A basic green sand control program.
The green properties tests that appeared last month in Part 1 of this series quantify the physical/mechanical properties of the sand such as strength characteristics. Structural properties are also commonly used. Structural properties tests provide information on sand composition, such as the amount of clay and the amount of carbonaceous material in the sand. (Moisture could also be considered a structural property because it is the quantity of water in the sand.)
While the green properties tests should be run daily as often as practicable, the structural properties tests may be conducted weekly unless the variation in the process indicates the need for a greater testing frequency.
Two clay tests usually are performed--AFS or 25 Micron Clay, and the Methylene Blue Clay Test.
AFS Clay is a weight percent measurement of particles less than 20 microns. It is performed by washing a sample of molding sand to remove the clay, drying and reweighing the sample to determine the percent loss. This procedure is called a clay wash. AFS Clay particles include the live clay plus dead clay. AFS Clay also includes a portion of the fines and of the organic materials in the sand that wash out with the AFS Clay in the clay washing procedure.
In this test, a dried representative sample of molding sand is placed in a tall-form 1000 ml beaker. The sand is then pretreated for five minutes with water and a dispersing agent in a high-speed mechanical stirrer to loosen the clay before performing the clay wash. After this five-minute pretreatment, the beaker is positioned on a device that will automatically wash out the clay.
The washing cycle consists of filling the beaker to a level of 6 in., agitating the sand as it fills. The sand is allowed to settle (10 min), and then the first 5 in. of water are siphoned off. This cycle is then automatically repeated. After two 10-minute cycles, the same process is repeated, but with five-minute settling periods before siphoning and refilling.
The first 5 in. of water siphoned off contain the suspended AFS Clay particles. AFS Clay particles are those less than 20 microns. The sand quickly deposits as a sediment on the bottom of the beaker. The AFS Clay particles fail to settle at a rate of an inch per minute and they remain suspended in the water, which is then siphoned off. In this way, the AFS Clay is removed from the sand based upon rate of sedimentation.
The AFS Clay wash can take several hours with high clay molding sands, but the washing device performs the cycle unattended.
Once the test is complete and there are no more suspended AFS clay particles, the inch of water remaining in the beaker is decanted off through filter paper. The filter paper and beaker's contents are completely dried in a laboratory oven at 220-230F for several hours.
After drying, the sand and any residue in the filter paper are transferred to a balance pan to be reweighed. The AFS Clay content is calculated from the weight loss.
25 Micron Clay
The 25 Micron Clay Test provides an alternative to the autoclay method of performing a clay wash. This test can be run much faster than the AFS Clay Test.
In the 25 Micron Clay Test, the clay is separated from the sand based upon particle size rather than by rate of sedimentation. A 25-micron sieve basket and an ultrasonic scrubber are used to perform the separation.
The sample is prepared exactly as described for the autoclay method as far as drying, weighing into a tall-form 1000 ml beaker, and washing the sample with a high-speed mechanical stirrer and a dispersing agent for five minutes.
Following this, the 25-micron sieve basket is positioned in an ultrasonic scrubber. A jet hose provides a steady stream of water. The beaker's contents are carefully rinsed into the 25-micron sieve basket.
The ultrasonic action facilitates washing of the sample and prevents blinding of the screen. Particles greater than 25 microns are retained in the sieve basket, while particles less than 25 microns are removed. (AFS Clay, by definition, includes particles less than 20 microns, but the methods are comparable. Slight differences can result if the sand contains particles--such as woodflour--that are large and light. These particles will be retained in the 25 micron method, but are removed in the autoclay one.)
During the washing, the effluent from the ultrasonic drain tube is sampled with the tall-form 1000 ml beaker and observed. Early on, the water will appear murky. A few minutes later, it will look cloudy as the clay is removed. When the effluent is finally crystal clear, the wash is complete. This usually takes only about five minutes.
