Latest from Molds/Cores

Optimizing New Sand and Core Sand Additions in Foundry Green Sand Systems

Aug. 17, 2005
The old rule for new/core sand additions is 300 lbs. per ton of metal poured, but that approach needs to be re-evaluated ? and made more exact ? for current molding methods and materials.

In order to understand the effect of core sand and the various binder systems on green sand properties and castings, the new sand's contribution to the dilution must be established. The objective of this research is to create a model where different amounts of new sand and core sand are added to system sand, so that the green sand’s properties and casting results can be evaluated. Initial work by the Green Sand Committee of CISA has shown that certain levels of new sand can be added, along with new clay/additives, to maintain consistent green sand properties. The study also found that there are critical levels where major changes in the green sand properties occur.

The old rule of 300 lbs. of new sand per ton of metal poured needs to be re-evaluated, because it may not be appropriate for current molding methods and materials. It is the goal of this project to establish a number that is more precise in terms of percentage or pounds for new/core sand additions. The number should also provide a standard test for all metals, sands, and binder systems. By effectively analyzing the effects of new- and core-sand dilution on green-sand systems and optimizing these properties, casting defects and problems could be reduced resulting in improved efficiency.

Presently, there are no guidelines to determine new or core-sand additions. Some research has been performed on core-sand dilution, but no distinction or separation of the sand versus the binder has been established. When new-sand addition rates were evaluated in the past, metalcasters were using different core-binder materials and shakeout rates. There have been numerous developments in metalcasting processes and materials since previous research was performed. In general, we use less chemical binder by percentage today and formulations have been designed so that more core-sand dilution occurs by volume. Many chemical binders are produced with additives that also improve shakeout. With these improvements much more core sand is introduced into the green-sand system. Under continuing environmental pressure, chemical binder manufacturers have developed more “environmentally friendly” chemical binder systems.

There is no current literature that states the effects of new- or core-sand dilution on the system sand.

There is no current, proven green-sand test that determines the level of core-sand dilution in the system sand. Many foundries do keep track of the core weights and make informed assumptions on dilution, based on their own criteria. But, there is no clear distinction between resin-bonded sand and virgin sand performance as a green-sand additive. It is not clear what effect older binder systems — much less newer ones — have on the system sand. By analyzing the effects of new- and core-sand dilution on green-sand systems, and optimizing these properties, casting defects and problems may be reduced, and casting efficiency may be improved.

New and core sand are additives of green-sand systems, but their effects have not been measured quantitatively. In order to understand the effect of core sand and the various binder systems on green sand properties and castings, the new sand’s contribution to the dilution must be established. The objective of this research is to create a model where different amounts of new sand and core sand are added to system sand, and the green sand properties and casting results can be evaluated. It hoped that this project will establish a more exact number in terms of percentage or pounds for new/core sand additions. It should also provide a standard test for all metals, sands, and binder systems.

Phase I research includes development of baseline data for various levels of new sand additions by weight to green sand. Phase II evaluates similar additions of core sand by type and weight. Phase III includes analysis of casting quality with both new and core sand additions. This includes varying the new- and core-sand dilution rate above and below recommended rates. The resulting data will be used to develop a simulation model to predict new and core sand requirements based on variables such as sand-to-metal ratio, amount and type of binder/cores used, and metal type. This information also will be used to develop a standard test method.


Testing evaluation

Phase I of the project aimed to evaluate the impact of new-sand addition on a traditional green-sand molding system. In order to complete this project, a prepared sand mixture was evaluated in a laboratory environment. This green-sand mixture was prepared with 8.8 pounds (4,008 grams) of return molding sand from an iron foundry in Wisconsin (round grain silica), and 0.17 pounds (78 grams) of Preblend used on the foundry’s molding line. The sand mixture was “dry mulled” for one minute and water was added to the prepared sand mixture for four minutes in order to meet the targeted compactability of 40% (±2%). This prepared sand mixture was evaluated for percent moisture; specimen weight in grams; percent compactability; green compressive strength in pounds per square inch; permeability; wet tensile in newtons per square centimeters; dry compressive strength in pounds per square inch; cone jolt toughness (represented in number of jolts); friability as a percentage of methylene blue clay; and loss on ignition as a percentage. The sand properties were tested at four-minute and eight-minute mull times.

In order to evaluate the impact of “new sand” on a prepared green-sand system, a series of prepared sand mixtures were produced. The testing laboratory prepared sand mixtures with the same 9-lb (4,086 grams) batches, except the laboratory removed an incremental amount of return molding sand from the Wisconsin foundry and replaced it with the desired quantity of new base sand used in that facility (round-grain silica). The mixtures were mulled again for four and eight minutes (as indicated) with the same testing completed on the prepared sand to the desired compactability of 40% (± 2%). The first test run included addition rates of the new sand from 10 to 60%, in increments of 10%. Obviously, the final sand mixture was prepared with a predominance of new sand versus molding sand. The testing team felt that this is not uncommon in large automotive foundries with high levels of core-sand dilution. In subsequent test runs, percentages from 2 to 35% were tested to ensure sufficient addition rates were studied.


