[Trade Journal]
Publication: Journal of the American Ceramic Society
Columbus, OH, United States
vol. 8, no. 9, p. 547-554, col. 1
ORIGINAL PAPERS
SOME OBSERVATIONS IN THE CASTING OF HEAVY
ELECTRICAL PORCELAINS ¹
BY K. H. FRITZ AND A. L. DONNENWIRTH
ABSTRACT
A summary is given of experiences had in the manufacture of heavy and intricate electrical porcelains by the casting process. Some of the subjects discussed are selection and test of clays, mixing of casting slip and its control, and several details of the methods used in casting the ware. Special emphasis is placed on the limits established for viscosity and specific gravity of the casting slip and the general proposition that the slip must be within the limits set for viscosity regardless of specific gravity.
We have been casting heavy and intricate electrical porcelains for about two years. The process is undoubtedly more difficult to control and more sensitive to variations in clays than any other in the whiteware industry.
Variations in Casting Behavior of Clap
Successful casting depends primarily on the proper choice of clays. Most all clays will cast, but their manner of casting is radically different.
We have fast casting clays and slow casting clays among both the china and ball clays. As a class the ball clays cast much slower than china clays.
Among the fast casting clays there are those which build up a firm solid wall from the mold to where the slip is casting and thus show a distinct change from slip to solid material. Others build up a firm cast adjacent to the mold but gradually the wall becomes flabby and soft as the slip is approached. The wall appears to be well cast but when the supporting molds are removed, it often is so weak due to its semi-solid condition that it collapses. The former type of fast casting clays are the most desirable but the latter can also be used just as successfully where the sections to be cast are not very heavy. We have had very little experience with slow casting clays for with them we have been unable to cast heavy sections. Possibly they could be used for lighter pieces.
The classification of clays into these types can be easily made by the various methods which have been proposed, most of which employ a standard test piece to be cast and noting the casting behavior. To determine the cause of a clay's peculiar casting behavior is a more difficult matter. Grain size is a most important factor but it does not tell the whole story. We find that clays show considerable variation in casting behavior with no appreciable difference in grain size as indicated by settling. There are evidently other factors such as soluble salt content or a change in the nature of the colloidal material which affects this.
The Bingham Plastometer, like the settling test, will usually classify clays as to their casting value. We find that as a rule clays which show a low plasticity according to this test are the best casting clays. Plasticity tests do not, however, indicate the sometimes serious variations in a clay which alters its casting behavior.
We have read the suggestion of control of the hydrogen-ion concentration as a means of casting control. We have done some work on this test and are familiar with it. This test will undoubtedly give a measure of the amount of salts necessary for the slip, but we do not see how it can reveal a variation in the casting behavior of a clay which is not eliminated by a change in the amount of salts added. We will describe a variation of this kind in more detail, but it is well to emphasize at this time that this is the most difficult problem with which we have to contend in casting, and which so far has not been solved.
Relation of Design to Casting Rate
The designs which we are called upon to make limit us to clays which cast fast and solidly and which have low shrinkage in the mold. We have cast pieces where the wall varies from five inches to eight inches and in weight up to five hundred, pounds, and we believe we can go further although it has not been necessary up to this time. For such work, the ball clay content must be kept as low as possible. We use no more than what is necessary to give the needed unfired strength so that the pieces can be handled without breaking. Eight to twelve per cent accomplishes this. As the ball clay is increased, the time of casting increases and the maximum thickness of section that can be cast is decreased. We have been able to finish the pouring of the largest pieces eight hours after it was started and the molds can be removed twenty-four hours after the pouring was begun. Such a schedule indicates the importance to us of low shrinkage of the body while it is in the mold, and this is another factor which is considered in our choice of clays.
Preparation of Slip
In preparing the clay slip for casting, all the ball clay, china clay and scrap is placed in the mill, with sufficient water and sodium silicate for the entire charge, and ground for forty-five minutes. The amount of water and silicate is determined by the amount used in the previous charge, the water being measured by means of a meter and the silicate accurately weighed. To this is added the feldspar and flint and the entire charge is ground for thirty minutes more.
