A native of Minneapolis, Paulson is a 1961 graduate of the University of Minnesota with a degree in mechanical engineering. He has designed a range of food-handling equipment throughout his career, from peanut transfer systems for major confectioners to dedusters for the breakfast cereal segment. In 1987, Paulson founded Lancaster, PA-based Pelletron, which collaborates closely with the University of Pittsburgh engineering department's small-business incubator program. Graduate students did much of the prototype testing of Paulson's new elbow design, and he expects to work with them and Professor George E. Klinzing, vice provost for research, in verifying results and establishing the design's limitations.
Paulson: Thirty years ago, I studied the dynamics of changing direction in a pneumatic system and found that if a particle hits another particle, depending on the speed and angle of impact, direction will be altered in a particular way. It's similar to striking an eight ball with a cue ball: if you do your mathematics, you can direct the eight ball in the desired direction. In the case of an elbow, if you control the shape of the chamber to cause a particle to react with another particle, it becomes a fluid-like directional change. Particles collide with other particles, and that results in less wear on the elbow walls and reduced particle and pipe damage.
The elbow I designed for Hammertek in the early 1980s could convey diamonds without wearing out the elbow; letting product wear on product was a great benefit. It also ensured that the first particle in was the first particle out. But the 45° rotation of particles is a potential problem. The directional change caused wear to occur near the outlet. With this new design, I wanted a 90° angle change so that dense-phase to dilute-phase transition occurs in the elbow and moderate acceleration out occurs. When solids flow stops, the elbow immediately cleans itself out. Lower friction means a lower pressure drop occurs, and there is less elbow wear. Actual pressure drop will depend on the solids-to-air ratio and inlet velocity condition, but my educated guess would be a pressure reduction in the range of 15-40% for each bend.
FE: What followed the theoretical work?
Paulson: To verify the design, we took our sketches to a firm that generated a three-dimensional prototype in SolidWorks. The computer image is then sent to a printer that makes a plastic model you can hold in your hand. Next, we made molds, created clear elbows and sent them to the University of Pittsburgh, where they created a video of products flowing through the elbow in the way we intended them to flow.
FE: How did you arrive at the elbow's unorthodox shape?
Paulson: Design and development is a two-part process, and the design considerations were to change particle direction, cause minimal damage and reduce high-pressure resistance. That led to a geometric analysis to come up with a solution. We knew that enlarging the elbow area would reduce pressure. By calculating the mass, flow and vectors, we came up with the shape.
Next, we made a model and found we had enlarged the area too much to have an effective clean-up phase. The second design wouldn't purge, and that led to the third and final design. The shape is a function of the Bernoulli effect, the principle that increased velocity results in decreased pressure, and vice versa. Instead of an elbow, everything is happening in the heel or what looks like a swollen ankle.
FE: Is deterioration of a pneumatic system's infrastructure primarily a maintenance issue, rather than a food safety concern?
Paulson: If a metal elbow deteriorates, that means metal is being added to the finished goods, and that can't be good. Of course, maintenance and product damage also are concerns. I once worked on a dense-phase transfer system for whole peanuts. Conveying a peanut presents many problems; there's lots of oil, and the nut is fragile. If there's too much friction, you wind up with peanut butter in the pipe.
There was a sugar conveying system in a bakery years ago that relied on a polyurethane flapper on the diverter valve to a scale hopper. The bakery had to replace those valves a few times every week. One day, a couple of technicians were having lunch at a nearby diner and they could smell burning rubber. "That happens sometimes when we're toasting bread," the waitress informed them. Suddenly they realized where their polyurethane was going.
That's what makes pneumatic conveying fun.
FE: How much space is saved compared to a conventional elbow's radius?
Paulson: It is considerable. With a long radius elbow, changes in direction occur by a factor of eight to ten in relation to the diameter of the pipe. With a four-inch diameter pipe, that means up to 40 inches are needed. With our elbow, the directional change takes nine inches, or about 75 percent less space.
There's also the issue of regaining conveying speed. With a long radius turn, you need at least 15 feet of pipe to get back up to conveying speed. Because this uses fluidization, you can bolt two elbows together and still be at the right conveying speed immediately after. That allows the designer of a pneumatic conveying system to reroute the piping wherever he wants in a plant.
FE: Pellbow was designed with plastic pellets in mind. Can you adapt it quickly for food handling?
Paulson: The fabrication materials are the only difference. Molds have been released to the pattern makers for fabrication in cast iron, nie-hard, aluminum and 3/16th inch stainless steel, typically 304 stainless to withstand caustic cleaning. The foundry that will make the elbow specializes in precision valves and fittings, and they'll fabricate elbows to fit pipes anywhere from 1.5 to 8 inches in diameter and in a variety of styles: flanged, plain ends, and so on. We're a 12-person corporation specializing in manufacturing and engineering services. It's much more efficient and inexpensive to outsource manufacturing to experts in that area.
FE: When did you first become involved in pneumatic conveying?
Paulson: I worked for a company in the 1950s and '60s that designed the first system to deliver flour from a hopper to mixers. We didn't have computers to regulate delivery amounts, so we used a system of timers and drums with pins to regulate cycle activity. The drum would turn one notch at a time for each batch. It was Flintstone engineering, but it worked.
The company eventually was sold and I moved on after 13 years. I spent four years managing R&D on pneumatic flow systems for the Flo-Tronics division of Brown and Root in Houston (now part of Halliburton). Brown and Root had 85,000 people and two mainframe computers, and all bids and projects had to go through those mainframes. It was terrible. Today, all the computing you need to do is done on your desktop, and efficient manufacturing is accomplished by outsourcing functions to experts to handle them.