Small radius heat induction coil connected via coaxial cable?

[Plasmateur]

Hello, I am in need of something I believe to rather specific.

A small radius heat induction coil. I currently have access to a 13.56MHz, 300W of power with a tuning circuit. I am able to match for certain working coil configuration, however I can only match when the radius and/or turns of the working coil are not within the parameters I need.

I'm a little confused by the matching network as well. It employs two tunable capacitors. The first one from the source to ground and the second on in series with the working coil. I'm confused because the working coil is shorted to ground, so I'm not sure how the math works out for matching the impedance of the load. Does the load factor in radiation resistance?

I need to use coaxial cable for this application in short pulses as well. I was able to heat a 1mm diameter tungsten wire to white hot conditions in about 1 second without damaging the coaxial cable I had previously used - but the working coil that I constructed isn't of ideal dimensions.

Is there something commercially available for this application?

What If I terminated the working coil into a 50Ohm power resistor? Would this make the match easier to achieve? Thank you.

 

[GeoffreyF]

Impedance at frequency is not the same as DC resistance.  What would appear to be a DC short is common to see in RF circuits.

Impedance will change when there is a metal object for the coil to heat.

Why do you believe your inductive heating coil is appropriate for the RF source that you have?

If you are asking these kinds of questions, you are quite possibly out of your depth for working on a device that heats things up.

[coppercone2]

what are you doing with it/

i am curious because i have been thinking about building one for a long time but i think its because its just a popular project. i had a few control system ideas i wanted to build and test at least, let alone get into power electronics, but i keep asking myself why do this... i just see people doing brazing and heating bolts with them, but a torch is fine for home use. casting is not really too appealing either because the complex shapes are either useless or cosmetic without really advanced finishing equipment, and even simple shapes need a well equipped machine shop to clean up. and if you use softer metals you don't get strength increases from extrusions and stuff. i thought alloy chemistry might be useful but you need mad equipment to study crystal structures and i don't know if anything remotely useful would ever come out of it.

for reference a 2000c commercial vacuum induction kiln that is capable of casting 30cc of metal has a power consumption of like 5kw. scary electronics.

for some reason the whole metal casting and alloy making scene makes me feel a tad nihilistic. unless its custom vacuum tubes. maybe casting custom alloys for vacuum tube use in tiny amounts. or special wires.

[Plasmateur]

Impedance at frequency is not the same as DC resistance.  What would appear to be a DC short is common to see in RF circuits.

Right, I'm aware impedance vs resistance. I'm trying to understand what you're saying here. If one end of the working coil, or "heating coil", is connected directly to ground, it is not correct to say shorted to ground? What's the best terminology to use then and how would that be described on a circuit diagram?

Impedance will change when there is a metal object for the coil to heat.

I'm aware of this as well.

Why do you believe your inductive heating coil is appropriate for the RF source that you have?

Where did I say that?

I have an RF source with a matching network. The matching network doesn't appear to be appropriate for the particular coil I need to fabricate. The matching network on hand works for other coils I fabricate which are outside the dimensional parameters I need. I need to have higher frequencies (MHz) than what are typically used in inductive heating applications (kHz) because I am attempting to heat wire - meaning  smaller skin depth. 

If you are asking these kinds of questions, you are quite possibly out of your depth for working on a device that heats things up.

And making these statements to someone who is trying to gain a better understanding of any topic isn't helpful. Why would anyone bother to learn new subject matter if they are initially "out of their depth" when first encountering it. What an absurd thing to say.

I've already used this device to heat wire, but under non-ideal working coil dimensional parameters. The wire heats up when I achieve a match - the reflected power is minimized. I'm trying to gain a better understanding of the theory behind the matching network and how it relates to what I'm doing.

Should I just stop attempting to learn?

[Plasmateur]

what are you doing with it/

i am curious because i have been thinking about building one for a long time but i think its because its just a popular project. i had a few control system ideas i wanted to build and test at least, let alone get into power electronics, but i keep asking myself why do this... i just see people doing brazing and heating bolts with them, but a torch is fine for home use. casting is not really too appealing either because the complex shapes are either useless or cosmetic without really advanced finishing equipment, and even simple shapes need a well equipped machine shop to clean up. and if you use softer metals you don't get strength increases from extrusions and stuff. i thought alloy chemistry might be useful but you need mad equipment to study crystal structures and i don't know if anything remotely useful would ever come out of it.

for reference a 2000c commercial vacuum induction kiln that is capable of casting 30cc of metal has a power consumption of like 5kw. scary electronics.

for some reason the whole metal casting and alloy making scene makes me feel a tad nihilistic. unless its custom vacuum tubes. maybe casting custom alloys for vacuum tube use in tiny amounts. or special wires.

