US2657113A - Foundry composition and method - Google Patents

Description

Patented Oct. 27, 1953 FOUNDRY COMPOSITION AND METHOD Thomas E. Barlow, Jackson, Ohio, assignor, by mesne assignments, to International Minerals & Chemical Corporation, Chicago, Ill.

No Drawing. Application February 13, 1950, Serial No. 144,037

6 Claims.

My invention, relating generally to foundry practice, more particularly concerns not only foundry compositions, typically foundry cores, but as well, a method of producing the same.

One object of my invention is to provide a method both of producing foundry core compositions and of producing foundry cores therefrom, which method is at once simple, rapid, certain and predictable in its results, and while appreciably minimizing the manipulative time heretofore required, demands only a small number of procedural and manipulative steps, requiring only modest labor demands of but limited skill.

Another object is to provide a foundry core composition which can be quickly and rapidly produced from materials which in themselves are inexpensive and readily available throughout the country, which is relatively free from stickiness or tackiness during production, and this with moisture content appreciably lower than heretofore employed, by the use of which composition both the surface of the core itself and the resultant cast metal surface are improved, and in which, upon the inclusion therein of the usual quantities of added binder, appreciably greater tensile strength and hardness are achieved in the finished core, or alternatively smaller quantities of customary binder additives such as commercial core oil or commercial resin core binders will produce entirely satisfactory strength and hardness in the finished core.

A further object of my invention is to provide a foundry core which can be rapidly and readily formed, and which when produced, not only displays the required high green strength and moldability, but possesses required hardness of the finished core, and this with achievement of excellent surface detail in the poured casting without necessity of employing a core wash; and in which, without detrimental results, a moderate quantity of argillaceous substances may be included along with the clean silica sand, for increased green strength.

Other objects and advantages of my invention will in part be obvious and in part more fully pointed out hereinafter, during the course of the following description.

My invention accordingly may be seen to reside, not only in the combination of substances, mixtures of materials and composition of ingredients, but as well, in the several manipulative steps, and in the relation of each of the same to one or more of the others, the scope of the application of all of which is more fully set forth in the claims at the end of this specification.

As conducive to a more thorough understanding of my invention, it is to be recalled at this time that the production of foundry cores interposes a number of problems which are unique unto themselves, and which are encountered only in this phase of foundry practice. In short, the production of foundry cores is much more difficult than in the production of foundry molds in general. Now, it is well recognized in foundry practice that unfortunately, ordinary silica sand alone does not give rise to a core sufliciently hard or strong to answer the requirements which it must satisfy, either in its green or in its final baked form. Thus, the practice has arisen, starting some years ago, of employing a drying oil, a thermal-setting resin, or a thermal-setting pitch as an additive to the sand, in the form of a binder.

Such binders which impart desired baked strength, however, do not in themselves give rise to requisite green strength. Accordingly, to these binders are frequently included additives of which starch, dextrin, glue or other drying-type glues and binders are typical. All such additives to the silica sand are extremely expensive when compared with the cost of the sand itself, and when measured in the light of the use to which the mold is put.

Some practical difiiculties as concerns the physical properties of foundry core compositions have been introduced as a result of the recently growing reliance on mechanical devices known as core blowers. The two somewhat inconsistent requirements exist for use in such machines that while the sand must have cohesiveness or green strength sufficient to maintain its proper shape during handling, at the same time it must be free from adhesion to the core box or mold, i. e., it must display minimum stickiness. Moreover, core sand for blowing operations must display ready flow qualities under normal operating conditions.

An important object of my invention, therefore, is to remove or to minimize in substantial measure the disadvantages and defects heretofore confronting the art, and at the same time to approach in practice the maximum theoretical characteristics of core binders under prevailing industrial practices. Another important object is to eliminate or at least to minimize the necessity for coatings or core washes. In one sense, my invention comprises a mode of improving existing binders and of increasing the commercial utilization and ease of handling them.

And now having reference to certain typical 3 embodiments of my invention which I now set forth simply by Way of illustration, it may be noted that in general the binders heretofore employed have the effect of setting or hardening upon exposure either to air or to elevated temperatures, or both. They are frequently called thermal-setting for that reason. 'It'isan imthermal-setting :binder which is required. To

achieve successful blending of the :non-thermal --with therthermal-as'etting binders, and to :insure .co'mpatability of my non therma'l setting binder :both with the thermal-setting binder and the wvaterscontentof the -sandplemploy an emulsifying :agent and an aromatic solvent for the :therino-plastic ihydrocarbon.

