A power point presentation I made a few months ago to the group Lunch with a Glass Artist.
It is 33 slides long.
A power point presentation I made a few months ago to the group Lunch with a Glass Artist.
It is 33 slides long.
Let’s think about moisture and large bubbles from under the glass. It is not the water, but the gasses created by the decomposition of materials that can cause the bubbles. There are other causes of large bubbles too. The most common causes are discussed here.
The usual explanations are:
Shallow depressions in shelves can cause large bubbles. Occasionally,
the shelf can be damaged in various ways causing scratches or dings in the
shelf. Air can be trapped in these depressions. And it does not take much
volume of trapped to be a problem. The heat of kilnforming causes the air to expand.
As the glass becomes less viscous with increased temperature, the pressure from
the expanding air forces the glass upwards. The amount of air and the amount of
heat work combine to create bubbles from simple uprisings to large thin walled
or even burst bubbles.
There are some things that can be done to detect and avoid
bubbles from forming. It is possible to screed powdered kiln wash over kiln
washed shelf. This gives pathways for the air to escape. It does leave a more
marked bottom surface than kiln wash.
Using 1mm or 2mm fibre paper allows air from under glass. You can maintain a relatively smooth surface with Papyros or Thinfire over the fibre. Even Thinfire or Papyros on its own will allow air from under the glass.
Checking for depressions can be done by spreading kiln wash
powder over shelf and drawing a straight edge over the shelf. Depressions will
be shown by the presence of the powder. It can also be done with powdered glass
frit.
Any particle resistant to kilnforming temperatures holds the
glass up while it is forming so creating an air space. It is important to
ensure the shelf is clean as well as flat. Small pieces of grit or dirt that
are resistant to high temperatures will hold the glass up from the shelf enough
to create a bubble – small or large depending on the temperature. Vacuuming the
shelf before adding anything to the surface before each firing is important to
bubble free results.
This is sometimes cited as a cause of bubbles. If so, the heat would need to be very localised. This is possible if the glass is very near elements. In general, the temperature is equalised at a distance equal to the width of the elements.
A wide variety of
glues are used in kilnforming. Those available to enthusiasts all burn away
leaving gasses between layers. These gasses - if trapped - can thin the glass
below as well as above the glue’s position. This will give the impression that
the bubble has come from between the shelf and the glass. Most often the bubble
forms between the glass layers, pushing a bubble only into or through the top
layer. The solution is to avoid using glue or minimise it and place it only at
the edges.
Organic materials can be a problem. When you are using a
large or thick fibre paper sheet under a piece of glass, occasionally the
gasses from burning out of the binder can be great enough to create a bubble. Although
normally, this only leaves a grey to black mark on the underside of the glass.
Vermiculite boards need to be fired before use, as they contain significant
amounts of binder.
Inclusion of organic materials such as leaves, twigs, or bones,
leads to bubbles. Very long soaks below the softening point of the glass are
required to allow the organic material to burn out of the objects. The time required increases from an hour for
leaves to 24 for bones.
Moisture is very often cited as the source of bubbles. It is
possible that the steam from water may be trapped in shelf depressions, or the
areas held up from the shelf. And anytime there are no precautions to allow the
air from under the glass, or between sheets bubble formation can be promoted.
If adequate precautions are taken (flat shelf, clean shelf, bubble squeeze) the
moisture will evaporate before the glass is hot enough to form a seal around
the edges and trap any steam. It is another good reason for moderate ramp rates
at the beginning of a firing.
Of course, if there is a lot of moisture there can be
problems. Simply applying kiln wash in four coats does not leave enough water
in the shelf to be a problem.
If you have washed the kiln wash off a mullite shelf, there
will be a lot of water in it even after it feels dry. Then it does need to be
kiln dried before use. To avoid breaking the shelf you need to fire slowly to 99°C/210°F
and soak there for a couple of hours with the vents open or lid propped up a
little to allow the moisture out of the kiln.
