Science Fair Projects for Hot Glass

Rev. 2001-08-09, 2005-11-09, -12-03

Back to Sitemap

Corning Museum of Glass | Resource on Glass | Simple Glassmaking

Being certified as a teacher, I am not about to give you projects on a plate, but will try to give you enough information to make a reasonable start on your own. At each stage, there may be choices that involve higher cost and lower cost.

In these descriptions, the words CORE EXPERIMENT are used when the idea is to test the fundamentals of science involved. These are often the cheapest, but also require the most adept understanding of the science involved. Doing these core projects provides cheap tools for later projects. BASIC EXPERIMENT are those that simply test the essentials of what is going on, to gather data to demonstrate that the science works. These normally require purchase/availability of moderately costly tools of some kind.

MEASUREMENT - Although it is possible to do projects related to gas flow and pressure and tie them to glass, most of these are so vaguely connected as to allow me to leave them alone. But temperature is a whole different matter, as knowing the temperature of glass is absolutely vital to reliably making good glass objects repeatedly. Voltage is tied to glass because most measuring devices for temperature actually measure voltage.

  1. A thermocouple is two dissimilar metal wires connected (welded) at one end and mounted so that end can be put in the heat to be measured. (See RTD below) When the tip is heated (or cooled) the thermocouple effect produces a small voltage that can be measured or amplified.
    CORE EXPERIMENT 1 - Build a sensitve analog voltmeter from scratch and calibrate it for further use. Most physics books will discuss making a galvanometer.
    OR - Get a good, inexpensive, digital volt meter that will measure millivolts.
    BASIC EXPERIMENT 2 - Create and test, using a sensitive voltmeter, several thermocouples using different metals. Thermocouples usually have the different metals welded at the tip, but simply cleaning the ends with a file, heating red in a flame and pounding the two with a hammer on an anvil-like surface will usually work for a good enough mechanical and electrical connection. A number of metals have been tested for good thermocouple effect and are used in standard devices. Wire is available for these metals and should be included in testing. Read about why some metals are better at different temperature ranges and how thermocouples fail. A site with information and a catalog with a lot of information is http://www.tcomega.com 1-888-TC-OMEGA
    CORE EXPERIMENT 3 - The thermocouple effect depends on the change in temperature along the wire and normally can not be measured with a single metal because if a wire is heated in the middle, there are equal and opposite effects moving away from the flame which produce no net voltage. If, somehow, a connection could be made to the heated point that did not involve the thermocouple effect (to any great degree) it could be possible to measure or use the effect in one kind of wire. A safety device for flame detection does exactly this by using the fact that a high temperature flame is conductive to test if the flame is on - the hot flame grounds the end of the thermocouple wire and the effect produces a voltage back to the cold end of the single wire.
  2. An RTD (Resistance Temperature Device) or a Thermister are devices that measure temperature by carefully measuring the change in resistance that occurs in almost all materials when heated or cooled. RTD's use a lot of fine wire, often platinum, chosen to give a very even resistance change over a wide range of temps. Printed circuit versions can be made on ceramic material. Platinum is very nice (and expensive) for high temp measurements. Thermisters are solid state devices that have a great change in resistance (10:1) over a much narrower temperature range (0-300F), which is nice because they are cheap. Not used in glass work because too low temps.
    BASIC EXPERIMENT - Using a simple reliable heat source like an oven or an electric fry pan, measure the resistance of lengths of wire of various materials (iron, copper, nichrome, brass, stainless steel, aluminum, etc.) Test first to get some idea of how long a length is needed to get a measureable change in resistance. Since longer lengths will be involved, solve the problem of keeping the coils separated/insulated.

HEATING - Heat can be provided by flame or resistance to electricity (and chemical reaction and nuclear power - neither used in glasswork.) See the notes on insulation below to help control the heat. Flame has the problems that it requires venting to provide air for burning and to remove exhaust products (besides safety factors of escaping fuel and carbon monoxide.) Electrical has problems at the higher temperatures involved in glass working as cheaper materials reach their working limits.

