Over the course of a year and a half, in 1979-1980, I designed and supervised the production of a 30×36 molding modernization project for Spokane Steel Foundry—despite Mt. St. Helens’ best efforts to interrupt its final few days of installation. My concept for this project was based on modular design. First, however, maybe a brief basic description of how steel casting are produced in green sand molds, and what Spokane Steel Foundry’s process was, when I went to work for them.

The first thing a skilled pattern maker needs to do, back then and still today, is to make a full-sized, three-dimensional model of the part to be made. It is made in two halves and each half is mounted on a wood or metal plate. They then are mounted on a pair of molding machines, followed by sand with a binder being packed over them. After the pattern is removed from the sand and the two halves of the mold are mated together, the cavity that is left is the outer shape of the part to be made. The assembled mold has a channel through which molten metal is then poured to fill the cavity in the sand. After the molten metal cools and solidifies, the sand is removed and the excess metal is trimmed off to produce a finished casting. These finished castings are then heat treated to produce the properties required for their use.

Before our production modernization project started, each half of a mold was produced on one of two automatic molding machines. These machines were kept without modification. These halves, which could weigh up to 300 pounds, then rolled off the machine and a person would use a small crane to lift the bottom half and place it on a level roller conveyor. He would then pick up the top mold half, roll it over and place it on top of the bottom mold half. The completed mold would then be picked up by a person-operated traveling bridge crane and set on an open concrete floor with others, and, when enough steel was melted and alloyed, the molds would be filled by the pouring crew again using the overhead traveling bridge crane. 

After the poured molds cooled, the overhead crane would again pick up the sand-filled boxes one by one and place them on a vibration grate to remove the sand from the casting. The casting would then be removed by another small crane and sent off to the finishing department and the empty steel molding boxes would be returned to the molding machines holding area.

This handling system involved several employees and a lot of crane time. It was rather slow and inefficient.

That’s what the semi-automatic, modular-powered, mold-handling, sand-removal, and molding-box return system design was meant to improve.

What was mentioned previously, but not described in detail were the first three machines I designed and had built locally. 

The first machine was a hydraulically powered roller conveyor that automatically moved the bottom half of the finished mold from the molding machine and into an automatic mold closer. 

The second machine powered the finished top half of the mold into a hydraulically powered rollover machine and transferred it into the automatic closer, which lowered the top onto the bottom when the operator actuated it. The finished mold then rolled out for the crane to pick it up and move it to the pouring floor.

As I said, thankfully everything worked to this point.

I then designed and had a large shakeout machine built and tested before I could connect the semi-automatic hydraulically powered roller conveyor system that distributed the molds to separate lines on the pouring floor. This eliminated the need for the overhead crane to be used. 

After the molds were poured full and cooled, they would then be powered into the shakeout machine where the sand fell into a below-the-floor return system. The casting was removed to go to finishing and the empty molding boxes automatically returned to the molding machine for reuse.

When finished this automated mold-handling system increased production from 10-12 molds per hour up to 30 and the number of employees required to operate it was reduced by two. Additionally, I proved that the new, expensive molding boxes that the other vendor said were required were actually not required. 

In the end the total cost for designing, building all components, and installing them was approximately $350,000.  That was $300,000 less than the previously quoted price.  Additionally, the $100,000 cost for new molding boxes was avoided.

All things considered, this was a fun and exciting yet anxiety-filled project, and probably the most rewarding project in my foundry career.

As an aside, a few months after this was completed, I was contacted by the company that originally quoted this system to Spokane Steel Foundry and asked if I would be interested in managing the modernization of one of their client’s foundries in Mexico on a name my own price contract. Gracefully I declined.

Ken Kaiyala
5-18-2024