A check is made at this time, before the mill is emptied, on the viscosity and specific gravity and corrections are made, if necessary, to make the slip suitable for casting.
Testing of Slip
The viscosity is measured by the flow type viscosimeter and is expressed in the number of seconds required for 100 cc. to flow through an orifice 7/32" in diameter. We do not use a lid on this viscosimeter, which enables the test to be made more rapidly; but on the other hand a slight error is caused due to the difference in pressure, as the height of the slip decreases. By always filling the viscosimeter to the same point, however, the results are correct for comparative purposes.
For specific gravity determinations, we use a small necked bottle which is calibrated so as to express the weight in ounces per quart. This we find, to be much more accurate than a pint or quart measure.
The viscosity of the slip, before it is emptied into the cistern must be between 40 and 65 and the specific gravity must be 60.2 ounces per quart. These limits were established after some experimenting and also upon observations made in the casting shop, which will be described later.
The slip in the mill, after being checked, is passed through a thirty-mesh lawn into a storage cistern, where it is kept in motion continuously by an agitator.
Limits of Viscosity and Specific Gravity
Every morning the slip in the cistern is again checked before using. The viscosity must lie within the limits of thirty-five to fifty-five. It will be noticed that these limits are ten points lower than the limits set for the slip as it is checked in the mill, but we find that by agitating the slip over-night, the viscosity is lowered to this extent. This is due, probably to the fact that the grinding time being so short, the point of maximum deflocculation is not reached until the slip is further agitated. As an illustration of this, some time ago we made up a slip having a weight of fifty-nine ounces per quart and a viscosity of forty. Blunging for eight hours additional reduced the viscosity to thirty-three, all other factors remaining the same. Continued blunging for sixteen hours reduced the viscosity to twenty-seven, but blunging for a longer time had no effect.
When we first started casting, a slip weight of 60.5 ounces per quart was maintained, and we were of the opinion that we could not change this weight, especially decrease it without encountering trouble such as excessive shrinkage. Everything went along fine for a time, until a change was noticed in that we could not maintain the proper viscosity, regardless of the amount of silicate used. The slip acquired a very sluggish appearance, the length of time required for casting increased and the slip would hang up in the mold, which prevented solid casting, causing what is termed "balling." In casting very thick and heavy pieces, we found on removing the side walls and core that the piece of ware would be so weak that it would sink down under its own weight indicating a soft flabby cast which, however, appeared solid.
We decided it was necessary to use a lower viscosity and there was only one way to accomplish this, and that was to lower the specific gravity. By reducing it to 60.2 ounces per quart, the viscosity dropped to within the limits and the trouble just described disappeared. The reduction in specific gravity was not sufficient to give any shrinkage trouble, and we have therefore used this weight since that time. This experience led us to believe that the viscosity is more important than the specific gravity and that the specific gravity should be reduced, if a satisfactory viscosity cannot otherwise be obtained. We are using this theory as a basis of our slip control and so far have had very satisfactory results. We believe that it is desirable to have a viscosity as low as possible at all times at a given specific gravity, but it is necessary that we set a lower limit because of settling. A viscosity below thirty-five will cause the heavier particles to settle more or less. In other words, we seem to encounter this before maximum deflocculation is reached.
It is, of course, understood that the specific gravity value should be set as high as possible, without exceeding the viscosity limits, on account of lower shrinkage and faster casting time. The important point, however, is that we reduce the specific gravity, if necessary to bring the viscosity to the proper value. When this is necessary a decrease of from .1 to .2 ounce per quart is usually sufficient to lower the viscosity from ten to twenty points.
Amount of Electrolyte Required
The amount of electrolyte required, often varies from day to day and over a period of six months fluctuated between 800 grams and 2200 grams. This is to be expected, however, and is probably due to a variation in the soluble salt content of the clays used.
