Research home page

Electric furnace

     I find it occasionally necessary to heat treat small metal or ceramic parts in an environment where the temperature can be quickly ramped to a desired value and instruments can be easily connected to the part being treated. I have a large muffle furnace that I use for firing/heat treating things, but it is expensive to run and hard to get instrumented. I was curious what could be done with the odds and ends kicking around the lab; these are the results.

     The heart of my small furnace is a vintage electric toaster element. ( Eagle #335 Fit's All toaster element, circa 1956 ) I bought a pair of these from a junk dealer on the old Canal St in Manhattan, and I'd be surprised if I paid more than a dollar or two for them. Each element can run separately at 110V RMS at about 5A RMS, for about 550 watts of power. The design is nichrome ribbon wrapped around mica sheets. They look well built and rugged; your mileage may vary depending on the toaster you scrounge to build your furnace.

     The body of the furnace is firebrick, I picked up a few from a local beige box hardware store that shall remain nameless. I tried several configurations, and found that the flat "pizza oven" shape gave the best performance in terms of peak temperature. There is a reason ovens look like that (grin). The size of the furnace cavity is 4.5 inch x 4.5 inch x 1 inch.

     The controller is an old autotransformer; flip the switch and dial in the voltage to get the element as hot as is needed. The brick in the bottom left part of the picture is the furnace roof. An additional piece of mica sheet is shown which can be used to cover the electrically active heating element while the furnace is in operation.

     In the picture below the mica sheet has been removed and a thermocouple probe is used to probe the furnace temperature. The furnace rests on a steel plate with one inch legs, these are sufficient to allow operation without damaging the tabletop.

     The following temperatures were measured at the indicated autotransformer settings. Note that full scale on the autotransformer is 150V RMS, so a scale reading of 80% is actually 120V.

     Measurements were taken in the center of the furnace cavity with the mica insulator covering the element. Using the thin piece of mica reduced the temperature by about 4 percent.

     The temperature variation from the center of the furnace to a point one half the distance to the wall was about five percent. Operating the furnace with the mouth open as shown makes the front somewhat cooler, but it is easier to get probes and instruments in and out. The time constant of the furnace on cooling is about 15 minutes.

     To date I've run the furnace up to 630C without problem, 800C seems within reach with this autotransformer. I could use both heater elements for double the power but I prefer having a spare in case of failure.

     Total cost of the furnace, minus the autotransformer, is about 10 bucks. Total running cost is about 10 cents an hour.

Research home page