this is my first post but have been around for a while.
I want to build a melting point apparatus.
I plan to use a heating cartridge (about 200 W) but need to control the power input.
A Variac would but really need to control the heating rate.
Would a PID controller work?
ScienceSquirrel - 3-2-2013 at 15:35
You can do it with a big test tube filled with mineral oil.
To stir you want to use a coil of wire or if you want to get sophisticated a flea and a magnetic stirrer.
A thermometer of an appropriate range with a sealed capillary filled with the test substance is strapped to it and placed in the oil.
Heat very slowly with stirring using a small colourless gas flame.
If you want to be really traditional, puff at a long stemmed pipe and stroke your beard while doing the mp! mr.crow - 3-2-2013 at 18:41
Well a PID controller would be great for setting a desired temperature. For a melting point you want a slow constant temperature increase, not a
feedback loop.
To control the power to a heating element use a standard dimmer switch in an electrical box. This also works great for heating mantles.chemrox - 4-2-2013 at 13:19
Squirrel describes an alternate to the Thiele tube. Thiele tubes can be had from ebay and most science supply stores. Their use is a time honored
way to get melting and boiling points. The Thiele tube obviates the need for a stirring "flea" by providing constant flow of the hot oil due to
convestion through the tube. However, to avoid the messiness of mineral oil and improve precision a melting point apparatus is preferred. However,
you can't do boiling points with (most) melting point machines. Besides this is clearly a DIY adventure, like building a radio.veganalchemist - 4-2-2013 at 13:39
I work at a local Uni and the undergrads use tubes filled with oil for melting points.
I want to build something a bit more sophisticated, where I can control the temp rate. watson.fawkes - 12-2-2013 at 10:05
Well a PID controller would be great for setting a desired temperature. For a melting point you want a slow constant temperature increase, not a
feedback loop.
PID controllers can be used for a ramp, though the math is a bit more involved. The error
term is calculated from the target temperature as a function of time, not simply as a constant. That said, it's desirable that the thermal mass of the
sample holder be rather larger than the sample, to make it easier to get a consistent ramp, and also that thermal conductivity be high, so that
temperature consistency is good.
