Compatibility

What does compatibility mean?

Compatibility is the ability of two pieces of glass to be fused together without cracking, breaking or showing signs of internal stress. Such pieces are  said to be compatible: to be a ‘good fit’.

A major factor affecting this is coefficient of expansion (COE), but some authorities consider that mismatching viscosity curves also play a part; especially when dealing with pieces of glass of different colour; or a single piece incorporating multiple colours..

Coefficient of Thermal Expansion (COE, or CTE)

Almost all materials expand as they are heated and contract as they cool. COE is a measure of the percentage change in length of a material with unit change in temperature. It means how much they grow when heated and, more importantly, how much they shrink when cooled. Glass with a high COE will expand and contract more than glass with a low COE.

COE, and compatibility in general, is of concern only when two pieces of glass are to be fused together. It is of no concern whatsoever when slumping or bending a single piece.

If two identical pieces of glass (cut from the same sheet) are fused together they will contract at identical rates as they cool. There will be no pulling or tugging of one piece relative to the other. Stress will not develop.

However, if two pieces of different COE are fused together they will contract at different rates, so stress will develop. Should the stress be great enough, the glass will break.

Consider the two pieces of glass in Fig 1A, the top piece having a higher COE than the one beneath. As they are heated the high COE piece will expand more than the other. When they fuse together as at B the high COE piece will be larger than the other. As they cool, the piece which shrinks the most (high COE) will be put under tension, (strain, being stretched or pulled apart) whilst the one which shrinks the least (low COE) will be under compression (being crushed, squashed) as at C.

The amount of expansion and contraction is quite small, and the difference between commonly used glass of different COE is even smaller.

A 300mm long piece of Bullseye (COE90) will expand by about 1.3mm between room temperature and 500°C, whilst a piece of Spectrum (COE96) by less than 1.4mm. The difference between the two would be less than 0.1mm: almost too small to be detected by eye but sufficient to shatter the piece. Some of the compatability tests involve differences in shrinkage as small as 1/100^th of a mm.

When the pieces are grossly incompatible then the one under tension (higher COE) will usually crack to relieve the stress. Sometimes the shock can be so great that both pieces will break: the second layer breaking as a consequence of the release of energy in the first.

With lesser degrees of incompatibility the tension can be released by minor cracking.

Where there is only minor incompatibility, and the stress developed is not great enough to cause the piece to crack, additional stress at some later time can cause the piece to break. This can be due to incidents such as a knock, or uneven heating from a heater or from the sun shining through a window, or any similar circumstance.

Detecting stress

Significant incompatibility is shown by breaking or cracking of the glass. This is evident and requires no test to detect.

Minor stress not causing a fracture can be detected by polarized light.

What is polarised light?

Fig 2

Light is myriad electromagnetic waves oscillating in random directions. A particular light ray oscillates in only one direction, somewhat like flicking a rope and causing a ripple to run along it, as in Fig 2. If the rope is flicked upwards the ripples will be up-and-down and its axis of oscillation may be said to be vertical. If flicked sideways it will undulate like a snake with the axis of oscillation horizontal; and similarly for light whose axis lies anywhere in between.

The distance between two oscillations is called its wavelength, and the wavelength determines the colour of the particular light ray. What we call white light consists of myriad rays of all the wavelengths, and thus colours, visible to the human eye. Each of the individual light rays, of whatever colour or wavelength, can be oscillating about any axis.with each ray oscillating only about its own particular axis

Polarised light consists of rays of all the different visible colours which are oscillating about only one common axis. It is produced by passing a light beam through a polarising filter, also known as a diffraction grating; so that all rays other than those whose axis is in a particular plane will be blocked.

Fig 3

If a polarising filter is placed in a light path, then only those light rays whose axis aligns with the slots will get through; all the other rays are blocked, as at filter A. If another polarising filter is placed after the first, but with its axis at right angles to the first, then no light rays should get through; as at filter B

This arrangement, using two polarising filters and a light source is known as a Polariscope. The term Polarimeter is often used but that term more accurately describes more sophisticated laboratory equipment. We will use polariscope.