The sample is then dried with an infrared dryer in less than 10 minutes. After drying, the sample is reweighed and the weight loss is used to calculate the 25 Micron Clay Content.
If the MB (live) clay is subtracted from the AFS/25 Micron Clay, the difference is a ballpark figure of the amount of dead clay. As the difference becomes larger, this indicates a buildup of dead clay, ash, fines, etc. in the sand system. If the buildup continues and is not diluted with new sand, permeability of the system may be reduced (which could lead to gas defects). Also, dead clay has a lower refractoriness than silica and the metal can fuse with dead clay resulting in penetration defects.
If the difference between AFS/25 Micron Clay and MB clay is small, this indicates little buildup of dead clay. This is not necessarily a problem, but it can be associated with brittle sand, since this sometimes occurs with insufficient mulling due to a high influx of core or new sand and bond.
Performing the clay wash by the autoclay or 25 Micron Clay methods allows for the determination of AFS or 25 Micron Clay content. The clay wash also serves as the sample preparation procedure for the screen analysis. Once the molding sand is washed by the AFS Clay or 25 Micron methods, a screen analysis can be performed.
AFS Grain Fineness
The screen analysis determines AFS grain fineness and sand distribution. The fineness of a molding sand can change as it is used in the casting process. Weak sand grains break down and generate fines. Dust collection removes fines and causes the sand to become coarser.
If a core sand is used and has a fineness different than that used as the new sand, the core sand will cause a change in the fineness of the molding sand when it becomes mixed in at shakeout. Many factors can cause the sand fineness to change. The screen analysis provides a tool for monitoring these changes.
The fineness of the sand affects casting quality so it must be monitored and controlled. Coarse sands and sands with a narrow distribution (sands having mostly the same size grains) have low surface area, low binder requirements and high permeability. If the sand is too coarse and there are large voids between the sand grains, it is possible for the metal to mechanically penetrate at the mold metal interface and produce a rough casting surface finish.
If the sand is fine and has a wide screen distribution (sand grains of many sizes), it will produce a smoother casting surface finish. But if the sand is too fine, permeability may be reduced to a level where the gases produced cannot escape readily and gas-related defects such as porosity, pinholes, blows and misruns are likely to occur.
In the screen analysis, a standard set of testing sieves is arranged in a stack from coarser to finer. The sample is weighed and transferred to the top of the sieve stack. The sieve stack containing the sample is then placed in a sieve-shaking device and a timer is set for 15 minutes of agitation. The sieve-shaking device simulates rotary and tapping motion of hand sieving, but in a more consistent mechanical manner.
After agitating, the weight of the sand retained on each of the sieves is recorded. Once the screen analysis is completed, the AFS Grain Fineness Number and Screen Distribution may be calculated. The screen number of a sand is the number of adjacent sieves that contain at least 10% sand.
The screen analysis is an important tool for monitoring changes in the grain fineness and distribution of the sand as it is recycled.
It also is used sometimes to monitor the fineness of incoming shipments of new sand. The determination of fineness of new unbonded sand presents a special problem. When a screen analysis is run on unbonded sand, it is necessary to go through a sample reduction procedure. Unbonded sand has a natural tendency to segregate. The coarse particles separate from the fine particles.
If a small sample of sand is taken from a large segregated sample of sand, the screen analysis will depend upon where the sample was taken with respect to the bulk sample (i.e., whether it was taken from an area where mostly coarse grains were concentrated or where mostly fine grains were concentrated).
To obtain a sample that truly represents the fineness of the entire bulk sample, obtain a sample of the molding sand using AFS sampling procedures (see AFS Mold & Core Test Handbook). Then, reduce the sample using a sample splitter. A gate type splitter is recommended.
To use a sample splitter, the hopper is filled evenly with the sample to be reduced. The splitter gate is opened and the sand is allowed to flow through the chutes. In this process, the sample is halved.