Recently we received a car load of china clay which appeared to be no different from any previous shipment, but on using it we encountered quite a bit of trouble. It was possible to bring the slip within the limits for viscosity and specific gravity, but the casting behavior was changed decidedly. It had a sluggish appearance, cast slower and hung up in the mold somewhat, making it difficult to obtain a solid cast in the heavier designs. It resembled the slip we were using when we were trying to maintain too high a specific gravity. This condition was improved slightly by maintaining a viscosity very close to the lower limits, but could not be entirely eliminated. Upon the arrival of another shipment of clay, the trouble disappeared.
There was evidently some difference between these two lots of clay, but up to the present time we have been unable to find a test which indicates it. A plastometer test or a settling test, as mentioned previously, showed a slight difference but not sufficient to segregate the clay which gave trouble from that which we know was satisfactory. It is a serious condition and we hope that a solution or helpful suggestions are offered by those who have had similar experiences.
Method of Casting
Open top molds are used entirely. They are so superior to the funnel type that no other choice is possible. The slip is supplied by the continuous pump circulation method. This has been very satisfactory and seems to have no effect on viscosity. A ten-foot length of eight-inch pipe or a tank of some kind must be inserted in the line between pump and outlets to eliminate pulsation. The pipe lines are cleaned every evening with compressed air, to eliminate any chance of clogging.
Both core and drain casting are done, although the bulk of the work is by the former method. To facilitate release of the cores, an application of talc or compressed air is used. The former is more simple but not quite as effective and cannot be used when the cores are large and the taper not great.
For designs of variable thicknesses where wide sections occur in the lower part of the mold, it is necessary in pouring to fill the mold only in part, at first, thereby casting the wide section before any casting is done on the narrower sections above. Additional pours, must of course be made, before the top of the preceding pour has become too hard to unite.
We have tried to give a general idea of our casting methods, bringing out difficulties we have had and are still having, as well as those things we believe we have fairly well in hand. We have incorporated this in-formation in a process specification which has been issued to the shop for their guidance and which cannot be changed without the ceramic engineer's approval, insuring definite control. We expect that it will have to be revised, perhaps frequently in the future, for casting is still new with us and we hope that as we progress, further improvements will be obtained.
THE WESTINGHOUSE HIGH VOLTAGE INSULATOR CO., LATROBE., PA.
Discussion
E. H. FRITZ: We have been unable to find any test that will indicate the variation which we sometimes (not very often) encounter in clays that we use.
W. L. SAMPLE: Do you take into account the difference in temperature of slip from time to time? That would give a different viscosity reading. In our plant we have quite a difference in temperature of the slip. On a cold day the slip will be quite cold and vice versa.
A. L. DONNENWIRTH: We do have variation in temperature. It varies from about 65° to probably 85°. This has a slight effect on the viscosity, but we take that into consideration.
W. L. SAMPLE: Do you have a graduated chart that shows that?
A. L. DONNENWIRTH: Yes.
CHAIRMAN MCAFEE: Speaking of differentiation in clays, do you make a viscosity test on clays or do you make viscosity tests on your slips? There might be some advantage to you in differentiating that.
E. H. FRITZ: The plastometer test referred to in the paper is something similar to that.
R. A. HORNING: That does not indicate this variation that you have. I thought a viscosity test was made on the clays before sodium carbonate was added.
R. W. HEMPHILL: We tried out the clays separately by casting them instead of this other test where they are put in a tub and are shaken up. Does that make any difference: to test out the china clay and see how long it takes, how much water, and if that varies any?
E. H. FRITZ: That might indicate the difference but what we are also after is why we have that difference. What is in the clay that causes that difference? If we know that we can test for it quantitatively and possibly be able to make adjustments for it.
B. W. PARMELEE: Perhaps Mr. Fritz has not taken into account the organic content of the product.
E. H. FRITZ: Probably not, but we have not had any trouble which we could trace to the ball clay. The china clay we use is English china clay and contains practically no organic material.
