Needed to do this to keep the current in spec for the caps.

There are several online inductor calculators each one gives me a different answer but some are in the same ball park. One calculator says #20 wire it good for 180 watts. I remembered from my experence building coils 40 years ago the more compact the coil is the high the uh value. I have a box full of cardboard coil forms so I will built this.

Inductance 0.05 mH
DC Resistance 0.11 Ohms
Wire Gauge 20 AWG
Wire Diameter 32 mils (1 mil = .001 in)
Coil Length 0.5 in
Coil Inner Diameter 0.63 in
Coil Outer Diameter 0.88 in
Average Turn Diameter 0.71 in
Wire Length 10.81 feet
Copper Weight 0.03 pounds
Turns 58
Levels 3.71
Turns/Level 15.63

Should be close enough.

 

[gary350]

I sold all my test equipment years ago including an RF generator that I could use for this project. All I have now is a Volt Ohm meter.

I was reading that the MOSFET operates best at a certain frequency an RF generator lets you find the freq your circuit likes best.

 

[ronv]

I sold all my test equipment years ago including an RF generator that I could use for this project. All I have now is a Volt Ohm meter.

This project will be a little bit complicated and most errors will "Pop" the FETs. Without any test equopment it will be difficult to test the individual parts before turning on the high power stuff. I'm concerned that you won't be happy later on, but it's up to you.

I was reading that the MOSFET operates best at a certain frequency an RF generator lets you find the freq your circuit likes best.

Yes, the capacitor and big inductor form a resonant circuit that greatly increases the current in the work coil, but due to tolerances in the caps and inductors you need to search for the exact frequency of resonance. We can do this with the 555 almost as easily as the generator but we have no way to tell when we have hit the sweet spot except to see something get hot..

 

[gary350]

I have a lot of experence with 555 timers they are temperature sensitive. If this induction heater has to stay on frequency don't count on a 555 timer to do that once it warms up the frequency will drift way off. Fire it up on a cold day and it will way off in the opposite direction. I have never tried a heat sink or fan on a 555 timer wonder if that would work? They are not very accurate above 20K the frequency goes up very fast then it is hard to adjust but that problem could be solved with a fixed resistor connected to an adjustable resistor.

I could buy some test equipment again on ebay they have some pretty good stuff at good prices. Now that I am retired I might be building a lot of cool fun projects. What test equipment do I need?

I have another idea. I have a good electronic friend from college that has test equipment. I can make all the parts, assemble the project then mail it to him. Maybe he can get it adjusted and running.

 

[ronv]

Well lets get the system together and see what you think. A scope would do it.
Most of the time if you have temperature problems with a 555 circuit it is because of the capacitor used in the timing circuit. It's easy for them to vary a few percent over temperature, but if you use a cog or npo they are quite stable. See this write up in the spec sheet.
https://www.mouser.com/Search/Refine.aspx?Keyword=c410c102j1g5ca

Attached is a timer circuit that should work. I'll add the circuit to keep both fets from being on at the same time later.

Do you have what you need for the power supply?

 

[gary350]

Well lets get the system together and see what you think. A scope would do it.
Most of the time if you have temperature problems with a 555 circuit it is because of the capacitor used in the timing circuit. It's easy for them to vary a few percent over temperature, but if you use a cog or npo they are quite stable. See this write up in the spec sheet.
https://www.mouser.com/Search/Refine.aspx?Keyword=c410c102j1g5ca

Attached is a timer circuit that should work. I'll add the circuit to keep both fets from being on at the same time later.

Do you have what you need for the power supply?

I was planning to use the house wall electric 120 VAC outlet as the power supply. A bridge rectifier and a filter cap I will have 170 VDC. I have 34 caps that are 76,000. MFD rated 200 VDC peak 250 VDC and several smaller ones. I may have to buy a bridge rectifier the correct size I have a lot of diodes I'm not sure what size I will need for this project. I have a bridge that is no longer marked and I don't remember what the rating is. I have 2 diodes rated 600 amps 600 volts with a 20 lb aluminum heat sink. I have about 20 diodes from 100 amps to 300 amps I will have to check the number to know what they are. The smaller diodes I'm not sure about some have no numbers but I do have a package of 1N4004 diodes rated 1 amp 1000 volts.

 

[ronv]

You need an isolation transformer or it will be a death trap. It won't be cheap, but maybe on ebay? It needs to be at least 5 amps RMS.

 

[gary350]

You need an isolation transformer or it will be a death trap. It won't be cheap, but maybe on ebay? It needs to be at least 5 amps RMS.

I have several transformers, what do I need?

 

[ronv]

It would need to be 120v in 120v out and it will be big! Something like this:

https://www.google.com/products/cat...V5GT7_KGM_JiQK2sITbDQ&sqi=2&ved=0CI0BEPMCMAQ#

 

[shortbus=]

That is what the two microwave transformers were doing in your post #4 of this thread, making an isolation transformer from them .

 

[ronv]

Good idea!

Here is the timer and the delay to keep both FETs from conducting at the same time.

Note you also need a 12 volt supply. A wall wort will do.

 

[gary350]

Good idea!

Here is the timer and the delay to keep both FETs from conducting at the same time.

Note you also need a 12 volt supply. A wall wort will do.

I think 1 MOSFET is a much simpler design.

2 microwave transformer for isolation.

A variac to start at with low voltage about 5 volts to make adjustments.

What about a tickler coil feedback oscillator circuit to make the MOSFET go on/off at the frequency of the tuned circuit???

 

[ronv]

]I think 1 MOSFET is a much simpler design.

Much easier, but alas it doesn't work.

2 microwave transformer for isolation.

Good idea!

A variac to start at with low voltage about 5 volts to make adjustments.

Another good idea!

What about a tickler coil feedback oscillator circuit to make the MOSFET go on/off at the frequency of the tuned circuit???

Don't know about that one. Do you have a circuit in mind?

 

[gary350]

This is the circuit I had in mind.

**broken link removed**

 

[ronv]

Think about it this way. If the inductors were just a wire the 470NF capacitor would just charge up to 180 volts because there is nothing to discharge it.

 

[gary350]

Think about it this way. If the inductors were just a wire the 470NF capacitor would just charge up to 180 volts because there is nothing to discharge it.

OH.........now it makes sense. One MOSFET charges the capacitors the other MOSFET discharges it.

 

[gary350]

You need the bottom MOSFET so you have a path to ground on the oposite half cycle.
Use 12 of these caps in parallel for the resonant circuit.
https://www.mouser.com/ProductDetail/Cornell-Dubilier/940C30S47K-F/?qs=gSIQ2CbtXsYGKLy6248XRQ==
Make the matching inductor 50Uh.
You can use a single .47 Ufd of the same type for the second cap.
Try out the calculator for the inductors and let us know what you come up with.

The system is a few hundred watts the fets are a very small part of that power loss.

I am ready to start building this circuit. Is 1/4" copper tubing large enough for the work coil or should I use 3/8" copper tubing?

**broken link removed**

 

[ronv]

I would think 1/4 inch would be enough for 80 amps.
Maybe you could post the whole thing so we can take a look at it.

 

[gary350]

I would think 1/4 inch would be enough for 80 amps.
Maybe you could post the whole thing so we can take a look at it.

\

I will order the MOSFETs and Capacitors I hope they arrive before I leave.

I can't build this without a circuit to keep both MOSFETs from being on at the same time. I was wondering if a solid state relay would work. One NO the other NC use the 555 to turn them both on/off at the same time.