I need to heat a small wire to white hot temperatures in a vacuum for an experiment I am attempting. Way outside the realm of brazing or soldering or joining different metals together. I was already able to do this with only 300 watts of power @ 13.56MHz, but the heating coil dimensions aren't ideal for the application.

I don't believe I will be able to use a helical coil. The wire I am heating will be electrically attached to an oscilloscope further down the line. Off axis components of the H field from the induction coil relative to the wire it is attempting to heat might cause current to run down the length of the wire and back into the oscilloscope. At high power, this might cause some problems!

[coppercone2]

what are you trying to determine with this experiment/ materials physics constants maybe question mark.

btw shorted means that you basically substitute one thing for another. like if a switch closes its shorted. i guess people say the capacitor creates a short, but it is kind of confusing. but its not proper anyway, because the switch is actually closed.

i would say don't use the word shorted unless you are referring to a state change, other wise refer it as a connection. even saying a switch shorts something is kinda hokey because when most engineers hear short they think of a failure mode. it seems to be a term that emphasizes low impedance connections, but its almost like an action verb used to make a story interesting, but really, it should be 'connected directly to ground by solder joint/crimp'. it is weird slang. it means short circuit.

but it is useful to say shorted in a trouble shooting sense, i.e. you shorted digital signal one to ground. people will know that you used a alligator clip to connect something to ground and its considered fairly proper. but some people tend to associate shorts with visible releases of energy. if its a digital signal it might be more clear to say you jumpered it to ground, wheras if you put a screw driver across bus bars you shorted them, since there is a spark and a pop. when i think of jumpering i think of delicate microelectronics work, when i think of shorting i think of jamming a screw driver into something with safety goggles on.

but again, unless you are doing it as a field repair method, you should not really short things out with a screw driver, more like make a resistor adapter to slowly drain a circuit to prevent surge current damage.

also something that comes to mind for some reason is a motor, you can say it is connected to a short circuit, or a capacitor is connected to a short circuit i.e. during storage of a big electrolytic cap, but you would not say the motor is shorted or the cap is shorted because it creates ambiguity about the integrity of the motor. if you say something is 'stored shorted' that will probably not create ambiguity either, in most cases, since the context of storage and knowledge of capacitors implies it has been connected to a short circuit for safety during storage, but again you need more context because it might imply someone put a short circuit damaged capacitor on the shelf. but even that is ambiguous, because it does not really say what it is. it could be a capacitor that went open circuit because it was short circuited due to mechanical shock on the conductors or a capacitor that internally completes a short circuit because of film damage. short damaged vs short stated.. or is it just less reliable because it was shorted in the past and its current integrity is unknown

i also have a feeling if you are told to 'short something' or 'short it out' the first one can have an association with safety i..e turn off power and leave a shorting link just incase, where short it out is slightly more inclined to make me think the guy wanted me to make a big spark. like a kind of cave speak 'it out' means 'summon a spark from the circuit to trigger a protection mechanism or make something destructive and stressful happen'. then you also have spark it, which means you kinda mess with it to get a spark but try to break the circuit fast enough to beat thermal and mechanical time constants associated with serious damage. or just to create a small emp/electrical disturbance that seems like a reasonable field test for some kind of stupid problem that is cost prohibitive to really test. the idea being you play some not super low impedance circuit like a guitar. also possibly refereed to a 'hot wiring'. not because of how it is actually done to steal a car, but how it is portrayed in movies, where they kinda spark it a bit.

[1uk3]

Should I just stop attempting to learn?

No! I read it like the typical kiddy that wants to build a 1337kW 42Ghz induction heater. The "shorted to ground" threw me as well

What If I terminated the working coil into a 50Ohm power resistor?

I dont' quite understand what you mean by that. Could you do a quick sketch?

So the problem is that you have a smaller coil and therefore the tuning capacitors are not large enough to bring the resonance frequency down to 13.56Mhz right?
Maybe we could do some calculations if you provide information like coil dimensions and type of tuning circuit(brand, name, configuration, capacitor values,...)

