These two pieces are fired at cone 6. The base transparent glaze is the same (G2926B Plainsman transparent). The amount of encapsulated red stain is the same (11% Mason 6021 Dark Red). But two things are different. Number 1: 2% zircon has been added to the upper glaze. The stain manufacturers recommend this, saying that it makes for brighter color. However that is not what we see here. What we do see is the particles of unmelting zircon are acting as seed and collection points for the bubbles (the larger ones produced are escaping). Number 2: The firing schedule. The top one has been fired to approach cone 6 and 100F/hr, held for five minutes at 2200F (cone 6 as verified in our kiln by cones), dropped quickly to 2100F and held for 30 minutes.

This was a fast firing. The glaze is G2934, a silky matte. But that does not mean it is pinhole-prone, it has good melt mobility. The clay on the right is Plainsman Coffee Clay. It contains 10% raw umber, that generates plenty of gases during firing. The centre one is Plainsman M390, not normally difficult to fire defect-free. The left one, M332, should be the worst, but is the best! What is needed to fire these without pinholes? The drop-and-hold and slow-cool C6DHSC firing schedule. It is extra effort to program your kiln controller, but well worth it. If you don't have a kiln controller then by a little experimentation you can develop a switching pattern to produce the same effect.

After a customer experienced blistering with a transparent glaze on a terra cotta body, suspicion was raised that the batch of frit of bad. Ferro even admitted there was an issue. But we decided to test, sampling about 20 of the 50 lb bags and combining that to make these tests. Upper left is a GLFL test, if compares the melt flow of an old batch of frit 3195 with the questioned one, fired at 1750F. Although the questioned batch does run a little more, a more important difference is in the bubble development (look closer), perhaps there is some fluorine contamination. A GBMF test compares the two samples on the bottom left (10 gram balls melt downward onto a sample tile). Again, the questioned batch is melting and bubbling a little more. We made two glazes: Cone 6 GA6-B (top right) contains 20% of the frit (the left one uses an old frit batch). The cone 05 G1916QL1 glaze (bottom right, it contains 60% of the frit) uses the questioned frit and works well on both the white and red bodies. So, although it might not be working for some, we determined that the frit is OK! Although it is possible that only certain bags of the batch were bad, that seems unlikely given the way frits are made, all bags in a batch should be the same.

Ravenscrag Slip, by itself, produces a silky transparent glaze at cone 10R. It is an excellent base to which to add colorants and modifiers. This is a simple addition of 10% iron oxide (Ravenscrag Slip already contains 2% iron, making about 12% total Fe2O3). It produces a stunning crystalline fired surface on these two porcelains. We can call this a "Beyond-Tenmoku" (crystals happen because of more iron or a slow cooling rate than a Tenmoku). The 12% iron dissolves in the glaze melt during firing, but during cooling in the kiln, the extra 2% precipitates out to produce these surfaces. The iron also acts as a flux in reduction firing, greatly increasing melt fluidity. Take that last statement seriously: The iron is a flux, the glaze will melt much more. So just adding iron oxide to a glossy transparent will wreck your kiln shelves when it runs down off the ware. Ravenscrag Slip does not melt-to-glossy, it has just enough feldspar to fire to a durable surface, making it a more stable host for the iron addition.

Drying cracks are opportunistic. They will initiate inside sharp acute angles. The sharper the angle the greater the chance of crack. By doing this procedure after trimming you will deny a crack a place to start. Of course even drying is still important, the water content of a handle should now be allowed to get ahead of that of the main body of the mug at any time. In the pictures on the right, two tools are being used to compress and round the angle at which the handle meets the wall of the mug.

This bag is very small, this material is very dense and heavy. The primary use of this material is obvious: For making welding rods. However notice the bag bottom is marked "Ceramic Rutile" (with a batch number). Why would a product intended for making welding rods be used in ceramics? The answer is very interesting.

This is M340S with G2934 matte white outside and G2926B glossy white inside. Consider what can go wrong. White matte glazes tend to crawl. No wait - they love to crawl! This would happen on every single mug unless I add CMC gum to make it a base coat to adhere well to the bisque (the tin oxide version is worse than zircopax). The clay has granular manganese added to produce the speck, if accidentally over-fired, even half a cone, it will bloat. And the clay body: The outer glaze is ugly on dark-burning clays. And it is drab on porcelains. It does not even look good on this same body if the speckle is not there. Another difficulty: Controlling the degree of matteness. I blend in about 20% of the glossy, otherwise it would fire too matte. And the firing schedule: PLC6DS - its drop-and-hold step is critical, without it the surface would be full of pinholes. Another problem: If the kiln is heavily loaded and cools slower than the programmed ramp-down, the surface will be too matte. Finally, glaze thickness: If it is too thin it will look washed out and ugly. Too thick it will bubble and look pasty.

These are porcelain and stoneware mugs. The glaze is G2571D (based on the fritted G2571B version of G2571A). This project started with calculations to source boron from a frit (instead of Gerstley Borate), MgO from talc and a frit (instead of dolomite). These moves enabled eliminating raw silica from the recipe. This produced a finer silky texture and better melt fluidity (for hosting colors). Adding rutile and zircopax produced a great bamboo. But what about variegating using titanium instead of rutile (it contains no iron so colors should be brighter)? Look what happens with this red stain! The titanium has done it's job even a little too well (creating a slightly more abrasive surface). Cobalt and titanium also worked. Imagine if other stains would produce dramatic results like this, each with the amount of titanium tuned to it!

Stains can work surprisingly well in matte base glazes like G2934. But they perform differently in a matte host glaze. The glass is less transparent and so varying thicknesses do not produce as much variation in tint. Notice how low many of the stain percentages are here, yet most of the colors are still bright. A good reason to minimize stain concentration is to avoid leaching. We tested 6600, 6350, 6300, 6021 and 6404 overnight in lemon juice, they passed without any visible changes. It is known that MgO mattes, like this one, are less prone to acid attack that CaO mattes. A down-side to the MgO-matte-mechanism is that chrome-tin stains do not work (e.g. 6006), high CaO content is needed in the host glaze to develop the color. The inclusion stains 6021 and 6027 work very well in this base. As do the 6450 yellow and 6364 blue. And the 6600 produces an incredible gunmetal black. The 6385 is an error, it should be purple (that being said, do not use it, it is ugly in this base). The degree-of-matteness can be tuned by blending in some G2926B glossy base.

This glaze, G2926B, is our main glossy base recipe. Stains are a much better choice for coloring it than raw metal oxides. Other than the great colors they produce here, there are a number of things worth noticing. Stains are potent colorants, the percentages needed are normally much less than metal oxides. Staining a transparent glaze produces a transparent color, it is more intense where the glaze layer is thicker, this is often desirable in highlighting contours and designs. If you add an opacifier, like zircopax, the color will be less intense, producing a pastel shade the more you add. The chrome-tin maroon 6006 does not develop well in this base (alternatives are G2916F or G1214M). The 6020 manganese alumina pink is also not developing here (it is a body stain). Caution is required with inclusion stains (like #6021), the bubbling here is not likely because it is over fired (it is rated to cone 8), adding 1-2% zircopax normally fixes this issue.