Once the sand flows through, the pan on one side of the unit is removed and the sand is placed back in the hopper. When the sample flows through a second time, one quarter of the original sample will be delivered to each pan. Repeated passes will produce eighths and so on. This process is repeated until the sample is reduced to the size required for testing.
The weight for the screen analysis should be about 50 grams. Smaller samples can cause higher weighing error and larger ones can cause inefficient sieving due to overloading on certain screens. For efficient screening, no sieve should retain more than 35 grams in the final analysis.
Methylene Blue Clay Content
The Methylene Blue Clay Test allows measurement of "live" clay in molding sand. Live clay is capable of acting as bonding material. Molding sands also contain "dead" clay, which has been so close to the mold metal interface that it was exposed to temperatures high enough to destroy it to the point where it can no longer rehydrate and contribute to bonding.
AFS and 25 Micron Clay, as previously described, are measures of the sum of the live and dead clay. In controlling foundry molding sands, it is important to know how much of the total clay is live clay. The Methylene Blue Clay Test is used for this determination.
In the MB test, a sample of the molding sand is subjected to ultrasonic or boiling action with a dispersing agent to disperse the clay and break up fines, clusters, clay balls and heavy clay coatings. This is necessary because the methylene blue dye is adsorbed to the clay rather than absorbed by the clay.
With adsorption, the dye is electro-chemically attracted to the ion exchange sites on the surface of the clay particles. It will not absorb, like a sponge, into lumps of clay, but rather it will attach to the surface.
After dispersion, methylene blue dye is added and the sample is stirred for two minutes to give the dye a chance to be adsorbed by the clay. A drop of the clay/dye mixture is then placed on a filter paper. If the clay adsorbed all the dye, the drop will appear as a dark blue dot on the paper with a ring of water/dispersing agent around it. Methylene blue dye is then added in 1 ml increments and stirred for two minutes after each addition. A drop is placed on the filter paper until the clay can no longer adsorb any methylene blue.
At this point, the excess unadsorbed methylene blue will appear as a halo surrounding the dot. If the halo persists through an extra two minutes of mechanical stirring, the test's endpoint has been reached. The number of milliliters required to reach the endpoint determines the percent clay in the molding sand. By the AFS procedure, bentonites usually adsorb about 5 ml of methylene blue per percent clay, so an endpoint of 40 ml corresponds to a clay content of about 8%.
It is important to know how much live clay is in the system so corrections can be made if green strength is low. If green strength is low and the MB clay level is also low, more bond must be added to the system. If green strength is low but the MB clay level is satisfactory, the muller may need maintenance.
If the MB clay level of the system is kept to a minimum, the cost to maintain the sand system will be lower, and the sand should be flowable. If it is too low, it may not be possible to draw pockets in patterns and the mold may not have enough strength to hold together on automated mold handling systems.
If MB clay is too high, the molds will be stronger, but difficult shakeout, poor flowability and higher system cost from unneeded clay and sand carryout are problems. Since the sand's moisture requirement will be higher, it may be more difficult to obtain desired compactibility in minimum mulling time.
Effective clay can be determined from a chart (available from AIMCOR or Fischer/Dietert) based on green strength and compactibility. Effective clay is the portion of the live clay activated by mulling and contributing to bonding. The ratio of effective clay to MB clay is mulling efficiency:
Effective Clay(*) x 100 = Mulling Efficiency MB Clay
* Determined from charts based on green strength and compactibility
For example, if the MB clay (live clay) of system sand were 8% and the effective clay were 4%, mulling efficiency would be 50%. The other 4% portion of the live clay would be latent clay (live clay that is inactive).
Mulling efficiencies of 60% are typical for ferrous systems, while 90% is typical for nonferrous ones. This is because ferrous metals are poured at higher temperatures and more clay is burned out. Thus, more clay is continually added, and clay requires several passes through the mixer before it becomes activated by mulling. In nonferrous systems, less clay is burned out, so less clay is added. Since the same clay is recirculated, more of the total amount of live clay becomes effective.
