Diagnosis of Fractures

Knowing what has happened to your piece when it is broken or cracked is important to developing your skills as a kilnformer.  Most of the knowledge about diagnosis comes from looking carefully at the cracks and the shapes apparent in the flawed piece.

Breaks in the Kiln

Breaks in fusing at tack or full fusing levels in the kiln are generally of four kinds.

Breaks with hooked ends

Breaks that go across the whole piece, with a hook or significant curve at each end, usually indicate an annealing problem. The slight hook seems to result from inadequate annealing. The break will have sharp edges as it occurs as the glass is entering the brittle stage.

Multiple breaks in a crazed pattern

Crazed glass – similar to the cracks in ceramic glazes - usually indicates the glass has stuck to the supporting materials. These materials can be shelves or moulds. It is a sign there was not enough separator present between the two surfaces.

Breaks following the edge of glass pieces

Breaks that skirt around colours or pieces of glass almost always indicate a compatibility problem with the glass pieces chosen.  In severe cases the crack will be all around the incompatible pieces of glass as though it is trying to escape the base layer.  Sometimes the break will be from side to side, but skirting the incompatible glass.  These breaks will have sharp edges as the compatibility problem only becomes apparent on the cool.

Breaks from side to side following the line of glass pieces is not an infallible indicator of incompatibility, though.  Glass which has varying levels or thicknesses can break alongside the thicker areas, even though the glass is compatible. Often the break will be rounded due to temperature differentials in the glass on the heat up.  As the glass continues to get hotter, glass pieces on top - or strongly contrasting colours - can heat as such different rates that the stress overcomes the strength of the glass.

Of course, this kind of break can be sharp because the break occurred during cooling.  In effect, this appears to be an annealing problem when it really is a problem in matching the scheduling with the annealing requirements of a complex piece.  You need much longer soaks and slower cooling on tack fused pieces than on flat fused ones.

These two contrasting causes of a break means that you need to think about how the glass is layered.  One is to do with compatibility and the other to inadequate annealing due to the complexities of the layup.  They also tie up with the fourth cause of breaks.

Breaks can also follow the edges of inclusions.  This of course, indicates incompatibility.  All metals are incompatible, but if thin and not excessively large in relation to the piece, the glass is strong enough to contain the stress.  When the metal or other inclusion is too large, strong, or thick, the glass will break or show cracks around the inclusions.

Broken and separated lower layers

Sometimes people will open the kiln to find the lower layer of a multi-layer piece has broken and separated a small distance.  This is the fourth kind of break. This break will most often be a nearly straight break from edge to edge.  The broken edge will be rounded but the top layer(s) will have the expected profile.   This is an indication that the heat up was too fast not allowing the lower layer to achieve the same temperature as the top. 

This most often happens where there is an exposed lower layer (which gets hot) along with areas on top that get equally hot, but not the glass underneath.  Glass is a poor conductor of heat, so the upper layers "shade" the heat from the glass below.  The temperature difference between the two can be great enough to break the base glass apart but leave the top intact.  You know this was on the heat up because the layers of glass could move independently when the base broke and moved under the upper layers.  The glass was not hot enough to be sticky yet, so it had not reached lamination temperatures before the break.

Rounded vs. sharp edges

In addition to the location of the breaks, the condition of the edges is important in diagnosis of the cause of the problem. The accepted rule is that rounded edges mean the break occurred during the heat up.  Sharp edges occur during the cooling.  This is most often the case (but see the conditions for slumping). For flat pieces breaks that occur on the heat up will be rounded to some extent.  In a full fuse, usually the edges of the break will be rounded similar to the outside edge.


Cracks on the bottom surface

Sometimes the broken pieces will recombine either partially or all along the line.  There may even be a full recombination leaving only a crack like appearance on the bottom.  This recombination also will be the case where there was where only a partial break or crack in the early stages of firing. It leaves a smooth top surface, but a visible crack on the bottom. That means there is only a marginal reduction required in the scheduling of the initial rate of advance, as the temperature differentials were not great enough to break the piece completely across.

