Jill's vision included using tools found in a home workshop. Now my idea of what should be in a home shop is skewed from the norm. My father is a sculptor and by the time I was a teen his studio had welders, both gas and electric, a table saw, and a drill press along with a huge amount of hand tools, clamps, and interesting materials. These days my shop contains lathes, milling machines, drill presses, thread tapping machines, welders, hand sanders, hand grinders, metal band saws, wooden band saws, a horizontal band saw, and stationary belt and disc sanders. The list could go on for several more lines. My challenge would be to scale back the beater build plan to more modest tools. After agreeing to the class I began to suffer a mental sweat. How do I keep the cost down and what materials from local stores would make a working machine? I remembered hearing a talk by Mark Landers, a paper artist from New Zealand and the builder of the Critter paper beater. He is a clever and resourceful man and he described using the motor from a discarded washing machine to power one of his early beaters. When I was a young guy I did things like this too, but nowadays I am more likely to buy a new part for the job. Building both toys and machinery for years I have had to consider safety as a primary design criterion. An open frame motor from a washer may be a cheap solution but would have safety issues over an enclosed motor. I mentally walked the aisles of Home Depot thinking, what could I use for the beater roll, or the shaft for the beater roll, or the blade material? I could find things to make a tub and covers but for other important components I didn't see an immediate answer. As for tools, I felt that we would need to go beyond a hand drill and a hand-held saber saw, and should at least include a good table saw and drill press. Brian Queen volunteered to create precise computer drawings and became my sounding board for the project development. He is a resourceful designer; and I appreciated being able to bounce ideas back and forth with him. Before I delve into the details of what we made in the class, I would like to outline the various parts of the beater and what we expect them to do for us. The four major components of the Hollander beater are the tub, the roll with its blades, the bedplate (the stationary blades that the roll blades work against), and a power source. The first Hollanders made in the 1600s had wooden tubs built from planks of wood like a barrel. Tubs over the years have been made from materials such as metal, fiberglass, Plexiglas, PCV plastic, concrete, or tiled concrete. The outer wall of the tub is shaped like the letter "O" with some type of central wall, creating a raceway for the water and fiber to circulate around and around during the beating process. Sometimes the central wall is also an "O" shape leaving room inside the "O" for some of the machinery of the beater such as bearings, pulleys, and drive belts. On other beaters the center of the tub is just a single standing wall. This is the design we used for the class. When the single wall is used all the machinery of the beater such as the bearings (which support the roll and the drive pulley) are located on the outside of the beater tub. In the tub, there is a hill or bump shape called the backfall directly behind the spinning roll. On the side of the "hill" next to the roll, the radius of the roll is mimicked with just a little gap between the rotating roll blades and this hill. The other side of the hill angles down to the tub floor. Generally the height of the rise of the hill is about half of the roll diameter. The purpose of the backfall is to create a pump effect in conjunction with the spinning roll. By forcing the slurry up and over the hill a current is created which causes the pulp to rotate around the tub. If you were to do an experiment and run the beater without the backfall you would find that the beater roll would just churn away or cavitate without moving any water and fiber around the "O"-shaped tub. The tub's job is to contain the slurry and to allow the best possible circulation rate so that the slurry of water and fibers interacts most often with the roll and the bedplate blades. On many tubs you will notice that the roll side of the "O" shape is wider than the opposite side. This is done to create a higher pressure of slurry at the opposite side to bring the pulp back faster between the roll and bedplate blades. On some tubs the floor of the tub is raised where the pulp exits the roll and then angles down to a lower height right before the incoming side of the roll, again to maximize circulation. During construction of our beater in the class Howie Clarke pointed out that putting a wood fillet or radius where the tub walls meet the tub floor would also aid in circulation. Another beater part that serves to increase circulation is the roll cover. Its first purpose is to keep the slurry from being slung out of the tub by the spinning roll, but it also functions as part of the pump design. As the pulp is flung over the backfall, it hits In addition to computer diagrams for the workshop beater, Brian Queen fabricated laser-cut miniature parts for a model of the beater. Photo: Michelle Wilson. The author's initial sketch of the home-built, hardware-store beater. Courtesy of the author. the roll cover and rebounds down at an angle in a direction that produces the maximum current possible. The beater roll is a cylinder (or sometimes an open frame of discs holding the blades) with a shaft passing through the flat ends of the cylinder. On the curved surface of the drum are blades of metal that are attached to the drum as if they were radiating out from the center of the drum in a pattern like the paddlewheel on an old steamboat. The roll serves two purposes. First its spinning