As 'has already been suggested, my invention "has applicability .to 'all' -three known types of thermal-setting binders,includingthe oils, resins and pitches. Within'each of these threeclasses ."the modifications possi'ble are-too numerous for ready discussion. They have, however, in common the :general characteristic that they set or "harden in such manner as to *bond the sand to- :gether in :ahard stable mass through the application of air or heat, or both. My first discusfsio'n, therefore, will be directed'to the -treatnient of the oiltype of binder.

It .in'ay be stated without fear of successful contradiction that the' core "binders most commonly in use-at the present time are-the oils.

Th'e'se'binders are either forme'd entirely of, or

contain, "an appreciable amount 'of polymerizing "or'drying oils, suchasftung-oil,soy o'il, and similar "thermalor air-setting organic oils, Since they display little or no ability to'bond sa'ndat room temperature, that issince in 'foundry parlance "they display no "greenk strength, these oils are I commonly 'admixed with cer'eals, dex- .:trin," s'tarch,clay'or other adhesives. 'While these additives iinpart requisite green stren th, there "are certain undesirable features 'attendant upon their use, as will be' pointe dout. Moreoverj-since these drying oils oxidize to at least some extent at room temperatures they display "a certain amount of stickiness, i."e., absence enem 'fiow qualities. n

Now, I'have discovered that when suchbinders, either alone or when admixed -'-'with substances imparting green strength thereto, are combined in proper manner with a"the'rmeplastic or "nonthermal-settinghydrocarbon, eitherve'getableor mineral, "and "when this "last-mentioned hydrocarbonissuspended;along with' a' suitable emuliiying agent, *in proper: carrier, this 5 charaeteristic stickiness is'no longer evident.

Perha'ps the 'most active or" impor tanting'redh entin m new additiveisthe non thernial s etting hydrocarbon-resin. gl for this "purpose "I employ 'a "petroleum base resin having jafrn'elting or sbftnine ipf enrbr m itlr ark-re 1 decomposition point of approxiiriately 600 B. Its ductility or plasticity at room temperatures is 4 such that it is not activated by aromatic solvent or by bonding agents, such as drying oil, thermalsetting resins or any solvents which are normally used in introducing such binders into core sand. This latter point is most important, for should the thermoplastic hydrocarbon resin be plastiicized or activated bythe fsolver'itsgit yvill produce a continuous phas'e'a nd"thereby' dominate the physical properties of the core. Because of the non-thermal-setting qualities of the resin, therresultantcore; in which its physical propertis'p'redoininatevviIl be soft and weak.

.It1-is iinportaIIt -that the resin employed have ra r'r'iie'ltingfir'iointf below that at which drying oils terminal-setting resins become set or hard.

jlhat is the melting point must not exceed 400 F. --Eor, if this requirement is not satisfied, then the.thermo-plastic resin would quite likely enter intoihe bonding phenomenon. Or, because of its thermoplastic characteristics, it would detriir'ientally reduce the hardness and strength of the baked-core. With a melting point below the setting"orhardening-temperature of the thermalsetting resins, however, the thermo-plastic resin melts and serves to coat the sand grains before "hardening takes places, an'din'this manner efiectively removes'itself from the binding phenomenon. Thus removed, the'melted thermoplastic 'resin'does not detract' irom' the bakedcore properties. in this regard, Vanadiset'is found to be *eminently'satisfactory.

VaIla diSetdS a non thermal setting petroleum pitch which'i's substantially free of products that are volatile=-be1pw--about 550 F. It is black in the mass, fractures conchoid'ally and gives a brown streak on porcelain. "The specific gravity at 77 F. is'abou't 1.1 2 andfthemolecularweight about '3200. It "softens at ITO-F. to 340 and is soluble in carbon disulphide, benzene and other aromatic solvents. Van'adiset essentially consists of asphaltenes 20% to 70%, 'asph'altic resins 5% 'to 15%, and oil constituents 26% to '51%,-"with carhenes notover about 1%. Chemical analysis reveals that it contains about 21% "to-641% unsaturated hydrocarbons, withthefixed Carbon cofitent amqlll'iting-to about 36% to 18% and the hydrogen content-'-about*9% or less. The ash content is' 'l'ess than ofwhich-van'a'dium *is i the principal constituent. (see, United States Patent $2,409,437 "and the Biilltin'of October -1, 1 947, of the Wilson carbon company, Inc., "60 East 42nd stlet'Nw"York 17, New York.)