C. W. PARMELEE: What is the effect of time on that slip, a slip which is made today and not used for twenty-four hours or more? Have you noticed any material changes in the viscosity conditions or other conditions?
D. H. FRITZ: A minimum viscosity was reached after sixteen hours. But that would not necessarily be the same. You would reach maximum deflocculation sooner at one time than another.
A. S. WATTS: Were these results obtained from china clay which came uniformly from the same mine?
E. H. FRITZ: Yes.
A. S. WATTS: A short time ago the statement was made to me that some English clay miners at the present time are bleaching their china clay with a dilute solution of blue vitriol. I am wondering whether it could be possible that the presence of some bleaching agent would account for such a thing as this in regard to the behavior of the clay.
When I was in England I saw a lot of extremely blue water in the bottom of the clay mine, but I assumed that it was just the normal color of clear spring water intensified by a certain amount of clay in suspension, and I rather question whether there is anything of that sort used in the bleaching of English clay. I do not pretend to know anything about the bleaching that is done in English china clays, but I raise the question in order to elicit information if there is anyone here who can tell us anything about it.
R. REIF: We had a carload of English china clay shipped to the plant. It was chalky and had a great tendency to crumble in my hand. We received another car some time later, same clay, same concern, but this clay was fatty and greasy. The chalky stuff would not work but the fatty stuff would. They had practically the same fineness. Can you explain it?
F. SCHRAMM: Did the clay have the same moisture content?
R. REIF: There was a difference of moisture content of one per cent.
E. SCHRAMM: Was there any chance of the one lot having been frozen?
R. REIF: Both came in about June or July.
G. GOODWIN: Do you use the china clay in the raw state or do they put it through a filter press?
A. L. DONNENWIRTH: We use the china clay in the raw state.
E. H. FRITZ: Would it be possible that contamination of the clay by ocean water on the way over might have something to do with that variation?
A. S. WATTS: There would not be very much opportunity for contamination unless the compartments become damp with sea-water.
E. H. FRITZ: We know of cases where there has been leakage in the boat.
A. S. WATTS: That may be possible. The sodium chloride would act very definitely, but I had never taken that into consideration because such an impurity would be abnormal.
A few years ago a large shipment of clay was put into the hold of a steel compartment vessel. The compartment had not been properly cleaned and a certain amount of iron rust was there which caused much argument.
CHAIRMAN MCAFEE: The point was brought out in the discussion on the English versus the American clays that at one time Mr. Brian found them slushing the ball clay down chutes into the boat, but he did not men¬tion whether they used sea-water to slush it down or fresh water. However, I do not believe that is a common practice. It might possibly cx plain some variation in ball clay in case they use brackish water for that purpose.
E. H. FRITZ: For the casting process we have not been able to get an American china clay that will cast the very thick sections that we have to east. It is possible to cast thinner sections with the American china clay but not sections over four inches. We have had no difficulty whatever with American ball clays. We use no English ball clay in our cast body.
A. S. WATTS: Some of the English china clay plants are filter pressing their clay and some of them are still putting it on top of the driers, a tile floor where the water is dried out. There would be a certain appreciable difference. If you were buying from a concern that was drying part of the clay by one process and part by another, that would undoubtedly introduce an error
H. GOODWIN: On the whole the English china clays are fairly moist when they leave the mines. Some of them are kiln dried and they are apt to get a little drier than those otherwise dried. However, I do not think that any absorption would take place which would affect the body.
There is no possibility, except in rare cases, of any salt water getting into the clay, because if there was leak in the boat of any amount the clay would be "soupy" when it was delivered and you would know there was something the matter. The boat people on the whole take very good care to keep the clay dry. They even go so far as to discontinue discharging when there is the slightest amount of rain at the docks.
1 Presented at the Annual Meeting of the AMERICAN CERAMIC SOCIETY, Columbus, Ohio, Feb., 1925. (White Wares Division.)