Loss on Ignition
Organic materials sometimes are added to copper-base and ferrous systems to act as a cushion to reduce expansion; to provide a reducing atmosphere to prevent chemical reaction-type penetration (in ferrous systems); to provide sand peel; to increase flowability; and to lower hot strength and increase hot deformation of western bentonites.
Cereal, which is sometimes added in small amounts to ferrous systems and in larger amounts to steel molding sands, acts differently than most of the other organic additives in that it decreases flowability and increases hot deformation while maintaining hot strength.
For sands that contain organic additives such as seacoal, woodflour and cereal, the LOI test (sometimes called the combustibles test) is an important tool. In this test, the organic material in the sand is measured by determining the weight loss after the sand is fired and the organic materials burn out.
There is some error in this test due to water of crystallization in the clay, carbonates and other impurities in certain base sands that contribute to the LOI. Also, metallic materials such as chromite sand and residual metal in the sand produce an error by gaining weight during the test as they convert to oxides. Gas evolution tests provide a faster and more accurate alternate, but the LOI test is more commonly used.
Recommended crucibles for the LOI test are low-form, quartz open-dish crucibles. This aids in thorough combustion by allowing high surface area exposure of the sand while firing. Fused silica crucibles also are used sometimes. These crucibles are heavier, more rugged and are recommended for nonferrous sands or any sands that have a tendency to crack quartz crucibles because they contain materials (tramp metal, etc.) that can melt and burn into the quartz crucibles during the test.
Note that both types of crucibles will hold a dried 50 g sample, while exposing a large surface area of the sand to the heat of the furnace. This is important to reduce weighing error and to ensure thorough combustion.
For the LOI test, the furnace should be preheated to 1800F. If the crucibles are new, they must be prefired for about one hour. This is necessary because new crucibles may lose weight the first time they are fired and cause an error.
To begin the test, a representative sample of molding sand is dried at 220-230F. The sample must be dried so the weight loss includes only the organic material in the sand and not moisture. The empty crucible's weight is recorded, then about 50 g of the molding sand are transferred and the combined weight of the crucible and sample is recorded. Weights are recorded to hundredths of a gram. Following this, tongs are used to place samples in a furnace heated to 1800F. They are fired thoroughly to constant weight. On a 50 g sample of clay-bonded molding sand, this usually takes two to three hours.
When the samples have reached constant weight, they are carefully removed from the furnace with tongs and placed on a refractory pallet to cool to below red heat. The crucibles are then transferred to a desiccator. The desiccator contains silica gel which absorbs moisture and prevents the sample from picking up moisture from humidity in the air while cooling. After cooling, the samples are reweighed and the weight loss (in percent) is recorded as the LOI at 1800F.
In general, for sands used for ferrous castings, if combustibles are high, the sand will have a lower expansion tendency and the casting will have a good surface finish. If too high, excessive smoke and fumes, and gas-related defects such as porosity, coldshuts and misruns are likely problems.
If the combustible material is low, the cost to maintain the system will be lower, the sand will have a lower moisture requirement, and will produce less gas. If too low, however, difficult shakeout, poorsand peel, burn-on, burn-in, penetration, sand losses, hot tears and expansion defects (such as veining, rattails, buckles and scabs) are possible problems.
The Volatiles Test
An optional and useful structural properties test is the volatiles test. It is similar to the LOI test, only the sample is heated at a lower temperature in a covered crucible for a relatively short period. The volatiles test differs from the LOI one in that it measures only a portion of the combustible material. Volatile material is the portion of the combustible material that comes off quickly at low temperature. It is the fresh, or live, combustible material.
The volatiles test at 900F must be run very carefully and with considerable attention to detail. If run according to the following procedure, it is very reproducible.
A muffle furnace is heated to 900F and the heating chamber is allowed to soak at this temperature for at least an hour before running the test. Next, an empty volatiles crucible and cover are placed in the muffle furnace and preheated for at least 15 minutes. This preheating of the crucible is necessary to minimize weighing errors by weighing the crucibles and sand while still warm.