From "a purely theoretical position, and in order' to impart-maximum flow qualities to the 'sarid arid maximumireedo'm from stickiness-as well as'to keepthe moisture content of the'sand toa minimum,*l*wouldprefer to employ 'for the theijmo-plastic additive -a*biii'der comprised of the 'the'rnioplastic hydro-carbon. From a practical standpoint, however, and from necessity, I introduce along therewith both a'solverit and an'einiilsifier. Thissolvent serves asacarrier for the thermo-plastic resin and permits the same t'o be'readilydistri-buted"through the sand mixture. The solverit-employedfrnust be com: patible' with the resin which I u'se; and through experience I find aromatic petroleumsolvents to be preferable. have,-'however, usedkerosene arid ininer'ahspirits with s ome 'success. Since itis quite *likely' that the solvent itself probably detracts to someeX-tentfrorn themaximmn benefits "attendant upon the 'ga' ddition of thermoplastic resin, I 'i'riaintain the quantity or" such solvent' at afminimuinconsistent with obtaining workable viscosity and fluidity in the combination, thesolvent amounting only to some as a minimum.

Moreover, since water is the moistening element usually employed in foundry compositions, particularly core sands, my additive must not only be compatible with thermal setting resins and thermal-setting pitches, but as well, reasonably compatible with water. To satisfy this requirement, I employ an emulsifier. To this end, I incorporate in the additive up to 5% by weight of a soap or a combination of tall-oil and ethylamine. Other than to insure the compatibility of the thermo-plastic resin with both water and the drying oils and thermal-setting resins, the emulsifier does not enter into the action, and has no direct bearing on the core eifect obtained. Similarly, stearines, lubricants or humectants may be included in the basic additive to improve certain results thereof. These, however, are in no sense a substitute for all or any part of the basic additive.

A simple and typical embodiment of the basic additive is as follows:

Percent Vanadiset resin about 31 Aromatic solvent about 66 Soap (12% diethylamine88% tall-oil) about 3 In treating foundry core sand, the sand first is mulled with core oil, this amounting to some one to six quarts for 1000 pounds of silica sand. Along toward the end of the mulling cycle, I introduce my new additive in the amount of onehalf pint to two quarts for the proportions of sand and core oil noted. The mulling is then continued until thorough mixing is had and the treated core sand is ready for use.

When my new additive is employed in small amounts, say from one-half pint to about one quart per 1000 pounds of core sand containing a core-oil mixture, either with or without other additives imparting green strength thereto, a marked improvement is displayed in the ability of the sand to flow into position in the core box. Moreover, a marked decrease is observed in the formation of sticky films on the core boxes, on the tools, and on the person of the operators. The presence of such additives is found to prevent plugging or closing of screens, vents and other portions of equipment used for blowing the core sands into the core box. All of these advantageous features sum up to the fact that effective use of blowing machines can be achieved with foundry compositions produced according to the practice of my new invention.

As an illustrative embodiment, in a 1000 pound batch of sand wherein I employ a linseed-base drying oil, together with cereal or corn flour to impart green strength thereto, with water admixed thereto I find that the inclusion of onehalf pint of my new additive, comprising a thermo-plastic hydrocarbon in the presence of an emulsifier and a solvent, permits a reduction of one pint in the amount of drying oil used, as W811 as the elimination of the approximately onehalf pint of kerosene which was formerly required as a lubricant or release agent. That is, where three pints of drying oil were heretofore required, by my practice this can advantageously be reduced to two pints.

Moreover, the necessity of cleaning or washing the core boxes with kerosene or lye to remove the adherent sand is completely eliminated. Finally, the average weight of the blown core I find to be increased approximately 6%. 'This indicates a substantial improvement in density without noticeable increase in cost. Improved surface detail is observed. This increase in density in turn apparently is attendant upon the improved flow qualities of the sand following upon the inclusion of my new binder additive. Smoother cores are observed to result, with accurate attention to surface detail, and with minimum penetration of the molten metal into the face of the core. The necessity for dipping or washing the cores is minimized and the smoother casts resulting are accompanied by a reduction in the number of rejects occasioned by improper dimensioning of the cored sections of the castings.

