Heat Up Events

This is based on Graham Stone’s work with float glass. The temperatures are applicable to float glass, and so need to be adjusted for any other glass, but illustrate the principle of how heating temperatures affect the glass. Temperatures in degrees Celsius.


10-250 Slow rate heating up. Risk of thermal shock. Venting often done in this phase.

250-500 Medium rate heating. Risk of thermal shock diminishing.

400 + Many glasses now tolerate fast heating up ramp rate.

550 Glass surface beginning to soften slightly

600 Safe from thermal shock above this temperature

610 Glass bending slightly, picking up texture.

680 Glass begins to stick to itself. Tin bloom becomes iridescent.

690 Fusing glasses reaching their softening points.

715 Glass beginning to stretch. Tack-fired pieces adhered by now.

720 Subtle devitrification and iridisation burn off becoming a factor with some glasses.

730 Softening point of float.

750 Edges no longer sharp. Tin bloom stretching becoming "frosty".

760 Tack fuse range for fusing glasses.

770 Float glass fused, but still "sitting up".

790 Trapped air can cause bubbles under sheet glass at this temperature.

800 Full fuse for most fusing glasses.

820 Fused float glass nearly flat.

825 Full fuse for float glass. Devitrification more pronounced.

850 Glass flowing.

950 Glass soft enough to "rake".

1000 Approximate liquidus temperature.



Based on Firing Schedules for Glass; the Kiln Companion, by Graham Stone, Melbourne, 2000, ISBN 0-646-39733-8, p24
Post revised 5th March 2014

Stretch Marks in Slumping

Occasionally a slumped piece will develop faint lines beginning about half to two-thirds of the way from the centre and radiating toward the edge.

My experience leads me to think that these marks come from the glass moving too quickly at too hot a temperature. The glass softens as it reaches its slump point. If the temperature is taken above that, the glass conforms to the mould and then begins to slide downwards. The mould is by its nature not perfectly smooth and so the high points make marks on the glass as it moves.

This is re-enforced by the fact that the glass at the centre of these slumps does not have those marks. It deforms less than the edges of the piece and so (at whatever temperature) does not get so marked as the sides and edges.

To avoid these stretch marks you need to slump at the lowest possible temperature and ensure the glass is the same temperature throughout by the time it gets to its slumping point.

Temperature

Finding the lowest temperature for the slumps in a particular mould requires experimentation and observation. A simple curve – circular, oval or rectangular – requires less heat than one with a flat bottom and much less than one with angles. For a simple curve you can set your slumping temperature at say 620ºC with up to an hour soak. The important element to remember is that each shape and curve of mould will require different schedules. To determine this you need to make observations.

The glass for these two moulds requires different temperatures or  schedules. The back one will conform to the mould at a lower temperature than the front one due to the simpler shape and larger span of the back one.

From about 600ºC you need to make periodic observations of the progress of the slump. Note the temperature at which the glass begins to move – the reflections in the glass will begin to be curved. This is the minimum temperature you can use for this span and thickness of glass on this mould. The length of time required to get a complete slump may be so long as to make using this temperature impractical.

Slump not quite complete

Now observations need to become more frequent – possibly every 10 minutes or less. When you reach a temperature where the glass is visibly distorting, it is time to cease the temperature advance and begin the soak. Record this temperature and continue to observe, recording the time it takes at this temperature to fully slump. Continue to the anneal.

Inspect the piece when cool. If you have the result you want, you have the temperature and soak time needed for this thicknesses and size of glass on this mould. Record this information. If it is not fully slumped you can try either extending the time (if that is practical, it is the best option) or increasing the temperature on another piece. This increase should be by no more than 10ºC, so that you do not over fire the piece.

Glass conforms to the bottom of the mould

Of course, it is possible that the piece was slumped at too high a temperature as evidenced by stretch marks, mould marks, uprisings in the centre, distortions on the edges. Then you need to reduce the temperature on the next slumping of a piece of the same dimensions. Start with 10ºC less than your first piece, and programme the same amount of time. Observe, record and inspect as on the previous one.

