> Case Studies > Glass Grinding Case Study

Centrifuge Takes Center Stage in Plant Expansion

Glass Grinding

By Michael Spellman, president of IGE Solutions,
with Tom Forsythe, LMI Glass

Glass Digest, July 2001

Early in 2000, Larry Methvin, the founder and owner of LMI Glass, asked Tom Forsyth, LMI's mirror department supervisor, to design and build the fabrication department at the company's new 80,000-square-foot plant in Ontario, Calif. LMI Glass is a midsize manufacturer and installer of mirrors, showers, wardrobe doors and ventilated shelving for residential homes and apartments in California and Nevada. Methvin assigned each of his 3 department supervisors the task of designing the most efficient systems possible for their respective areas. Forsythe seized the opportunity to build a system that would eliminate many of the traditional problems associated with glass fabrication departments, while increasing its productivity and profits.

"I wanted to build the 'ultimate system' for glass fabrication," he recalled. "My number one goal was to change and improve the way we handle our glass grindings. My other goals were to make the department more efficient by utilizing automated equipment and processes, and by creating a better work flow for materials and people."

The Greatest Challenge

The LMI mirror department processes (polishing and beveling) an average of more that 150,000 square feet of glass and mirror every month. The new department layout would need to accommodate 4 polishers, 2 beveling machines, and a central coolant filtration system.

Like many glass fabricators, Forsythe had grown increasingly frustrated with the problems and inefficiencies created by the glass grinds generated in the polishing and beveling processes. The beveling machines' three 300-gallon tanks had to be drained every 3-4 weeks, requiring four employees to spend half a day or more digging out the 18 inches of rock hard glass sediment. Two additional 150-gal tanks for the polisher required cleanout even more often, with three employees spending 2-3 hours digging out 12 inches of glass sediment from these tanks every other week. These numbers equated to approximately 300 and 200 hours per year of downtime for the beveling and the polishing lines respectively, if the cleanouts were done during normal plant working hours. Cleanouts done outside normal working hours usually required paying overtime so, regardless of the approach taken, tank cleanouts were costing LMI a lot of money.

Even between regular cleanouts, the coolant would get so dirty that the machines would have to be shut down to clean out clogged coolant manifolds and plumbing, and to replace prematurely worn diamond and resin wheels. In addition, dirty coolant can create quality problems, forcing the fabrication equipment to be run at slower speeds, and it generally makes a mess on the machines and the surrounding production area.

Forsythe knew that a better coolant filtration system would minimize the frequency of diamond wheel replacements and that his coolant would also last longer. He was determined to minimize or eliminate these problems in the new facility.

The Consulting Team

Forsythe researched all of the known filtration methods for glass coolant, but nothing had the potential effectiveness of a centrifuge. He had been introduced to US Centrifuge (USC) and the idea of a central centrifuge filtration system, by IGE Solutions, Inc., USC's exclusive North American agent. Although he quickly saw the benefits of this type of system, and he also realized that USC had the expertise to evaluate his needs and help develop the best solution. A team was formed that included Forsythe from LMI, Scott Behrens, CEO of USC, Brian Cooper, president of USC, and Danny Seals from US Centrifuge. Rounding out the team were Michael Spellman and Andy Hilditch from IGE Solutions, and Wayne Cox, president of Cox Commercial Industrial (CCI), a local industrial engineering and contracting firm.

"It is critical in a project of this type to have a team of consultants that are genuinely concerned with designing a system that is going to meet your needs today and well into the future," said Forsythe. As the team began work, three critical areas were addressed: the type of centrifuge, the arrangement of the machines, and the design of the coolant handling system. Forsythe was impressed with how well everyone worked together throughout the entire project.

The Centrifuge Solution

The centrifuge, the heart of the entire system, was the most critical decision.

With more that 400 systems in operation, USC already had a good reputation for building innovative applications. Adding to its reputation as a leader in the centrifuge industry are its patens, such as the Centri-Lock Clutch, and the LSD Smart Machine technology, and other innovations such as its Cake Sensitive Plow Sensor.

But even more than advanced features, Forsythe required a system that was capable of processing the dense, heavy glass grinds being generated.

"The USC machines are extremely reliable, simple, and low maintenance designs, but they have not sacrificed efficiency or top-end performance as some manufacturer's designs have done." Forsythe said, "They can truly handle the concrete-like grindings that my bevellers and polishers produce." Tom was further convinced of his USC choice when he heard a previous glass customer state that his centrifuge had run 24 hours a day, 7 days a week for 1.5 years, and that the unit did not need to be overhauled and the bearings did not need to be replaced.

After a careful study of LMI's needs, IGE solutions and US Centrifuge recommended the large 262 SupraMatic model, which has a maximum separation force of 1,500 G's and is capable of "filtering" up to 60 GPM (Note: USC states their flow rates in terms of their true filtering capabilities, not flow-through rates).

The Efficient Floorplan

The next step was to determine the physical layout of the beveling and polishing machines. Forsythe wanted to produce the most productive workflow possible for both materials and people, while creating an efficient central coolant system. The final design was a basic in-line arrangement, but with some unique considerations implemented.

