Lyle W. Olson, a 30-year veteran in food process and packaging engineering, embarked on a project a decade ago to turn potato processing's least-saleable waste into a value-added commodity. Olson identified a flash dryer more efficient than rotary kiln and other dryers and designed a system around it. Now he is demonstrating the system for some of the potato segment's leading processors and exploring other opportunities for the technology.

A graduate of the Milwaukee School of Engineering with a degree in mechanical engineering and an advanced degree from Minnesota State University, Olson has been involved in numerous design, implementation and optimization projects. He was maintenance manager at Golden Valley Microwave Foods in Eden Prairie, Minn., and an engineer with Pillsbury Co.'s Green Giant division in Le Sueur, Minn. He developed the first fluid bed dryer for animal feed while serving at Mankato, Minn.-based Hubbard Milling Co.

Currently a project engineer at Sebesta Blomberg & Associates Inc. in suburban Minneapolis, Olson recently took time from his regular duties to discuss his waste-recovery work.

FE: What is the waste-recovery potential in potato processing?

Olson: The rule of thumb is that 50 percent of the potato goes out the back door as finished product. Peel is a third to half of the waste, depending on the product being made, and white waste is 30 to 50 percent.

In french fry production, white waste is probably the most valuable fraction, and many processors keep it separate from the other byproducts to sell directly as cattle feed. My focus has been on the culls, peels and other filtered material. Processors have less than a day to either have that material hauled off or process it into a saleable commodity. It's an excellent business opportunity.

FE: How do processors typically deal with this waste?

Olson: One option is to spread it on fields, though some states severely restrict this option. In Minnesota, for example, waste has to be transferred in leak-proof trucks, adding more costs to the manpower and machinery expense of disposal. Another option is feed, but cattle are the only livestock that can consume potato waste. One major processor raises cattle so it will have feed lots to dispose of potato waste, a strategy that puts them in two businesses requiring different skill sets.

The cattle feeder often controls the cost of getting rid of the waste. A major processing plant might use 1 to 1.25 million pounds of potatoes a day and generate 250 tons of waste. One processor I know paid $12 a ton to have potato waste hauled away. Since 89 percent of it is water, that's $3,000 a day to haul water.

FE: What are the features of the dryer you selected?

Olson: To some people, a flash dryer connotes a ring dryer, but a flash dryer has some uniquely different features. A ring dryer might have a 12-by-12-in. duct extending 30 to 50 feet high and looping back down to form a ring. Hot air is pumped into the duct, wet material is introduced, and air velocity pulls the material up, drying it before it descends. It requires a lot of velocity, a lot of capital cost, a lot of energy, but it works well on materials that are a consistent size, density and moisture content.

If you're working with particles that vary considerably in size, a ring dryer isn't going to work well. Potato waste exhibits significant variation in size, so a ring dryer isn't suitable. With a flash dryer, residence time is measured in seconds. Extracting moisture quickly is the critical aspect.

FE: How is rapid drying accomplished?

Olson: It's a function of mechanical agitation and high-velocity airflow. The agitation causes some grinding and reduction in particle size, which helps the moisture in the inside of the particle to migrate from the center to the surface, where it evaporates. Particles then are suspended in a hot air stream to dry.

Three companies make this style of dryer, including an Australian firm that has manufactured them for the last 20 years. They have sold about 200 units in Asia, South America, Europe and Australia, drying a wide variety of materials, but there aren't many units in America, so people aren't familiar with the technology. Variations on the dryer's design are now available from two Minnesota companies. Our firm developed processing steps that are applied prior to the actual drying.

FE: Does faster drying mean higher operating temperatures?

Olson: We operate at temperatures under 600 F degrees. A rotary drum dryer might operate in the 800 to 1,200 degree range. That translates to energy savings of as much as 20 percent. The capital cost of the dryer is lower, and the footprint is quite a bit smaller.

There can be a problem with the formation of starch balls in horizontal rotary drum dryers. With flash drying, the short residence time and the drop that occurs in air temperature reduces the opportunity for starch balls to form. In drum dryers, there is more time for moisture and heat to react and form starchballs.

I've come across cases where the balls were getting to be the size of golf balls. Operators then had to grind the balls up and reintroduce the material to the feed. You have to wonder about the moisture contained in the center of those balls and its potential for developing mold in the feed.

FE: What are the economics of your process?

Olson: Our tests show that it takes 1,100 BTUs to extract 1 lb. of water from potato waste. We don't have firm data for conventional rotary drum drying, but it probably is in the range of 1,600 to 2,000 BTUs per pound of evaporated water. If you recycle some of the exhaust, a 10 to 15 percent reduction in BTUs is possible.

A significant advantage of flash drying is that you have a homogenous piece coming out of the process, which opens the door for using the finished material in feed pellets for nonruminating livestock. One of the best potential markets for this material is poultry. Protein levels are similar to those of corn, but potato waste also contains amino acids that feed manufacturers have to add to corn. We ran assays of some of our dried materials, gave them to feed manufacturers and asked them what they would be willing to pay. They indicated a premium of $10 to $20 a ton more than corn was likely, provided processors could produce it in sufficient quantities.

The problem is that the price of corn has dropped to the $2-$3 a bushel range. At today's prices, this material might fetch $71 a ton. That's a $27 loss compared to what it would be if corn was in the $3-$4 range.

FE: What would it cost to install your system?

Olson: The dryers alone would be $250,000 to $500,000, and a large processor would need four of them. When you factor in a dedicated building and other costs, it could cost $10 million for a company processing a million pounds of potatoes a day. Higher corn prices would certainly help the economics.

FE: Are other byproducts candidates for this process?

Olson: We have done tests in other areas, though I'm not at liberty to discuss them yet. We also have applied for funding from a state program to use food waste and other materials to generate steam and electrical power. A cogeneration plant needs a steady, continuous source of fuel, and food plants often operate around the clock and could use steam to cut their energy loads. Self-generation of power isn't rocket science. It's feasible, and we hope to demonstrate it and give processors confidence that they can achieve some energy independence.