This process shows why it is important to have a kiln with observation ports to be able to follow the progress of your work. In some ways, it is more important to have observation ports than whether the kiln is front or top loading, coffin or clam shell opening. But that is by the way.

Heat

The second important element in avoiding stretch marks is to enable the glass to be at the same temperature throughout its thickness. This involves the concept of heat work.  In general terms it means you can achieve the same result by putting the heat in fast and at a high temperature or slowly and at a low temperature. The “slow and low” approach allows more control and allows the glass to be the same temperature on top as on the bottom.

It is important to heat the glass slowly and steadily all the way up to the slumping temperature. The temptation to increase the temperature rapidly after the strain point needs to be resisted. Getting the top too hot can at the worst, cause a split on the bottom of the glass as the tension from slumping glass on the top splits the stiff glass at the bottom.

This means there is no need for a soak at the strain point, nor a speed up in the rate of advance up to the slumping temperature. Exactly the opposite is indicated. Choose a rate of advance for the glass according to its thickness – at 6mm a rate of 150ºC will be adequate. Maintain that rate of advance all the way up to the slump temperature. This also is required when you are making observations to determine what the slump temperature should be. The moderate rate of advance all the way to slumping temperature ensures the whole thickness of the glass is at the same temperature.

Heating the glass slowly to enable all of it to be at the same temperature, allows the glass to change shape at the lowest possible temperature and avoid picking up so much of the mould texture. The glass at the edge and upper sides is in contact with mould longer than central parts as it changes shape and slides along the surface of the mould at elevated temperatures. The lower the temperature used with a long soak, means that the glass is less likely to slide along the mould and so adds to the avoidance of stretch marks.

Tack Fusing Considerations

1 – Initial Rate of Advance

Tack fuses look easier than full fusing, but they are really one of the most difficult types of kiln forming. Tack fusing requires much more care than full fusing.

On heat up, the pieces on top shade the heat from the base glass leading to uneven heating. So you need a slower heat up. You can get some assistance in determining this by looking at what the annealing cool rate for the piece is. A very conservative approach is needed when you have a number of pieces stacked over the base layer.  One way of thinking about this is to set your initial rate of advance at approximately twice the anneal cool rate. More information on this is given in this entry. 



2 – Annealing 

The tacked glass can be considered to be laminated rather than fully formed together. This means the glass sheets are still able, partially, to act  as separate entities. So excellent annealing is required.

Glass contracts when it's cooling, and so tends to pull into itself. In a flat, symmetrical fuse this isn't much of a problem. In tack fuses where the glass components are still distinct from their neighbours, they will try to shrink into themselves and away from each other. If there is not enough time for the glass to settle into balance, a lot of stress will be locked into the piece that either cause it to crack on cool down or to be remarkably fragile after firing. In addition, in tack fusing there are very uneven thicknesses meaning it is hard to maintain equal temperatures across the glass. The tack fused pieces shield the heat from the base, leading to localised hot spots on cool down.

On very difficult tack fuses it's not unusual to anneal for a thickness of four to six times greater than the actual maximum thickness of the glass. That extended cool helps ensure that the glass has time to shift and relax as it's becoming stiffer, and also helps keep the temperature more even throughout.

So in general, tack fused pieces should be annealed as though they are thicker pieces. Recommendations range from the rate for glass that is one thickness greater to at least twice the maximum thickness – including the tacked elements – of the whole item. Where there are right angles - squares, rectangles - or more acutely angled shapes, even more time in the annealing cool is required, possibly up to 5 times the total thickness of the piece.

It must be remembered especially in tack fusing, that annealing is much more than the annealing soak. The soak is to ensure all the glass is at the same temperature. The anneal cool over the next 110ºC is to ensure this piece of different thicknesses will all react together. That means tack fusing takes a lot longer than regular fussing.

3 – Effects of thicknesses, shapes, degree of tack

The more rectangular or pointed the pieces there are in the piece, the greater the care in annealing is required. How you decide on the schedule to use varies. Some go up two or even four times the total thickness of the piece to choose a firing schedule.

A simplistic estimation of the schedule required is to subtract the difference between the thickest and the thinnest part of the piece and add that number to the thickest part. If you have a 3mm section and a 12mm section, the difference is 9mm. So add 9 to 12 and get 17mm that needs to be annealed for. This thickness applies to the heat up section as well.

