Induction Heater Simulation

[sabu31]

Hi all,

I tried to simulate, induction heater based on royer circuit.
However, the circuit is not oscillating.

What could be the issue?

 

[andre_luis]

Not everybody here have the LTspice application instaled.
Consider posting a schematic picture aswell.

 

[sabu31]

Hi all,

I am attaching the schematic.

 

[andre_luis]

An oscillator circuit consists of an amplifier stage and a tank selective attenuator stage, in series.
I can't distinguish how the above LC net performs such a job.

 

[sabu31]

An oscillator circuit consists of an amplifier stage and a tank selective attenuator stage, in series.
I can't distinguish how the above LC net performs such a job.

Thanks for the reply. I am trying to simulate schematic given in Induction Heater

 

[dick_freebird]

I recommend you use NPNs rather than power MOSFETs,
because the power MOSFET is likely quite capacitive and
needs a big voltage swing to toggle, so harder to kick into
oscillation (BJT needs much smaller signal).

Also recommend that you separate the oscillator from the
power drive. The push-pull Royer is going to spend a lot
of time in inefficient regions, a bang-bang bridge driver
with a tank-locked oscillator (you would like to sit on the
resonant frequency no matter what varies, so make the
heater coil the dominant "tank"?) would let you optimize
each, to its strengths.

 

[FvM]

Power supply should be DC, not AC.

@dick_freebird These MOSFET push-pull induction heater modules are available in various variants e.g. at aliexpress, basically working.

 

[Easy peasy]

Why is the power source a sine ? it should be DC.

--- Updated Nov 10, 2022 ---

p.s. strictly speaking it is not a Royer, it is a Mazzilli, as the mosfets pull down the other's gate.

 

[sabu31]

Power supply should be DC, not AC.

@dick_freebird These MOSFET push-pull induction heater modules are available in various variants e.g. at aliexpress, basically working.

I am giving 12V DC with 1 V ac ripple. I thought pure DC was not initiating oscillation. What could be issue with it not oscillating.

 

[FvM]

O.k.,1V sine with 12 V DC offset should be o.k. but is unlikely to help starting oscillations. If uic option doesn't help, you need a small asymmetrie in resistor values or a "kick start" pulse. I didn't yet check in LTspice.

 

[FvM]

The problem is rather trivial.

Your MOSFET type is "NMOS", with a few mA Idss and very low gm. Can't oscillate.
Why not choose IRFP4668 used in the link?

 

[sabu31]

The problem is rather trivial.

Your MOSFET type is "NMOS", with a few mA Idss and very low gm. Can't oscillate.
Why not choose IRFP4668 used in the link?

Thanks FvM for the reply. There was no IRFP4668 in LTPSICE to simulate. So used generic model.
Now I have replace one mosfet with AON6290 the current is now in 800 A peak with short duration width. What should be changed to increase pulse width.

 

[FvM]

There was no IRFP4668 in LTPSICE to simulate.

It's included in LTspice built-in library since many years. NMOS generic model is a small MOSFET as used in logic ICs, no power MOSFET. You may use it to design a milliwatt heater with respective component values.

Now I have replace one mosfet with AON6290

Take two.

 

[sabu31]

It's included in LTspice built-in library since many years. NMOS generic model is a small MOSFET as used in logic ICs, no power MOSFET. You may use it to design a milliwatt heater with respective component values.

Take two.

Hi FVM ,

I tried with two such devices instead of one. But I am getting huge spikes in both voltage and current. Is is due to the anti-parallel diode not being present. Please let me know if any thing is wrong in the simulation. Also if there any literature references for same.

 

[BradtheRad]

Do L1 L2 L3 share a common flux field (and common core)? Typically it's a center-tap transformer (where DC supply goes to center tap).

Not sure since you want the flux to heat nearby metal?

 

[sabu31]

Do L1 L2 L3 share a common flux field (and common core)? Typically it's a center-tap transformer (where DC supply goes to center tap).

Not sure since you want the flux to heat nearby metal?

All the inductors are wound on different toroidal cores