Inductive Heating Circuit Options

Thread Starter

DC_Kid

Joined Feb 25, 2008
1,072
I am in process of designing a more flexible inductive heating tool. The main goal is to have easy ability to change frequency of the inductive element. Most of the simple circuits are LC tanks, but they are limited in frequency.

I was looking at simple H bridge configuration, but there is the risk of shoot-through when one pair is going off while the other pair is going on.

Another option I thought of is to use two inductor coils with each coil 180deg from each other magnetically, driven by independent MOSFET. One idea to switch them (flip-flop mode) is to use the source-sink ability of the 555, which means one N gate and one P gate MOSFET with 555 running 50% PWM with variable frequency.

Thought?
 

DickCappels

Joined Aug 21, 2008
10,153
Would this arrangement be equivalent to driving a center-tapped transformer (tap toV+) with two transistors in push-pull fashion?

That would suggest that the two windings are magnetically coupled (at least they are in my mental picture at the moment).

That sounds like a good solution except you will have two windings to instead of the more usual one winding.

When push-pull driving a transformer a little dead time between the turn-off of one transistor and the turn-on of the other is often needed to prevent conduction overlap.
https://www.microchip.com/forums/m215207.aspx

There are power supply controllers with dead time between gate pulses. The ancient TL494 comes to mind. The Microchip ATTINY25 controller has pulse width modulation outputs (you can leave them fixed) with programmable dead time.

You can probably couple the gates of the MOSFETs to drains of the opposite MOSFET or couple to the drain circuits with a timing transformer (or pulse transformer if you wish) if the drain voltage is too high for the gate of the opposite transistor to deal with. With some careful design and adjustment of the levels and timing be able to make a multivibrator with acceptable overlap.

Maybe something like this:
1642580699414.png
 

MisterBill2

Joined Jan 23, 2018
18,167
Is this for induction heating/hardening? Or is this for an induction heating stovetop?
Induction with opposite phased coils will certainly alter the magnetic flux a great deal, possibly even cancelling it.

And what is the benefit of changing the frequency of the drive, aside from maybe being able to have more flexibility of coil configurations?
 

Thread Starter

DC_Kid

Joined Feb 25, 2008
1,072
Post #2 - yep, the "shoot through" is an issue in H bridge, especially if fall time is bigger than rise time. To avoid this issue I propose two coils being driven independent of each other.

Post #3 - Induction heating of parts for either hardening, annealing, or loosening. Frequency changes allows to tune for different metals, and, depth ("skinning"). I am trying to model two coils in LTspice.
 
Last edited:

DickCappels

Joined Aug 21, 2008
10,153
Can you sketch out the two coils NOT coupled by magnetic fields? If they are in the same plane or along the same axis and not shielded from each other they will be coupled.

In post #2 I pasted the wrong circuit. I thought you were working your way toward something like the topology below:
1642603544430.png
 

Thread Starter

DC_Kid

Joined Feb 25, 2008
1,072
The two coils are would side-by-side (weaved), but the amps are controlled by independent FET's, one Pfet and one Nfet.
Tie the FET gates, control gate charge using PWM (VCC down to GND) so the gates can charge and discharge.

So, in essence, two coils, but only one layer on a tube.
 

shortbus

Joined Sep 30, 2009
10,045
So, in essence, two coils, but only one layer on a tube.
From what I have seen with doing what you are trying you would use a center tapped coil, not two separate coils.

Also I think your not understanding the induction heating principals. The difference of frequency is for the size of coils, not the material being heated. The material in the coils is only linked magnetically not electrically. And even coil size/diameter is pretty forgiving. One thing I did read about home built induction heaters is that most of them are run at too high a frequency.
 

Thread Starter

DC_Kid

Joined Feb 25, 2008
1,072
A CT where each half is on/off is logically same as just two coils. How each half turns on/off is TBD.
With LC tanks, the physical coil size/turns is critical for tank resonance.

I have an idea using a folded CT coil. I will draft it and then post.

Skinning, or eddy depth, is in the workpiece. Material type and depth wanted will determine frequency used.
https://www.asminternational.org/documents/10192/1849770/ACFAA5C.pdf
 
Last edited:

MaxHeadRoom

Joined Jul 18, 2013
28,617
I am in process of designing a more flexible inductive heating tool. The main goal is to have easy ability to change frequency of the inductive element. Most of the simple circuits are LC tanks, but they are limited in frequency.
Thought?
Most of the HF heating systems I have been involved in were designed around specific applications, i.e. Frequency, jig design, nature of product to be heated etc.
Trying to cover all bases may result in loss of efficiency?
 

Thread Starter

DC_Kid

Joined Feb 25, 2008
1,072
Most of the HF heating systems I have been involved in were designed around specific applications, i.e. Frequency, jig design, nature of product to be heated etc.
Trying to cover all bases may result in loss of efficiency?
Agreed.
I am not building dedicated HF induction heater for big run of anything, more like a hobby tool where I can easily change frequency with twist of knob, or swap coils to fit a specific object. A little lost efficiency is acceptable.

a 1" x 1" 5turn air coil is about 400-425 nH, but I don't see where AC analysis will apply because on each half cycle of PWM one coil is not being used at all (no amps).
 

Thread Starter

DC_Kid

Joined Feb 25, 2008
1,072
I think you need to reread the section on Skin Effect in your link again. It says in a "higher scientific way" what I said in my second paragraph.
I not quite following. I only mentioned variable freq in context that depending on certain wants/needs, freq choice is an option.

The depth of heating depends on the frequency of the ac field, the electrical resistivity, and
the relative magnetic permeability of the workpiece.
 
Top