and bonding power in the raw state are not required of the clay. Provided it is ground sufficiently fine to form an intimate mixture with the sand, the physical properties of the raw clay are unimportant.

According to the kind of sand used and the proportions of sand and clay used in the mixture, a vitrifying or a nonvitrifying fire clay should be used. In general the softening point of the clay should be high. Vitrification and softening point determinations can be made as indicated in earlier sections of this paper.

Clays in Steel Molding-Sand Mixtures

Requisite Properties of Clays.-Plastic fire clays are used extensively in sand mixtures for molds for steel castings in both green-sand and dry-sand mold making. The function of the clay is to act as a bonding material, alone or in conjunction with other binders, such as molasses, and thus to render the mixture plastic when wet and enable it to hold its shape when partially or wholly dry. The requirements of a clay for this use are that it be readily miscible in a very intimate manner with large proportions of sand and a 'small amount of water, that it have considerable bonding power when wet, and that it shall not form mixtures that vitrify at the temperatures at which steel is poured. For dry-sand molding, a certain amount of bonding power in the dry state is required, but practically every clay that has sufficient bonding power when wet will be satisfactory in this regard when the molds are dry.

Slaking Test-In order to be readily miscible with sand and water in the sand-mixing machines in use in steel foundries, a clay must either be finely ground or slake down rapidly to a fine powder when wet. A quick slaking clay does not need to be finely ground for this class of work, but a slow slaking clay must be very finely ground and thoroughly blended dry with the sand if an intimate mixture is desired in the wet composition. The slaking properties of ground clays can be determined in the following manner :

Mixtures of the clay to be tested are made with potter's flint in the proportions of 1 to 1 by weight. These are mixed with water and molded into cubes 1 inch square and dried first at atmospheric temperatures and finally at 110 degrees Cent. The cubes are then placed on a piece of 1⁄2-inch mesh wire screen and immersed in water at room temperature. The time required for the whole sample to slake and settle through the screen is taken as the time of slaking of the sample of clay. Several determinations should be made and the average taken as representative of the time of slaking of the clay.

Clays vary greatly in the readiness with which they slake In the testing of 28 clays at the bureau of standards it was found that the time of slaking varied from 52 to 128 minutes. Fat, sticky clays require much longer time for slaking than short, sandy ones. However, as stated above, the rate of slaking of fat clays can be accelerated by fine grinding.

Bonding Power Test.-The bonding power of clay in wet mixtures of clay and sand varies so greatly with the relative amounts of clay and sand, the amount of water, the size of grain of the sand or sands, and the method of making the test specimens, that it seems necessary for each foundry to make tests of this property on the particular mixture employed in that plant. The clays cannot be tested alone for the strength of some clays is increased by the addition of sand while that of others is decreased. The moduli of rupture of the dried mixtures cannot be taken as an indication of the bonding power in the wet condition of the clay. It has been found that certain kinds of clays, such as the adobe clays of the Southwest and the loess clays of the Middle West, have high moduli of rupture in the dry state but practically no bonding power in the wet condition. On the other hand most clays that have high bonding power when wet have considerable strength when used in dry sand mixtures.

The bonding power of a clay in a wet sand mixture is determined in the foundry laboratory of the bureau of standards by making a shearing strength test as follows: Bars, of the mixture to be tested, 12 inches long by 1 inch square

in cross section, are molded in a snap flask on a smooth glass plate. The sand is first gently rammed with the thumb and forefinger of each hand and then firmly rammed once with a round wooden bar from one end to the other. Then the flask is removed, the sand bar being left on the plate. The bar is then gently shoved lengthwise over the plate at about the rate of 1 foot a minute until it breaks off. The weight of the portion of the bar breaking off is the shearing strength per square inch of the mixture. Several determinations can be made on the same bar, and several bars should be tested in order to get a fair average.

In this test it will be found that fat sticky clays, especially the fine-grained ones, have greater bonding power than short, sandy ones. This agrees with the fact that it has been found practicable to use a smaller percentage of finely ground fat clays in clay-sand mixtures than of sandy clays.

