Technofile: Ian Currie’s Grid Method


Glaze Behavior and The Grid Method

Ceramic artists love to collect glaze recipes. We fill notebooks and save scraps of papers full of recipes we hope to try someday. But glaze recipes are only useful to us if we understand the raw materials they are made up of.

If you see a glaze that you think would be great to use as the main glaze for all of your functional ware, you need to be aware of the many variables that determine whether you’ll get a desired result. Possible variables include kiln type, firing method, where your raw materials are mined, and what type of clay body you are using. A glaze that works for one potter may not work for another. Say you test this glaze in a 100-gram batch and it fires well on a test tile. Then you mix a 5000-gram batch, and use it on an entire kiln load, but this time it fired with rough spots and was thin and dry. Assuming that the glaze was weighed, mixed correctly, wetted to the specific gravity of the original test, and fired to the same temperature/cone, we must consider the raw materials used in order to understand what went wrong and how to fix it.

To help us get a lot of glaze results in one firing without mixing 35 individual glaze tests, the late Ian Currie developed a technique called the Grid Method. In short, it is a method that segments test tiles so you can mix up just four base glazes and by combining them, discover 35 glazes, and understand what is happening in each one. The alumina and silica are systematically varied throughout the tile, while the proportions of the flux materials, one to another, are the same in all of the 35 tests.1

RO Groups and Raw Materials

We know that glazes are a balance between three essential groups of materials: a glass former (also called the RO2 group), which generally is silica, fluxes (also called the R2O, RO group), which reduce the high-melting temperature of silica, the intermediates (also called the R2O3 group), include a viscosity agent (stiffener), which keeps the glaze from running off the pot and is generally alumina (Al2O3)  and boric oxide, which is both a flux and a glass former. Learning the characteristics of these groups and how they interact in your glaze will allow you to control the results.

Screen Shot 2015-09-17 at 3.08.11 PM

Silica and alumina are very reliable. Systematic variation of alumina and silica is central to the Grid Method, and is largely responsible for its success. The method is organized so that one is able to isolate the variables and therefore highlight cause and effect. Currie explained that, “It gives precise control and understanding of things like color response, maturity, crazing; glaze surface phenomena such as mattness, shininess and orange-peel surface; opalescence, opacity, color-break phenomena, etc.”

The fluxes also influence a glaze’s look and surface. A glaze often contains a mix of fluxes. Depending on this percentage of the flux mix in your overall glaze balance, the surface produced may be glossy, dull, crazed, fluid, stiff, smooth, or dry and many other surface expressions.

Screen Shot 2015-09-17 at 3.33.37 PM
Screen Shot 2015-09-17 at 3.33.44 PM

Your choice of flux and the added oxides they contain can greatly influence the outcome of the glaze surface and the hue of the coloring oxides you have added. Here is a common example: A glaze recipe calls for calcium oxide (CaO) and the source you chose is dolomite, which includes calcium oxide, but also contains magnesium oxide (MgO). The base glaze is a white matte and you have painted a decoration under the glaze surface with cobalt oxide. The fired glaze results in a matte white but your brush decoration turned lilac, not blue, as intended. That is because magnesium promotes lilacs and pinks when combined with cobalt. Test the recipe again with wollastonite as the source for calcium oxide, and your cobalt decorations fire blue, but your surface is a bit glossier, because additional silica was also added to the mix. A third test with whiting as the source for calcium oxide yields blue decorations and a matte surface because whiting doesn’t bring extra silica to the mix (1). The lesson to take from this is that, depending on the source used to provide the fluxing oxide you want, it may contain other oxides that will influence the glaze result.

There is a basic group of 12 glaze oxides (see sidebar on page 102). It’s essential to recognize each oxide in the group and know which subgroup they belong to (flux, intermediate, glass former), in order to understand the grid of tests you will be making, and later, to make the glazes work for you. Becoming oxide literate gives you the ability to alter glazes to your specific needs because you know how the oxide percentages behave in the presence of other oxides.

Screen Shot 2015-09-17 at 3.33.55 PM

Calculating a Blending Chart

The Grid Method can be used to put together a glaze base with an understanding of what influences oxides in a glaze balance have, their potential, or to find all the glazes in a family set.

Currie’s website has a calculation page (, constructed by his son Hamish), where you can type in the fluxes from your recipe, omitting the silica and kaolin, and the program will calculate all of the 35 glazes to test on your tile including the varying silica and kaolin/alumina percentages in each one (2). (The example is one of my favorite glazes for decorative surfaces.) After the calculation is complete, you will have the recipes for the 35 tests (3). You can test each one separately if you like, or to test all 35 you can do so by starting with the recipes for the four corner glazes and mixing the combinations by volume.

The following diagram (4) shows what the glaze recipe is for each grid square. It will be the same for every new glaze grid you make. Here you can easily see the percentage of flux materials, and how the kaolin and silica are systematically added.  Create a blending chart for mixing the 35 combinations and highlight the four components on your blending chart with four different colors (5). Color coding the blending chart is helpful in order to fill the test cups with the correct amounts of each corner glaze. Next, fill in the blending chart with the appropriate volume of each glaze.

