At a slightly different scale this problem has been worked over quite thoroughly by the RepRap community - heating element connected to a tubular melt chamber. A rough overview in order of successfulness (did I just make up a new word?= ) Wrapping uninsulated element wire with various barriers (PTFE tape, barbecue paint, Kapton, fire putty, etc) Using _insulated_ element wire Using a heat transfer block and using a power resistor as heater element Using a heat transfer block and using commercially available cartridge heater State of the art at the moment seems to be the Prusa nozzle which is an aluminum block with sensor and cartridge heater clamped to a tubular melt chamber. May require a bit of machine work, but should scale easily to paraffin in a 3/4" tube. Also noteworthy for this discussion is flexible "sheet heaters". Easily flexible enough to wrap this size of tubing. -Denny On Fri, Jun 7, 2013 at 7:19 AM, Spehro Pefhany wrote: > At 04:52 PM 06/06/2013, you wrote: > >On 2013-06-06 12:08 PM, Dwayne Reid wrote: > > > 2) Consider using the copper tube as a shorted 1-turn secondary on a > > > transformer. In other words, wrap however many turns of insulated > > > wire you need around the pipe and feed with enough high-frequency > > > current as you need to bring the pipe up to the desired temperature. > > > > > > Note that the primary wire does NOT get hot, so space it away from > > > the copper tube if desired. Better yet, wrap the copper tube with > > > insulation, then wrap the primary coil over top of that insulation. > > > > > > Details left as an exercise for the student . > > > >I was considering this, and if done right, would be ideal. But it would > >complicate the electronics. With the NiCr wire, I could get away with a > >MOSFET and switch it to control temperature. With the induction heating > >method, I'd need at least an H bridge and potentially complicated PWM > >control to shape the output signal into a sine wave. > > > >Also, how would I go about calculating the inductance, coupling, number > >of turns, frequency, etc...? I'm assuming this is far from an ideal > >transformer and losses will be great. > > Sure, since that is where the heating comes from. > > If you're interested, have a look at the work of Tim > Williams- he has built rigs that work up into the kW range. > > http://www.mindchallenger.com/inductionheater/induction10.html > > If you want to buy a resistive heater, something like shown on page 509 > here: http://www.watlow.com/products/heaters/cable-nozzle-heaters.cfm > would be good. For even better coupling, buy a wound MgO-insulated > coil and braze it to the outside of a relatively thin (to minimize > the heat capacity) copper tube. Stainless sheath is fine for your > low-temperature application. > > Hmm... since you say you want fast heating it seems inevitable that a > failure of the controller would cause the flash point of the wax to be > exceeded. Suggest some kind of reliable safety agency approved thermal > fuse (cutoff) in the circuit if this is a concern. > > --sp > > > -- > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .