Core sand control avoids brittle sand casting problems.
Studies using wet tensile, friability and cone jolt toughness tests showed a high core sand influx can cause brittleness in green sand. Chemical and physical effects are the two essential mechanisms involved in rebonding.
Chemical effects of condensates occur when the pyrolysis products of the core binders (seacoal or other organics) condense in the mold during the casting process. These materials are absorbed into the bentonite in the green sand, causing a deterioration of the bentonite. Wet tensile tests(1) of pyrolyzed coldbox, hotbox and shell core sands confirm the bentonite absorption. Deterioration is evident in wet tensile tests and is most serious in coldbox, less so in hotbox and lowest in shell.
Chemical acidity or alkalinity effects can also occur from changes caused by certain binder systems and some base sands.
Physical effects on rebonding can occur when a carbonaceous residue is left on core sand grains. Studies were made of the rebonding properties of phenolic urethane coldbox core sand using four different degrees of thermal exposure to evaluate rebonding properties. Conventional tests were made, including those for friability and cone jolt toughness.
These sand grains with the four degrees of thermal exposure are pictured in Figs. 2, 3, 4 and 5. Figure 2 shows particulated sand with no thermal exposure; Fig. 3 illustrates the material after exposure to heat in an oxidizing atmosphere; Fig. 4 shows the material after exposure to heat in a reducing atmosphere; and Fig. 5 depicts phenolic urethane coldbox core sand that was calcined.
The studies showed good sand rebonding properties in Figs. 2, 3 and 5, providing ample clay rebonding to the sand surfaces. Figure 4, however, shows brittle sand that was evident in friability and cone jolt toughness tests. It is covered with a carbonaceous residue caused by the same incomplete binder combustion found in a mold cavity.
When core sand grains with a carbonaceous residue enter a green sand system, they can weaken the clay bonding mechanism(2) by preventing bentonite from adhering to the sand surfaces. There is little the foundryman can do to prevent chemical effects and their interaction, but he can counteract their rebonding effects.
Solvents and other organic materials in the binder are the sources of carbonaceous residue. The more reducing the mold atmosphere, the more the tendency for carbonaceous residue. Several actions can reduce the formation of these residues:
* make sure cores are vented and tightly connected to mold vents leading outside the mold;
* for coldbox cores, use plant air heated to 140F to purge solvents from the cores (air should also be dry, -65F. The general rule for purge time is 10 times gas time);
* pouring the metal hotter and faster reduces lustrous carbon formation.(3)
Green sand composition/condition/system-related factors can also cause brittleness and/or compound their effects. Moisture/clay factors are more often responsible for brittleness than are physical and chemical effects. They include:
* low clay level;
* low moisture level;
* hot sand;
* lumps in the sand (core or green sand);
* high input of new material in the system, resulting in a fast sand turnover, leading to short mulling cycles and too little time for clay/water activation;
* system sand level too low;
* sand system capacity that is too small;
* low compactibility;
* insufficient fines level in the sand.
The sand system design has a major impact on sand quality. Silo size, screening, bond control, dust collection, sand temperature and throughput rate significantly affect the sand properties and consistency.
Return sand silos should hold sand sufficient for at least two hours of processed production sand and should be kept full. Periodically, they should be checked for a buildup that severely reduces their capacity.
Charging and discharging of silos is important. When hot sand is charged into cold bins, condensation occurs, causing sand buildup. Sand should be charged into silos at different loading doors to maximize capacity and even out return sand variations. Sand feed from the silo should also be evened to prevent a last-in/first-out effect.
Screening is not efficient if done after mixing with spill sand and water additions. Lumps should be reduced to grain size before adding sand back to the system. If the volume of sand to be screened is large, a portion of the system sand (10%) can be put through a two-step screening process using a finer 8 mesh screen.
A sine wave effect is often observed in the sand properties within a sand system. This effect is aggravated when adjustments are made in response to sand test data without proper consideration being given to the time lag between tests and the sand cycle time. Only one adjustment should be made per cycle. When any adjustment is made, further adjustments must be avoided until the sand circulates one full cycle and the results of the first adjustment can be evaluated.
Rather than adjusting the bond based on results of periodic laboratory tests, anticipatory control of the sand system may be used. Bond is added to:
* coat new material additions;
* replace deactivated bentonite;
* replace material lost to the dust collection system.
The loss from dust collection should be relatively constant unless the dust collection system is not being maintained or is not functioning properly. The bond addition can then be calculated based on the amount of burnout expected for a given sand-to-metal ratio. The ratio changes, depending upon pattern scheduling, product mix and the amount of bond that must be added to coat the new material additions.
For systems where the sand-to-metal ratio is fairly constant and the largest variable is the amount of core sand dilution, the bond addition actually can be adjusted based on the core sand dilution. If core sand input is anticipated to increase, the bond addition should increase and be made prior to the actual core dilution. This is so the system will contain an adequate amount of dispersed and water-activated clay by the time the increased core sand influx occurs.
Clay is not water-activated fully or dispersed on its first cycle through the mixer. Clay requires several passes, depending on the mulling efficiency.