The polariscope

There are numerous ways in which the filters can be arranged to detect for the presence of stress in glass. One arrangement using a studio lightbox and two filters is shown in Fig 4. Instructions for making your own filters are given in 31. Making a polariscope.

Fig 4

The first filter is placed on the lightbox and the fused test piece is placed on top, as in Fig 5. Only polarised light will pass through the bottom filter but, other than for a dulling of the intensity of the light, little difference will be noticed.

Fig 5

When the second filter is placed over the top as in Fig 4 there will be an immediate dulling of the light, depending on the axis of the second filter relative to the first. If the axes are aligned then most of the light will pass through; if at 90 degrees to one another then no light should pass. Rotating the top filter will cause the intensity of the light to alternately increase and decrease.

An area of clear glass without stress will have no effect on the light rays passing through the filter layers. Similarly for fused areas in which no stress is present.

In an area containing stress, light rays will be rotated about their axes by the stress as they pass through, so that they will no longer be aligned with the rest of the polarised light arriving at the top filter. If the top filter is aligned so that it blocks all the light passing through unstressed areas, then this misaligned light should pass through the top filter and show up as a brighter image.    

Detecting minor stress is a talent which comes with experience and it requires much perseverance to become adept. 

Remember, a polariscope can only show up stress in the glass. If the glass is cracked then the stress has already been relieved and there will be no other indications relating to the stress which caused the crack. Of course, there may be other weaker stress areas remaining, but sighting the first major stress crack is all one needs to confine the piece to the scrap bin.

Stressometer

A device using a torch and filters is sometimes available from Art glass suppliers and is called a Stressometer. Instructions on sourcing and making your own device can be found in 31. Making a polariscope (stress detector).

Fig 6

Polariscope from sunglasses

If polarizing film is not available, another option is to use the lenses from a pair of ‘Polaroid’ sunglasses in the manner described above and as shown in Fig 7

Fig 7

Not all sunglass lenses are polarising filters. To test, hold one above the other in a light beam and rotate one with respect to the other. There should be a decrease and increase of light passing through.

Testing for compatability

This is done by fusing two test pieces together and examining them for stress when they cool. It is best done with a smaller piece placed on top of a larger piece, so that any stress will develop around the edges of the smaller piece. This makes it so much easier to detect than when the two pieces are the same size.

If there are cracks visible the pieces are obviously incompatible. The stress has been so great that cracking has occurred to relieve the stress.

Testing colours for compatability

Much kilnforming involves the fusing together of a number of pieces of different colour. Whilst they may be labelled 'compatible', and bear the same 'compatibility label', good studio practice is to verify that is so. Better to do so before making up a work using much expensive glass and having it break.

Detecting evidence of stress in two coloured pieces fused together can be difficult. The task is simplified by placing small pieces of each colour on a larger piece of clear. If each colour is compatible with the clear then it can reasonably be expected that each colour is compatible with the other. Many colours can be tested at the same time by this method.

Fig 8

If all the samples are free of cracking then, using a polariscope, examine them for signs of stress around the edges of each small piece. If the pieces are incompatible there will be stress indications either immediately inside or outside the boundaries of a coloured piece, depending on whether the coloured piece has contracted more or less than the base piece. This can range from an opaque band to a slight halo effect, showing stress which is not severe enough to crack the glass: now. What it may do at some future time is another matter.

If there’s no halo effects, no cracks, no other signs of damage, then the pieces may be considered compatible.

Strip test

Another method is to cut very narrow strips of the two glasses and fuse them together. If the strips are narrow enough, minor incompatibility will cause the strips to bend as they cool. Gross incompatibility will cause the piece with the higher COE (shrinks the most) to crack.

String test

A similar test without using a kiln involves the use of a torch or lamp, as used for bead making or lamp working. Small pieces of the two glasses are held together and heated in the flame.

