How To Build and Operate
Shaker Tables
Part – I Table Fundamentals By Gary Weishaupt Rev 11-4-2009
Shaker Tables
Table Fabrication – Part I Introduction For a small-time claim operator or a successful multi-site prospector who has been fortunate enough to find a significant amount of good paying raw materials the first piece of large equipment they typically begin to investigate is the so-called ‘Shaker Table’. In the initial stages of developing a site a very simple Miller Table will usually be effective for separating a fairly significant amount of concentrates but the problem with these water tables is that they are relatively slow. The typical progression most miners make in their separation endeavors beginning from day one is standard panning of the concentrates. Once you start accumulating a quart or two of cons per day most people go out and buy any one of the numerous ‘bowls’ or ‘wheel’ type devices and these work well for small batch work. Once you begin to recover a gallon of cons every day then the ‘Poop-Tube’ and Miller Tables come into play and if you are fortunate enough to gather more than a gallon of concentrates daily a Shaker Table is almost a necessity if you want to have any time left over from processing to actually work your site. Gold mining is an incredibly time consuming undertaking and anything you can do to save time will just you make you more money in the long run. Unfortunately most things we do to save time also cost a lot of money. Even a cheap basic plastic ‘Blue Bowl’ system with accessories costs around a hundred dollars. The more sophisticated ‘Desert Fox’ spiral wheel system runs around three hundred and the even more sophisticated ‘Gold Genie’ will set you back almost five hundred dollars. None of these bowl systems are fast enough for production work even though they all work very well for the part-timer who only has to worry about processing a quart or two of concentrates every day. Once you start exceeding this quota it’s time to look into faster methods of operations and these include the use of a Miller (water) Table and/or Shaker Tables. For the recovery of ultra-fines both devices are normally used in conjunction with one another. Both of these devices are extremely easy to build and you can most certainly build even a relatively sophisticated Shaker Table for less than two hundred dollars.
Shaker Table Historical Development Almost anybody who has been shown how to make gold separate from the black sands and climb up the side of a pan by tapping on the pan edge has already used a miniature version of a Shaker Table. This phenomenon has been known to miners since pans were first invented and is one of the standard techniques miners are first taught. 1
Shaker Tables As early as 1820 engineers had already understood that a bare sluice at a shallow slope running without riffles was extremely effective in separating light and heavy materials but there wasn’t any effective way to ‘stop’ or ‘divert’ the heavy materials until some guy applied the old pan tapping routine into the operation of a bare sluice and the so-called percussion tables were born. These devices were sometimes called ‘rocking tables’, ‘gravity tables’, ‘jerking tables’, ‘bumping tables’ and most commonly ‘percussion tables’. Today they are almost universally referred to as ‘Shaker or Shaking Tables’. As far as is known, Robert Stagg designed the first successful table used for a large-scale operation in 1828, specifically for concentrating lead ore. Over the next fifteen years all types of table patents were issued based upon his initial concept. Most of the early devices however were incredibly complex and expensive to build, difficult to tune and operate and not to effective in their general operations. In 1844 a German engineer named Peter Rittinger thought to himself that he could take advantage of this remarkable ‘bumping’ effect but on a much more efficient scale and greatly simplified in operation. In other words he invented a machine that was cheap and easy to build and operate. His invention was called the Rittinger Percussion Table and almost every other Shaker Table made today is based upon his original design ideas, which were genius in their simplicity.
Figure 1
2
Shaker Tables Figure 1 illustrates a side view of a Rittinger table looking up towards the feed box. Note the multi-lobed cam. The ‘spring’ was simply a long piece of thin wood mounted vertically at the opposite end of the machine. The table deck is supported on rods from a framework that surrounds the entire table. None of this development in the early days was done with the recovery of gold in mind but was instead directed primarily at Coal mining, which is where the real money was in that period of time. Ironically the commercial success of the Rittinger machines however came from the gold and silver fields as miners adopted the new ‘Tables’ in droves in almost every country of the world. The real beauty of Rittingers device, unlike those of his predecessors, was that it could be built virtually anywhere there was good timber by almost any half way skilled carpenter. The Rittinger tables were so successful that several manufacturers openly copied the design concepts and marketed their own even less expensive devices. The most successful among these were the so-called ‘Gilpin County Bumping Tables’ from Colorado as seen in Figure 2. Note the suspension rods holding the decks and double-lobed cam and large coil spring. These machines may look incredibly primitive but they were actually very efficient.