I have discovered that the inclusion of thermoplastic hydrocarbon in proper manner in accordance with the teachings of my invention appreciably diminishes the detrimental stickiness of the core composition as heretofore observed. Probably because of this, I have further observed that when my new composition is employed, there is no longer necessity for frequent cleaning of the blowing equipment and of the flasks to remove the sticky film formations. This advance in the art assumes considerable importance when it is considered that I materially increase the rate of core production by decreasing both the number and duration of non-productive steps and the amount of labor employed in connection therewith. Moreover, absence of stickiness is accompanied by appreciable reduction in the plugging or obstruction of vents in core blowers and the like. That this last feature involves appreciable improvement in quality of product is evident when it is considered that when carried to a certain point obstruction of these vents prevents the proper manufacture of a sound or solid core. This is due to the fact that the free egress of air from the flask through the vents is blocked. The trapped air creates back pressure which, building up upon flow of sand therethrough, eventually prevents the sand from filling the core-box cavity and conforming to the outline of the core.

It has long been recognized in the art that the pick-up or absorption of water by the baked core, an undesirable characteristic of baked core composition, sharply reduces the strength and hardness of the core. In this respect I have observed a further highly important advantage inherent in the use of my new composition in that the finished core formed of such composition displays a strongly water-repellant characteristic. This is probably due to the fact that the thermo-plastic resins which I use are in themselves essentially water-repellant, in the absence of the proper emulsifying agent and organic solvent. Thus, in the green state, after packing about the mold, it is true that these core compositions are waterabsorbent. However, the emulsifying agent and organic solvent are driven off and destroyed during the baking process. The thermo-plastic resin retained in the core following the destruction of the emulsifying agent and organic solvent I find imparts its water-repellant characteristic to the baked core in highly satisfactory manner.

Among the advantages attendant upon my new practice may be listed the reduction of undesirable gas formations. Moreover, it appreciably enhances the economy of the core formation in that it frequently permits a substantial reduction in the amount of drying oil or thermal-setting binders employed. It is highly significant that this decrease in amount of thermal-setting binders can be achieved despite the fact that portion to the thermal-setting resins than'in the case of drying oils, in order to achieve satisfactory results. Probably this is due to the in herent stickiness of the thermal-setting binder. When employing say a 62% solution of phenolresin in water as the thermal-setting resin, I have achieved excellent results by admixing approximately one-half pint of 31% solution of thermal-plastic resin with one quart ofthe water solution of phenol-resin. This is added to 1000 pounds of clean silica sand. The foregoing is to be taken as merely illustrative of the general range of proportions. It is by no means intended to be limitative.

The permissible range of ingredients in my core composition is quite broad, ranging from some one pound of the thermal-setting resin and onehalf pound of the thermal-plastic resin up to about thirty pounds of each for 1000 pounds. of sand. So too, the permissible relative ranges of the two ingredients may widely vary. The lower limit of the thermo-plastic resin depends upon the stickiness of the thermal-setting resin involved. If the stickiness of the latter is high, then a larger amount of the thermo-plastic resin must be employed. The upper limit is determined somewhat from a practical standpoint. In general, it is the point at which the thermo-plastic hydrocarbon predominates the core mix, and so predominating, reduces the hardness strength of the core below certain critical values.

The thermal-plastic resins do not in themselves impart adequate baked strength or other baked properties, and I do not believe that the use of the thermo-plastic resin would be practical from a commercial standpoint where it would be present in an amount greater than the quantity of thermal-setting resin employed. As an upper limit then, the composition would comprise 50% thermal-setting resin by weight, and 50% thermal-plastic resin. I find, however, that no apparent advantage attends upon using more than a minimum quantity, the amount needed to limit stickiness within a workable range, since the thermo-plastic binder is intended simply to permit utilization of thermal-setting binders with maximum effect, and is not in itself intended to serve as the strength-imparting binding element. This amounts to about by weight. I will discuss, however, at a later point herein, one technical exception to the foregoing.

I have suggested hereinbefore that one compromise solution of the problem of the increased stickiness resulting from high moisture content in the presence of thermal-setting resins has been to employ release agents, which have the effect of minimizing this stickiness. The necessity of this practice has been deplored by the art, because these release agents impart undesirable physical properties to the baked core, or at least, detract from the desired good qualities thereof.

