A Low Cost Tester of Glaze Melt Fluidity

Description

Use this novel device to compare the melt fluidity of glazes and materials. Simple physical observations of the results provide a better understanding of the fired properties of your glaze (and problems you did not see before).

Article

There are many complex and expensive instruments designed to observe and measure the goings-on in firing kilns. Generally this type of equipment is expensive and measures absolute physical properties that can be quantified easily. However glaze melt flow is like clay plasticity, it is more subjective and not so easy to quantify. It is best measured comparatively, that is, one specimen directly compared with another. That being said, each glaze does have a unique melt flow patterns over a range of temperatures, these can be like a finger print. Melt fluidity testing can be done using inexpensive methods and devices.

I would like to submit a general-purpose testing method for many glaze melt properties that is both inexpensive and easy to use: The GLFL test. So many factors related to the melting, solidification and physical properties and defects of fired glaze surfaces are related to melt viscosity. Thus a test that provides information about this has the potential of being very valuable.

Before going on, I will give credit where credit is due. This is not an original idea. I have seen this device described in industry literature to compare melt properties of nepheline syenite and feldspar. Also, I was sent a very nice dual-flow mold by Hugh Nile at Sterling China (it had the initials IMC embossed on it). I am aware that other industries also use similar devices. However I want to take it to the next level by clearing documenting its advantages and a procedure to use it.

I have made a rubber master mold of the one described herein and can making working molds for others. If you would like one please see the bottom of this article.

Testers that do not work well

Small or steep angle testers: Although I have messed with smaller sizes in the past I have now seen the light. They just do not work as well. You need a large enough reservoir, and long enough flow ramp at a shallow enough angle to get repeatable and sensitive tests.

Inclined tile testers: Some companies prepare a lump of the glaze to be tested and glue it to one end of a tile using a slurry made from the same material. While this will often work it is problematic with compounds that shrink a lot or those lacking dry hardness. The former could crack off and the latter may crumble off. I'll leave it to your imagination what might happen if pieces or the whole sample rolls into contact with a kiln element.

Photo not found
or not published

The Dual-Flow Large Tester

This is shown in the picture. It is 13.5cm high while standing (5.5 inches). The long runway is at less than a 45 degree angle for extra sensitivity (there are actually two orientations for two different angles). One of the big advantages of the dual tester is that it can be employed for side-by-side testing of two specimens (e.g. one alongside a benchmark). It is amazing how close you can match the melt fluidity of two materials using this method.

This device is cast in a plaster mold using a mix refractory enough to resist warping if walls are cast thin (in production situations flow testers should be made from the same clay that ware is made from but if such is too vitreous you can reduce the feldspar content somewhat. See below for more information on the slip recipe. I usually bisque fire these testers for extra strength. The reservoir accepts a 10-12 gram ball of material that you can just drop right in. These balls are easy to make by dewatering the glaze or material slurry on a plaster surface to the right working consistency and then rolling the ball in your hands, drying it and shaving material off to achieve the right weight. (thus the glaze does need to have enough plastic ingredients to enable this workability or you need to add some bentonite to impart it).

I have defined a procedure for this test in the testing area of this site. As noted in the procedure there, for repeatable results it is important that your testers be the same thickness, made from the same clay, fired at the same rate of rise and to the same temperature, and the ball sample must be the same dry weight each time.

In case you are not yet clear on how this tester is used: Two glazes are compared by dropping dried balls of each into the reservoirs at the top and the whole thing is fired to the desired temperature (with a tile below to catch any glaze that runs right off the end of the runway). During the firing, the glazes flow down the runway according to melt development, melt surface tension (and other factors like bubble development).

What this tester can show you about glazes:

G1215U vs. G1215W glaze flow test
These recipes have the same chemistry but the 1215U uses frit to source the MgO and CaO. This demonstrates that it is not just chemistry that determines melt flow. Raw materials are crystalline and have different melting patterns than frits (which have already been melted and reground).