Figure 2
Most experts agree that the functionality of the old original Rittinger or Gilpin tables is hard to beat even today but keep in mind that during the industrial revolution a man could become a millionaire just by improving on other peoples ideas and this happened when Arthur Wilfley came out with his version of a ‘Bumping Table’ that was finally patented in 1895. 3
Shaker Tables
Wilfley’s table was specifically designed to process Lead/Silver ores in the Colorado mines and while based upon Rittingers concepts his table differed in many ways, especially the drive mechaninism, and is still being used today in modern mines all around the world. The Wilfley table for all practical purposes has become the defacto standard by which all other Shaker Tables are judged even though the design is over a hundred years old. Another table worth mentioning is the ‘Cammett’, which was produced, in the late 1800’s. This table pioneered so many ‘modern’ elements of Shaker Tables that it would be at home competing against any of the currently produced machines but it utilized the original Rittinger drive system. The Cammett tables utilized router-incised riffles in lieu of raised riffles. They invented an automatic classifying feed system that distributed the particles pre-sorted onto the tabletop, which significantly enhanced their effectiveness. These were also the first machines to use roller bearings under the tabletop frame and to build composite lightweight non-warping decks. Basically they were decades ahead of their competitors from a technological standpoint. Ironically the Wilfley Company sued the Cammett makers saying their design infringed on their patents. Eventually the Cammett group became cash strapped trying to fight the court battles and sold their entire operation and inventory to Wilfley. In a strange quirk of fate the courts finally decided that Wilfleys patents were not violated in the first place but by then it was too late and these great machines were never again manufactured. Readers can do their own research on the historical development of Shaker Tables as there is a considerable amount of printed material available on the subject in old Mining Engineering books but their time will most likely be better spent if they concentrate on the design development of the Wilfley Tables in particular. These tables are still in production today and are constantly being refined but even the old original 1895 design is extremely effective. Most of us small time miners will be most interested in the smaller modern tables, sometimes called Lab Tables. These units range in size from 24x48 to 17x40-inches. Examples of these machines are the U-Tech RP-4. Deister makes an excellent lab table called the model 15-S and I’ve posted a video of this unit in operation at the Canadian Prospectors Forum. There are several makers of these small tables and it is not uncommon to see the same table marketed under different trade names. For example Keene sells the U-Tech RP-4 under their own model number ST-1 and the Deister table is sold under other brand names as well. Figure 3 illustrates the U-Tech RP-4 table with its optional stand. Note the three-point adjusting system. These tables sell for around $2500 and according to the maker can
4
Shaker Tables process about 800 pounds of raw material per hour. Most users report that the figure is wildly optimistic and that it should be more like 400 pounds per hour.
Figure 3
Figure 4 depicts one of the small Wilfley tables made by Motive-Traction and this one is set up to be portable with a self contained re-circulating water supply.
Figure 4
This particular unit was custom made for a Geologists and the umbrella has sockets built in all around the table so it can be moved during the day to provide constant shade for the 5
Shaker Tables operator. It’s pretty slick but I’d add some beverage holders as well for those hot summer days. This is about the best thought out table I’ve ever seen but you can build one at home without a huge outlay of cash. Angus MacKirk makes a very small table called the ‘Orofino,’ shown in Figure 5, that sells for around $1500 but in my opinion it is really to small to be very practical for a small time production operation but it probably would work well for a weekender or hobbyist type application. This particular table is unusual in that it operates from a 12-volt power supply and is set up to run in re-circulating mode for remote locations. It is relatively new however so there hasn’t been a lot of user feedback posted at any of the discussion boards. It looks to be an extremely well engineered piece of gear but we’ll have to wait and see how they work out. The only drawback I can see at first glance is the very small size of the deck in comparison to other portable units.
Figure 5
6
Shaker Tables
Theory of Operations As mention earlier all Shaker Tables operate on the principal that materials of different specific gravities can be separated or isolated from one another if they are subjected to an impulse force (bump). A rapid series of bumps accelerates the separation process and tends to cause particles of similar specific gravity to become grouped. To understand this phenomena at a scale larger than simply tapping on the edge of a gold pan you can take a long piece of wood positioned at a very shallow angle and place your raw concentrates at the low end. Then take a hammer and continually give this piece of a wood a good blow and over time you’ll see that the material begins to become segregated with the heavier particles migrating up the incline. This is exactly how the earliest percussion tables operated. Some smart engineer realized that the process could be improved upon if you simply added another force to the table that operated in a direction that was perpendicular to the hammer blows and the perfect secondary force came from a sheet of flowing water that continually washed sideways across the table. When you combined the two forces, one a direct longitudinal blow, with a perpendicular water flow the particles now followed a path, based upon their specific gravity, which was in a parabolic curve. This curved path was about three times longer than the old previous straight climb up the board so the particles were exposed to the forces far longer and as a result had more time to separate so the isolation became much more refined and graduated. This phenomena and the processes that capitalize on it have not changed or been improved upon in over a hundred years. Instead the development has been towards creating machines that can better take advantage of these physics. The point I’m trying to make is that you’ll probably never be able to improve upon the basic operational nature of these tables but that you can build and develop the ‘mechanicals’ to an amazing extent. This is good news for us budget minded types as it means we can build low-cost tables that perform as good, perhaps even better, than the very expensive commercial models if we pay attention to the details and are willing to spend time in learning how to properly operate the devices. Unfortunately Shaker Tables, like Miller Tables, are not the type of device you just unpack and start using on day one with 100% efficiency. There is a significant learning curve involved. Over the years I’ve had a chance to look at a lot of shaker tables, both home-brewed and commercial products. I’ve also had a chance to design several commercial tables and in my opinion almost all of the lot, including my own renditions, are unnecessarily complicated. I can understand where the commercial builders have gone a little overboard in attempting to stretch the functionality of the products but home based fabricators seem to have become caught up in trying to emulate the big guys so in this article I’ll try to go back to the basics so that we can all start with a level playing field. 7
Shaker Tables
Table Orientation From the earliest of times table builders and operators referenced the orientation of tables to correspond with the function of a table and this practice continues today. It’s handy if we’re all using the same nomenclature when talking about tables as it helps to eliminate confusion. Figure 6 illustrates a typical table and we’ve added some text characters around the four quadrants.
