Over the last few years Gaffer® Glass has evaluated dozens of different soda lime and lead based studio furnace glasses from the viewpoint of expansion and viscosity characteristics. Our job as color bar manufacturers is to somehow fit with quite a large bandwidth of expansion coefficients out there. Unlike people who work with Pyrex, Bullseye, Spectrum 96, or Moretti for example, there is no acknowledged “master glass” that color bar makers can tune their glasses to. We therefore decided to formulate our own “master glass”.
Our research, both theoretical and physical, showed that the bell curve of linear expansion coefficients, melted in studios in the USA, Australia, New Zealand, Japan and England, peaked at around 96x10-7 (0-300oC) with the outer limits ranging from 94-99.5.
Viscosity data, based on Lakatos’et. al factors,¹ also showed very wide variations at melt and working temperatures, ranging from 1290oC (2354oF) at log 2 (a measurement of viscosity in poises at upper fining temperatures), up to as high as 1430oC (2606oF). Not that we are suggesting that people are necessarily melting at those temperatures, but for the purposes of comparing apples with apples, the Lakatos calculations are instructive. The bell curve for the temperature at log 2 viscosity peaked at 1370oC (2498oF), with some commercially available batches in excess of that.
Gaffer decided to set out designing a clear batch from scratch that would meet certain guidelines we felt were important and have the batch made available commercially by being manufactured by Philips (Code #3300) and Spruce Pine Batch in pelletized form. The criteria we considered important for a studio glass (not necessarily in order of priority) were the following:
The cost of energy remains one of the larger overheads in running a glass studio. Far too many batches, made commercially or privately, have not paid enough attention to viscosity characteristics. The viscosity of the melt at upper fining temperatures determines how quickly the seed is eliminated from the melt. A discreet amount of some chemicals can make a big difference to how fast a glass will melt and fine out.
The Gaffer batch formulation has focused on bringing the theoretical log 2 temperature down to 1325°C (2420°F) (Spruce Pine 87 is 1380°C (2516°F) by comparison) and working temperatures down to a more comfortable 1080°C (1976°F) range. Tests conducted on a 150 kg (330 lb) capacity freestanding pot furnace, have shown a gas saving of 10% when run at 1080°C (1976°F) compared to 1150°C (2100°F). This represents a considerably lower energy bill and less wear and tear on refractories. Pellets show a 15% faster melting rate than free flowing batch. The lower energy savings this batch achieves could come close to paying for it.
This formula, apart from having low working temperatures, also has a reasonably long working time.
It is remarkably fast to reheat, has good body and moderate surface tension. Whether trimming with shears, jacking in, shaping or knocking off the punty, it is a very forgiving glass.
It shows little tendency to devitrify and works well under the lamp.
We recognize that compatibility standards are set for us by our competitors and already widely used commercial batches. There is no point in being in our own universe. Trident seal tests2 with Gaffer® batch, melted under controlled time and temperature conditions, show zero strain with Spruce Pine 87 and East Bay Regular for instance.
We have melted this glass at 1400°C (2550°F) for eight hours after the last load (compared to the standard melt temperature/time of 1310°C/2400°F for six hours) and have lowered its COE by only around 0.5 points, showing that it is a forgiving glass re volatization. It shows very little drift after a week at working temperatures, perhaps ½ a point at most.
Expansion: 0-300°C: 96.0 x 10-7
25-300°C: 98.5 x 10-7b>
Anneal Point: 486°C (907°F)
Strain Point: 439°C (822°F)
Many of the batches we have analyzed have shown scant attention to durability, while others seem somewhat over the top. Alumina is the most effective cheap agent to make a glass more durable, but too much leads to a glass that is more difficult to melt. At melting temperatures, 2% Al2O3 can push up the melting temperature for the same viscosity by as much as 35°C (nearly 100°F).b>
Gaffer batch contains a much smaller amount of alumina, but supplements it with zinc oxide, which is as effective as alumina in resistance to water, steam and acids, but lowers viscosity/temperatures at the same time. Testing by Integrix3 shows its durability to water and acids, lies comfortably within industry guidelines for container glasses.
Pelletizing the batch has obvious benefits, not only melting faster, but cutting dust to virtually zero, leading to longer superstructure refractory life and lower dust levels in the studio generally. It should be noted that some of the new batches coming from Glasma and Philips contain Barium carbonate in quite large percentages, which is not only toxic, but at amounts over 4%, quite corrosive on pots.
Gaffer batch is relatively non-toxic, containing neither Barium nor Lead oxides. We decided against fluoride additions, which, while useful for lowering viscosity, are volatile at melt temperatures and speed up the corrosion of pots and superstructures. Fluorides also contribute to toxic and corrosive flue emissions. Gaffer batch proves to be a low impact glass from a corrosion and toxicity viewpoint. It is equally suitable for either electric4 or gas firing.
1. Viscosity/temperature data is compiled using factors given by Lakatos T., Johanssen L., Simmingsköld B., Viscosity-Temperature Relations in glasses composed of SiO2-Al2O3-Na2O-K2O-BaO-PbO-B2O3-CaO-ZnO-Li2O in the Compositional Range of “Crystalline” Glasses. Glasteknisk Tidskrift. 34 (3) 1979. 61-65
2. We rely on the trident seal test for all compatibility or thermal mismatch tests. It is far more accurate than thread tests or dilatometers. See Hagy H.E., The Trident Seal- A Rapid and Accurate Expansion Differential Test. J. Am. Ceram. Soc. 62 (1972) 60-62
3. As tested by Integrex Testing- a laboratory division of Owens Corning Inc. Tests performed on freshly melted glass fined after the last load for six hours at 1310°C (2390°F) in a free standing pot.
4. We advise active air flushing around the interface between the power input and the elements or electrode connections.