I have found, however, that an important advantage attendant upon the proper inclusion of thermo-plastic resins in the sand composition is that the commonly used release agents such as stearic acid, oleio acids, or fatty acids ofhigh molecular weight no longer detract from the physical properties of the baked core. On the contrary, and quite apart from the common expectation, I find that such release agents, when incorporated into foundry compositions in which a thermo-plastic resin is properly included often results in an actual increase in the desired qualities of the finished core. In no sense, however,

and.

do they comprise substitutes for the basic ingredients of my new composition.

To illustrate the practical value of my invention, it may be noted that I have already successfully and commercially placed my invention in use in an operation which had theretofore been typical of the practice employing thermalsetting resins. Bythe proper incorporation of my thermo-plastic resins therein, I find that-for any given moisture content, cores of greater strength and hardness are produced than had hitherto been possible. Particularly are the desirable properties of such core sands enhanced when moisturecontents of below 3% are employed. Moreover, although moisture contents below 3% are desirable for core-blowing techniques, so thatincreased flow qualities are obtainable, I find that blown cores can be readily produced according to the practice of my invention with moisture contents as high as say 4.2%, and this without the usual stickiness and lack of flow qualities ordinarily attendant upon such elevated moisture content.

In the practice of my invention, I find that the ready-flow qualities of my new foundry composition effectively increase the rapidity of core formation, and thereby achieve a corresponding reduction in production costs. Further, important reduction in operating costs attend upon the elimination from the composition of the release agents which counteract stickiness, an elimination which, desirable but not essential, has heretofore been virtually impossibleof achievement. Also eliminated in substantial measure are the use of counteracting adhesives, which serve as green strength binders, and which materials in themselves represent a substantial element of cost. The phenomenon of wetter feel. of the core sand embodying my new composition for a given moisture content suggests a greater com-- patability between the water and the thermalsetting resin. This is a possible further explanation of the success of my invention.

I have mentioned hereinbefore that the thermo-plastic resin improves the finish detail of the resultant casting. Further to improve this finished appearance, I may, where desired, introduce additional carbon into the foundry composition, thereby completely eliminating the necessity for coatings and core washes, and this without detracting one whit from the value of the new process. In point of fact, my new composition actually permits the use of certain forms of carbon which previously could not be used because of their tendency to soak up the binder, such as drying oil, or otherwise to detract from its effectiveness.

If handled properly, the thermo-plastio hydrocarbon which I employ may be used as the additional carbon material, without detracting from the effectiveness of my invention. In this instance, the amount of thermo-plastic hydrocarbon may actually exceed the amount of thermalsetting binder such as phenolic formaldehyde or urea-formaldehyde. This is the technical exception to which I have hitherto referred.

Thus, it will be seen that there are provided in my invention a method and composition in which there are achieved many new and unforeseen advantages. When my composition is incorporated in a sand mix along with a thermalsetting or air-setting binder, eflicient mulling is obtained in absence of excess moisture. The core sand may be readily blown into position about the core pattern, giving sharp definition. The sand flows. freely. without. stickiness3 Necessity of cleaning the blowing; apparatus andthe. mold-is materially diminished. Use or-release agents is diminished; Allthese and many other highly advantageous results attendupon the practice of my-qinvention, and this while obtaining castings of" sharpdefinition and close dimensions; without the necessity of facing washes and-with diminished laborcosts.

Once the broad. aspects of my invention are disclosed, many embodiments. thereof will readily suggest themselves to those-skilled in the art; Accordingly, I- desire the foregoing disclosure to be considered as simply illustrative, and not by wayoflimitation.

I claim".

1. Method of, producing a foundry core sand composition comprising. admixing with sand a sufiicient' quantity of binder possessing thermalsetting qualities to impart required'strength and hardness to the baked core imparting improve-d flow qualities-:w-hile.relieving stickiness. and the amount of required moisture-by introducinginto the: mix a. binder: additive, the binder additive consisting; essentially ot 90% to 30% Vanadiset resin-binder, 10% to:70%- aromatic solventand effective amountsup to 5% ofsoap emulsifier, thervanadisetresin. binder amounting to' from to-50% of: the combined'weight of the two binders, and mixing the additive thoroughly with the sand and binder.