blades act against a set of fixed blades (the bedplate) that is located below the drum in or on the floor of the tub. The interaction of the spinning and fixed blades cuts and macerates the fibers into fibrillated bits as the cellulose travel between the blade faces. The second job of the roll is to act like a paddlewheel, or a pump, to move the slurry around and around the tub. The blades of the beater are blunt-faced, and look to have an edge 90 degrees to the side of the blade, but in fact the blade edge follows the radius of the roll. The only sharp aspect of the blade is the leading edge which, when rotating with the roll, will be the first part of the blade to contact the bedplate blades. This sharpness is sometimes blunted to create different pulp qualities. Bedplate blades are arranged closer together than the roll blades, generally the width of a blade space between the blades. These blade faces also appear blunt, and on a closer look will have a concave shape matching the radius of the roll. On a small beater we might have seven to ten blades in the bedplate that interact with the spinning roll blades. The pounding or macerating of the fibers is a primary function of the beater. The action of the roll blades against the bedplate can also produce a small degree of cutting. This is achieved by setting the bedplate blades at a slight angle in relation to the roll  blades. The greater the angle, the greater the cutting action will be. A good way to think about this is to imagine a dissembled scissor. If you were to try to cut a piece of fabric by pushing the blades straight at each other, it would be unlikely that you would cut anything. But if the blades are joined with a screw and meet at an angle they would slice the fabric without any trouble. Furthermore, the cut happens more easily in the crotch of the scissor where the angle of the blades' meeting is greater. By setting the bedplate blades to a small angle such as 6 degrees we reach a compromise between too aggressive a cut and no cut at all. Early Hollanders were powered by water wheels. The water wheel or now the electric motor needs enough power to drive the roll without bogging down. For the workshop beater I bought a ¾-horsepower, totally enclosed, fan-cooled (TEAE) motor, singlephase, with a strong start-up torque—a feature needed when starting against a loaded tub of unbeaten fiber. The motor has manual thermal protection, meaning that it will turn off when overheated, but will not restart until a reset button is pushed. Also, pulleys and drive belts must have enough grip to drive the roll through tough beatings. For any beater there is an optimum speed for the roll blades to pass over the bedplate blades. Spinning the roll too fast will produce cavitations with the pulp slurry rather than good beating action. Too slow a roll speed will unnecessarily lengthen the beating times. We set the speed by adjusting the ratio of the small motor pulley to the large roll pulley. For the workshop beater, we would shoot for 400 to 500 rpm (revolutions per minute). In comparison, a larger beater with a bigger diameter roll requires fewer rpm to maintain the same surface speed of the roll blade passing over the bedplate of a smaller beater. Finally we need a way to hold the spinning roll and to adjust the distance between the roll and bedplate in order to control the pulp characteristics as it is being beaten and to finesse the final qualities of the pulp (short or long fibers, over or under beaten, etc.). In the small laboratory test beaters built by Valley (now Voith) the roll is held by two fixed-in-place bearings and does not go up and down. Instead the operator moves the bedplate that is attached to a long arm hinged under the beater tub. To vary the beating pressure on this machine, weights are hung on one end of the arm which levers the bedplate more tightly against the spinning roll. Another method is to lift and lower the roll equally on both sides using a gearing system. And a third system, which we employed for our beater, involves attaching the roll bearings to two long arms that are parallel to the sides of the tub. The arms pivot at one end, allowing the roll to lift and drop over the bedplate. In this system, weights hung on the end of the arms control the beating pressures. In order to keep the construction simple for the workshop, I decided we would make the beater tub from 1-inch-thick Baltic birch plywood. The two curved ends of the tub would be made of faceted flat boards. The roll would be supported by two plywood arms hinged at the tail end of the beater tub. It is important to use technical-grade plywood glued with waterproof adhesive. Unfortunately Home Depot or Lowes only carries furniture-grade plywood. Even when buying plywood from a specialty supply house, I suggest an empirical test: throw a scrap of the wood into a pail of water for a few days to see how it holds up. In planning for the class I was stuck on how to construct the roll. Then one day I had an epiphany. I decided that we would make the roll out of a stack of discs cut from the same plywood we were using for the tub and insert metal blades into the wood using some table-saw tricks to create the slots for the blades. If the roll were solid wood it could provide enough support for the pounding of the blades into the drum of the roll as the pulp passes between the roll and the bedplate. The next vexing problem was to come up with a way to hold the blades into the wood. I had some techy ideas, but in the spirit of the class, I decided that we would simply glue the blades in place. I was concerned that the different expansion rates for the wood and metal may lead to failure over time, but I researched epoxies and found one that claims to be shock and vibration resistant. In addition, we would drill holes through