  1. BASIC EXPERIMENT - Show both the resistance heating effects of electricity and the dangers of under sized wires by making a short (2-3') extension cord from light speaker wire (24 gauge) or light bell wire (22 ga), plug and connector. To provide a load that will heat/damage the wire, use a toaster or small room heater. To protect the other wiring, use a power strip with a builtin circuit breaker. When the wire heats and burns though, it is likely to drop hot wire below it, so be sure it is over concrete or other fireproof material. Don't do this experiment indoors unless good ventilation is provided!!! Besides the hot metal, there will be smoke from the insulation. Record the smoke and flame with a camera. Plug in the heat source, check the wiriing, plug in an extension cord to the thin wire and MORE
  2. MORE

INSULATION - Keeping heat in the right place is the key to efficient use of fuel and of melting glass with high temperatures. This can be done with a very high temperature torch and a small amount of glass, which is the method of lampworking.

Insulation for glass working is referred to as refractory because it has to stand high temperatures. The two most common forms of insulation are insulating fire brick and ceramic fiber blanket. Each is amazing for how cool they are on one side while the other is torched to white heat. Both can be bought from Refractory suppliers in the Yellow Pages of the nearest metropolitan area. These supply to the high temp heating industry including ceramics, metal treating, foundry, and brick. Normally each product must be bought in at least 1 box quantity which will cost $60-90, but if only a single student is pursuing projects, asking may result in brick from a broken box or blanket leftovers. Caution: Blanket has some of the characteristics of asbestos and should be handled and cut with care; ask for the MSDS or look in the internet for it.

  1. Insulating fire brick (IFB) is a brick with lots of air space in it made of a special clay. It normally has to be bought by the box of 25 bricks from a refractory supplier. It is possible to make IFB by mixing straw in with the clay while shaping the brick, the straw burning off when the brick is fired, but that is more ceramics than glass and I will leave it to the student who is interested.
  2. Ceramic fiber blanket is a space age material that is bought by the roll from refractory suppliers. It was developed the the space shuttle and is a white non-woven fiber mat of fibers like spun Corning ware. Different densities and temperature ratings are available.

MORE

REFRACTORY - MORE

POTS & CRUCIBLES - Glass must be melted in something (except in outer space) and that something must not melt and must not contribute either trash nor added ingredients to the glass. Trash can include fragments of the pot, small or larger. Modern commercial crucibles are brittle and must be heated carefully over several hours because, although they stand very high heat for months and shed little into the glass, they handle thermal shock very poorly.

When factories were making glass bottles and windows by hand, the large pots were made at the factory from local clay and were used for a single large batch of glass (500-1000 pounds) that would be blown out in a week and then the pots were removed and crushed to make grog for the next batch of pots. MORE

GLASS CHARACTERISTICS - Glass (or more precisely glasses as they are many substances) has two important testable properties - Relative COE (Coefficient of Expansion) and Sag Point. To demonstrate the first, a torch is needed, while a kiln with good control is needed for the second.

One way COE is tested by making a thread of the two glasses to test, pulling the molten glass into a long piece. As it cools, if the COE's of the glasses are different, the thread will curve toward the glass with the lower COE. MORE

Sag point is important because it is used to figure the annealing point - the temperature at which glass has to be held to allow relief stresses put in while working before cooling at a controlled rate is allowed. The sag point is found buy cutting (about 12") long thin slices from the edge of sheet glass or pulling 1/4" rods of molten glass. These are supported at the ends and placed in a kiln. The temperature is raised slowly and held at 5 steps for five minutes until the glass just begins to bend. Normally, the approximate sag point is found by raising the temperature more quickly and observing the temp then setting a new arrangement with the temp about 50 lower and bringing it up slowly. Different colors of the same maker can have different sag points. These MORE

MELTING A BATCH OF GLASS - It is relatively easy to melt glass (bottles or window) and blow some using clay pots, insulation and a powerful burner - IF you don't want to keep the blown glass (no annealler), don't care how about fuel economy (wasteful in the short run, so no long runs) and can be careful with black iron (gas) pipes that will get very hot, don't care about the pots, which will probably crack on cooling, and don't care about the quality of the glass (bubbles and trash from the pots.)

I am working on getting and testing a simple glass recipe of sand, limestone, and pot ash (or equivalent) that can be melted and produce some (probably fairly cruddy) glass, just to show what is needed. MORE

 

Contact Mike Firth