Hands-on experience at Armstrong Mold.

Armstrong Mold Corp. is a small, short-run foundry producing aluminum, zinc and magnesium alloy castings. Though it uses scarcely 1000 tons of sand a year, its experience offers proof of the benefits of using sand reclamation regardless of foundry size. Located in Syracuse, New York, Armstrong specializes in castings ranging in size from that of a pinhead up to three-foot cubes.

With its production space at a premium, the foundry's sand molding area is compact. Sand equipment consists of a twin-screw sand mixer, an improvement over the original system that charged sand into an overhead hopper by using a dump bucket installed on a forklift.

There was no intention to recycle the sand when the equipment was installed, because annual sand use was so limited. The foundry uses a proprietary polyol-urethane binder with silica sand to make molds and cores with a few shell or plaster cores for special applications. After shakeout, the used sand was simply hauled away and dumped. With time, however, instead of being classed as good clean fill, the used sand was reclassified as a hazardous waste. Eventually, permits had to be obtained to dump the sand, even in an approved landfill, and disposal costs began to rise sharply. Justifying reclamation

When the local landfill closed and spent sand disposal costs rose to more than $20/ton, and haulage fees to a new landfill exceeded $25/ton, the cost of disposal exceeded the cost of new sand. Soon, large, unsightly piles of used sand collected behind the plant waiting to be trucked away. Burdened with rising costs and frustrated with the general problems associated with waste sand disposal, the foundry's management began to review the potential of sand reuse.

Many obvious difficulties surfaced during the reclamation review, but chief among them was the fact that the foundry's small throughput made it difficult to justify the capital investment for a reclamation system. Lacking any sand-handling equipment and without adequate space for new equipment, the problems seemed insurmountable.

The foundry had no mechanical shakeout because many of its castings are fragile nor did it have any need for a baghouse or dust control equipment. The plant's electrical power and compressed air supplies were operating at capacity. A low roof would not permit building upwards and digging pits in the floor was expensive and generally undesirable due to a high water table.

With technical and maintenance personnel also limited, it was decided that outside expertise was necessary. In early 1988, a foundry engineering consultant was retained to review the situation and advise the company.

The chemical binder used by the foundry to produce cores and molds for the low-melting-point alloys does not burn out during the casting process, with the result that the sand to be processed existed in large, hard chunks. it was decided that the optimum reclamation method would involve a combination of attrition (mechanically breaking down the lumps) and calcination (burning the residual binder off the sand grains). A three-ton test batch of refuse sand was processed in this way to determine the type of reclamation equipment needed and to evaluate the end product. The test, using the three steps noted, was satisfactory, and the decision was made to go ahead.

The foundry's sand throughput was so low that the smallest commercial units available would far exceed its needs. Fortunately, at that time, a small but complete reclamation system became available. It had been used as a demonstration pilot plant by a sand reclamation equipment manufacturer. Even with this, capacity of the equipment was three times greater than the foundry's needs, and its electric power requirement dictated that it could not operate while the foundry was operating. But the excess capacity proved economical because it allows the foundry to reclaim used sand in a relatively short time at night or on weekends when off-peak electric rates are in effect. System Operation

In operation, sand enters the lumpbreaker, where lumps are reduced mechanically to original grain size and screened to remove tramp metal and other foreign materials. The sand is then conveyed to a hopper for delivery to the calciner. Because the sand is now in its original granular form, a fluid bed unit can be used with high thermal efficiency. The calciner throughput is controlled to ensure adequate binder burn-off with maximum fuel economy. Sand binder content is checked by a loss-on-ignition (LOI) test. Some of the clean but hot sand grains fracture during the reclamation process, which changes the grain size distribution due to a build-up of fines. To counteract this, the sand is passed through a cooler-classifier to restore the original grain-size distribution, and the reusable sand is conveyed pneumatically to storage.

Installing the equipment offered a challenge due to the space constraints, but it was completed in just over three months. Placing the lumpbreaker in an unused entranceway and locating the output hopper outside the building, made it possible to assemble an equipment grouping compact enough to require only a single dust-collection unit. The control panel was designed so that the system can be run consistently with a minimum of operator intervention, and care was taken to set up an appropriate preventive maintenance schedule. Operating Results

The sand reclamation system has exceeded expectations. The reclaimed sand is very similar to new sand by analysis, but in use it has proved to be superior. Probably due to slight rounding of the sand grains, it was found that flowability was better, 20% less binder was required and sand quality has been entirely consistent. So far, almost all make-up sand has come from the sand piles behind the building, contributing to the system's overall savings.

Actual operation of the reclamation equipment is simple, and it routinely runs unattended from Saturday noon to early Monday morning. A straightforward preventive maintenance program is a key part of the system, reducing downtime to a minimum.

The equipment was purchased for $160,000, and installation brought the total cost to a little over $200,000. All elements of sand recovery that were operative at the time of installation would have shown a payback of about 2.5 years, but there have been further developments. Recovered system fines (approximately 10%) are used in the foundry's molding plaster mix, representing a double cost saving because disposal costs are eliminated and fine sand purchases made unnecessary. More significantly, however, are the landfill cost savings. Currently, landfill charges have escalated to over $90/ton and haulage fees have risen to more than $36/ton. The payback period under current disposal rates would be less than two years. Looking Back

In retrospect, a belt or bucket conveyor would probably offer a more maintenance-free operation than the screw conveyor now in use. In addition, the calciner unit has had problems with its porous ceramic membrane, which has been replaced several times. The equipment manufacturer, recognizing the problems encountered with the porous ceramic membrane, has since incorporated a new, patented distributor tube fluidizing principle. The new design appears to have eliminated the problems, which were believed to have been created by thermal and mechanical shock due to the intermittent operation of the unit.

Most sand foundries should find it much easier to justify the capital investment for a sand reclamation plant than did Armstrong Mold because of their higher sand usage. Initially, a thorough study of the problem is recommended before purchase of any equipment. Employing a professional to design and oversee installation is a wise course for the average shop. It can save money and time.

Foundries handling higher-melting point alloys should also have less trouble justifying reclamation. Since their sand losses are low, it is possible to use olivine or zircon sand, for instance, to further improve casting quality. There is evidence that even silica sand is dimensionally more stable after thermal reclamation, so that scabbing and rattailing defects are reduced.

Even if waste sand disposal imposes no immediate problem, it is probably safe to assume that it will at least become more expensive, so the trend must favor reclamation of as much material as possible.

This article was originally presented at the 1991 APS Sand Reclamation Conference.