Figure 6
The end of the table labeled ‘A’ is typically called the ‘Drive’ end or ‘Motive’ end even if the mechanicals are mounted underneath the frame. The Board on this end of the table is the ‘headboard’ and it just keeps water and feed material from dripping off the end. The long side of the table labeled ‘B” is the ‘Feed’ side and typically includes the feed trough and water distribution system. This is high side of the table. The far end labeled ‘C” is called the ‘Concentrate’ end but on some tables a small portion of the concentrates may be collected just around the corner for a few inches on side ‘D’ The side labeled ‘D’ is the ‘Tailings’ side but middlings and waste also are collected along this side. Side ‘D” is always slightly lower than side ‘B’. In most instances the concentrate end, ‘C’ is just slightly higher than end ‘A’ but the primary slope is always from the feed side to the tailings side. 8
Shaker Tables
Table Design Basic Assembly Parts Shaker tables consist of four basic parts plus a ‘foundation’ which is usually any substantially heavy and massive solid object that can absorb the vibration forces transmitted by the table itself with are considerable. In the old days the foundation was a massive stone or concrete pour. Nowadays the foundation is roughly akin to the foundation you’d make for a small house. If you’re setting up you own table at home the foundation will most likely be the concrete slab in your garage, workshop or shed. For a portable table the foundation will be some kind of relatively heavy steel ‘stand’ and it really does need to be heavy so these stands will typically be made from thick-walled steel rectangular tubing like 2x3x.185, .25 or even .375 wall material. Even then such a portable frame or stand really needs to be temporarily anchored to some kind of foundation but the mass of this foundation depends to a large extent on the size of ones table. I cannot over emphasize how important it is to have the primary table stand anchored into something that can absorb the vibration transmitted down from the tabletop since any ‘slop’ in the system is just lost motion that degrades the effectiveness of the tables basic operational nature. A good 85% of all table problems can be traced back to inadequate anchorage or lack of mass in the undercarriage. A permanently mounted table assembly will be supported by what I call the Table Frame, or table stand that is typically a completely separate fabrication apart from the ‘foundation’ even for a portable table. Figure 7 shows the substantial stand for the little RP-4 table. On top of the table stand will be what I call the Table Carriage and this is the fabrication that contains the suspension system and the support system that the Deck (tabletop) frame attaches to. The drive mechanism is usually mounted to the table stand or foundation and not connected to the suspension carriage except by the drive rod or drive link. The tabletop (deck) is mounted to the deck frame with hinges and adjusting bolts so the tilt can be changed as needed. The last major component is the drive system that consists of a motor with a linkage system, rollers or cams to move the tabletop back and forth. There are other smaller components like the raw material feed system, water system and launders but the major pieces, the most complicated and expensive to build are as just listed. There are minor variations in the exact configuration of the parts but this isn’t too important at this stage.
9
Shaker Tables
Figure 7
Table Carriage The table carriage is the most expensive and complicated component of a Shaker table and is generally built as a completely separate unit. The carriage frame will support the deck frame, house the suspension system and launders and provide attachment points for all the small ancillary bits and pieces that make a table ‘work’. As a result is can become quite complex and this is the primary reason it is usually a standalone entity. Sometimes this carriage will be called a sub-frame or a suspension frame. Using the RP-4 again for illustrative purposes the Carriage is shown with the red arrow in Figure 8. What can’t be seen in this snapshot is the sub-frame that carries the deck itself but we’ll get to that assembly later on in the article.
10
Shaker Tables
Figure 8
The most important part of the Carriage assembly is the table suspension system as this is what allows the table deck to move.
Suspension System There are several different ways to mount the deck frame to the carriage frame that will permit the table to move back and forth but basically all methods fall into one of two categories. The first category is that comprising various ‘rail’ systems while the second is that comprising various ‘link’ systems.
Link Systems Link systems are the cheapest and easiest to implement but there are some potential drawbacks to their use that need to be considered. The biggest potential drawback with any link type system is that there is a certain amount of vertical motion transferred to the tabletop since links travel through small arcs as they cycle through their motion. The longer the link is or the shorter the horizontal range of motion, the shorter this vertical movement component.
11
Shaker Tables Most of the time this vertical motion of the table doesn’t seem to affect the operation but running at high speeds with an extremely short throw can cause a table to behave more like a vibrator instead of a percussion device. For most applications however links work just as well as rail systems. Figure 9 illustrates one method of using very long links that holds the vertical motion to a minimum.
Figure 9
These link types can be made from 1x2 hardwood and the pivot points can be equipped with nylon bushings to reduce wear. Figure 10 illustrates an interesting link design since it incorporates two tables on one frame. Links don’t necessarily have to be pivoting arms as solid bars, tubing or rods can be used to suspend a table frame from an overhead structure. The ends of the bars or rods are mounted with beveled washers to permit the same type of movement seen in more conventional link systems. This was the method of suspension used on many of the early tables as illustrated in the woodcut of the Gilpin Table from Figure 2.
12
Shaker Tables
Figure 10
Links can also take the form of flat ‘sheets’ or ‘strips’ of material that have some elasticity so that they bend under the applied force from the drive system. A good example of this link type is seen in Figure 11 where ‘Drifter_046’ from the Canadian Prospectors Forum built his own table using strips of nylon as the suspension system. The very popular Gemini tables use two diaphragms of nylon as the suspension links. The strip type links in the photo are those four white colored pieces of nylon that run between the table base and the deck. The length of the strips will vary according to how flexible the material is so there is no hard and fast rule about this element of the design. Notice in this table project that the ‘stand’ for the table is an old heavy welding table that’s bolted into a concrete slab. In this particular case there really isn’t a ‘carriage’ proper as the suspension strips serve the dual purpose of supporting the deck frame and raising it above the stand to provide clearance for the drive mechanicals and other accessories. This is a very cost effective system for constructing a table that can realistically process about 400 pounds of raw material an hour. That’s a table with only a 24 by 48-inch deck surface. This same system could be scaled down if you happen to have an old welding table sitting idle in the backyard that’s smaller in size.
