How to make glaze test tiles and keep track of the results

Start with the right tile

A test tile only predicts your pots if it is made from the same clay. Linda Arbuckle's studio handout on test tiles says to use the appropriate clay body and calls 2 by 4 inches a handy standard size. Potclays, the UK pottery supplier, suggests that beginners prepare six to 10 tiles from the body they plan to use, and notes that tiles can be flat, curved, textured, or shaped like the piece you intend to make, because each form reveals different behaviour.

Give the tile features that make a glaze show its character. Arbuckle recommends an incised texture to show glazes that break or pool, and suggests a fork rather than a pin tool, whose lines are too small. Jessica Fong's ceramics text for the University of the Pacific recommends tiles with both a flat area for color and texture and vertical sections that show how the glaze flows. If you want to hang tiles on a board later, Arbuckle reminds you that holes shrink in drying and firing, so make them bigger than you need.

Arbuckle also suggests keeping a stock of bisqued tiles on hand, so there is always something ready to glaze. Her handout ends with the line that every kiln should have a test or two in it.

Mix small, measured test batches

Glazy's help pages recommend 50 g or 100 g of dry material for a first test, a 1 to 2 kg batch once a test looks promising, and 5 to 10 kg for production. For 100 g of material they suggest about 50 to 60 ml of water, less if the glaze has little or no clay in it.

Small batches need an accurate scale. Glazy advises one that reads to at least 0.1 g, ideally 0.01 g with a capacity over 100 g, and recommends sieving most glazes through 120 mesh or finer. It also suggests writing the glaze name and recipe on the container in permanent marker, which keeps the jar, the tile, and the notebook pointing at the same thing.

Apply it thin and thick on the same tile

Application changes the result almost as much as the recipe. Potclays notes that a thinner application may look more translucent, while a thicker one may crackle, run, or develop a deeper, richer colour, and asks you to note whether you brushed or dipped and how many coats went on. Fong's text recommends applying glaze in a gradient from thin to thick.

Arbuckle's method gives you both on one tile. Wax the back of the tile and the bottom quarter inch, up to an incised line, so the foot stays bare. If you glaze your pots by dipping, dip the tile the same way: dip the entire tile, wait a moment for it to dry, then dip the right half a second time. The result shows a regular and a thick application side by side, and how far the glaze moves in firing.

Fire tiles upright, and catch the runs

Arbuckle is emphatic that tiles should be fired upright, propped against a brick. Fired flat, a tile only shows what the glaze does on a flat surface, not whether it pools, striates, mottles, or runs on a vertical wall. If someone else is loading your tiles, she suggests leaving a note asking for upright firing.

An unknown glaze can run further than expected. Digitalfire's Tony Hansen fires his glaze flow tests with a tile underneath to catch overrun, and the same habit is a cheap way to keep a runny test off a kiln shelf.

Placement matters in fuel kilns. For gas firing in reduction, Arbuckle recommends two tiles, one near the top and one near the bottom, marked Top and Bottom, because heat and reduction usually vary through the kiln. Hansen also suggests preparing samples to fire one cone higher and one cone lower than your working temperature, which shows how forgiving a glaze is.

Label the tile and keep the record

The most repeated advice is to put information on the tile itself. Arbuckle brushes the glaze recipe onto the back in iron oxide and water, and spells out why: so often someone finds an interesting tile and its owner cannot locate the recipe notebook. Potclays lists stamps, carving, underglaze pencil, or a marker for writing a test code on each tile before firing, with the details noted elsewhere.

That code points to a fuller record. Rhonda Willers, writing for Ceramic Arts Network, describes a spreadsheet or dedicated studio notebook of tests, a labeling system of numbers and letters when there is too much to fit on the back of a tile, and notes on the application method, final kiln temperature, and cones used, with photographs of the results when she is being thorough. Potclays adds the firing temperature, ramp speed, soaks or holds, and cooling schedule, warning that even slight differences in these can dramatically change the result. Glazy suggests recording the type of clay, the firing date, and the actual cone reached.

Write results down the day you unload. Fong's text says to record observations immediately after unloading the kiln, including unexpected effects. A photo taken then, next to the tile code, does more than a description written a week later.

If you would rather keep that record on your phone, FireCard stores a photo of each test tile tagged with its glaze, clay body, cone, and atmosphere, and lets you filter the whole collection by any of them. Glaze recipes live in their own notes, and each firing is a separate card with its cone and schedule segments.

Common test tile mistakes

Testing on a different clay. Arbuckle points out that because fluxes and colorants interact, even clear glazes are not all alike, and testing over a range of slips helps you find the right one. Test on the bodies and slips you actually use.

Firing flat. It hides pooling and running until the glaze is on a tall pot.

One coat only. Without a thick section you will not know how the glaze behaves where it collects in a foot or a groove.

No code on the tile. Tiles outlive notebooks and get separated from them. Mark every tile before firing.

Recording the cone but not the schedule. Two cone 6 firings with different ramps, holds, or cooling can give different surfaces, which is why Potclays asks for all of them.

A test tile shows how a glaze looks and moves in one firing. It is not a food-safety or durability test. Digitalfire's Tony Hansen notes that a glaze visibly attacked by cutlery, stained by coffee, or altered by lemon juice deserves suspicion, and he treats leaching as something to test for separately.

Frequently asked questions