A thread is drawn from the pair of samples so that they fuse together as they are being drawn and then allowed to cool. As they cool they will stay straight if compatible; either bend or snap if incompatible, with the piece with the higher COE (the one which shrinks the most) being on the inside of the curve, or breaking.

This method is generally confined to hot glass workers or to bead makers, as they are the ones most likely to have the equipment which enables them to use the technique with minimum risk.

Where stress may show

Usually, when fusing two pieces of roughly equal thickness, the one under tension (the one which shrinks the most) will crack.

Sometimes the reverse can occur. If a thin, high COE piece, is fused to a thick base glass with a lower COE (the thick piece is under tension and the thin piece is under compression) then  the piece under compression could be crushed. This latter is more likely to occur when fusing stringer or shard to a thick base glass.

The greater the difference in COE between the pieces, the more certain and immediate will be the fracturing.

It is generally held that a difference of one in COE can be disregarded, although this isn’t always the case.

Just to make life more interesting, sometimes pieces with the same claimed COE won’t fit, whilst at other times pieces with different COE will work OK.

Blessed by the kiln god? Who can say.

Compatability of frit threads & paints

Irrespective of the thickness of the materials, the same forces will be at work whenever two pieces of glass having different shrinkage rates are fused together.

Small pieces of coloured frit or thread fused to an incompatible base sheet may well survive the stress, often because it is ‘almost a good fit’ and the stress is too weak to crack anything.

The degree to which such a mismatch is acceptable will depend on circumstance. If a small blob is sitting proud on the base sheet and a stress halo surrounds it: then it could well ‘pop off’ at some future time.

If, however, a thin thread is well embedded in the base sheet and shows only the slightest signs of stress at a couple of localised points: then it could well survive intact, or develop only one or two fine cracks across it.

A film of paint can be subject to stress if incompatible with the base glass.Should the paint film shrink more than the base glass then extremely fine cracks may appear in the colour layer. These can be extremely difficult to detect, often only in a strong light and after much twisting and turning to get the light to glint on the edges of the cracks.

Does this mean that the paint is faulty? Not necessarily.

If the paint is on float one should remember that the COE of float varies widely, so a paint which doesn’t work so well on one piece could work well on another.

Do these cracks make the piece itself faulty.

Again, it depends on circumstance. A few minute stress relieving cracks may well be acceptable. Consider the crazing which often occurs in the glaze (glass) on pottery pieces and which is sometimes acclaimed as an inherent feature.

Of course, if the particular paint regularly shows defects it makes sense to not use it with that particular base glass. Who wants to be continually firing faulty or potentially defective pieces. However, there could be other glass on which the paint may work perfectly well.

Small pieces of frit or colour, even though they exert only a small force, can sometimes pull flakes off the face of the base glass: take a small piece of base glass with it.

Consider how ‘glue chip’ is made: the glass is coated with hot animal glue and when the glue cools it pulls chips off the face of the sheet: the result of a combination of good adhesion between glue and glass and vastly differing COE.

Concerned about a fused piece. Not too sure if it’s good or bad?

One can put it through the dishwasher, or put it in a plastic bag and into the freezer. That treatment often sorts out the good from the not-so-good.

Considering the wide range of COE of float, and the relative infrequency of crazing problems with paints on float, then the ‘A difference of more than one in COE should be avoided ’ rule doesn’t appear to always apply here.

In much of this one must rely on ones judgment. Hard and fast rules can go only so far.

Entrapments in fuses

Incompatible frit or other thin coloured glass will often work if fused between two layers of compatible clear: because the force exerted by or on the thin piece is too weak to crack the outside sheets.

Similarly with paint. A paint film on the surface may crack ’because it has room to move’ and is a potential hazard because it could fly off the surface. If entrapped between two base layers the thin film is much less free to move, and has nowhere to go if it did break up.

The technique of painting between layers of thin glass is used by many prominent artists: sometimes building up panels four or more layers thick and giving a sense of depth to the piece.

Of course, they would choose their materials carefully, test them thoroughly and limit themselves to using what they know will work best. They wouldn’t use glass from a multiplicity of sources: nor buy it a bit-at-a-time.