13
Shaker Tables
Figure 11
Strips can be made from virtually any material that is flexible but strong including plywood or thin pieces of a good hardwood like maple. Steel strap stock is used on several commercial tables. Keep in mind that the vast majority of tables will only have a maximum 1-inch of total travel; that’s .5” to either side of vertical when your running course material but only .5” of total travel when running fine materials. For this reason you don’t need a huge amount of flex in the strips and in fact they should be relatively stiff and take a fair amount of force to get them deflected. By the way you don’t need to limit yourself to only using four pivot type links or strip type links since larger heavier tables might require 6 or even 8 links in order to operate properly and carry the table weight. The big advantage of using the strip type links is that there is virtually no friction or wear in the motion system so you can get by using a much smaller motor with minimal torque than would otherwise be necessary. As mentioned earlier the downside to links of any type is the slight vertical motion transferred to the deck but with respect to ‘springy’ strip type links there is also the potential problem of oscillation and vibration caused when the strips are deflected from vertical and bend and then spring back to vertical again as the deck is moved. Figure 12 and 13 illustrate the motion of strip type links during deck movement.
14
Shaker Tables
Figure 12
Figure 10 illustrates the deck at rest, just prior to being moved by the drive system and Figure 11 shows how the strip type links deflect under force as the deck is pushed in one direction.
Figure 13
If the strips were not anchored at both ends of the deck they would simply bend in nice smooth parabolic arcs but since the deck is restrained both fore and aft the strips have to deflect into two arcs; one at the deck connection and the other at the frame connection. When the table is pulled in the opposite direction with the next stroke of the drive arm the strips will have a tendency to oscillate slightly as they release tension and this vibration may be transmitted back to the table deck. Simply wrapping the strips with duct tape or some type of elastomeric foam jacket can dampen a good part of this slight vibration if it becomes a problem. Longer strips are more problematic in this respect than shorter ones.
15
Shaker Tables Rail Systems In most instances rail systems are a little more precise than link systems and they can more easily be built to handle heavy loads. I also think that they will hold up better over time before needing an overhaul. There is no vertical motion whatsoever applied to the tabletop when using rails. Rail type system can range from something as crude as a couple of flat boards that rub against each other as the table moves, to something very sophisticated, incorporating precision linear bearings running on machined rails. You can improve on the flat board system by adding Teflon buttons or nylon facing to reduce friction but you will need to add some ‘hold-down’ tabs to keep the table from lifting during motion. The hold-downs can be as simple as pieces of aluminum angle or mirror clips running inside a saw kerf. Rail systems can also consist of some hardware store casters bolted to the table support frame that roll inside some aluminum or steel channel sections screwed to the deck frame. A slightly fancier system would use a set of rollers that run inside a ‘U’ type track so the table could not lift. Johnson Hardware makes some good tracks and sliders for this type of application and they are inexpensive. Another simple rail system could be made using standard drawer slides available at any builders supply outfit. These work extremely well even on relatively large tables if you use the heavy-duty commercial slides. A rail system that I personally like to use consists of grooved rollers that run longitudinally in between a pair of rods or tubes. This type of roller system can be set up with several different pairs depending upon the equipment weight and the rollers themselves can be oriented either vertically or horizontally. You can purchase precision stainless steel rollers and rails for this type of system but they are very expensive. The last arrangement I designed for a customer cost a little over five thousand dollars and in my opinion was massive overkill. The same system can be duplicated for around sixty dollars and if you’re careful with your fabrication, it will be almost as precise as expensive units. The cheap version however will wear out every few years and most certainly doesn’t have the ‘bling’ of the fancy model. As with most home-built gear it’s a tradeoff between the Levi wearing crowd and the Calvin Klein groups. The first thing most people think about when contemplating a rail type suspension system is that they’ll need some expensive ball or roller bearing gear but history has proven conclusively that such bearings do not in any way perform well when cycled over short distances. Roller bearings are designed to handle loads, forces and cycles that are long in 16
Shaker Tables duration and go through numerous complete revolutions. If they are simply moved back and forth repeatedly for short distances they rapidly wear out and begin to bind. For this type of short-cycle movement you really need bushings instead of bearings. Bushings are ancient little pieces of gear and the oldest were made from a wood called lignum vitae, teak or Ironwood since these woods were incredibly dense and contained a tremendous amount of natural oils. Modern bushings are made from bronze alloys, many of which have been impregnated with oil hence the term ‘Oilite’ which has become a trade name. Oilite bushings are perfect for rail type Shaker Table suspension systems and also link type designs.
Deck Riffles Riffles are not always necessary and the old original percussion tables didn’t use riffles to begin with and even some of the modern manufacturers offer smooth decks so don’t be afraid to build a smooth deck to begin with as riffles can always be added later as you gain experience. No matter what type of table you have, whether it’s a top of the line commercial model or some home-built contraption about 90% of how effective that table is comes down to operator skill. This is the thing that a lot of people simply can’t understand and it’s also a reason why so much ‘junk’ is sold to newcomer prospectors who expect miracles from some kind of machine that will take the ‘brainwork’ out of mining and processing.
Table Deck Ideally a table deck is constructed to be a light as possible and still retain its shape and remain flat since a light weigh table reacts to the slightest motion much faster than a heavier deck. About 90% of table development in the last century was involved with finding ways to make decks lighter and flatter. Today almost all decks are made from fiberglass with a sandwiched foam or balsa wood core as this is the lightest system we’ve come up with so far. Milled aluminum is the second most popular deck arrangement especially for the smaller Lab type tables. Most homebuilt tables are made using plywood as the deck and this is perhaps the primary reason many of these tables do not perform as well as could be expected since they are simply to heavy and contain to much mass to function properly.