First, the 6 beveling and polishing machines, which were various models from different manufacturers, were lowered into the floor to crate a consistent, worker-friendly height of 30 inches for the main glass working areas. As workers move glass sheets from one machine to another, they do not have to raise or lower it to accommodate the next machine. The lower working height also reduces the physical labor required by LMI's employees, especially shorter employees, for whom the machines were obviously not designed.

Another aspect of this design is a concrete coolant trough that runs under all of the machines, leading to the centrifuge filtration system. This trough is sloped towards the central filtration system, and thereby uses gravity to return the dirty coolant to the centrifuge. The trough also houses the pipes that deliver clean coolant back to the machines and is concealed with steel plating, which is easily removed for cleaning or maintenance.

The lowered machines and the coolant trough work together to create another interesting aspect of this system design. A steel grid was used around the lowered machines to create and even and safe work floor.

The concrete under the grid, on which the machine is resting, is sloped towards the centrally located coolant trough.

The Ultimate Coolant Handling System

The combined consulting team of US Centrifuge, IGE Solutions and CoxCommercial Industrial determined that LMI required a 3,500-gallon coolant system. The complete system consists of a collection pit, a processing tank, the centrifuge and a clean coolant return tank.

The collection pit is a 7-foot-deep circular pit that the coolant trough empties into. Special jets at the bottom of the pit are designed to keep the glass grindings in suspension. Brian Cooper, president of US Centrifuge stated, "Our system works contrary to the traditional logic of settling tanks, we actually want to keep the grindings from settling, so we can filter them out. Keeping these heavy glass grindings in suspension is actually one of the biggest challenges with systems like this, but we have a lot of experience in this area and we have some trade secrets for doing it successfully."

From the collection pit (see above), the dirty coolant is pumped into the large above-floor processing tank. This is where US Centrifuge's Centri-Flow technology is utilized to condense the glass grindings, yet keep them from compacting, in the conical bottom of the tank. Scott Behrens, CEO of USC stated, "Although it worked from day one, we have spent more than a decade refining the Centri-Flow system for the glass industry. The basic concept is that it utilizes a series of tangential jets, and a custom-engineered, internal weir to speed up and slow down the motion of the coolant within the tank, based on the requirements of the specific application."

From the conical bottom of the processing tank, the dirtiest coolant is transferred to the centrifuge. The centrifuge continuously separates the glass grindings from the coolant, and senses when it needs to clean itself. During a cleanout cycle, the centrifuge quickly stops spinning, and a powerful motor engages blades that scrap the compacted glass grindings from the inner walls of the centrifuge, dropping the waste into a trash bin below. During the cleanout cycle, the dirty coolant bypasses the centrifuge and is sent back to the collection pit.

The clean coolant from the centrifuge is discharged into a return tank (see below) that utilizes a unique piping manifold for the lines returning coolant back to the beveling and polishing machines. As in the other tanks, this return tank also has a series of special jets to keep the coolant in motion. Since the entire coolant handling and filtration system runs 24 hours per day, at night or anytime the machines aren't running, the clean coolant is simply recirculated back to the collection pit.

Wayne Cox, president of CCI, was an important player in building this system. Forsythe said, "Wayne's company did an excellent job of interpreting US Centrifuge's design requirements and making it happen."

Expectations Were Exceeded

Tom Forsythe's tank cleaning and dirty coolant nightmares have been totally eliminated. Tom stated, "This system has far exceeded where we thought we would be in the end. At the 8-month mark we emptied all the tanks and inspected the entire system. To our surprise, we only had a thin film of sediment on the bottom of the collection pit and a few inches of sediment in the cone of the processing tank. We made some minor adjustments to the jets and we think the system can now easily run for an entire year without emptying the tanks. Our goal is to do a yearly 'inspection', but we don't anticipate that cleaning will actually be required."

Forsythe isn't the only person at LMI excited about the new system. Steve Irons, LMI's maintenance engineer said, "The centrifuge, as well as the whole system, is actually very simple, so it is easy to operate and requires virtually no maintenance. The whole system has exceeded our expectations, and we had high hopes to begin with."

Although Forsythe's payback calculations have not been finalized. He expects "it is going to be very quick." Describing the many costs savings he was realizing, including some that he had not expected, he said, "IGE Solutions thought we were getting our diamond wheels from someone else, because we hadn't ordered any in such a long time. We used to replace diamond wheels every 3-4 weeks, but we are now running for 'months' between replacements. Coolant use has also been dramatically reduced. I'm anticipating that I will only use one 55-gallon drum per year, instead of the 6-8 drums we used to use. Machine downtime has been completely eliminated and the massive amounts of labor that were required for tank cleanouts has been reduced to one person simply emptying a trash can each morning."

In reflecting on what made the system so successful, Forsythe is convinced that a key component was his willingness to listen. In his research, he determined that the majority of the unsuccessful centrifuge systems sold into the glass industry failed because the company refused to listen to the centrifuge manufacturer's recommendations.

"These companies purchase a centrifuge like it is a commodity piece of machinery and then they set up a coolant handling system using their own judgement instead of the centrifuge manufacturers advice and assistance." He said, "A company like US Centrifuge has been doing this for many years and if glass manufacturers want to achieve filtration performance like I have, they need to listen to these manufacturers."