Testing results

Evaluation of primary sand testing properties: Most foundries use compactability testing and moisture content as the primary quality-assurance control. The tests can be evaluated “in-line” with controls at the mulling and molding areas or in the laboratory testing facility. With this concept in mind, the comparison of these properties is critical. The relationship of the moisture requirement to the desired compactability is critical. Figure 1 compares the compactability measurement of the prepared sand mixture (after four minutes mulling) versus the moisture requirement to meet 40% (±2%) with the varying amounts of the new sand addition. (See Figure 1 on p. 47.)

The comparative data shows that as the quantity of new sand increases the moisture level required to produce the desired compactability (40%, ±2%) decreases. This results in an increase in the compactability-to-moisture ratio. This is logical because the addition of new sand does not have the oolitic layers on the regular molding sand, which requires an increase in moisture content to meet the desired compactability.

It has been documented in the AFS publication, Back to Basics: A Green Sand Primer, that a “well mulled green sand for high-density molding will often display a compactability-to-moisture ratio of 10.5 to 12.1. The test data in this study reveals this to be true up to the 40% level of new sand addition. Considering the desired compactability-to-water ratio referenced and the information determined in this investigation, the sand addition has an impact on molding-sand properties.

Evaluating “secondary” testing properties: A typical foundry will rely on certain green-sand testing methods used daily in their process control areas. These include (but are not limited to): green compressive strength, specimen weight, permeability, and methylene blue clay tests. For this testing protocol, the investigators held the methylene blue clay content constant so that there was little or no impact on this investigation. Therefore, the comparison of data found in Figure 2 (see p. 52) is a comparison of the impact that additions of new sand have upon the green compression strength, specimen weight, and permeability.

To gain a better understanding of the specific data developed in the laboratory investigation, Figure 3 (see p. 52.) is a graphic presentation of this information gathered in the testing laboratory when the specimen weight evaluation was completed.

This series of testing evaluations was not unexpected. It is logical to assume that as the new-sand addition increases, the specimen-weight density should increase. The green compressive strength and the permeability should have changed accordingly.

Evaluating additional testing properties: A green-sand testing laboratory has many “tools” available to help understand the characteristics of molding sand. In this investigation the following additional tests were conducted: wet tensile properties, cone jolt toughness, friability, loss on ignition, and dry compression strength. These test methods have been investigated at length and referenced in AFS Transactions and other publications. Figure 4 (see p. 53) is a representation of the resulting data at multiple mulling considerations (four and eight minutes.)

As stated previously, the data resulting from the testing evaluations were not unexpected. The wet tensile test is an excellent method for evaluating the molding sand “just below” the surface. Any time there is an increase in the sand’s wet tensile properties, it is a “good thing”. When evaluating the cone jolt toughness properties, the test is a good method for determining the “toughness” of the molding sand (or simply the ability to “draw” a deep pocket of sand.) Similarly, an increase in this physical property is a “good thing.”

When evaluating results of the friability test, note that the lower number, the better are the effects. When new sand is added to molding sand, it has been observed in actual foundry conditions to increase the friability percentage. This investigation observed a decrease in friability, and this should be considered “desirable.” The loss on ignition test was a good indicator of the sand dilution study itself. As expected, as more new sand is added to molding sand, the loss on ignition should become lower.

The reason for this decrease in the loss on ignition is that the oolitic layers that occur on molding sand contribute to the material’s volatility. For this investigation, the final property evaluated was the changes in the dry compression strength. The increase in the dry compression strength can be related to the bonding characteristics of the clay itself. Figure 5 (see p. 53), Figure 6 (see p. 58), and Figure 7 (above) illustrate some of the changes in properties that occurred at various new-sand dilution rates.


New sand addition study conclusions

The foundry industry consumes a large quantity of new sand. Regardless of the type (round-grain silica, lake sand, etc.), it is important to understand the impact that large quantities of new sand will have on the molding sand properties. In general:

  • Moisture content should decrease.
  • Compactability-to-moisture ratio should increase.
  • Specimen weight should increase.
  • Green compression strength should decrease slightly.
  • Permeability should increase.
  • Wet tensile properties should increase.
  • Cone jolt toughness should decrease.
  • Friability should decrease up to 30% new sand additions.
  • Loss on ignition should decrease.
  • Dry compression strength should increase up to 30% new sand additions.


Future testing

As stated, the CISA Green Sand Committee will continue the work on this project with particular emphasis to the dilution of core sand into green sand. In order to understand the impact that “core sand dilution” has on green-sand molding properties, it was important to study the relationship that new sand has on green sand simply within the parameters of “dilution.” Many times core-sand dilution is blamed for negative effects on the molding sand, without considering the simple concept of the dilution of sand into the molding sand itself. This investigation was designed to supply information to our understanding of dilution by any source (new sand or core sand) of materials into molding sand.

Sara Joyce, Mark Ziegler, and Vic LaFay are members of the Casting Industry Suppliers Assn. Green Sand Committee.

Sara Joyce is vice president of technical support for Badger Mining Corp. Mark Ziegler is director of foundry technical sales with Unimin Corp. And, Vic LaFay is vice president of research and technical development with Hill & Griffith Co.

The authors express thanks to CISA and its member companies for their contribution to this project.