[coppercone2]

i wanna build a one thousand three hundred thirty seven kilowatt tesla coil that runs at forty two gigahertz. post some schematics please

https://hamwaves.com/inductance/en/index.html#input

use this link to model your required coil size please

[T3sl4co1l]

What kind of coil geometries would be acceptable to your application, and what L and R parameters would they have?

That is the question, as far as being able to tune it properly.

The two-capacitor network acts as an impedance divider, which when the capacitive reactance is canceled out, also acts as an impedance transformer.  Usually there is an inductor in there, as well.

The easiest way to demonstrate this working (given that familiarity with reactances is a prerequisite), is to picture a resonant tank: a capacitor and inductor in parallel.

At resonance, reactances cancel, and the parallel equivalent impedance is infinite (like an open circuit), and the series equivalent impedance is zero (like a short circuit).  Suppose we connect a resistor in series with the loop: the series equivalent is then just that resistor (at resonance).  Or suppose we connect one in parallel: same thing with the parallel equivalent.

What if we do both series and parallel resistors?  We have some choice of which places to put them, which gives the four permutations of L-match networks.  Suppose we measure the series equivalent impedance, with a resistor in parallel with the capacitor.  That is, a circuit of (port)--L--(C || R)--GND.  What is the resistance at resonance?  (But first, what is resonance in this circuit?)

As it turns out, the resonant frequency is slightly different, by an amount on the order of -1 / (R^2 C^2).  For non-precision cases, and Q > 5 or so, this is negligible, but it is worth noting that the frequency changes.  (This is more obvious with exaggerated values, say R < sqrt(L/C), and the transient response rather goes "thump" instead of "dinggg"!)

So that's easy to approximate out, and we're left with the resistance question.  As it happens, the R value is inverted relative to the resonant impedance sqrt(L/C).  That is, the equivalent series resistance measured is around Req = L / RC.

Consider what happens at resonance: the loop current rises, and the voltage (across the inductor or capacitor) rises.  Both rise proportionally, the ratio being the resonant impedance, sqrt(L/C).  We can start with a low voltage, apply it in series to an LC tank, and get a large voltage on the resonant node -- "Q multiplication".  If we load that resonant node, the Q is spoiled and the voltage drops relatively rapidly, which is to say, it is a high impedance node -- we have transformed the source impedance.  It works inversely, starting from a high impedance and making a low impedance, just the same.

This is the basic, qualitative way in which the tuning network operates.  To match any impedance, high or low, two L-match networks are glued together back to back; which puts two inductors in series so only one physical component is needed.  That leaves two variable capacitors, in effect, one to match the source and the other to match the load.

L-match calculator to play around with:
https://www.daycounter.com/Calculators/L-Matching-Network-Calculator.phtml

Tim

[Plasmateur]

Should I just stop attempting to learn?

No! I read it like the typical kiddy that wants to build a 1337kW 42Ghz induction heater. The "shorted to ground" threw me as well

What If I terminated the working coil into a 50Ohm power resistor?

I dont' quite understand what you mean by that. Could you do a quick sketch?

So the problem is that you have a smaller coil and therefore the tuning capacitors are not large enough to bring the resonance frequency down to 13.56Mhz right?
Maybe we could do some calculations if you provide information like coil dimensions and type of tuning circuit(brand, name, configuration, capacitor values,...)

I might drop the on hand 13.56MHz source I have and just build one based on this thing
I'll have to contact them to see what frequencies this can get up to without overheating the transistors.

As for the working coil, it's a one turn coil about 10mm in coil diameter with a wire thickness of 1.5mm, so that should give 12.4nH of inductance.

The matching box has two capacitors. One in series and one in parallel. It is shown in the attached picture with the exception being the load is replaced by the working coil inductor.

C2 - 900-30 pF tuning range
C1 - 100 - 20 pF tuning range

Let's say I remove C1. Then I just have an inductor and a tunable capacitor in parallel, then by taking the maximum value of C2 and the inductance of my working coil, I believe I have a resonance at 47.6MHz. So yes, I need higher capacitance as you suggested.

I'm not sure how to combine C1 and C2 to have an effective capacitor in parallel with the working coil.

[Plasmateur]

What kind of coil geometries would be acceptable to your application, and what L and R parameters would they have?

That is the question, as far as being able to tune it properly.