Table Motion System Many first-time Shaker Table builders are under the mistaken notion that the movement of the deck is a simple back and forth motion created by using an eccentric mounted on the table motor that drives a control arm attached to the table deck. This is a completely 17
Shaker Tables false notion and if you build a table based upon these principals all you’ll end up with is an ‘oscillator’ and this won’t work as you might expect it to even thought it may appear to be working. Shaker tables operate on the principal that particles can be separated based upon their specific gravity when subjected to a sudden ‘jolt’, ‘bump’, momentary ‘impulse’ or rapid acceleration of the surface they are placed on, and that this separation is most effective when the particles are forced to climb up a slight incline as lighter materials fall back downhill. There are actually two different schools of thought about table motion. One school holds that the table needs to be pushed foreword, towards the concentrate end, with a smooth slightly accelerating movement and then suddenly and very rapidly pulled back in the opposite direction. This is motion that cam drives provide. It is akin to the old magicians trick of pulling the tablecloth out from under the china. The other school of thought believes that the deck needs to be pushed very rapidly forward and suddenly halted in a small jerk as the deck is pulled backwards at a slightly slower velocity. This motion is akin to what you do when you toss a load of gravel off the end of a shovel. There really isn’t a noticeable ‘bump’ or ‘impact’ in this system but merely a rapid change of deck direction. Neither of these two types of motion are in a any way uniform as would be the case if you were using a simple eccentric or crankshaft to drive the table. Both of these motion concepts seem to work equally effectively and are most usually called the ‘bumping type motion’ and the ‘jerking type motion’. Some tables are adjustable so that you can change between the two motions and in some circumstances combine the two motions to a certain limited extent. The popular Gemini tables incorporate this type of adjustability. Figure 14 is a time-motion diagram for a Wilfley Table. The three curves represent different stroke settings.
18
Shaker Tables
Figure 14
Notice that as the stroke increases towards its maximum extension the time duration becomes accelerated. For example it takes .04 seconds for the table to travel one half of an inch but then only another .02 seconds to travel the next one half inch to maximum extension and then there is a sharp peak at the reversal point which is the ‘jolt’. A simple way to think of this table motion is that the deck needs to be pushed forward slower than it is pulled backwards and that there is a slight instantaneous jolt at the point where the deck transitions from the push stroke to the pull stroke. In the old days the correct motion and movement was supplied by simple wooden ‘cams’, an example of which is shown in Figure 15 from an enlargement of an original Rittinger drive system. You can buy modern steel cams from a variety of machine part manufactures as well as ‘eccentric shafts’ that serve the same purpose but the costs may be prohibitive except where you’re building a commercial level machine. We’ll discuss some simple alternatives for cam fabrication later in the article.
19
Shaker Tables
Figure 15
Cam drive systems are still used on some tables today and even though the concept is very old the system is very effective and extremely easy to build. Figure 16 is a motion diagram that I took from an old cam I saw in a mining museum. It isn’t perfectly accurate since I did the measurements with a tape measure but the curve is still representative of the motion imparted by cams used in old ‘Bump Tables’. The motion of the cam starts at the right side of the chart where the bumping beam is at rest and then as the cam is turned the bean is moved a slight amount. The range of this movement is charted as shown. You can see the forward motion is gradual until the end of the lobe cycle when the beam slips off the end of the lobe and the return movement is almost instantaneous represented by the heavy vertical line at the left of the chart. I personally like the simplicity of a cam drive system for a home built table and a full size pattern for a three-lobe cam from a modern table is included at the end of this article. The big advantage of cams is that they can be made from thick plywood and faced with thin strips of nylon so their construction doesn’t require any fancy tools beyond a scroll saw and a drill. In addition you can do some very creative things with cams to change the ‘timing’ and acceleration/deceleration curves if you want to get deeper into design work.
20
Shaker Tables
Figure 16
Unlike the cam-operated percussion tables the Wilfley tables utilized an ingenious linkage system to provide motion to the deck. This system is adjustable in that you can change the stroke velocity fairly easily. Figure 17 is an old woodcut of a Wilfley drive system.
Figure 17
The mechanism may look complicated but in reality it is very simple and consists of a differential link system that converts rotary motion to linear motion. It is loosely based on the Peaucellier-Lipkin linkage patented in 1864. Figure 18 is a simplified schematic stick diagram of the motion this linkage provides.
21
Shaker Tables
Figure 18
If you have access to a lathe and milling table you can build these drives with a tremendous amount of precision and adjustability but in this article we’ll only provide plans for something average folks can build in their garages with simple tools.
Table Fabrication – Part II Prototype Fabrication So far we’ve discussed the history and theory of shaker tables and the basic suspension and drive system but in this next section we’ll be implementing some of these ideas during the construction of prototype tables. I’m a very strong advocate for building prototypes for any kind of project for many reasons but the most important is because you can catch your design flaws and construction mistakes before you get committed into a full-blown expensive fabrication project. For something as finicky as a shaker table you might actually end up building more than one prototype before you finally come up with something that works as good as the commercial models. That is the objective, to make something that is at least as effective as the commercial versions but at less expense. Some people have the mistaken notion that the point of building stuff is simply to save money and that is true up to a point but why build something cheap that doesn’t work to well?
