Tiny induction heater

[leadinglights]

I am trying to make a very small induction heater to replace a resistive heater which delivers between 20W to 40W. This should be a simple problem but one principle need is for the heater circuit to be as small and light as possible. I will fill in the details of where this is to be used at the end of this message, but at the moment the questions are:-

Are there any very good tank capacitors in physically small sizes?
While I can work out the voltage across the capacitors from high school physics and can work out de-rating for frequency and temperature from datasheets, there seems to be little consistent information on how to determine the necessary voltage rating in the real world. Are there any good general rules for this?
Temperature control would be easiest by switching on and off the induction heater circuit and the existing use involves switching a resistor at several times per second. Would doing this with an induction heater lead to white smoke and bitter disappointment?
In several of the smallest oscillator circuits the tank is fed from the +ve supply through two ferrite cored inductors. Are there equivalent to matching inductors? Would there be any problem having approximately one meter of wire between each inductor and the tank?

Details of why I need the small size.
This is to be used to replace a resistive heater in the plastic extruder (hotend) of a 3D printer. The temperature at which the plastic is laid onto the print can vary by as much as 8C in a couple of millimeters during a print. It is my hope that an induction heater will be able to deliver a controllable temperature to the very tip of the extruder. As the hotend moves quite fanst and changes direction frequently it is very desirable to keep the moving mass as small as possible.

There is some discussion on the RepRap forum along with a CAD impression of the heater (I haven't yet succeeded in putting a picture in this forum)

Mike

[Mads Barnkob]

I do not think this is the best approach, a mazilli/ZVS/royer induction heater. You got no control over current in the basic circuit, other than on/off which often is a critical point in this circuit as a wrong startup switch position could make it explode if its switching at a point of low voltage and in liniar mode etc

I do not have any particular experience with building small royer IHs, but it is going to be high frequency, you can not avoid that. I have had a lot of traffic to kaizerpowerelectronics.dk from a Russian model ship builder forum, perhaps you can google translate and find something interesting there? https://www.shipmodeling.ru/phpbb/viewtopic.php?f=10&t=18078&start=105

For you application, where you say that you have a large temperature drop in a few mm of plastic material getting heated, I am first and foremost thinking that the mass of the head is too small, if it can not store enough heat to transfer without getting cooled too much itself.

To gain the fine control you are asking for, I think you should look at LCLR induction heater topology where you can vary the PWM signal to control the heat with a PI controller, this is more complicated than the Royer, but I honestly think you risk getting too high temperatures with a simple circuit like that. But that is only my rough estimate, I would love to see you experiment with it

Royer tank circuit voltage is pi * Vinput, double that figure with switching spikes when it is at its worst.

I see no catastrophic problem with a long wire from the inductor to the work coil/tank circuit

[petespaco]

The 20 watts to 40 watts that you want to transfer to the hotend nozzle will be VERY dependent upon the metal from which it is made.  I have found it much harder to transfer power to brass and copper (non-ferrous) metals than those containing iron, for instance.

[leadinglights]

I agree that a well designed resistive heater with a good thermal reservior and proportional control would be hard to better. Having said that, the makers on the RepRap community are forced to use available materials and components - typicaly a resistor or heater element mounted in a rectangular brass or aluminium block. Commercial versions are available but these are copies of those evolved within the community.
Most improved methods that I have been able to think up have needed facilities such as the ability to sinter alumina on to a nichrome element - generally unavailable to the back room tinkerer. It is just within the bounds of possibility that an inductively heated nozzle would be a more practical option.

Plus inductive heating is much cooler 

Sectioned view below shows one of my hotends with a copper block and air delivered near the nozzle tip. When there is maximum plastic flow and cooling near the time the temperature of the plastic can easily be 8C lower than at lower flows.

 Also a possible view of the new induction heated version. Dark grey material within the coil is carbon steel or feritic stainless steel, lighter grey is austenitic stainless steel


Mike

[Justin]

I'd take a look at soldering irons that use induction heating.

[leadinglights]

It is the wonderful Metcal RF heated soldering irons that made me think that induction heating may be worth looking at, even the best super quick resistive soldering irons take much longer to get up to heat.

Mike

[Mads Barnkob]

It is the wonderful Metcal RF heated soldering irons that made me think that induction heating may be worth looking at, even the best super quick resistive soldering irons take much longer to get up to heat.

Mike

The Metcal MX-500P induction heating soldering station is not as simple as a Royer induction heater, check this schematic and description of how it works: http://scopetechniques.com/Metcal/MX-500P-11.pdf