Annealing glass

Annealing as a process

Annealing is a simple process. It involves lowering the temperature of the glass through the transition zone at a sufficiently slow rate that the temperature gradient through it will be sufficiently low that unsafe stress levels will not develop.

When making sheet glass it is a continuous process but in kiln re-firing processes it is done in two steps:

1. Holding the glass in a fluid state at a constant temperature for a time sufficient to allow the temperature throughout the piece to equalise; and then

2. Cooling it through the transition zone and into a solid state, at a rate sufficiently slow that excessive stress will not be locked into the piece. 

Most glass used for kilnforming, both float and rolled, is ‘soda lime glass’ with a coefficient of expansion (COE) ranging from about 80 to 100.

Individual glass makers, in their product data sheets, state the temperature values for annealing and other points in terms related to specific viscosities.

Here are some values gathered from recent glass makers literature. All values are in degrees C.

The values stated here were gathered from printed or website sources available at time of writing, in late 2008. They could have changed subsequently.

There’s a high degree of preciseness implied in the above numbers, possibly due to the way they are calculated. As they relate to glass viscosity values, they clearly refer to the temperature of the glass itself, and not to the kiln temperature.

On the other hand, other given recommendations, such as anneal soak and cool rates and times, are often based on manufacturer preference, as much as on technology.

Here are some recommended process times taken from the same sources.

Note that all are for fusing two layers of 3mm glass.

It is difficult to conceive that the minor differences in formulation of glass, with near identical COE, could require such a wide variation in process times.

It is much more likely that some experts are conservative, whilst others are more adventuresome in how they specify time and temperature. Logically, anywhere in between will also work, so you could argue that it is almost a case of ‘pick a number’.

Multipoint annealing

You may not be aware that some of the glass types mentioned above are also designed to be fused together, (such as Bullseye transparent and opal). The widely differing anneal and strain points in such an instance can present the kilnformer with interesting challenges when writing firing patterns.

As a general rule you should anneal soak for the hotter glass and terminate anneal cool below the strain point for the cooler glass. 

The indicated kiln temperature at which the anneal soak is initiated can be rather arbitrary. It shouldn’t be too far below the annealing point given by the glass maker, otherwise the soak time will need be extended to compensate. It certainly won’t hurt if it is higher.

It makes good sense to put the bottom value of anneal cool below strain point, to allow for temperature gradients across the glass. How far below can also depend on the cooling rate of the kiln. 

Note: As makers recommendations will usually refer to firing their certified compatible glass range, one can’t go very far wrong in following them: certainly until one works out where they have been overly conservative. 

Cooling rate

The cooling rate is set purposely to avoid building in too high a temperature gradient across the glass. The thermal conductivity of all soda lime glass is similar, so no great concession need be made for the colour or mix of pieces (as distinct from heating rates where dark glass absorbs heat faster than light colours, or clear).

The rate at which the glass loses heat, is determined by the rate at which heat flows out of the kiln to the atmosphere, so the loss from the glass should be uniform over the entire exposed surface.

If you regularly fire different types of glass whose makers specify different ‘annealing points’ it isn't necessary to have different firing patterns for each of them.Rather. design a pattern which has regard to the different values and which offers a sensible compromise. 

It’s a good idea to keep changes to firing patterns to a minimum, as any change is an opportunity for human error.

From the information detailed above it is possible to select values for anneal soak temperature and time, and for anneal cool rate, which will suit more than one of the glass types you are firing.

It is also fairly obvious that personal choice, belief or company policy all play a significant role in what is recommended for annealing; Keep in mind that its also your responsibility to interpret the advice provided by the manufacturer.