Slumping with moulds

A 2400mm x 1200mm sheet of thick Optiwhite glass slumped over entwined bodies. The models are in the background.

An ambitious undertaking involving numerous positive and negative moulds of ever increasing heat resistance; and a project to be contemplated only after what follows here is well and truly mastered.

Moulds can take many forms and be made of many different materials. They are discussed at length in 59. Moulds. Here we are looking at some of the techniques of using moulds and the way glass behaves in some situations.

Shallow slumping

Slumping into a mould often makes use of a cut-out mould on a shelf or hearth so that the glass will slump and settle on the shelf. The drop will usually be between 10mm and 25mm (3/8” and 1”) and is often determined by available materials from which the mould can be made.

Consider the platter shown being fired in the Basic Kilnforming section of the site.

Fig 1

This was 6mm float glass slumped into a 25mm deep fibreboard mould. It was fired to 780 degrees C with a soak of 30 minutes in a Riley GS-2 Top-hat kiln.

When the area of the vertical surfaces is added, the total surface area is now much greater than it was before the firing. The skin has had to stretch and glass has had to flow from one place to another to accommodate the changes.

From whence did it come, and where did it go?

Fig 2A

Fig  2B

The above figures show a strip of the same glass as was used for the platter slumped with a 25mm drop: as if it was a slice  of the platter.

They highlight the small area in which the "borrowing" and "flowing" takes place. Glass has flowed from the inner part of the top flat so that it is little more than 4mm thick at the lip but there has been little or no effect at the outer edge. The vertical section is thinner at the top than the bottom, but nowhere as thin as it would be if it couldn't seriously narrow the width of the strip. Note how much the strip has been narrowed at the bottom of the bend. In the full platter it couldn't do this, so the wall would be thinner again. 

Fig 3A

Fig 3B

Whilst not strictly true, it may help to understand what is taking place if the glass is considered as having two components; the skin and a gooey inside which flows independently of the skin. In Fig 3A, the hatched area marks the skin which must stretch to create the increase in surface area. It bends at the top edge of the mould and swings ever more toward the vertical as the firing progresses. 

At the same time the 'gooey stuff' is flowing in to try and maintain glass thickness in the stretched area.

The area at the top bend is where reduction of glass thickness will be the greatest, and that will be greater the more the depth of drop. This imposes a limit on the minimum thickness of glass which should be used in such situations.

Take-up of mould detail or defects

This test piece was slumped over pieces of lightly battwashed insulating fire brick (IFB)  and has taken up every minute crevice and indentation in the brick at the top bend: more so at this point than anywhere else. Not only can irregularities here show through the glass but they can also create points of possible failure.

It is important to pay particular attention to the surface finish of a mould at this point. A slight radius and smooth surface is best for the glass. Fill with battwash and rub smooth.

Distortion of edges

When slumping into moulds the edges of the piece can be pulled inwards. This is particularly the case with square or rectangular shapes or where there is a length of straight outer edge.

Fig 4

Fig 4 shows an undersize piece of glass fired on the mould used previously so that the end flats would be of different widths.

The 50mm wide long sides showed no pulling-in, and this conforms with the bestowed wisdom that a flat width about twice the drop will prevent the pulling. The 35mm flat on the right shows some pulling in whilst the 15mm width left edge has really been pulled down the hole, even though the length of unstiffened edge is quite short.

The flat strips between the unslumped areas act as stiffeners, breaking up the length of edge where pulling can take place. The width of glass on the outer flat needed to resist pulling in can be influenced by the ‘adhesion’ of the glass to the mould. A firm indented surface to the mould would be better able to ‘hold back’ the glass than would a loose powdery surface on which the glass could easily slide.

Placing a weight on the flat strip of glass can improve the adhesion to the mould and reduce ‘pull-in’, but this can also introduce a number of other problems, including;

• the Imprinting of a pattern from the weight into the top surface of the glass.

• Possible thinning of the flat strip

• Insulating the flat surface and reducing its heat uptake, with possible adverse effect on edge development.

 Undercutting.

Sometimes a thin mould is supported on props to allow a deeper drop to be made, as in Fig 5.

Fig 5

This is often done to allow the use of thin ceramic ring moulds to be used to form dishes with a deeper slump. The props can be as short as a few millimetres or can be quite long. Whatever the height of the gap, there is always a risk that the glass can flow sideways and create an undercut; making it difficult or impossible to separate glass and mould. The longer the soak the greater the risk. 

This method could have been used when making the platter in Fig 1 but was rejected because of this very real risk. Better to make the mould solid so that undercutting could not take place.

It is preferable that moulds be solid and that the sides be vertical or slope slightly inwards toward the bottom and that any imperfections be filled with battwash, or be repaired with a filler before being battwashed.

