Here's my INHT (induction heater). The actual Lua script is at the bottom of this post.

Induction heater conducting SPRK will affect metalic materials by heating them if they are withing 20 pixels from the nearest powered INHT pixel. It works by causing a random pixel within a 20pixel radius to have its temperature increased by 100Kelvins (if it is an "induction heatable solid"). These solids are METL, IRON, BMTL, and RBDM. However it has no affect on semiconductors NSCN and PSCN. If the solid gets so hot as to melt, then it still can be induction heated but the efficiency goes down to 50Kelvins each time it is "hit" with induction heating effect. This applies for both the molten state of metals (LAVA with CTYPE = METL, IRON, or BMTL) and also for elements that are liquid metals LRBD and MERC. It has no effect on conductive waters (WATR and SLTW) even though these conduct electricity. If it is a powdered metal (BRMT) then it is only is only 25Kelvins per hit.

So here's a simpler and better explanation I think, so you can understand more how the code makes it work.
During each frame of the TPT simulation, each pixel of INHT performs a check on a random pixel within a 20 pixel radius of the INHT pixel in question. If it finds an acceptable solid at the location checked then it adds 100Kelvins to the temperature of that pixel. If it finds an acceptable liquid it adds 50Kelvins. If it finds powdered metal (BRMT) it adds 25Kelvins. Because metals are good thermal conductors, the temperature doesn't stay at that assigned value but gets "spread out" through the material by diffusion (thermal conduction). This causes the material to appear to be being evenly heated (not just one random pixel at a time). So solids appear to be heated very fast. Liquids appear to be heated at a medium speed. Metal powder appears to be heated slowly. Because the energy is being transferred "magnetically" (though technically TPT doesn't have magnetism simulation), the actual INHT doesn't get very hot at all (only slightly warm from the SPRK passing through it).

Because all metals get white hot from induction heating, INHT is best attached to a power source via PSCN or NSCN (I intentionally made it so semiconductors didn't get hot, so there'd actually be a way to power my element without destroying the power connector).

I'm not sure if this is true for real induction heaters, but this is based on my intuition that solid metals are the best conductors so more current can be magnetically induced and get hotter. Molten metals have their atoms in a liquid "jumble" and so would be less organized so electrons would be harder to move via magnetic force. Powdered metals while solid, have very poor electrical contact between grains so the electrical eddy currents that need to be magnetically induced would be minimized because there is no single large solid conductor for them to flow in. Least current flow means lowest level of heating (in fact actual electromagnetic transformers and inductors have laminated cores which consist of separated metal plates to reduce eddy current heating in the core, and thus reduce wasted energy transfer).

The color I picked is designed to resemble copper, the most common element used for various wires in electrical devices including those "enamel coated" wires used in electromagnetic components such as transformers and inductors (which is what would be used as the magnetic field source in an induction heater).

Here's the Lua code to turn DMND into INHT.
tpt.el.dmnd.properties=16416
tpt.el.dmnd.color=0xffaa55
tpt.el.dmnd.name="INHT"
tpt.el.dmnd.description="Induction Heater. This device when being fed by SPRK will heat nearby metals (but not semiconductors or water)."
tpt.el.dmnd.menusection=1
math.randomseed(os.time())
function inht_update(i, x, y, s, n)
if tpt.get_property("life",i) > 0 then
scany2 = math.random(-20,20)
maxscanx2 = math.sqrt(400-(scany2^2))
scanx2 = math.random(-maxscanx2,maxscanx2)
elemtype = tpt.get_property("type",x+scanx2,y+scany2)
celemtype = tpt.get_property("ctype",x+scanx2,y+scany2)
if elemtype == tpt.el.iron.id or elemtype == tpt.el.metl.id or elemtype == tpt.el.bmtl.id or elemtype == tpt.el.rbdm.id then
tpt.set_property("temp",tpt.get_property("temp",x+scanx2,y+scany2)+100,x+scanx2,y+scany2)
elseif elemtype == tpt.el.lrbd.id or elemtype == tpt.el.merc.id or elemtype == tpt.el.lava.id and (celemtype == tpt.el.iron.id or celemtype == tpt.el.metl.id or celemtype == tpt.el.bmtl.id) then
tpt.set_property("temp",tpt.get_property("temp",x+scanx2,y+scany2)+50,x+scanx2,y+scany2)
elseif elemtype == tpt.el.brmt.id then
tpt.set_property("temp",tpt.get_property("temp",x+scanx2,y+scany2)+25,x+scanx2,y+scany2)
end
end
end
tpt.element_func(inht_update, tpt.el.dmnd.id)

Though I'm not counting on it, I'm hoping that when the developers see this and try it out via my Lua script, they will be so impressed with the idea presented here (and by how well it actually works in my script) that they will decide to make an official INHT element in TPT.

keperitan:

I'd like to point out that induction heaters can heat water. Have you seen an induction stove?

That works because the metal pan has currents electromagnetically induced in it. These currents heat the metal pan, and the hot pan boils the water.

Turban-of-Terror:

This would be good with materials like the conductors that only conduct above or below a certain temperature.

Actually those are semiconductors because most heat sensors are semiconductors. As such I didn't program it to be able induction heat them. The only elements that can be heated with this are:

Solids (heated rapidly)
METL
IRON
BMTL
RBDM

Liquids (heated at a medium rate)
Molten METL
Molten IRON
Molten BMTL
LRBD
MERC

Powders (heated slowly)
BRMT

Turban-of-Terror:

I see what ya did there sonny. My bad, for reasons I don't feel like explaining I have been out of it all day and I just forgot the word I was looking for. Why wouldn't it affect semiconductors though? I would think that that would be its only true good purpose in TPT.

It is a powerable heater that doesn't require contact to heat. Put anything you want to heat in contact with the metal it's heating and it will heat that substance. Semiconductors are poor conductors in and of themselves (only useful in small quantities in diodes, transistors, and computer chips where current doesn't have to flow very far). I think just a few inches of silicon wire has several ohms of resistance in real life (just like carbon). So it would not work very well to support the needed eddy currents that flow in metal to heat it up in an induction heater. Electromagnetically induced eddy current can be HUNDREDS of amps but the voltage between any 2 points of the metal object that the current is flowing through is less than one volt, so the resistance would have to be less than one milli-ohm for this to work due to ohms law resistance=voltage/current, and only metal has such low resistance within objects even as large as things like the pots and pans that you'd want to heat on an induction stove for example.

Yes I actually try to base my elements on my scientific knowledge, not like "unrealistic elements" as used in the official TPT software releases.



Also you are gonna need a way to get current to the heater to power it, cause METL and other metals will melt when heated by this so I left semiconductors as not heatable by my device so you won't destroy your power cable to the heater just by running the heater. So basically don't use anything other than PSCN or NSCN to power INHT, otherwise it will melt the wire powering it.