If glaze ingredients shift in particle size or chemistry and thus change the melt, it will be immediately evident either by the flow reaching further down the runway or by a change in the character of the flow. This information is valuable in quality situations since it is so hard to guage glaze fluidity by simple observation of a normal thin layer on glazed ware or a test tile.
Information from a flow tester is valuable when adjusting the recipe of a glaze for other things like thermal expansion, color, material substitution for no-longer-available materials, etc. while trying to maintain the same fluidity.
This test helps to rationalize discrepancies between between what the chemistry of a glaze indicates should happen and what actually does happen in the melt. Often glazes of very similar chemistry will have different flow properties due to factors related to mineralogy and physical properties of materials. For example, the tester shown here is two glazes with the same chemistry, one sources CaO from calcium carbonate, the other from wollastonite.
Often there is a need to maximize the amount of SiO₂ in a glaze. For example, this test can demonstrate if changes made in the chemistry of a glossy glaze to reduce its thermal expansion also produce a more fluid melt. If so, the SiO₂ can be increased, producing an even lower expansion. Likewise, this test will demonstrate if more silica can be tolerated if adjustments to produce more hardness also yield more fluidity.
A flow test does not just show the degree to which the glass is melting, it also reveals the surface tension. Melts of high surface tension meet the tester are 90 degree angles whereas those of low surface tension spread out. The better the melting of a high surface tension glass the narrower the flow will be. In transparent glazes the flow will be white because it is not releasing the entrained bubbles. Very low surface tension melts meander down the runway in a thin layer. Knowing about this can really help analyze the cause of crawling, blistering and clouding problems.
Ball milling time: By extracting samples from your mill at regular intervals, firing, and comparing the degree of flow you will be able to assess the mill's effect on glaze maturity and melt development.
Because the glaze is so thick in a flow tester, bubbles resulting from products of decomposition within the glaze will be evident by the character of the thick flow and in the broken cross section (bubbles can even disrupt the melt flow).
The glazes ability to wet the surface of the clay is evident by the angle at which the leading edge and sides of the flow meet the runway surface.
These testers are great educational tools. This one, for example, shows the impact that a simple addition of opacifier has on glaze flow.
Changes in properties like opacity or tendency to crawl, blister, pinhole, crystallize, craze or shiver, develop entrained bubbles or boron-blue clouding are often amplified by this test. The flow provides for an opportunity to see a very thick layer of your glaze and this can reveal differences not noted in thinner layers. Glazes which do not necessarily run on ware may run very badly on a flow tester, this indicates a lack of SiO₂ and Al₂O₃ and is a warning for susceptibility to cutlery marking and leaching. Since so many glaze defects are either related to melt viscosity or revealed by a thick flow, monitoring this property is important.
This device can even be used to help determine the optimal firing temperature, by experience you will know the fluidity of glazes that perform well. In addition firing a glaze in a flow tester at a range of temperatures may reveal that fluidity begins to increase more quickly through a narrow temperature range. After all, what is more significant to determine the freeze-point than flow of the glaze melt?
This test is ideal for the development projects of under and overglaze colors. These need to be less fluid than glazes. Different stains require differing amounts of flux in the host to get the same degree of flow, using this type of test a line of products can be made that all have the same melt flow properties.
The character of the melt flow, especially when compared over a range of temperatures, is unique to each frit. It is possible to identify them and even spot equivalent products from other manufacturers by using this test.

Raw Materials Testing

Most companies can readily test clay materials for use in bodies and glazes using physical testing methods that require a minimum of equipment. But it is not so obvious how to compare and test fluxing materials like feldspar for consistency. One can just trust the particle size and chemistry information provided by the manufacturer for each shipment and compare numbers. But what is the actual relationship between these numbers and the consistency of product on a production line? Can you trust the numbers anyway? The tester is an elegant simple alternative. It accurately shows melting power, color and impurities, you need to see two feldspars side-by-side to see how sensitive it is (see pictures at bottom for an example).

Product Development

Many ceramic products are tuned to melt to a certain extent to achieve their function. For example, an engobe needs to have a stiffer melt than a glaze, but much more maturity than the underlying body. Likewise, a ceramic printing ink must have a specific degree of melt fluidity, enough to adhere or melt to a smooth hard surface, but not so much as to bleed into the covering or underlying glaze. Melts used for bonding purposes likewise need to develop enough glass to bond, but not so much that fired geometry cannot be maintained. A standard and a test can be evaluated side-by-side using this tester. If the melt is not fluid enough, then it can be fired higher, or a percentage of frit can be added.

Taking Photos

Since these fired testers are quite large, storing them for future reference can be a problem. Taking a picture of them and scanning it onto the computer for archival purposes makes more sense. Make them at least twice as large as the ones shown here and they should still take less than 100kb of memory. You may find that making the testers from an off-white, grey or even tan body might be better to prevent washed-out results when taking photos. Also, have plenty of side lighting so that gloss is highlighted.

Slip Recipe

A good starting recipe is #L2540, it is 50% ball clay, 25% feldspar and 25% silica. This does not cast quickly but the pieces have good green strength and the clay will vitrify around cone 10-11. For a more refractory mix replace some of the feldspar with kyanite, calcined alumina or some other non-plastic high temperature material. You will need to know how to mix and deflocculate a clay slip, search in this library for the word "deflocculation" for an excellent article on understanding the casting slip mixing process.

Getting a Tester

We have provided detail pictures of our mold so you can make your own. We are planning to add 3D geometry to enable printing plastic shell molds for rubber masters (from which you can pour working plastic molds). The GBMF test is an alternative to this one, although not as accurate.