I have another question, hope some one will help me.
I'm working with a full bridge inverter coupled to an LC series resonator through an impedance-matching transformer for induction heating applications.
The inverter switches in ZVS just above the resonant frequency, about 450 kHz and delivers a sinusoidal current.
The transformer is made with an ETD59 core and 3F3 ferrite, and it has 8 turns at primary and two turns at secondary (coupling ratio 4:1).
The system is supplyed by the rectified main (230Vrms - 320 Vpk) and the actual power delivered to the probe through induction coupling is about 3 kW.
I need to increase the output power by reducing the turns ratio of the trafo.

Questions:

1 - I'm not sure if I can reduce the turns at primary or if it's better to increase the turns at secondary: how can I calculate if we are close or far from saturation ?

2 - How can I calculate if the ETD59 is still a suitable core or if I have to go for a larger core ?

Thanks for supporting

1) If you are operating at 450 kHz with the 3F3 material core loss would be more of a concern than saturation, operating close to saturation would not be possible at that frequency without major core loss heating problems, most likely the core is operating well under the saturation level.
2) A larger core requirement would depend on the magnitude of the power increase and the design margins in the present transformer. A 20% increase might be possible but a 50% would most likely not be possible.

Hi Davide, I have some free of charge design software from Mag Inc that I can send to you if you want to get in touch, it will allow you to experiment with different materials and configurations?

Hello Jim, the trafo currently runs at full power (3kW) and its surface reaches about 60°C at steady state. It is not too much, but I think we cannot go over (I need twice as much power). How can I calculate the correct trafo?

Thank you Peter, just let me know how can I get the software. I will try with it and let you know the results.

At the outset, I must admit I don't have any practical experience with LLC resonant inverters - so you will have to take my comments in that light. Conceptually, I do believe that reducing the primary turns would:
1) Reduce the magnetizing inductance and hence impact the resonant frequency
2) Induce an earlier onset of saturation.

For these reasons, increasing the secondary turns should be the preferred option.