After the preheating period, the covered crucible is removed with tongs from the furnace and placed in an opening in a refractory brick pallet. The crucible is allowed to cool for 10 minutes.
At the end of the cooling period, a representative sample of molding sand (about 60 g) is put in a forced air dryer and dried at 220-230F for five minutes. This time and temperature are sufficient to dry molding sands to constant weight without loss of volatile material.
The warm crucible and cover are then transferred to a balance that weighs to 1/100th of a gram and the weight is carefully recorded. Following this, about 50 g of sand are weighed into the crucible, the crucible is covered, and the combined weight of the crucible, cover and sand is recorded.
Using the tongs, the covered crucible containing the sample is transferred to the muffle furnace. The crucible must be placed directly under the tip of the thermocouple, where the temperature control of the furnace is most accurate. Note that only one volatiles test should be run in a furnace at a time.
The crucible containing the sample is left in the muffle furnace for an hour (plus or minus one minute) at 900F. When the crucible is removed, it is put in an opening in a brick pallet and a timer is set for 15 minutes. During this time, the cover is not disturbed. The sample should remain covered until the final weighing is completed.
After 15 minutes of cooling on the brick pallet, the crucible is transferred to the balance, and the final weight of the combined crucible, cover and sample are recorded.
The crucible and cover are checked before discarding the sample and brushing out the crucible. The top of the crucible and cover may be coated with a silvery layer of lustrous carbon. If lustrous carbon is present, this observation is recorded along with the results. Lustrous carbon, if not present in excessive amounts, provides the casting with a nice surface finish.
Lustrous carbon residue must be burned off before the crucible is used for the next test. This will burn off when the empty crucible is preheated for 15 minutes at 900F before the next test.
The volatiles test should be run twice and the results averaged. When the tests are run according to this procedure, the two values should agree within 0.2% volatiles.
The volatile matter in the sand forms a reducing atmosphere around the casting and provides it with a nice surface finish. If volatiles are too high, excessive smoke and fumes, and gas-related defects such as porosity, misruns and coldshuts may be a problem. Lustrous carbon defects also may result.
The volatiles test, when run at 900F, provides a distinction between carbonaceous materials that vary in their volatiles content. For example, a sand containing seacoal and a sand with an equal percentage weight of woodflour would have the same LOI data. The volatiles test, however, will reveal that the sand with the woodflour contains five times the amount of volatile materials as the sand with the seacoal.
For sands containing seacoal, the ratio of LOI to volatiles reflects the amount of coked carbonaceous material. Seacoal is present in different forms in molding sand. Coked seacoal is seacoal that has lost its volatile material and is less effective. It will lose weight in the LOI test but not in the volatiles test. Thus, the volatiles data should be used to supplement the data from the LOI test by indicating how "fresh" or volatile the carbons are.
The tests described in this article are the key ones used for controlling green sand. The green and structural properties tests monitor uniformity of mix preparation and system control. Obtaining consistency of the values by controlling the additions and the sand system engineering are the first steps to sand control. Once the system is stable, the properties can be optimized.
Unfortunately, many of the tests for green sand are not timely enough to allow real-time control of the sand system, but they are still of great value in the overall scheme of control. In many cases, automated controls are placed on the sand system to control it on a real-time basis where possible. Laboratory sand tests, however, are still absolutely essential for routinely monitoring the results and the effectiveness of the sand system controls.
Casting quality is directly related to sand quality. To produce world-class castings competitively and to meet the ever-increasing demands for higher quality and dimensional reproducibility, sand control is essential. And sand testing provides the key.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Reducing Casting Defects, Part 2; structural sand tests for determining sand composition|
|Author:||Krysiak, Mary Beth|
|Date:||May 1, 1994|
|Previous Article:||Chemically analyzing white iron.|
|Next Article:||The state of the cupola.|