Texture moulds are a form of bas relief in reverse. The texture of the
mould is the bas relief. The glass formed over the shapes is in negative relief.
The light is refracted through th
e back to give an image of bas relief although the surface is smooth.
Example of wave form texture mould |
The usual temperatures are too high. These moulds are an exercise in patience. The temptation is to fire higher than slumping temperatures to get good definition in the glass. However, a number of problems, especially bubbles, can be avoided by staying at the high end of slumping temperatures. This means the top temperature would be about 680C. To compensate for this low temperature, the soak needs to be three hours or more. To be sure the definition desired has been achieved, peeking near the end of this long soak is necessary.
Moulds that are produced with a rim around the edges can trap air and
create bubbles. The rim forms a perimeter dam to confine the glass. If the rate
of rise is quick to a high temperature, the edges can be sealed against the rim
before all the air has escaped. It is advisable to cut the glass for these
rimmed moulds a bit smaller than the internal dimensions formed by the rim.
Example of textured area surrounded by a rim |
Use of single
layers on texture moulds can lead to large, thin bubbles. This is most
prevalent when using high temperatures. Since the single layers tend to form
more slowly than an already fused two-layer piece, the temptation is to use
higher temperatures. The higher temperatures soften the glass to such an extent
that often bubbles form over the lower areas of the mould. Instead, low
temperatures with extremely long soaks should be used to allow the glass to conform to the undulations of the texture without dog boning or developing bubbles.
Of course, peeking will be required to determine when the texture is achieved. With
single layers, the surface will have greater undulations than with two layers.
The thinness of the single layer cannot fill the depression the way two layers
can.
Rapid rates to high temperatures can produce internal bubbles too. These
moulds have a multiplicity of hollows and depressions. Just as people are
warned about depressions in their shelves, the depressions in the texture
moulds can cause bubbles too. This means there are even more possibilities for
bubble creation than on apparently flat shelves. Long slow bubble squeezes are
required to allow air from under the glass.
Glass sometimes sticks to the mould. This is most often blamed on insufficient separator. Boron nitride is a good separator for these moulds especially if you go to tack fusing temperatures. At slumping temperatures, kiln wash will normally be sufficient. Both of these separators need to be applied carefully, as there are relatively steep slopes throughout the mould. Spraying needs to be done from at least four angles to ensure all the sides are covered.
Less often thought about is the draft of the shapes of the mould. If the slopes (draft) in the mould are too steep, the glass will “grab” the ceramic mould, because the ceramic contracts less than the glass when cooling. If shapes of the mould are steep and deep enough, the glass may even break as a result of this compression of the mould.
An example of some nearly vertical elements and a rim |
Other information is available:
Low Temperature Kilnforming, an Evidence-Based Approach to Scheduling, an ebook
In kiln forming, the moisture resulting from recently applied kiln wash is considered by some to be a cause of bubbles. The water in the mould will be evaporated by around 250°C/482°F in any sensible slumping schedule. At this temperature, the glass will not have begun to move, so the moisture can move out of the mould through any vent holes at the bottom of the mould, or past the glass as it rests on the edge of the mould.
The circumstance when a damp slumping mould could cause difficulties is when using an extremely fast rise of temperature. This is detrimental to the mould also, as the rapid formation of steam is more likely to break the mould rather than the glass. It is also unlikely to result in a good slump conforming to the mould without significant marking.
In casting with wet plaster/silica moulds water vapour can move toward the glass. Casting practice has alleviated some of the problem, by having an extended steam out before 200°C/395°F, or pouring the glass into the hot dry mould from a reservoir.
In pate de verre, the mould is most often packed while wet. The small particles normally allow any steaming of moisture to pass through, and so be dry at forming temperatures without blowing any bubbles.
Bubbles at the
bottom of the glass are much more likely to be the result of too high a process
temperature if the previous two conditions are met. This high temperature allows
the glass to slide down the mould. The glass is not plastic enough to
thicken and form a puddle at the bottom at most slumping temperatures. Instead,
it begins to be pushed up from the lowest point due to the weight of the glass
sliding down the sides.