22
Shaker Tables There’s also the possibility that you’ll end up building something that actually costs more than a product you could have purchased in some store to begin with. I don’t really see that as being a problem if your device works better than some mass-produced piece of gear but in most cases you can always build something better, for less money, than what you can buy and that’s what keeps us home-builders motivated. Keep in mind that in this section on prototypes we’ll be primarily concentrating on only building mockups of the ‘motion frame’ and one stationary frame that is typically the upper portion of the table carriage. We’ll go into more detail in later sections about frame construction.
Design Selection We’ve talked about several different types of shaker table designs and once you decide to undertake a project you need to make some decisions about what type of table you’re going to be building. A huge part of this decision will be determined by your budget, the availability of materials, tools, and garage space and even your particular fabrication skills and experience. For instance if you don’t know how to weld, that will somewhat limit what you can do unless you can pay to have somebody do welding for you. You have to be honest and realistic about the project. Secondly you have to decide what size table your really need. That’s assuming of course that you really need a table to begin with. A prototype table can be very small because it’s just a mockup of what the final product will look like but it’s best if the initial table is fairly close in size to what you eventually need to build. There is a point where small tables aren’t really very effective and in my personal opinion a machine with a deck much less than around 18x36-inches is verging on being a hobby sized device and to have something that can really do some serious work you probably should be looking at building a table that has a deck of around 24x48-inches. There is no sense in building something that will eventually be to small for your future needs. For the purposes of this article we’ll assume that you’ll be building something with a deck size that ranges from 18x36 to 24x48-inches give or take a few inches here and there. The smaller of this range can still do some serious work and I’d estimate than you could process around 200 to 300 pounds per hour depending upon the nature of the raw materials. The larger size could handle between 300 and 400 pounds per hour, perhaps a little more.
Prototype Materials I always build prototypes using wood structural members. There is simply no reason to waste good steel on a prototype. This isn’t to say that wood can’t be used for a finished table but that steel is much to be preferred for the long-term gear.
23
Shaker Tables Even for the initial mockups you really do need to use some halfway decent wood and if possible I try to use either Oak or Maple and I prefer the Maple since it’s easier to seal and waterproof. Good wood isn’t cheap so be prepared to spend some money at the lumberyard. Finding wood of almost any kind that is straight is about as easy as finding the Golden Fleece so be equally prepared to dig through tons of materials as you cull out the pieces that are actually useable for something as simple as a table mockup. Ironically I’ve found lately that I can find straighter pieces of material in cheap Pine or Poplar at my local Home Depot than what’s available in either Maple or Oak so take a look at any species that’s on the racks and look for the straightest lengths you can find regardless of wood type. Even for a crude prototype it’s fairly important to have all of the pieces straight and true.
First Prototype – Drawer Slide Suspension System Most people who what a shaker table want to build something as simply and cheaply as possible and to do this with a reasonable chance of success you could be looking at building a table that uses ‘drawer slides’ as the suspension system. This type of suspension does work fairly well and it is extremely easy to build but be forewarned that it is not very durable so annual rebuilding is usually needed. It’s also not as cheap as some would imagine. For our drawer slide prototype shown below we bought the cheapest slides we could find which cost $22.50. The wood, which was Pine, cost another $27.00 so just to build a small rudimentary base and slider frame cost $49.50 and this is when using bottom of the line materials. For a higher quality machine using commercial slides and good wood the cost would have easily doubled or even tripled. Even ‘cheap’ shaker tables aren’t inexpensive. Of course if you’ve already found enough pay material that warrants you building or buying a shaker table then money shouldn’t be much of a problem, which brings us back to the point of being ‘realistic’ about the project to begin with. Shaker tables are used to separate and concentrate gold from ore pulp or placer concentrates and if you don’t have either one then a table isn’t going to make your situation any better than using a gold pan. Shaker tables are just another tool in our gold processing arsenal but the gold has to be there to begin with in order to be processed. Ironically a shaker actually won’t even work very well to begin with unless there is a fairly significant quantity of gold in the pulp you’re running. Sometimes a water table would work better if the ratio of Au to rock particles is very low. Assuming that you do actually need a table then a ‘Drawer Slide Table’ might be a good starting point or at least one alternative design to be considered. Figure 19 illustrates a basic deck frame support and slider frame combined into a very simple package. 24
Shaker Tables This glides we used here were the cheapest we could find. For a better rendition you can purchase glides that don’t have the ‘stop’ ears and have much better and far more precise steel roller bearings but we just needed some glides to get a few pictures so we weren’t to picky. Even these cheap glides work remarkable well on this mockup, far better than I had imagined.
Figure 19
Figure 20 shows the top frame slid to the left so you can see how this type of system operates in principal. Keep in mind that the actual table deck movement is plus or minus around .5-inches on either side of center at the maximum range of motion so the displacement you see here is far more than you’ll have when the table is in working mode.
25
Shaker Tables
Figure 20
This type of suspension system is perfectly viable for a good working, easy to build table, if high quality slides are used and kept in proper repair. However it is not a suitable system for tables having a deck frame much larger than around 20x40-inches. Like most suspension systems however there will always be a certain amount of lateral (sideways) slop in the arrangement and this needs to be eliminated or at least reduced to a minimum by any means you can devise. For a drawer slide systems the easiest way to control unwanted sideways movement is to make some ‘rubbing’ blocks that bear on the moving part of the frame. These blocks will insure that the table only moves fore and aft with no sideways deflection. Figure 21 depicts the simplest of all rubbing blocks and that’s just a small piece of wood that barely ‘kisses’ the moving frame. These are installed at all four corners of the base frame and prevent the movable frame from shifting laterally during the fore and aft movement imparted by the drive system.