Force of Breaks

The space between the broken pieces shows the relative force that caused the break.  Greater space is related to more stress; lesser space or only partial cracks indicate a lower amount of stress. The amount of space indicates the degree of change required in scheduling. A small parting of the glass requires only a little (maybe 10% - 15%) reduction in the rate of advance.  Large spaces indicate that much slower rates of advance are required, and possibly a complete rethink in the scheduling of the firing.

Slumping breaks

Breaks in slumps are usually caused by a too rapid rate of advance. But this is not always the case.  The usual check of a sharp or rounded edge to tell when the break occurred does not work well at slumping temperatures.  The edge will be sharp whether it occurred on the heat up or the cool down because the temperature is not high enough to significantly round the edges.  The test must be different on slumps than that of sharp edges.  The test is related to the shape of the pieces. Take the pieces out of the mould.  If you can fit them together exactly, the break occurred on the cool down.  This usually will mean the anneal soak was too short and the anneal cool too fast.

Most slumping breaks occur on the advance in temperature.  The means of determining when the break occurred can be tested by putting the broken pieces together.  If they do not match exactly, the break occurred during the heat up.  This is based on the observation that broken pieces separated slightly in the mould by the force of the break on the heat up, and so will slump in the mould in slightly different ways from each other due to their positions.

Remember the blank for slumping is thicker than the original un-fused pieces.  This thickness requires a slower heat up than the original blank consisting of separate pieces.  In addition, the glass is supported at the edges of the mould which can allow the central area of the glass to heat faster than the edges, so further slowing the rate of advance is required.  These two factors of thickness and supports explain most of the breaks during slumping.

Splits in slumps

Sometimes the upper surface of the slump appears fine.  It is the bottom that exhibits a split or tear that does not go all the way to the upper surface of the glass. This is similar to the cracks on the bottom of a flat piece described above. It indicates the rate of advance was too - but only just - too fast.  The rate of advance has been quick enough to get the top heated and become plastic. But the lower surface is still cold enough that it is brittle. The weight of the upper softened glass begins to push down before the bottom has become hot enough to be plastic.  The force of the weight of the upper portion of the glass can be enough to cause the glass to separate because it is brittle, rather than move as the surface does. This split on the bottom but not the top indicates a slightly slower rate of advance for the thickness of the glass is required.

Breaks out of the Kiln

Breaks after the piece is cool

Breaks that occur days, weeks, months after a piece is cool can be impact damage, annealing or compatibility problems. 

Impacts

Impact breaks will be obvious in handling or moving other pieces near to the affected piece.  Usually there is evidence of impact by a small chip removed from the glass at the origin. The piece may or may not have been stressed to allow an easy break rather than a chip.  It is not possible to be sure of the secondary cause after the primary impact damage has occurred.

Breaks in warm glass

If the break occurs shortly after having been removed from the warm kiln, it is probable that the thermal shock to the glass has a contributory factor to incompatibility or inadequate annealing.  The diagnosis of the cause is the same as for breaks in the kiln - hooked for annealing and straight or following colours or inclusions for compatibility.

Breaks in cold glass

If the glass has been sitting undisturbed in a shaded place and suddenly breaks, the reason can be there was an incompatibility or that the annealing was inadequate.  There usually is not much difference in the breaks in a piece that has been cold for a long time.  If the break distinctly follows colours or pieces of glass, that would indicate a compatibility problem.  If the break crosses colours and thicknesses it is more likely to be an annealing issue.  But, as you can see, there is no certainty in this distinction as to the causes of breaks a considerable time after removing from the kiln.

Glass in strong light

Glass placed in strong sunlight that breaks can be incompatibility or simply contrasting colours being heated unevenly by the sunlight.  It is difficult to tell with certainty whether it is compatibility, annealing, or heat differentials that have caused the breakage.

Problem Solving

The essential purpose of problem solving is to prevent the same thing happening again. To solve the breakage problem, you need to think about the interrelationships between the various parameters – firing rates, soaks, cooling rates; and the ways in which the glass was set up.

Rounded edges

If the break is shown to be in the early stages of the firing, they most generally are caused by thermal shock.  They will generally be straight on an evenly thick piece.  If the piece is with variations in thicknesses, the line of the break may follow the thicker pieces. In both cases, you need to think about the rates of advance you are using.  If the separation of the edges is small enough that they have begun to recombine later in the firing, the rate of advance was only a little too fast.  If there is considerable space – say more than a finger width – the rate of advance was significantly too fast.

Sometimes the condition of the upper glass can give an indication of when in the firing the break occurred.  On a first firing, if the upper piece has broken together with the lower one, the break occurred after the pieces became sticky. This would mean the break occurred at or higher than laminating temperatures.  This is rare during the heat up.

If the break has moved small top pieces, it indicates the break occurred early in the heat up.  Sometimes the break will occur under the top piece.  Later it slumps and fuses into the space created by the break.  This also indicates a break early in the firing.  All these conditions indicate that the initial rate of advance needs to be slowed to avoid the thermal shock.  It does not indicate that soaks should be added at various stages up to the softening point of the glass.  Glass generally behaves better with steady, gradual inputs of heat rather than quick rises with soaks (although there are exceptions).

Sharp edged breaks

These occur generally on the cool down or after the piece is out of the kiln for a while.  If the break has occurred in the kiln, you should look at it carefully before moving it.  The relative location of the pieces can tell you some things about why.

Crazed glass normally indicates the glass has stuck to the supporting material – shelf, moulds, or other rigid materials.  This crazing may all still be in one piece, or slightly separated, sharp edged chunks.  These effects indicate there was not enough, or appropriate, separator for the process used.

The distinction between annealing and compatibility breaks is given above. 

Breaks all around a piece or pieces – looking as though they were trying to escape the base - clearly indicate an incompatibility problem.  You need to identify that glass and separate your stock of it from the rest of your fusing glass. 

Cracks that skirt pieces of glass can be incompatibility.  This is easiest to determine on flat pieces which have been full fused, or nearly so.  There is not a variation in thickness to complicate matters.  In full fusing, if the break skirts around a piece or pieces of glass along its path, it is likely caused by incompatibility between pieces and their base.

Breaks skirting pieces can also indicate problems with thickness, especially in tack fusing.  The more angular the tack fusing is, or the greater the difference in thickness, the greater the potential for an annealing break.  The annealing soak for tack fusing needs to be significantly longer than for a flat fused piece of even thickness.  Recommendations vary, but the anneal soak time needs to be at least twice the thickest part.  The anneal cool rate also needs to be half that for the the thickest area.

Breaks or cracks across the piece with hooked ends indicate inadequate annealing.  This will require some consideration to come to the appropriate length of soak and rate of the anneal cooling.  The anneal soak is about getting all the glass to the same temperature - top to bottom, side to side.  The soak is about temperature equalisation not just annealing.   This is shown by the Bullseye research on annealing thick slabs.  They discovered that a longer soak at a lower temperature can provide as good a base for the anneal cool as a higher temperature. The differences are that the soak at the annealing point can be shorter, but the annealing cool is much longer.

Annealing continues below the anneal soak - whether you chose the annealing point or a temperature below.  Bullseye uses a temperature about 30C below the annealing point.  This can apply to any glass.  Because the glass is cooler, a longer temperature equalisation soak is needed. But the anneal cooling range is shorter, making for a reduction in cooling time for thick slabs.

The point of this discussion is that when considering the solution to annealing breaks, you need to have a relation between the temperature equalisation soak and the rate of the anneal cooling.  If you have decided you need a longer soak, then you also need to reduce the rate of the anneal cool.  If you do not, you will still have annealing breaks or even thermal shock breaks, even with long soaks at or below the annealing point.

Breaks of slumped pieces

Breaks in slumping almost always appear to be sharp edged, unless you look carefully at the edge.  Fitting the pieces back together will give an indication of when the break happened.  If they fit, the break occurred upon cooling.  The anneal may have been inadequate, or the cooling too fast.  Unfortunately, in a formed piece, the curved hook of an inadequately annealed piece does not often show up.

If the break occurred early in the firing, the piece may still have sharp edges, unless you were firing at the upper end of the slumping range.  Here again the test of trying to put all the pieces back together is important.  If the pieces do not fit exactly together, the break occurred during the heat up.  This will mean that you need to slow the rate of advance for subsequent pieces.

“It hasn’t happened before” Scenario

Often people experience breaks even though the set up was very similar and the schedule was the same over several pieces.  There are two responses to this – “what did you change for the firing of this piece that broke”, and “you have been skating on the edge of disaster for a while.”  Glass behaviour is predictable. Since the break occurred when the setup was very similar, and the schedule was the same, something has changed.

The first thing to do is to test for stress. This means test before the piece is broken, as once the piece has broken most, if not all, the stress has been relieved.  You will need to construct another piece in the same way as the successful or the broken one – whichever you prefer.  Test the flat fired piece for stress. Remember to include an annealing test, so you can determine if the stress is compatibility or annealing related.  If there is stress in the flat piece, but not in the annealing test, you need to consider whether all the glass is compatible, or you need to slow the annealing cool for the larger test piece.

Next you need to consider what was different.  Review the differences in set up of the piece – colours, arrangement, thickness, volume of material used – everything that might be different at each stage of the layup.  Note these differences and review them one by one.  Could have any one element been sufficient to make the firing conditions different?  Could a combination of these differences have been significant?

Are there any differences in the firing schedule?  Have you made any little tweaks in the schedule? What is different?  Different times of the day, different power supply, plugs in or out, venting, peeking, different shelves (or none) – any small thing that could have introduced a variable in the firing conditions.

Further information is available in the ebook Low Temperature Kiln Forming.

Conclusion

Although breaks generally have only three causes – thermal shock, incompatibility and inadequate annealing – the diagnosis of which it is and how it was promoted is complex.  All three are forms of stress.  To problem solve, first attempt to determine the type of stress that induced the break.  Then attempt to determine the cause of that stress.

It is important in the early stages of a new kind of piece, or early in your fusing career to test for stress after each firing (although I fail in this often).  This will give you the information to progress to the next firing or to revise the conditions – glass or schedules – to remove the stress for this or subsequent pieces.

Volume control

Glass has a surface tension (viscosity) that draws the glass toward 6-7 mm thick at kiln forming temperatures. 

To test this out, prepare three stacks of glass squares.  They all should be the same size.  Record the measurements. Place them in a stack of one, a stack of two and the last of three squares.  Fire them to a full fuse.  Compare the sizes of the original to the fired. Note the expanded size of the three-layer stack, the same size of the two-layer stack and the reduced footprint, and dog-boning of the single layer.

Credit: Paul Tarlow

Glass in a single layer behaves differently from the thicker set-ups. When the glass is hot it begins thickening at the edges. The viscosity of the glass is drawing from both from the edge and from the centre.  This means the footprint of the glass is getting smaller. The result is needling. The glass retreats leaving small threads where the glass was held in the small imperfections in the separator’s surface. 

If you do not need a full fuse, you can reduce this needling effect. Reduce the temperature and extend the soak.  This means that the glass does not expand on the heat up so much, and the greater viscosity reduces the needling effect.

If you need a thick piece of a certain size, you need to dam the glass to overcome the tendency to expand.  With experience, you can get to know how much a three-layer (or more) set up will expand and cut the glass accordingly.  In this way, you can often do without dams. There will be some thinning at the edges and a rounding of the corners.

An excellent document on volume control is the Bullseye Tech Note 5.  


Note that this 6mm rule applies at normal kilnforming temperatures.

At higher temperatures, the viscosity is less so the glass will become thinner than 6-7mm.  My experience has shown that at around 1200°C the glass will spread to about 0.5mm thickness.  This is just to point out there is a relationship between temperature and viscosity, and therefore thickness. As the temperature rises, so the viscosity reduces. This relationship allows the glass to become thinner.  At normal kilnforming temperatures, the 6mm rule applies, at higher temperatures it does not.

Further information is available in the e-book: Low Temperature Kilnforming.

Dog boning in Slumping

Often even in shallow rectangular moulds the sides pull in during the slump.  To know what things to try to correct this effect, you need to understand why this effect is occurring.  These two pieces show the effect in different ways.

ebay 0916_slump_01

 This slump shows that even with thick glass the sides curve inwards even on shallow slumps.

theglassundergroundnj.org

This slump shows the interesting effect that the further up the piece you look, the greater the curvature. This relates to the greater amount of movement required by the glass to conform to the mould at the outer edges.

Why

During the slump of a rectangle or square the whole shape of the glass sheet is changing.  It is slightly stretching to form into the “hollow” of the mould, but it cannot stretch evenly all over, especially at the corners.  If you think of the analogy of Draping a piece of cloth into a rectangular depression, you will find it wrinkles up at the corners if you smooth it at the sides. This indicates the material is attempting to overlap there as it does not have a dart to take up the excess cloth.

This similar to what is happening to the glass sheet.  It is relatively thicker at the corners than along the sides.  Therefore, it does not slide down the mould at the corners as on the sides. It is simply thicker and is compressed by the movement of the glass at the sides.

Prevention

The question is how to use that knowledge to avoid or minimise the dog boning during the slump.  There are probably lots of methods, but three have occurred to me and others.

Add more material along the sides.  This involves fusing a piece with shallow arcs rather than straight sides.  This gives more material to counteract the dog boning effect when slumping a rectangle.  The difficulty is getting the proportions of the arc correct in relation to the length of the sides. You also need to ensure the arcs on the sides are not so much larger than the mould that they slump over the edge.  This means the whole piece will need to be cut smaller than the mould.

Remove material at the corners.  This takes the opposite approach.  To avoid the increased amount of glass at the corners, you remove some of it.  That is, you round the corners of the pieces to be fused. How much you will need to round the corners is a matter of experience, but is a shorter learning curve than cutting the edges in an arc.

Reduce the temp and increase soak time.  This approach requires less skill in cutting a shape.  It relies on giving the glass time to relax into mould with a minimum of stretch.  You need to find the lowest practical temperature at which to slump.  This will be the temperature at which you can first see the deformation of the glass in the mould.  Hold the temperature there for as long as it takes – possibly one or two hours. It is likely that you will still need some rounding of the corners of the glass, but only your experience will determine that, and if so how much.

Cold work the edges until straight.  This can be done by hand or by machine.

Further information is available in the ebook: Low Temperature Kiln Forming.

The 6mm Rule - Kiln Forming Myths 11

Glass always wants to be 6mm thick

This is true only at some temperatures.  

The surface tension or viscosity of the glass, together with gravity determines the extent to which the glass will thicken or thin.  The viscosity of glass is such that at high temperature tack and full fusing heats, the glass does tend to become 6mm - 7mm thick. This is taken advantage of in kiln forming to obtain rounded edges, and in making frit balls.  A single layer of frit up to about 10mm will become a round dome due the action of the viscosity and weakness of gravitational forces acting on a small mass. 

Larger pieces of single layer glass begin to shrink as the viscosity is great enough to overcome gravitational forces to allow thickening at the edges.  This causes dog-boning.   At the same time the glass is thickening at the edges, it is thinning in the interior allowing large bubble formation on thin pieces. It also is the cause of the needle points on thinner pieces at higher temperatures.  The glass is soft enough to conform to any imperfections in the surface and so be stretched thin as the main mass of the glass contracts. 

This contraction also applies to low mass items such as frit in casting moulds.  The glass particles contract to form a single mass of material, leaving some stuck to the mould. These pieces may be completely separate as tiny frit balls, or if attached to the main mass, a series of needle points on the edge of the finished piece.