Vitrification Test.-A clay-sand mixture that vitrifies or bakes to a hard mass at steel pouring temperatures is liable to give a high sand loss and castings that are hard to clean. A thick layer of caked sand will adhere to the castings. To determine the probability of this occurring with a given mixture. vitrification tests of the clay alone cannot be used. With the small amounts of clay employed in these mixtures vitrification is brought about by the actual fusion of a mixture of clay, the finest of the sand grains, and the surface layers of the coarse sand grains. The temperature at which this takes place bears no relation to the temperature at which the clay alone vitrifies. It is more nearly, but not directly, related to the softening point of the clay. It may be said that if a clay does not soften at steel melting temperatures, it will not give vitrified clay-sand mixtures. On the other hand if it does soften at or below steel melting temperatures it will or will not give vitrified clay-sand mixtures according to the kind of sand used and proportion of clay in the mixtures.

For the foregoing reasons it is desirable to test the claysand mixture to be used. The mixture for the vitrification test should be heated to the temperature used in pouring steel in the

foundry using the mixture. In other respects the test should be conducted as described in the section on refractory mortars.

Clays with high softening points are generally preferred for clay-sand mixtures used in making molds for steel castings. However, some clays with low melting points are used successfully. In tests conducted at the bureau on clays actually used in steel facing sand mixtures it has been found that the softening points varied from 1180 degrees Cent. (2150 degrees Fahr.) to 1600 degrees Cent (2900 degrees Fahr.). The clays with low softening points have great bonding power and are used in such small proportions in the mixtures that they do not cause vitrification.

The Place of Chemical Analyses in the Testing of Clays.Chemical analyses have little place in the testing of clays for foundry uses. The value of clays for these uses is dependent on the physical properties of the clays, such as ease of slaking, bonding power, vitrification behavior and softening point. The relations between these properties and chemical composition are so obscure that it is impossible to estimate them in a given clay from a study of either its ultimate chemical analysis or the so-called rational analysis.

No relationships have ever been established between composition and such properties as ease of slaking and bonding power. The relationships between chemical composition and vitrification temperature and softening point are very general. It is possible to say that a clay having a composition corresponding very closely to that of pure clay substance (39.8 Al2O3, 46.3 SiO3, 13.9 H2O) will have a high melting point and that a clay whose analysis shows the presence of large amounts of fluxing oxides, such as lime, soda and potash, will have a lower softening point. It is not possible to estimate the melting point with any accuracy in either case. The vitrification and softening points of a clay depend not only on its chemical composition but also on the state of chemical combination of the various elements, that is, on the kinds of minerals present, and on the size of grain of these minerals.

Discussion-Testing Clays for
Foundry Use

MR. R. F. HARRINGTON.-A number of years ago Dr. Moldenke introduced a method whereby the plasticity of clays could be tested by the use of a malachite green dye. Since then the laboratory of the Saunders & Franklin Co. at Providence has developed a method in which a crystal violet dye is used for determining the plasticity of clay. I have done quite a little work along these lines and have found the test well worth while and I would like to ask Mr. Staley if he has considered that method in his tests.

MR. H. F. STALEY.-The ceramic division of the bureau of standards experimented for several years with the use of dyes and as a result of this work decided that this test is not reliable and does not use it. The bureau found that impurities in the clay absorb the dye. It found also that if a man were attempting to select his clay by the dye test, he would not get results which he could not use in factories as well as he could by selecting clays according to his past experience in the use of those particular clays. There has never been a satis factory method devised for determining the plasticity of clays.

DR. RICHARD MOLDENKE.-There is no doubt that the dye test is not as satisfactory as it might be. Saunders & Franklin improved my own work. We were only trying to find something that foundrymen could use. I was wondering how many foundries in the country could make use of the information just given in the paper. They have no laboratories to go into the subject so deeply. The dye test is a simple one and I have always found it to work satisfactorily enough for the purpose intended.

Instead of buying good clay, many foundrymen dig clay. from their backyards, daub their cupolas and wonder why, after the seventh or eighth heat, they have to reline at the melting zone. This means money, work, lots of labor expense,