Ensure that the moulds are no more than damp before placing
in the kiln.
Firing for too long
or at too high a temperature will cause the glass to continue sliding down.
Having nowhere else to go, the bottom begins rising. This is the result of
the weight of glass pressing down onto the bottom, especially on steep-sided
moulds. This is a consistent experience across several kilns and with multiple
users.
Glass at low temperatures is affected largely by its weight and viscosity.
Thick glass will fall more slowly than thin, when using the same schedule. Thick glass takes longer to equalise the upper and lower surface temperatures. Since the lower surface is stiffer (has a higher viscosity) it will move less using the same heat up rate. This means slower rates should be used, or a significant soak just above the strain point will be required. This softening of the glass evenly throughout the rise to the top temperature is critical in obtaining even slumps.
It is possible to have glass slightly overhang slumping moulds if you use low temperatures. The glass has the appearance of behaving differently at these low temperatures than at fusing temperatures.
At low temperatures it cannot form exactly to the mould. It falls first in the middle. Because the glass is not very plastic, the edges rise up from the mould at first, because the weight there is not great enough to allow the unsupported glass to bend. The edges stay in line with the beginning of the bend in the middle.
At the beginning of
the slump the glass is not soft enough to stretch. It maintains its dimensions
as it falls. For deep moulds, the glass moves progressively to move over the
lip of the mould and begins to fall into the mould.
As the slump proceeds, the glass stretches very little and so the edges move further down the mould. The glass continues to slide down at the edges until the centre settles down onto the mould bottom.
During this slide
into place, the glass can become marked. This is usually most evident on back
of the upper portions of the glass where most sliding is happening.
During the sliding of the glass along the mould, it becomes more marked. The marks often look like stretch marks. And in many senses, it is exactly that.
At higher temperatures or longer holds, the glass softens more. At this point the uprising collapses and the glass begins to thicken at the bottom. It also thins slightly at the top.
The ramp rates
should be slow.
Using the lowest
practical slumping temperature gives the best results.
More information is available here.
This information shows you need to keep the slumping temperature to the minimum required. To find out what that temperature is, watch the slumping in stages in brief peeks (do not stare!). Look at the piece for a second or two every five minutes before you reach your desired temperature and at intervals throughout the hold.
If it has slumped completely at the beginning of the hold, you are firing too high. Reduce your temperature in subsequent firings and watch in the same way to find what the required temperature and time is. There is absolutely no substitute in slumping but to watch by peeking to learn what your mould and glass require.
To determine the temperature needed for your piece, use slow ramp rates – between 100°C to 150°C/ 180°F to 270°F. Set your top temperature around 630°C/1170°F for a simple slump of fusing glass. For bottle or window glass you will need a temperature closer to 720°C/1330°F.
It is necessary to observe the progress of the slump as you do not know the best slumping temperature. Start watching the glass at about 10-minute intervals from about 600°C/1110°F. There is not much light in the kiln at this temperature, so an external light is useful. You can also observe the reflections of the elements on the glass. When the image of the elements begins to curve, you know the glass is beginning to bend. You then know that is the lowest possible slumping temperature when using that ramp rate.
Hold for at least 30 mins at the temperature when the glass begins to visibly drop. This may or may not be long enough. Continue checking at 5-10 minute intervals to know when the slump is complete. If the glass is completely slumped before the soak time is finished, advance to the next segment. If not fully slumped, you need to extend the soak time. These operations mean you need to know how to alter your schedule while firing. Consult your controller manual to learn how to do these things. Stop the hold when complete and advance to the anneal.
In some cases, you may need to increase temperature you set by 5-10°C. You can do this by scheduling a couple of segments with 10°C/18°F higher temperature each and 30 minute soaks each. If you do not need them, you can skip them. If you do need the extra temperature, you have it scheduled already. You will know if you need the extra segments by whether the glass has begun to curve at the start of the first of the soaks. If it has not after 10 minutes, skip to the next segment. Once the new temperature has been reached, check for a curve in the glass. Again, if after 10 minutes there is no curve, skip to the next (higher temperature) segment.