However, the viscosity at full fuse temperatures is not great enough to keep thicker glass in its original shape.  So the effect of gravity on glass of 9mm or thicker overcomes the weakening viscosity force and the stack begins to expand. The extent of the expansion is the result of both viscosity (heat dependent) and gravity (mass dependent).

At lower temperatures, the viscosity is much greater.  This can be used for low temperature tack or laminating temperatures. The glass can be adhered with heat without distortion of the single layer, as the viscosity is so high the glass does not change shape, even retaining sharp edges, although stuck together.

At temperatures above full fuse the viscosity decreases further allowing the glass to flow.  This is used in casting, blowing, and various higher temperature processes, such as aperture melts and stringer formation.  Here the viscosity is low enough to allow gravity to make thin and elongated shapes.

There is a range of temperature above which glass will thin more than the 6mm – 7mm “rule”.  I do not know the exact correlation between temperature and thickness, but at around 1150°C  the glass will become only a little under one mm thick.  This can be seen from the results of kiln runaways. The glass that is melted onto the surface of the shelf is extremely thin, showing that the viscosity was so low that gravity was able to thin it to a fraction of what we think of as normal thicknesses.

The 6mm myth arises from the behaviour of glass at a specific heat range and is the result of the combined forces of viscosity and gravity.  Knowledge of how these interact can enable you to understand the outcome of various projects.  This knowledge of the forces can be used to help create the effect you want.  It also enables you to employ various means to counteract the natural forces of gravity and viscosity. 

More information is in the e-book: Low Temperature Kilnforming.

Large Bubbles

cambridgecreativenetwork.co.uk

Some times you want large bubbles, but when you don't, you need to know about the causes of, and ways to prevent, bubbles.

Causes and prevention of most large bubbles relate to volume control, layup, rate of advance and top temperature.

Volume control. We all know that glass tends toward being 6-7mm thick at full fuse temperatures. Any less volume and the glass thickens at the edges, so trapping air under the glass which will push up and through given enough time and temperature.

Layup. The lack of volume control is compounded by layups which do not allow the air to escape from under the edges of the piece.

Rate of advance. These two problems are compounded by asap, or even just rapid, increases in temperature. The glass softens quickly and the air finds it easier to push the glass up than to escape from under the edges. Slowing down is part of the answer.

Top temperature. A high top temperature softens the glass to the extent that any lack in volume control, layups that have hollows or traps for air, and rapid increases in temperature all allow the expanding air (and there will always be some) underneath the glass to push up and often through the glass.

These factors reinforce the Low and Slow mantra.

Other factors can promote bubbles, although the ones above are the most common. Debris between the glass and the shelf, or between glass layers can cause bubbles, given the right conditions. Small shallow indentations in the shelf can be the source of bubbles from underneath the glass. This can be identified by observing where the bubbles occur in relation to the shelf.

There are some things that can be done to reduce the likelihood of bubbles:

Internal bubbles

Avoidance

Heating

Rapid firing

Damaged shelves

Supports

Flip and Fire

Glue

Volume Calculations

When creating a casting, pot melt or other object from glass cullet or billet, you need to be sure you have a large enough volume of glass to fill the area. You can do it by measuring the volume or by calculating the weight. This note is about calculating the weight.

Filling a damed area with enough pieces of glass provides an illustration of volume control. To help make sure you have enough glass to fill the space, measure in centimetres to determine the area. For a rectangle, measure length by width in centimetres. For a circle multiply the radius by itself (radius squared) times 3.14 (pi) to get the area.

To determine the minimum volume required, multiply the area by 0.6 cm. This is the approximate thickness that glass takes up at full fuse. As the amount of time and heat that we normally give to the process is insufficient to allow the glass to fully flow, the glass will tend to be thicker in the middle when using pieces of glass rather than sheets. So you may wish to multiply the area by 0.7 cm (to make sure you have enough).

To get the weight of glass required for the space, multiply the calculated volume by 2.5 (specific gravity) to get the weight in grams. Divide by 1000 to get kilograms. If you must use pounds, multiply the kilos by 2.2, the number of pounds in a kilo.