2;.The: method of producing a foundry core sand-composition:comprising admixing with sand in the; proportions of l',000 lbs. ofsand, 1 to 6' quarts of thermal-setting; drying oil, imparting improved: flow' qualities- While relieving stickiness and. the amount of required1moisture, by introducing into the'mix /2 pint to 2 quarts'of Vanadiset resin. binder in. at least 10%- aromatic solvent. and-effective amounts up-to 5% of soap emulsifier, with .the .Vanadiset amountingto from 10% to 50% of the combined weight) of drying oil and Vanadiset,.and mixingthev additivethoroughly with the ,sandandbin der.

3. The methodof. producing a foundry core sand, composition. comprising admixing. in. the proportions of lilb. to. 30.1bs. of thermaLsetting resin, to 30 1bs..of Vanadiset. present. arc.- matic. so1vent1in the. amount 01 at least 10%, by weight of Vanadiset,.and. efiective. amounts. up. to

5% of soapemulsifier, WithLthfi, Vanadisetresin amounting; to from 110% to 5.0%; of the:-,combined weight oi resinaand Vanadiset, and adding the same to 1,000 lbs. of core sand: and thoroughly mlxmg;

4. A foundry coresand composition essentially consisting of the proportions: 1,000 lbs. of core sand; 1 to 6 quarts ofdrying. oil, pint to 2 quarts of Vanadiset resin binder' dissolved.- in 10% to 70% byweight of; aromatic solvent. and efiective amounts up to 5% by weight of. soap emulsifier; with. the-Vanadiset amounting. to :from 10% to of the'combined weight-of. drying oil and Vanadiset, said composition having improved-flow qualities, less stickiness and reduced moisture requirement ascompared to composie tions ofthe same ingredientswithout solvent and emulsifier.

5. A foundry core sand. composition consisting essentially of the proportions: 1,000 lbs. of silica sand; 1 to 30 lbs.. of. thermal-setting resin,, and /2 to 30 lbs. of Vanadiset resin binder presentin aromatic solvent in. the amountof 10% to. by- Weight of Vanadisetand effective amounts-up to 5% by weight of. soap emulsifier, with. the Vanadiset. amounting to from 10% to 50%. of. the combined weight of resin and Vanadiset, said composition having. improved flow qualities, less stickiness and reduced moisturev requirement. as comparedto compositionsot the same ingredients without solventand. emulsifier..

6. A binder for foundry coresands consisting essentially of thermal-setting resin. and. additive consisting of. to. 30%. Vanadiset resin,,10% to 70% aromatic solvent, andeflective. amounts up to 5% of soap emulsifier. consisting of ami'nor amount of. di-ethyl'amine. and. av major. amount of. tall-oil.

THOMAS E; BARLOW;

References Citedin the file of this. patent UNITED STATES PATENIS- Number Name Date 2g004g344 Quandt June" 11 I935 20.04 345: Quan'dt .Jun'e%11', 1935 2510133-30i De Laney- Dec. 7, 1937 2,422,118.: Meyer June 10, 194-7 2,444,413- Weston: July 6,1948 2,468,956? Burg May 3', 1949 2 5035088: Barlow: Apr..4', I950

Claims (1)

1. METHOD OF PRODUCING A FOUNDRY CORE SAND COMPOSITION COMPRISING ADMIXING WITH SAND A SUFFICIENT QUANTITY OF BINDER POSSESSING THERMALSETTING QUALITIES TO IMPART REQUIRED STRENGTH AND HARDNESS TO THE BACKED CORE, IMPARTING IMPROVED FLOW QUALITIES WHILE RELIEVING STICKINESS AND THE AMOUNT OF REQUIRED MOISTURE BY INTRODUCING INTO THE MIX A BINDER ADDITIVE, THE BINDER ADDITIVE CONSISTING ESSENTIALLY OF 90% TO 30% VANADISET RESIN BINDER, 10% TO 70% AROMATIC SOLVENT AND EFFECTIVE AMOUNTS UP TO 5% OF SOAP EMULSIFIER, THE VANADISET RESIN BINDER AMOUNTING TO FROM 10% TO 50% OF THE COMBINED WEIGHT OF THE TWO BINDERS, AND MIXING THE ADDITIVE THOROUGHLY WITH THE SAND AND BINDER.