Another way to estimate the schedule required is to increase the length the annealing schedule for any and each of the following factors:

·         Tack fusing of a single additional layer on a six millimetre base

·         Rectangular pieces to be tack fused

·         Sharp, pointed pieces to be tack fused

·         Multiple layers to be tack fused

·         Degree of tack – the closer to lamination, the more time required

The annealing schedule to be considered is the one for at least the next step up in thickness for each of the factors. If you have all five factors the annealing schedule that should be used is one for at least 21mm thick pieces according to this way of thinking about the firing.

4 – Testing/Experimentation

The only way you will have certainty about which to schedule to choose is to make up a piece of the configuration you intend, but in clear. You can then check for the stresses. If you have chosen twice the thickness, and stress is showing, you need to try 3 times the thickness, etc. So your annealing soak needs to be longer, if stress shows. You can speed things by having your annealing soak at the lower end of the annealing range (for Bullseye this is 482C, rather than 516C).

You will need to do some experimentation on what works best for you. You also need to have a pair of polarisation filters to help you with determining whether you have any stress in your piece or not. If your piece is to be in opaque glasses, you need to do a mock up in clear.

Glass Shifting on Mould

There are a number of things to investigate if your blank is shifting on the mould during firing.

Is there a heat differential?

Glass absorbs heat at different rates depending on colour and type meaning that one part may begin to move before another. The solution to this is to slow down the rate of advance to allow all the glass to gain heat at the same speed. It may also be useful to slump at a lower temperature.

There also may be a heat differential within the kiln. You need to run a check on the heat distribution of your kiln to be sure where the (relatively) hot and cold areas of your kiln are. Bullseye published Tech Note no.1 on how to do this.

Not perfectly balanced on the mould?

Glass can be placed just off square or level and that can allow it to start slumping unevenly. Measurements and observation can help to get the glass placed squarely on the mould. Also a small spirit level placed on the glass can tell you if the glass is level within the mould.

The mould may not be level.

The kiln, shelf and mould should each be checked for level in all directions. The kiln level can be established and can be assumed to be level until it is moved. The shelf level should be checked each time it is moved. The mould level should be checked each time it is used.

Is the glass overhanging the mould?

Glass overhanging the mould rim can hang up on some of the edges more than others. Check the rim of the mould for any rough areas and smooth them. If you do have glass overhanging, you should slow the rate of advance to allow the edge of the glass to tip up and begin to slide down into the mould. If the problem persists, make the glass blank smaller, or support the overhanging glass with a collar.

Is the glass heavier on one side?

The glass may be uneven thickness and so heavier on one side. The heavier area of the glass will begin to slump first and so promote movement of the whole glass in an asymmetrical manner. The solution to this is to fire slower and to a lower temperature.

Do you have a wonky mould?

The mould can be imperfect. So you need to check the mould for accuracy. I have a slumper that has one side lower than the other three. Being aware of this, I can place the glass so that it is still useable. Measuring the mould in all directions will help determine its symmetry.

If all these things have been investigated and the solution not found, it is possible to create a bevel on the bottom edge of the glass so that the edge sits in the mould at the same angle as the mould. This provides a larger contact point for the glass and mould than just a thin edge. This appears to allow the glass to move evenly during the slump.

Of course, a major solution is to observe the slump.  Peeking into the kiln at the beginning of the slump soak and frequent intervals after that will show if the piece is slumping evenly or not.  If it is uneven, you can put on the appropriate protective gear and with gloves on your hands, shift the glass to be set evenly in the mould.

The major solutions to avoid uneven slumping are:

• Avoiding the hot and cool parts of the kiln
• Making everything level
• Careful placement on the mould
• Slower rates of advance
• Lower slumping temperatures
• Observation

Shape of Aperture Drops

The shape of an aperture drop can be controlled by the speed of the slump. The speed at which the glass drops is a combination of heat and size of the hole. Patience is required.

