Understanding certain relationship in Induction heating circuits

Question

I will try to keep this short as I can and to the point, there is a bit of a back story, so bear with me.

For the last 18 months (on and off as life allows), I have been trying to develop an induction heating assisted machine for 2 reasons. 1 - To learn as much as I can by doing so (I want to actually make the circuits etc myself) and 2- To actually achieve what I want the machine to do, as I cannot get what I want anywhere else for a decent price and functions the way I want.

I have been researching induction heating circuits for quite a while now. But there is something I cannot work out. And I cannot find any specific info on the internet for exactly what I am asking.

Just a quick review of the machine I need to make: The machine will be called a Rifle Case Annealing Machine. It will incorporate an induction circuit to heat rifle case mouths to an annealed state (399C), so that the cases can be reloaded and fired again. This annealing relieves the work stress created by firing the cases. This is specifically useful for competition shooters, that require all their brass to be of the exact same hardness rifle round, to rifle round, and to get decent life out of the brass, as sometimes, specific calibres can get VERY expensive.

Now, I initially purchased a circuit from ebay - ZVS flyback driver circuit 1000w 12-48v (it came with a copper coil).

I had a play with it and a 36v 13.6a power supply.

Now just using this, I was able to anneal my small rifle (.204 Ruger) rifle cases manually, in about 25 seconds. This actually cooked the capacitors and blew them up, they got way to hot. Needless to say I learned something LOL. I then connected water cooling to the coil. I then replaced with the capacitors with some slightly lower value units, to try to increase the frequency with which the unit operated at. I did some very rough maths, and hypothesised that lower caps, would raise the resonance frequency. Apparently, to heat non-ferrous metals, you need a higher resonance frequency (not sure if this is true). I was hoping to increase the speed that the cases annealed at, as 25 seconds was far to slow. I did this, with not much noticeable difference, maybe slightly faster.

I then saw that someone gapped a ferrite core, and wrapped the work coil around it a few times. This made the ferrite core concentrate the switching magnetic field. https://youtu.be/axM177JrIBo

I tried this and I annealed the 204 brass within seconds!! Success!! Except incorporating a ferrite core (one I had to gap myself with a dremel and great difficulty) into the machine, and trying to water cool the whole coil was going to prove to be very difficult. There must be a better way.

Then I came across this video, where someone had actually made a very similar machine to the one I want.

https://youtu.be/EGfGV8xrfak

Now, my question is this:

In that last video, as far as I can tell, the guy that made it, has used water cooling (water bottle in the box), and has obviously used a normal copper coil (not a ferrite core concentrator).

How the hell did he get that brass to heat up so fast!?!?! Realistically, that is the speed I am after, with that size case. Smaller cases will anneal faster, which is no problem.

Is he just using more power? maybe 2kw power supply, running at 40 amps? Obviously he has designed the whole circuit himself, and kudos, as that is an awesome job. I have actually tried contacting him directly. I got a "read receipt" through the email, and no response :(

Has he used very low capacitors, to get the resonance frequency right for the non-ferros brass to heat fast? Has he changed the inductors, and some how got the heating speed from that? I have a feeling he is just using more power?!? I don't know.

I just cannot find any information, on how to speed up the induction heat rate, without putting more power into that system, and having to up the capacity all the components, to handle more power.....

Any ideas electro wizards?

Answer 1

Induction heating is not a resonance phenomenon. There is no one resonance frequency that will work better than any other. Rather, there is possibly an increase in magnetic coupling efficiency with increase in frequency, then a decrease with further increase of frequency as losses (skin effect, capacitive interwinding coupling, etc.) increase.

The major factors in how fast you will heat is the geometry of your magnetic system, the amount of current you input, and the magnetic coupling coefficient (which is frequency-dependent).

The geometry will influence the efficiency of magnetic coupling to the specific shape of object you're heating.

The eddy current induction in the object being heated is strongly dependent on the current in the induction coil. So if you have a coil with many turns, the effective current is multiplied, very much like the turns on a transformer winding. For maximum induction efficiency, you need to match the inductance of the induction coil to the electrical drive. Vdrive= 2πfLI. In an induction coil with few turns and no ferrite you will have small L, consequently I can be large for a small V.

If you are limited to the existing power supplies you have, then the major factor you can control is the number of turns of wire in your induction coil, and the core material. I would start out actually measuring your coil voltage and current under operation. It's possible that your power supply is poorly matched, and is actually current-limiting so that you are not supplying nearly as much power to the induction coil as the power supply is capable of delivering. In this case, increasing the number of turns in the coil will make a big difference in heating efficiency.