Induction heater capacitor selection

[fastline]

I am building a relatively small (300-600w) induction heater and trying to select capacitors for this purpose. From my research, low ESR caps are mandatory to reduce heating. However, I have yet to see any values at all to relate the "low ESR" and it seems that most of these devices are of very low duty cycle. I would want to design this one to run nearly continuous. I seem to be caught right between the hobby style builds with very limited duty cycle, and commercial hot rods using VERY expensive water cooled caps.

It seems polypropylene is common for these but I am just not sure how to estimate the bank to stay cool enough. I think it is obvious by the water cooled units that there is probably nothing I can do to eliminate the heating but staying tolerable would be great.

I will probably run around 80-150khz. going to have to play with it a touch.

 

[fastline]

Working with ferrite cores for inductors

As per my other thread, I am working on an induction heater unit and will need to tune the inductance and capacitance in the circuit to optimize performance. I am no EE and not familiar with how to calculate or predict the inductance by making cores. I was just going to use bare copper wire for R&D and then step up to lacquered wire later or maybe just dip them.

Can anyone help with this this?

 

[jpanhalt]

As per my other thread, I am working on an induction heater unit and will need to tune the inductance and capacitance in the circuit to optimize performance. I am no EE and not familiar with how to calculate or predict the inductance by making cores. I was just going to use bare copper wire for R&D and then step up to lacquered wire later or maybe just dip them.

Can anyone help with this this?

I think you need to step back more than a few paces and do some reading.

By "lacquered wire," you really should consider enameled wire and use it from the start. As for ESR, it is included in every capacitor datasheet; although, it may not be called "ESR" per se.

What size and voltage rating capacitors are you considering? Can you provide part numbers so we can help you interpret the datasheets?

John

 

[fastline]

https://www.electro-tech-online.com/custompdfs/2013/01/X2_B32921_926.pdf

This is one that I was looking at but will likely need 2-5uf total capacitance. Seems guys are stacking a bunch of these in parallel but I am just wondering if there would be a much better choice for this? For instance, I was looking at the old paper film HVAC oil filled caps. They have all the capacitance I need, cheap, and I guess per my ESR meter are pretty low ESR. The 12.5uf I tested was about .1ohms and I think it tests at 100khz which is where I would operate. I guess I should verify that.

I will need to play with it a bit but Just seems using 100 small caps is a lot of wasted time and space?

Re: inductor, I meant enamal wire. I have no idea why I keep saying lacquer. Why should I use it over bare wire from the start?

 

[ronv]

Here is an inductor calculator:
https://www.daycounter.com/Calculators/Air-Core-Inductor-Calculator.phtml

High voltage high current caps aren't cheap, but I think these have specs that you could use to see how many you needed to run in continuous mode.

https://www.mouser.com/ProductDetai...ha2pyFadujKVG%2b0t6h0gIyRLxewCP0GHkdX/Yw1rRQ=

 

[fastline]

Ron, I really appreciate your reply. The inductance calculation is one I needed.

Regarding the capacitors, those indeed look like audiophile level caps. I guess what I was trying to arrive at is the term "low ESR" but not sure if there is a table that defines that category per capacitance? I am also unsure if "low ESR" is enough to keep them from overheating. I know in commercial, purpose built cap banks, they are forced air cooled or even water cooled so I am a little curious. Is there any way to estimate the heating by knowing the currents, frequency, and ESR value of the cap?

 

[ronv]

I think if you dig down in the datasheet they publish esr and maximum current at 100Khz. I think I would model your circuit and see the measurements and go from there. Have you picked a circuit yet?

 

[fastline]

Here is an idea of what I want to try. A self oscillating circuit. I am certain open to other ideas and thoughts here. Just trying to keep it reasonably simple. I was going to push to 30-50V until I can find a power level I am happy with. My lab supply will only get me so far being able to do 30V and 10A. I am not sure if you can series those to make 60V of 5A. That might be interesting to test.


Anyway, was going to play with the C1 and L2 values and test heat pieces to get an idea which works best.

Probably my biggest concern (SMPS still largely confuse me) is the power supply. I want to use this as an excuse to make a power supply. It really does not need perfect voltage regulation for the coil and all other voltages that will need good regulation, I can just use linear regs.