the area of the blade where it is embedded into the roll in order to achieve a mechanical attachment as the epoxy fills the holes and bonds to the wood. (During the class we came up with an additional step of using an angle grinder with a slim cut-off blade to slice out a notch on each end of the blade that we could pack with the glue.) Next, how would we hold the wooden roll with the metal blades onto the metal roll shaft? I decided we would drill a ¼-inch hole in the shaft in two places corresponding to ¼-inch-wide grooves we would cut (using the table saw) into the face of two of the stack of plywood discs that are glued together to create the roll body. Into those holes in the shaft we would tap in two 3-inch-long, ¼-inchdiameter spring pins. Then we would coat the discs with glue and assemble the discs onto the shaft trapping the two pins in the precut slots. The bedplate would be simple to construct. We would sandwich together fifteen stainless-steel bars, alternating a high and low bar. The two outside bars would be stainless-steel angle to serve two purposes: the top edge of the angle would become one of our blades, and the lower edge would provide us with a flange to bolt the bedplate into the tub. The blades would be staggered so that when we set them in the tub at a 6-degree angle, the ends of the bedplate would be parallel to the sides of the tub. With the stack of blades in the correct configuration we would clamp them together and cross-drill them for three bolts to hold the stack together. We would then drill the bottom edge of the flange to bolt the bedplate to the floor of the tub. The bedplate would be inserted through a hole cut into the tub floor so that the blades project about ¼ inch above the tub floor surface. After an all-night drive from Brooklyn, New York (and running out of gas 8 miles from Arrowmont!), I arrived, just in time for the class, in one of my vintage cars, a 1955 Sunbeam Alpine, with trunk and passenger side filled with tools, parts, and materials. As people helped me to unload the car and set up for the class I was very surprised to see exactly who had signed up. Some were people whom I have admired for many years and whom we would all call master papermakers; other folks I was meeting for the first time. Our class was comprised of Timothy Barrett, Howard Clark, Peter Thomas, Michelle Wilson, Peter Sowiski, Tom Bennick, Nicole Donnelly, Jen Baker, May Babcock, Katherina Siedler, John Tyler, Vanessa Adams, Robert Possehl, Brian Queen, and Shannon Brock. After a short discussion of the overall battle plan we broke into groups to make the different parts of the machine. Brian helped direct the layout of the parts on the wood. Some went off to join the fifteen bars of metal into a bedplate. Others cut the wooden discs that would become the roll or the stack of curved wood pieces for the backfall. Comfortable on the table saw, Peter Sowiski cut the precise mitered pieces that became the curved ends of the tub. Howie Clark patiently assembled all the incoming parts onto the tub base. I was excited to show off my "eureka-moment" plan to attach the wooden roll drum to the shaft and to use a jig on the table saw to make the 24 blade slots in the roll drum. By the end of the day the parts we built started to look like a sketched-in beater after having worked as a group from 9:00 a.m. until 7:00 p.m. That night I was dog-tired and because I hadn't checked into my hotel the night before, I had lost my reservation. However the very understanding folks at the school were able to put me in a cabin on the property. It was a glorious sleep that night. Over the next two days we pushed to finish before the end of the conference. A really messy part of the job was buttering up the roll slots and blades with the epoxy and pounding the 24 blades into the roll drum. I seem to recall that it was Howie Clark and Robert Possehl who took on this portion of the work and who must have taken home a good sampling of the epoxy on their clothes. At the eleventh hour we flipped the switch and had a beater with a powered roll. We filled it with water and fiber, marveled at the nice pulp circulation, and triumphantly carried it up the hill to the banquet area to show off our work. Recently at my workshop in Brooklyn, I disassembled the beater, applied three coats of varnish (although it may have been better to use a clear marine-quality epoxy finish), reassembled it, and ran it for seven hours to grind the roll blades into a perfect fit with the bedplate blades. The machine held up through the rough grinding period. Next, Shannon Brock at Carriage House Paper did an 8-hour beating test with flax. I think we can call the machine a success. Time will tell how well the wood holds up and if the glue will continue to retain the blades. Circulation with ¾ to 1 pound of fiber is very good and the machine runs without much water leakage at those levels. I would like to add some spacers between the side arms and the tub to take up some side-to-side play in the arms; I may add some adjustable stops so the roll can just go down to the bedplate but no further; and I must build a guard to go over the drive belt and pulleys. In the end, it wasn't all that cheap to build. I spent $1,100 for new parts and materials. All of the parts came from three sources: my local specialty plywood supplier, Grainger, and McMaster-Carr. When you consider all the very skilled hands that worked on the beater in the three-day class and the several days of finishing work back in my shop I would guesstimate that it would take a single person at least a solid week to build. It was a great experience thinking through the project and working with everyone on the beater; and for the right person it could be quite rewarding to outfit the studio with a home-built, hardware-store beater.

‍