A low temperature slump will allow the glass to conform to the shape of the mould without softening so much that it takes up all the markings of the mould. That in turn means there are spaces for the air to escape from under the glass all the way to the slumping temperature as well as through the air holes at the bottom. It also gives the most mark-free slump possible for your shape.
If you are slumping at such a temperature that the glass has sealed to the mould, you are firing too hot anyway. Or put more positively, use a low temperature slump, that is, a slump at the lowest temperature to achieve the desired result over an extended period of your choice.
More information is available in the eBook Low Temperature Kilnforming available through Etsy or Bullseye.
“I sometimes slump at the same time as I do a tack fuse. Is slumping at this higher heat bad for the mould? “
Image credit: Creative Glass |
Mould
It is possibly not bad for the mould, but it does depend on your temperature and heat work. Ceramic moulds are typically fired to 1200° or 1300°C so higher kilnforming temperatures are unlikely to affect the moulds. The speed at which the target temperature is reached is of concern though. Ceramics have what is called quartz inversions.
Two of the constituents of ceramics – cristobalite and quartz – have significantly large expansions at 226°C and 570°C / 440°F and 1060°F. Rapid rises through these two temperatures risks breaking the ceramic mould. This is not the case with steel moulds, of course.
Any suggestions on how to avoid getting the oblong bubble under the neck of the bottle? This was my first try and I’m really happy with clarity, no devitrification in these.
I used this schedule:
Fahrenheit Celsius
300/1150/30 167/620/30
200/1370/20 111/740/20
400/1450/20 222/787/20
AFAP/950/60 AFAP/510/60
150/800/0 63/427/0
300/100 167/55/off
The bubble is kind of cool but not sure what it will do when I put it in a bottle mould.
To minimise the bubble, you need a bubble squeeze. There isn't one of sufficient length or at the right temperature in the schedule. The softening point of bottle glass is approximately 720C. Starting the bubble squeeze at ca. 670C/1240F and progressing slowly (ca.50/90F or less) to 720C/1340F may give a better bubble squeeze.
Also, the anneal soak is a bit low. Bottle glass and float glass both have annealing points of about 550C. You might make use of a lower annealing soak temperature to reduce the cooling time. It is usually possible to anneal 30C below the published annealing temperature. In this case that would be 520C.
There is pretty thick glass in some places due to the way the bottom and neck of the bottle form. You may want to extend your anneal soak to one for 12mm/0.5”. The soak time for this is 2 hours. The first cooling segment would be 55C/100F per hour to 475C/888F if you use 520C/970F as the annealing soak. The second cool segment should be at 99C/180F per hour to 420C/790F. And the final rate at 330C/600F to room temperature. It is important to include all three stages of cooling. The research for my book Low Temperature Kilnforming (Or directly from stephen.richard43@gmail.com) has shown that to get the best stress-free results use all three stages of cooling.
Bubbles at the shoulder of the bottle are common. The change in circumference of the bottle at the shoulder means there is a greater amount of glass to “compress”. Bottles with tapered circumference at the top of the bottle have fewer problems with creating bubbles. The abrupt change in size at the shoulder causes bubbles to be more common. A long slow bubble squeeze will allow the shoulder to form more closely in line with the neck.
There are other things you can do to
help avoid the bubbles. One thing is to insert a thin kiln washed wire into the
neck of the bottle. This gives a path for the air to escape and allows you to
pull it out, although a mark will be left. You could also think of
drilling a hole in what will be the underside at the shoulder to allow air out
to the shelf. It does not need to be a big hole.
Bubbles at the shoulder of a slumped bottle are a common problem. It results from the greater amount of glass that has to slump into the space. This leaves a cavity. Slower bubble squeezes can help, as well as various venting methods.