Rapid drops result from high temperatures. Rapid slumps cause a thinning of the glass at the shoulder where the glass turns over the inner rim of the aperture. The pattern is distorted and the colours are also diluted. And a relatively large rim is left around the fired piece.

A much slower rate of drop spreads the strain of the slump over the whole of the unsupported area of glass. This tends toward a bowl with a gentle slope toward the bottom, reduced distortion of the pattern, maintenance of the colour densities, and a more even wall thickness all over the piece.

The slumping temperature for a shallow angled slump is less than that used for normal slumps, and takes a lot longer – up to five hours typically. This means that observation is required at intervals, say every half hour.

A starting point for the slumping is around 100ºC above the annealing temperature for the glass. So for Bullseye and System 96 the temperature is about 615ºC. If after the first half hour, there is no movement, increase the temperature by 10ºC. Check again in another half hour and if the slump has begun, leave the temperature at that level and observe at the half hourly intervals until the desired slump is achieved. Otherwise, increase the temperature by another 10ºC with the check after half an hour, and repeat until the slump has begun. After you have done the first one of these with a particular size of aperture, you will know the temperature to start the slump.

The temperature you need to use is affected by the size of the hole. The smaller the aperture, the higher the temperature will be needed. But be patient. If the temperature is increased too much, you will get the thinning of the sides that you are trying to avoid.

Additional information on aperture drops can be found in this series.

Temperature Rise Rates

I am always concerned when people recommend soaks on the way up in order to equalise temperatures. If the soak is required because the ramp rate is too fast, there are breakages going to happen sometime - maybe not now, maybe not tomorrow, but certainly sometime. If you need that extra time, add it into the schedule. E.g., a ramp rate of 200C from 20C to 520C with a 20 min soak could also be written as 176C/hr from 20C to 520C - both take 2.833 hours to achieve the same temperature. A controlled heating rate is preferable to one or more rapid rates with soaks.

I am also concerned about very rapid temperature rises after the bubble squeeze. The controllers often cannot adequately control such rapid rises. The rapid rise also often requires a higher target temperature to achieve the desired effect. This can mean that it is easier for bubbles - large and small - to form and rise to the surface during the overshoot of the target temperature. Temperature increases are about heat work - the combination of temperature and time. This means that you can achieve the desired result in two ways:

1- fast rise to high temperature or

2- Slow rise to lower temperature.

The second strategy may also require a longer soak at the target temperature than the one with a fast rise to a high temperature.

The aim in kiln work should be to achieve the effect you want at the lowest practical temperature. This is because glasses tend to change their characteristics more at higher temperatures than at lower temperatures.

Cooling Events

This is based on Graham Stone’s work with float glass. The temperatures are applicable to float glass, and so need to be adjusted for a particular glass, but illustrate the principle of how heating temperatures affect the glass. Temperatures in degrees Celsius.


600 Common temperature for crash cooling toward. Glass beginning to "freeze".

555 Annealing temperature of float. Bungs in.

515 Approximate Strain Point of float.

535-400 Critical slow cooling down phase for float that overlaps annealing range.

400-300 Medium cooling down ramp rate.

300-10 Fast cooling down ramp rate. Cracking the kiln open possible.



Based on Firing Schedules for Glass; the Kiln Compainion, by Graham Stone, Melbourne, 2000, ISBN 0-646-39733-8, p24

Viscosity Changes with Temperature

This is based on Graham Stone’s work with float glass. The temperatures are applicable to float glass, but illustrate the principle of how viscosity changes in a non linear pattern with the increase in temperature. Temperatures are in degrees Celsius.

515 Viscosity 10145 poises (approximate strain point of float)
555 Viscosity 1013 poises
610 Viscosity 1010 poises
730 Viscosity 976 poises
850 Viscosity decreasing faster
900 Viscosity now 105 poises and falling

Based on Firing Schedules for Glass; the Kiln Companion, by Graham Stone, Melbourne, 2000, ISBN 0-646-39733-8, p24.

This shows that viscosity changes rapidly from the lower strain point (the solidification of glass) to annealing.  The change slows from the annealing point to full fusing, but changes rapidly after that.  This is an important factor to control in casting and free drops.

What is viscosity
Graph of the changes