26
Shaker Tables
Figure 21
If you look closely you can see two small red dots where screws will be driven to hold this block to the base frame. They do need to be removable. To reduce wear and friction you can face the rubbing block and the frame with pieces of nylon if needed. Figure 22 illustrates another type of rubbing block that kisses the steel slide on the movable carriage. Again you can face the block with some nylon or Teflon strips.
27
Shaker Tables
Figure 22
Figure 23 shows another method of controlling lateral play and this one involves using small rollers or casters.
Figure 23
These caster are installed at all four corners of the base frame and again, to reduce wear and friction you can add strips of nylon to the bearing section of the carriage.
28
Shaker Tables If you look in the ‘door’ section of almost any builders supply store you can find all kinds of rollers designed for patio doors and some of these are pretty slick with adjustable brackets and even some with spring loaded tension devices so you have to do some experimentation in this area to come up with ideas that are easy to implement. Always remember that the prototype-building phase is where you get to think out different ideas and concepts that suit your specific situation with respect to budget, tools on hand, raw material selection, and dozens of other factors not to mention your own personal skills and fabrication experiences. Sometimes you might be trying to copy an existing table design and other times you might be charting new ground so you need to be creative, inventive and open to improvisation. As one of my old bosses used to say, “If you’re not making mistakes then you’re not doing anything”. Some ideas will work and some won’t but don’t get discouraged as some of the very best pieces of mechanical engineering hardware have been invented out of pure frustration and desperation. As it stands at this stage of fabrication a little table frame and carriage like the one we just built is a good candidate for an entry-level table like the one you see posted at a lot of the discussion boards called the ‘Mongolian Shaker’ that is shown in Figure 24.
Figure 24
Second Prototype – Roller Suspension System 29
Shaker Tables Tables that run on rollers or casters should be about as easy and economical to build as the simple drawer slide type previously outlined. Roller systems however have the advantage of being usable on large-sized tables. Roller suspension tables can be basic, relatively simple contraptions or range up to extremely sophisticated designs using precision rollers in machined guides. It’s usually a matter of money which direction you decide to go. Even some of the more expensive commercial tables utilize standard urethane casters very similar to the type of wheels seen on modern skateboards. For most small-operation type table’s normal rubber casters are generally more than adequate. Please don’t try to use the cheap plastic casters since they simply are a waste of time and money. For a roller suspension system we’re basically just taking the old ‘drawer-slide’ design concept and recreating it at a much larger scale. For some people this system will actually be easier to implement. From an expense standpoint both systems are pretty much equal but it depends on the specific hardware you decide to purchase. As before we’re working with two basic frames. One is the stationary frame that is typically fastened to the table stand and the other is the ‘movable’ or ‘suspension’ frame that will carry the table deck frame. The first thing to do is to go out and buy four ‘fixed’, non-rotating, casters from the local hardware store. Usually you’ll end up with something that looks like the products show in the following snapshot.
Figure 25
The caster on the left is a fairly standard hard rubber roller that has an outside diameter of 2-inches. This size can be purchased is several configurations including urethane and even high-tech units with sealed roller bearings but the external dimensions are generally identical to the unit shown here. The roller to the right is just a one-inch version of the
30
Shaker Tables same product. These rollers are very affordable and last for years so buying something more sophisticated for a Shaker Table isn’t really necessary. As we mentioned earlier the motion of a Shaker Table isn’t really rotational so wheels and roller bearings aren’t the ideal supports for a simple back and forth motion but they do work well for several years without replacement in most situations. In use the casters are bolted onto some structural members on the carriage as shown in Figure 26.
Figure 26
Note that the casters are installed ‘upside-down’ so they serve as rollers for the upper part of the carriage, which is normally the deck support frame. Installed in this manner there is little chance of the wheel axles or the runner tracks getting clogged with dirt or sitting in water as there is a half-inch gap between the lower quadrant of the wheel and the galvanized mount. The 2-inch rollers work well even on small tables and for full-size tables you might install six to eight larger rollers to evenly distribute the weight of the deck.
Third Prototype – Flexible Strip Type Suspension System We’ve already mentioned and shown some examples of tables that use flexible strips of various materials as the suspension system. This system is very easy and very cost effective to build. There are several commercial tables on the market that use this system
31
Shaker Tables in their construction, such as the Gemini and RP-4, but there is one big drawback to the design that you need to be aware of. Tables that use flexible diaphragms or strips as the primary suspension component transmit a huge amount of energy back down into the carriage and frame assembly and for this reason the frame and anchorage needs to be substantially more massive than what you’d use to mount other types of table suspensions. If you’ve already built some of the other prototypes we’ve outlined you will immediately see what I’m talking about when you put together your first ‘flexi’ table as I call them. You can easily hold most prototype table carriages with one finger on the workbench as you cycle it through the throw motion without a lot of excessive vibration but if you try to cycle a ‘flexi’ table manually you’ll discover that it shakes your entire workbench all the way to the garage slab. This is not an exaggeration in the least amount. Even a small flexible strip shaker table needs a relatively massive frame and substantial anchorage into a fairly large mass of concrete. For remote location work Keene actually recommends that you pour a concrete slab to mount their little ST-1 Flexi table. This is simply impractical if you’re still in the testing and sampling mode at a small claim. Another problem with flexible type suspension systems is secondary or residual vibration and oscillation so usually on this type of table you need one or more adjustable spring loaded dampers, which in my opinion, just adds to complexity that most of us can do without. Some operators feel that this residual vibration is actually an enhancement and enjoy the ‘action’ these tables provide compared to a link or roller suspension table. If you can set this type of table up in a permanent or semi-permanent location and build a substantial steel support frame mounted on a good foundation they work extremely well and are almost completely trouble-free. Unfortunately this is not the type of suspension system that you can clamp down on the tailgate of your truck for quick and dirty test work. If you can build the support system and anchor it to a slab somewhere these types of tables are in all other respects very nice and extremely simple and economical to fabricate. This types of tables can be operated without anchorage if absolutely necessary but you’ll loose at lot of effectiveness as portions of the deck oscillations will be wasted as it’s transferred back into the stand and eventually into the ground or other surface the table sits on. It can be done, but I wouldn’t recommend doing it, unless there is no way to tie the table down.
