Glass Annealing Schedule Scaling Calculator

Anyone who has annealed glass has a schedule that works. What nobody has is the same schedule for a piece twice as thick, and reaching for the chart again usually produces a number that looks small enough to be wrong. It is wrong: the times on an annealing chart do not follow thickness, they follow something closer to its square, so a piece twice as thick is not twice the wait. This takes a schedule you already trust at one thickness and stretches every part of it by whatever relation your own source uses.

The thickness the times and rates below came from. Read it off the chart, the data sheet or your own log entry.
The thickest section, not the average. A thin dish with a heavy foot anneals as the foot.
Take two rows of your own chart at different thicknesses and read the exponent off them: divide the log of the time ratio by the log of the thickness ratio. This page publishes no exponent for any glass and the 2 in the box is a placeholder that makes the arithmetic visible.
From the schedule you already have. This page states no annealing temperature and no soak for any glass.
The slow leg after the soak. From your schedule.
Leave the span at 0 to skip a leg.
The last run down to where you open the kiln.
Heat-up, the working or fusing part of the firing, any hold at the top — whatever you count as fixed for this piece. Put 0 to see the annealing part on its own.
Optional. Electricity plus whatever you charge the kiln for elements and shelves — from your own meter and your own log.
Glass Annealing Schedule Scaling Calculator by ThicknessBuildFigure

Thickness does not go into the clock the way people read it

Put the defaults through it. A schedule written for quarter-inch glass — an hour at the soak, then 150 degrees at 100 an hour, 200 degrees at 200 an hour, 600 degrees at 400 an hour — comes to 5 hours of annealing. Ask for half-inch glass at the square, and every one of those numbers is multiplied by four: the soak is four hours, the first leg drops from 100 degrees an hour to 25, and the annealing runs 20 hours instead of 5.

Only the block marked fixed stays put. With three hours of heat-up in front of it, the firing goes from 8 hours to 23, and the annealing goes from 62.5 percent of the firing to 87 percent of it.

Why a slower rate and a longer soak stretch by the same amount

It looks like two different mechanisms and it is one. A soak is time, so scaling the time is the obvious move. A cooling leg is a number of degrees at a rate, and dividing the rate by the same factor gives exactly the same multiplication in hours, since the degrees do not change. So the soak and every leg move together and the ratio between them never shifts — which is why the table further down the page can be a single column of totals rather than a matrix.

The practical consequence is that there is no clever reordering that saves the time. You cannot take it out of the fast leg at the bottom, because the fast leg has been stretched by the same factor as everything else.

The exponent is yours, not ours

The box says 2 because something has to be in it, and the square is the relation most published annealing tables sit near. It is not a fact about your glass, and this page will not tell you that it is. The way to get the real one is to take two rows off whatever chart you actually use — say a thickness and twice that thickness — divide the log of the time ratio by the log of the thickness ratio, and put the answer in the box. If the chart gives 4.2 times for a doubling, the exponent is about 2.07.

Charts differ. A chart from a supplier with a wide annealing range and a chart written for a specific casting glass will not agree, and neither of them is a general truth about glass.

What the page cannot answer

It cannot tell you whether the piece annealed. That is a property of the glass, the actual temperature the kiln held rather than the one on the controller, how evenly the load sat, and how thick the piece really is where it matters — and the only way to find out is to test the work. The page also has no view on whether the thickness you entered is the right one to be scaling on; a plate with a thick foot is the foot, and a piece with a solid boss on a thin sheet is the boss.

Where it earns its place is in stopping the arithmetic mistake: reading a chart written for thin glass and adding a bit for a thick piece. Adding a bit is never the answer, and the page shows you by how much.

Questions people ask

How much longer does thicker glass take to anneal?

At the square, which is what most published tables sit near, the whole annealing part goes up by the square of the thickness ratio. Double the thickness and the soak and every cooling leg are four times as long. The page lets you put in the exponent your own chart implies rather than assuming one, because the relation is a property of the chart you use and not a fact this page can assert.

Why does the cooling rate get slower as well as the soak getting longer?

They are the same stretch seen from two directions. A leg covering a fixed number of degrees at half the rate takes twice as long, so dividing the rates by the scaling factor produces exactly the same multiplication in hours as multiplying the soak by it. Everything in the annealing part moves together, which is why no reordering of the legs saves any time.

How do I find the exponent for my glass?

Read it off the chart you already use. Take two thicknesses and the times against them, divide the log of the time ratio by the log of the thickness ratio, and that is the exponent that chart is built on. If a doubling of thickness quadruples the time, the answer is 2. This page publishes no exponent for any glass and the value in the box is a placeholder.

Which thickness do I put in for a piece that is not an even slab?

The thickest section that has to come through the annealing range, not the average. A shallow dish with a heavy foot behaves as the foot does. The page has no way to see the shape of your piece and takes whatever single number you give it, so the judgement about which section governs is yours.

Will this schedule anneal my piece?

That is not a question a calculator can answer and this one does not try. It stretches a schedule you already have by a relation you supply. Whether the original schedule was right, whether the kiln actually held the temperature the controller displayed, and whether the piece came out with stress in it are all settled by the glass data sheet, by the supplier and by testing the work.

Related