[SOLVED] How do you make a Q=0.7 inductor, very high power?

[mtwieg]

It's interesting that you're fighting to decrease Q, whereas I'm usually fighting to increase it...

What is your winding geometry like? A simple solenoid or spiral will give fairly high inductance. You should try making adjacent turns in opposite directions in order to cancel flux. In other words, wind a solenoid with bifilar pipes, making two tightly coupled inductors. At one end, short them together (but short the coolant path). The other end is your terminals. This should have very low inductance, while increasing the cooling capacity.

 

[chuckey]

Use a plastic tube through which water is passing and use a bit of nichrome foil as your inductor inside it. Wind a turn or two in it if required.
Frank

 

[ionp]

It's interesting that you're fighting to decrease Q, whereas I'm usually fighting to increase it...

What is your winding geometry like? A simple solenoid or spiral will give fairly high inductance. You should try making adjacent turns in opposite directions in order to cancel flux. In other words, wind a solenoid with bifilar pipes, making two tightly coupled inductors. At one end, short them together (but short the coolant path). The other end is your terminals. This should have very low inductance, while increasing the cooling capacity.

Thank you for your suggestion.
Yes, coming from a communications background into induction heating has turned a number of concepts on their head!

I would use a solenoid if I needed to increase the L, but I can get the same cancellation with a bifilar layout and running a linear path up and back. I can always increase the L later by cascading low Q elements, and with the amount of power (100-150kW) I have to dump somewhere the structure is certain to be unwieldy at best.

Getting a sufficiently low Q element is the primary challenge.
It would be a benefit if it could be moved when it was done, rather than sit out in the middle of the floor as built.

 

[FvM]

I believe you have heard sufficient suggestions to realize the test load.

 

[mtwieg]

Thank you for your suggestion.
I would use a solenoid if I needed to increase the L, but I can get the same cancellation with a bifilar layout and running a linear path up and back.

If you have the space for it, doing a simple straight bifilar pipe is good, but I'm surprised the inductance is still too high. Can you give us some more details on the geometry? You can get improvements by meshing more pipes in parallel with alternating current direction. If the spacing is tight, you should be able to get Q very close to zero.

An even more ambitious approach would be a coaxial structure, with coolant flowing between the inner and outer conductors. Of course construction would be harder, and you'd have to use an insulating coolant (DI water).

It would be a benefit if it could be moved when it was done, rather than sit out in the middle of the floor as built.

That's unlikely. Where is that much heat going to go? The radiator along is going to be quite large. You're looking at something the size of a large refrigerator, at least.

 

[c_mitra]

I would use a solenoid if I needed to increase the L, but I can get the same cancellation with a bifilar layout and running a linear path up and back. I can always increase the L later by cascading low Q elements, and with the amount of power (100-150kW) I have to dump somewhere the structure is certain to be unwieldy at best.

Getting a sufficiently low Q element is the primary challenge.

You need a low Q because your furnace load is likely to be variable; you do not need a low L because that will result into poor transfer of energy. You need to have a low Q for the whole system in the sense you have a rather broadband emitter.

100-150kW power induction furnaces are medium power, medium frequency for non-ferrous materials and you also need to have considerable distributed capacitance all along...