Creating a Grid Test Tile

Make a 5×7-inch ceramic tile from your clay body. Mark the tile with seven rows of five square impressions to create 35 equal squares.

Mixing 35 Tests

Mix and weigh the four corner glazes as 300-gram batches. Add water and sieve the glaze slurry of Glaze A a bit thicker than you normally use a glaze. Measure the volume of each corner glaze in a measuring cylinder. Start with Glaze A, which is the thickest because it contains the most kaolin. Add just enough water so that the wet glaze measures no less than 460mL and no more than 500mL (a little less than one pint). Measure and mix the remaining three corner glazes and add water sparingly so that each corner glaze has the exact same wet volume and none of them become too thin.

Set up 35 plastic cups in the same order as the blending chart/tile grid format. Label the cups and make sure the numbers correspond—Corner A (top left) is #1, Corner B (top right) is #5, Corner C (bottom right) is #31, and Corner D (bottom right) is #35. Fill in all five rows horizontally and seven vertical columns, with the appropriate numbered cups (from 2 to 34.) Note: The cups in the blending chart each have approximately 48mL of glaze.

At all times while filling the test cups, make sure the four corner glazes are stirred and not settled. C and D, due to their lack of kaolin, will need constant agitation or they will become thin and settled.


Glazing the Grid

Fill a syringe with the appropriate glaze volumes and add them to the cups. Be sure there are not any air bubbles in the syringe. Check off the blending chart as you work in order to avoid mistakes. When finished, view the cups at eye level. They should all have an even amount of liquid (not the amount of settled glaze, but the full volume of liquid). If any of them are not equal, use the remaining corner glazes to remix the cup.

Next, apply the glazes in the correct order on the grid tile. Break up the 5×7 grid of glaze cups with lines between them so that it is easier on the eye, to follow the grid pattern. Apply the glazes with the syringe. Grease the syringe with Vaseline or similar. It needs to be well lubricated for easy application and decrease any glaze squirting out onto another glaze compartment on the grid. Be sure to clean up any splatters, and re-apply. Squeeze a second application, after the first has dried, on half of the grid compartment so you can compare a thicker application (a thick application is approximately 2 cm, which can be checked with a needle tool).

If you would like to see how a white or transparent glaze looks with a coloring oxide, you can add a tiny dot of pure coloring oxide and water to the test grids.

Screen Shot 2015-09-17 at 3.29.50 PM
Screen Shot 2015-09-17 at 3.29.39 PM

Fired Results

Once your grid is fired (9), you will notice the glazes are drier with the increased kaolin (Corner A) and the ones with the highest amount of silica (Corner D) appear to be the ones with the best melt. A diagonal line from Corner D to Corner C reveals several melted and interesting glazes. Which ones are interesting, beautiful or have potential to you? What are their surfaces characteristics? Dry, stiff, runny, pooled, crazed, breaking, crystals, etc.? The tile is flat, so very runny glazes are not running downward, but if they are very runny, they could pool into the corners of the grid or craze. Retesting that singular glaze on an upright tile will give you even more information.

How are the glazes being altered as the kaolin level is being increased from the bottom of the tile and upward, and as the silica level is being increased from left to right? Remember that the Grid Method systematically varies the alumina and silica throughout the tile, and the proportions of the flux materials, one to another, are the same in all of the 35 tests.8

In conclusion, beyond discovering 35 possible glazes, the Grid Method can also help you find the ideal silica/alumina flux ratio for the cone, glaze surface, and color you want, ultimately giving you valuable information on how to adjust your glazes.

Thank you Ian Currie for your lifelong work and dedication to the ceramic community. Your large capacity for friendship and your brilliance has made a difference in my life, both in the studio and for inspiring me to share everything I discover about clay. You are remembered. Ian Currie’s book Revealing Glazes: Using the Grid Method is available at

 Alisa Liskin Clausen is an American-born ceramic artist with a BFA from Syracuse University. She has lived and worked as a potter in Denmark for the past 20 years, focusing her work on glaze development from local materials. She assisted and worked with Ian Currie in the early 2000s.

1, 8 Ian Currie, Revealing Glazes using the Grid Method, Grid Method Outline, page 30.

2, 3 Ian Currie, Revealing Glazes using the Grid Method, Blue-in-the-face Chemistry, page 13.

4, 7

5 Ian Currie, Revealing Glazes using the Grid Method, Standard Recipe Grid, page 32.

6 Ian Currie, Revealing Glazes using the Grid Method, 300 gram blending chart, page 102.

  • Cathy H.

    This article left me wondering how this technique would help solve the stated problem. How will making more test tiles solve the problem of scaling up from a small test batch to a large batch?

    “Say you test this glaze in a 100-gram batch and it fires well on a test tile. Then you mix a 5000-gram batch, and use it on an entire kiln load, but this time it fired with rough spots and was thin and dry. Assuming that the glaze was weighed, mixed correctly, wetted to the specific gravity of the original test, and fired to the same temperature/cone, we must consider the raw materials used in order to understand what went wrong and how to fix it.”

Leave a Reply

You must be logged in to post a comment.

Enter Your Log In Credentials
This setting should only be used on your home or work computer.

Larger version of the image

Send this to a friend