Single Ended Quasi Resonant Converter Design

[sabu31]

Dear All,

I want to know the design process for a single-ended quasi-resonant converter for use in induction cooking. I have gone through few app notes as attached. But have doubts regarding the design process regarding the selection of Lr,Cr. I have noted that C value is typically 5uF (for cooktops used in utility line).
What should be the value of Lr, R for the resonant frequency of 20Khz.
What other parameters can affect power level.

My input voltage is typically 24V-48V and need to it for operation at 1kW power levels.

Any reference/appnotes for the design process will be helpful.

Thanking you.

 

[BradtheRad]

With 5uF C value, you can get an idea of L value by applying the standard formula for LC resonant frequency.

If you want to conduct a certain Ampere level, calculate the C value which presents sufficiently low impedance at 20 kHz.
Use similar method to calculate L value so it permits sufficiently low impedance at 20 kHz.

 

[sabu31]

Thanks BradtheRad for the reply. After going through literature. I would like to know whether what would be the best approach for a reliable low cost system for induction heater using 24 V DC (1kW-1.5kW).
(i) Single ended quasi resonant converter (IGBT)
(ii) Half bridge series resonant converter (using IGBT)
(iii) Hard Switching System Using SiC (mosfet).
If there are any references for the same it would be helpful.

 

[BradtheRad]

Youtube has videos from experimenters who build induction heaters.

I lack hands-on experience with such high power levels so I can only bring up general concepts.

* Inductive heating works by creating eddy currents in your workpiece. By making frequency variable you can find the most efficient parameters. Varying frequency is an easy method to vary current through the inductor.
(Slow f greater Amperes. Fast f few Amperes.)

* By using minimum switching devices you minimize parasitic resistance.
A mere 1/10 ohm at 50 Amperes loses 5V which makes 250W.
Example, consider using half-bridge and bipolar supplies (simulated below). Current waveform through inductor is triangular or almost a sine.

 

[Warpspeed]

The design process is conceptually straightforward.
First define the load you are trying to drive power into.
Then design a suitable driver to match the dc power source to the ac load.
This applies to ANY power amplifier, not just an induction heater.

Its not possible to just specify capacitance and inductance without first knowing the operating frequency, the voltage, the current, and the required power.

 

[FvM]

I think the design would start with possible parameters of the induction coil. If you use an existing coil as template, you can try to scale the voltage keeping total ampere turns respectively flux. Scaling the existing Vdc = 300 V parallel resonant topology to low voltage results in about 1 turn and large resonant capacitor. Not impossible but unpleasant. Secondly I would consider to change the topology to push-pull ( possibly H- bridge) with LC series circuit which raises the resonator impedance level. A drawback is that the drive voltage must be varied with load to keep constant flux.

 

[FvM]

Regarding NXP application note, consider that it's an uC AN and focusses on control rather than power circuit. Don't expect detailed information about its design therein.

--- Updated Sep 11, 2022 ---

You might get some insights from this master thesis https://dspace.mit.edu/bitstream/handle/1721.1/100618/932637869-MIT.pdf
Also https://www.researchgate.net/public...roach_to_Support_the_Design_of_Induction_Hobs

 

[sabu31]

I think the design would start with possible parameters of the induction coil. If you use an existing coil as template, you can try to scale the voltage keeping total ampere turns respectively flux. Scaling the existing Vdc = 300 V parallel resonant topology to low voltage results in about 1 turn and large resonant capacitor. Not impossible but unpleasant. Secondly I would consider to change the topology to push-pull ( possibly H- bridge) with LC series circuit which raises the resonator impedance level. A drawback is that the drive voltage must be varied with load to keep constant flux.

Thanks FVM for the reply. The work in the MTech Thesis presents a full bridge circuit along with 4 diodes. There was a requirement for this work to be a low cost solution. Will half bridge push-pull with resonant work. You mentioned for low voltage systems, the capacitor needs to be high, why is this unpleasant? What about using hard switching.

--- Updated Sep 11, 2022 ---

There are two youtube videos for induction heating from DC which are at 1kw levels.

Is it possible to adapt the above systems to make itss power controllable through gate drive/ microcontroller