Fig 6

Oft times the use of a propped ring poses no risk whatsoever. Fig 6 shows a slump of a 600mm diameter 8mm thick float glass dish. The mould is a thick stainless steel ring covered with CF paper. The bare minimum of glass is resting on the mould to avoid bottom side blemishes and any change in line of the top surface. Firing was done at below 610 degrees C so there would be no change of surface finish and minimal change in edge formation. The glass was not intended to touch the shelf. Cold working of the edges would be needed to avoid frequent use of Bandaids. Life is full of these little compromises. 

Deep slumping

Is usually where the glass must drop a considerable distance and can take a number of forms. It can be:

• Into a deep or shallow bowl shaped mould such as a ‘wok’
• Through a thin cut-out ring or shape to create a ring-supported bowl with a lip
• Through a ring shaped mould as above but with the slump continuing until it settles on a shelf or other battwashed surface to create a vase-like piece.

In Fig 7, the glass is being slumped into a ‘wok’ shaped metal mould. The mould should be battwashed and supported in a ring or with small pieces of refractory material so that the top edge is perfectly level. One or two small holes should be drilled into the mould where the glass will last settle to allow trapped air to escape and prevent the blowing of bubbles.

Fig 7

The glass should be cut smaller than the full diameter of the mould and be installed so that it is perfectly horizontal in the mould. This can be done by making pencil marks at the appropriate distance down from the rim of the mould and levelling the glass to them. If the mould is level, then so will be the glass.

There is always the possibility that the glass will settle unevenly but this will usually be of no great consequence if the finished item is supported in a ring. Out-of-level slumping is of more concern if the mould has something in the bottom to give a flat shape to the bottom, as in Fig 8. This is often done using discs of CF paper or thin fibreboard..

Fig 8

Spun metal moulds can shaped to give a slight concavity to the glass, as in Fig 9.

Fig 9

The glass piece can be pre-fired to give texture and edge development before it is slumped into the mould. A rounded edge on the glass can also minimise hang-ups in the mould, as well as aiding the escape of air.

If a thin mould  is supported on tall props the glass can drop through a considerable distance, creating rimmed bowls which can be supported in wire frames. The glass supported on the rim will take up the texture of the rim but the unsupported area will remain clear; unless the entire piece is pre-fired to apply texture.

Fig 10 to come.

Smaller diameter moulds on longer props can be used to create vase-like pieces as in Figure 10 when the drop-thru is allowed to continue until the glass settles onto the shelf. 

In this instance several layers of float with spots of paint trapped between were fused and slumped in a single firing.

At other times the layers may be fused together, followed by a lower temperature slump. Much trial and error will be involved, whatever the technique used.

Slumping over a mould

Glass has a higher COE than many of the materials from which moulds are made. As a consequence, when glass is slumped ‘into’ a mould it will contract away from the inside as it cools. This allows quite deep slumps to be made, such as deep bowls: without risk of them jamming in the mould.

Such is not the case when slumping ‘over’ a mould.

Consider the shapes in figs 11 & 12. Provided the curve is gentle and the glass will be ‘sitting on’ rather than ‘wrapping around’ the mould, then a reasonably thick piece of glass should be able to withstand the stress due to it shrinking onto the mould as it cools. A 6mm thick piece may withstand what a 3mm piece couldn’t.

Sometimes round shallow dishes are made this way but more frequently the slump is over long narrow log-like moulds; making bowls this way is NOT recommended.

Fig 11

Fig 12

If the glass must be wrapped round the mould, then covering the mould with a 6mm or 12mm (1/4” or ½”) layer of ceramic fibre blanket can act as a cushion and provide crush ability to the mould.

Applying a spray coat of colloidal silica will give the blanket a crust which can support the glass in the molten state but which can collapse under shrinking pressure as the glass cools.

As can be imagined, maintaining the glass blank on the top of the mould can present difficulties.

One technique is to support it on props at the corners until it sags enough to retain itself in position on the mould; at which time the props can be removed and the slump completed.

 Fig 13

 Fig 14

Fig 15

These shapes can be made with 6mm float with paint decoration, or with two layers of art glass. In each case the glass is first fired at fusing temperature to set the paint, develop the edges and impart texture. The second firing can be at a lower temperature so that it will move more slowly, giving more time for observation and to remove the props. 

This work, like many kilnforming operations, can best be done in a Top-Hat kiln where easy access to all sides is assured.

Slumping temperature

Slumping to change the shape of glass can be done at a range of temperatures, from around 600°C  to above 800°C  (1110°F  - 1470°F) 

It should be stressed that there is NOT one correct slumping temperature for a glass, but rather a range from which the most appropriate number can be selected to best achieve the result one desires. Experience will show which works best for you.