In this approach to system control, the bond addition is not changed as a response to sand test data. These additions are based upon pattern scheduling or the product mix. The sand test data is used then as an important system of monitoring the effectiveness of the anticipatory control.
If the properties are erratic, variables exist in the system controls that are not being considered. Once the sand properties are constant, the anticipatory control is adequate. If it becomes erratic later, a change in some variable must be identified and corrected.
When there is a high core sand input, the clay level should be maintained slightly on the high side to enable already activated clay to coat the incoming sand, reducing the need for large bond and water additions. Excessive additions not only induce a sine wave effect, but also clay balling and lump formation. This leads to a moisture-starved system having low dry and hot properties.
When there is a large turnover of sand requiring large amounts of new material, the bond and water addition may be split so that part of the water addition is made prior to the return to the sand silos. This will allow for a tempering/activation effect. Clay takes time to absorb moisture and water additions must be controlled to avoid sand in the silos that is too wet.
Dust collection introduces another variable with a marked effect on bond requirements. High levels of inert fines and oolitic material (dead clay and ash) often accumulate in system sands, necessitating added new material to control the amount of fines and oolitic material. If excessive, they lead to low sand permeability and refractoriness and heavily oolitic sands are prone to burn-in, burn-on problems.
In many U.S. foundries, the molding sands are brittle and moisture-sensitive due to excessively low levels of inert fines and oolitic material. Part of the reason for this is increased levels of core sand dilution, which contribute to a high throughput of new material that prevents the buildup of inert fines and oolitic material. Also contributing is the increased effectiveness of today's dust collection systems.
The dust collection systems sometimes remove too much of the dead clay and inert fines (and also portions of the live clay and additives). It is beneficial to maintain a certain level of inert fines and oolitics because they help to hold moisture in the sand, making it less moisture-sensitive.
Materials removed by dust collection can be added back to the system in a controlled manner. As with all foundry sand components, problems can result from insufficient as well as excessive amounts. The optimum levels must be determined based on experience.
If dust is added back to the system, it should be tested to determine its exact composition. Testing also acts to monitor the ratio of the clay to the carbonaceous material. As the effectiveness of dust collection increases, the dust will contain higher percentages of the lighter carbonaceous material.
As production increases, sand temperature increases, especially in low sand capacity systems. Hot sand dries out as it is transported to the molding stations. If a mold is prepared from hot sand, the surface and edges of the mold cavity dry out faster. This leads to a friable mold and resultant sand inclusions. A sand cooling system should be included in the sand system design.
New Sand Relationships
If the clay/moisture relationship is good and consistent, clay will have a better chance of coating sand grains. The clay/moisture relationship and adequate mulling time for tempering and dispersing the clay are important for the success of a green sand system.
Excessive throughput of new material, whether from cores or new sand, should be avoided. As mentioned previously, the clay is not fully activated on one pass through the muller. The sand becomes conditioned as it cycles through the system. The longer the sand remains in the system, the more fully developed it becomes. Excessive input of new material leads to an underdeveloped, brittle system because the life of the sand in the system is shortened.
If there is enough sand input from cores to maintain a full sand system, then there is generally no need for a new sand addition. When the core sand input is over 300 lb/ton of iron poured, the system should be self-sustaining. If it isn't, leakage should be checked and remedied. If excessive amounts of green sand are being screened out, this sand should be conditioned and returned to the system. New sand should be added as a last resort to maintain a full system, or to control the fineness and distribution of the sand if it changes due to a coarser sand being used in the cores.
Particulated core scrap can be substituted as an addition in place of new sand but caution is advised. Particulated core scrap may have a binder content significantly higher than core sand that goes through the casting process and enters the green sand system at shakeout. If the binder on the particulated core scrap is one which could form lustrous carbon once it is exposed to heat, it may lead to a problem. Gas defects or the buildup of ammoniacal nitrogen or other chemical condensates.
Green sand systems tend to agglomerate. With a minimum of new material entering the system, it may be necessary to "deagglomerate" the sand as noted earlier.
Shakeout/Core Sand Dilution
System control is often complicated in existing sand systems by a shakeout that digests even large chunks of cores and returns all core sand to the system. The problem is aggravated further when the core dilution varies widely and the returning core sand influxes erratically into the system.
Ideally, the system should be designed so that it separates what is left of used cores from the system. The used core sand can then be metered back into the system in a controlled manner. The core sand could also be put through some type of reclamation process before returning it to the system.
1. D. Boenisch, "Effect of Cold Box, Hot Box, and Croning Cores on the Properties of Bentonite Bonded Molding Sands," Geisserei, 64 no 21, p 549-554 (1977).
2. L.J. Pedicini, M.B. Krysiak, "Coldbox Core Dilution in Greedsand, An Analysis of Effects on Casting Quality," AFS Transactions 90-50, p 151-159 (1990).
3. R.L. Naro and R.D. Tenaglia; "Formation and Control of Lustrous Carbon Surface Defects," AFS Transactions, Vol. 85, p 66-74 (1977).
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Sand Control|
|Date:||Mar 1, 1993|
|Previous Article:||Who's the real enemy anyway?|
|Next Article:||Effects of chemical surfactants on clay-bonded systems.|