The two-capacitor network acts as an impedance divider, which when the capacitive reactance is canceled out, also acts as an impedance transformer.  Usually there is an inductor in there, as well.

Yes. I removed the inductor awhile ago. Originally this was used for plasma generation, but I could never get the plasma to ignite with the additional inductor as well as the auxiliary capacitors.

The easiest way to demonstrate this working (given that familiarity with reactances is a prerequisite), is to picture a resonant tank: a capacitor and inductor in parallel.

At resonance, reactances cancel, and the parallel equivalent impedance is infinite (like an open circuit), and the series equivalent impedance is zero (like a short circuit).  Suppose we connect a resistor in series with the loop: the series equivalent is then just that resistor (at resonance).  Or suppose we connect one in parallel: same thing with the parallel equivalent.

What if we do both series and parallel resistors?  We have some choice of which places to put them, which gives the four permutations of L-match networks.  Suppose we measure the series equivalent impedance, with a resistor in parallel with the capacitor.  That is, a circuit of (port)--L--(C || R)--GND.  What is the resistance at resonance?  (But first, what is resonance in this circuit?)

As it turns out, the resonant frequency is slightly different, by an amount on the order of -1 / (R^2 C^2).  For non-precision cases, and Q > 5 or so, this is negligible, but it is worth noting that the frequency changes.  (This is more obvious with exaggerated values, say R < sqrt(L/C), and the transient response rather goes "thump" instead of "dinggg"!)

So that's easy to approximate out, and we're left with the resistance question.  As it happens, the R value is inverted relative to the resonant impedance sqrt(L/C).  That is, the equivalent series resistance measured is around Req = L / RC.

Consider what happens at resonance: the loop current rises, and the voltage (across the inductor or capacitor) rises.  Both rise proportionally, the ratio being the resonant impedance, sqrt(L/C).  We can start with a low voltage, apply it in series to an LC tank, and get a large voltage on the resonant node -- "Q multiplication".  If we load that resonant node, the Q is spoiled and the voltage drops relatively rapidly, which is to say, it is a high impedance node -- we have transformed the source impedance.  It works inversely, starting from a high impedance and making a low impedance, just the same.

This is the basic, qualitative way in which the tuning network operates.  To match any impedance, high or low, two L-match networks are glued together back to back; which puts two inductors in series so only one physical component is needed.  That leaves two variable capacitors, in effect, one to match the source and the other to match the load.

L-match calculator to play around with:
https://www.daycounter.com/Calculators/L-Matching-Network-Calculator.phtml

Tim

Tim. Thank you for taking the time to explain this. You are awesome.

What I am having a hard time understanding is that I don't have a load resistor, but an inductor in place of it. I understand how to solve for Q when there is a load resistor, but not when it is replaced by an inductor.

[T3sl4co1l]

What I am having a hard time understanding is that I don't have a load resistor, but an inductor in place of it. I understand how to solve for Q when there is a load resistor, but not when it is replaced by an inductor.

Simple -- your inductor isn't an inductance, it has resistance too!

The Q factor or ESR or EPR of it can be estimated from the geometry and materials, or measured in operation.  (Noteworthy, tungsten's resistivity changes wildly with temperature so the Q will change as well.  The figure measured when cold will not be the same at final temperature.)

I might drop the on hand 13.56MHz source I have and just build one based on this thing
I'll have to contact them to see what frequencies this can get up to without overheating the transistors.

Would guess a MHz is pushing it...
Hmm weird, there's some sort of driver chip in the middle.  Did they-- is it actually a class D driver, with analog feedback?  An attempt at a boosted "Royer" (actually Baxandall) oscillator as is commonly seen?

That's pretty awful... the problem is this: for the oscillator to start, it must start from a small-signal condition, amplifying noise until the dominant mode(s) take over, at which point the transistors begin to saturate, the amplitude limits, and stable (periodic) and efficient operation is had.

You can't just go from one to the other -- attempting to feed that small noise signal into some kind of comparator (if it's a gate driver IC, the typical input characteristic is a logic-level threshold with hysteresis), will result in the tallest (amplitude) signal taking over, which typically means oscillating near the maximum toggle rate of the driver itself.  Thus cooking the driver and its transistor.

This assumes it is actually a gate driver; even if it's not, the fact that there's no TVS diodes protecting the transistors is damning enough to me...

Tim

[coppercone2]

the calculator on the website estimates resistance and q