32
Shaker Tables Figure 27 illustrates a very rudimentary flexible strip type suspension system. In this case the strips are just pieces of 1/4 by 1.5-inch plywood. A single diaphragm of 1/8-inch plywood at each end would have worked just as well and been easier to fabricate.
Figure 27
Figure 28 shows the terminal connection at the individual strips. In this case I just used pieces of 1x.125-inch aluminum strap stock as compression plates but I’ve found through experience that small ‘U-bolts’ having the same width as the strips allows the material to flex in a much more natural manner. To prevent the strips from ‘climbing’ behind the compression clamps you can install a small ‘stop-strip’ or even two screws situated towards the ends of the strips at either termination end. My personal opinion is that ‘flexi-strips’ even though cost effective are more trouble than they are worth from a fabrication standpoint. While they appear to be incredibly simple to build and implement they are in fact somewhat complicated to get set up properly. If you’re building several tables it is probably worth the time and effort to develop a specific design that works well but for a one-off table it’s a lot of work compared to other types of suspension designs.
33
Shaker Tables
Figure 28
You can see in this snapshot how the aluminum strap simply ‘clamps’ over the suspension strip. You can always just use two screws through the strips at the connection points but by doing so you’ll eventually develop stress cracks in the strips plus a ‘solid’ connection hampers the ‘action’ of the strips.
Fourth Prototype – Link Suspension System Link suspension systems in my personal opinion are the cheapest and easiest to build and have the added advantage that they can be scaled up or down for almost any size of table from extremely small to extremely large. Like ‘flexi tables’ they have some drawbacks with respect to vertical deck motion but using long links can mitigate that to a great extent. Figure 29 illustrates the basic prototype parts for a links suspension table.
34
Shaker Tables
Figure 29
We have the base frame, which will be secured to the table stand, and a ‘mobile’ frame, which will carry the deck frame. The two frames will be separated by the ‘links’. For this mockup I made some short links out of scrap 1/4” by 1-inch aluminum strap stock. Keep in mind as with most tables the number of ‘support’ parts between the fixed and movable frames will vary according to the overall size of the table you’ll be building. For a small table 4 links will work while for a large table you will probably want to use 6 to 8 links. Since this is just a mock up I didn’t go to the expense of actually buying bushings and shoulder bolts for the link pivots and decided to use nylon bushing and standard bolts instead. The basic components after assembly are shown Figure 30 where you can see how the links support the movable frame that will hold the table deck. The lower frame will be bolted onto a table stand or other support system. As mentioned, these tables are extremely easy to build and in my opinion have the best performance, considering fabrication costs, of any other type of table.
35
Shaker Tables
Figure 30
Figure 31 shows the table with movable (upper) frame pivoting on the links to the left so you can get an idea of how the suspension system operates.
Figure 31
36
Shaker Tables The links on this mockup are 6.5-inches from center to center and as a result there will be a vertical differential of 0.023-inches as the deck moves thru one oscillation cycle. For a real table I typically use links having 12-inches between pivot points which reduces this vertical movement component to 0.007-inches which is less than that found on the RP-4 and Gemini tables. I think you could probably go as short as 9-inches and still have a well-behaved table but I’ve never tried it.
Deck Frame So far we’ve only shown the structural part of the upper (movable) suspension frame to keep things simple but on top of this support frame is another frame that actually carries the deck surface. This component is usually called the ‘deck frame’ for obvious reasons. Figure 32 illustrates a typical deck frame for a small (20”x40”) table. The deck surface, typically thin plywood, is screwed onto this framework.
Figure 32
For a mockup prototype we’ve used wood joined with ‘biscuits’ but for a production table the frame would be made from thin-wall steel or aluminum tubing. It needs to strong but as light in weight as possible. This frame is attached to the ‘movable’ frame with hinges so that it can be tilted as required for the type of materials that you’ll be processing. Normally the slope will never exceed 1/4 to 3/8-inch per foot so it’s actually very shallow but it does need to be adjustable if at all possible. 37
Shaker Tables In Figure 33 you can see the deck frame temporarily positioned on top of the upper portion of the suspension (movable) frame of the ‘Drawer Slide’ prototype we built. In this snapshot I’ve just shimmed it up into a tilted position for illustrative purposes.
Figure 33
The hinges would be on the underside, and between, the deck frame and the support (motive) frame at the edge that is towards the bottom of the photograph and two adjusting bolts would be on the opposite side.
Motive Power One of the most difficult tasks you’ll face when building a Shaker Table is designing the motive power system and finding just the right electric motor to drive the equipment. Larger tables are easier to do than smaller ones because for the bigger tables you can just use almost any one-half horsepower A/C motor and these are fairly well standardized with respect to frame mounting dimensions and available pulley selections. These types of motors come in several different ‘frame’ designations. The most common that you’ll find in hardware stores is called the NEMA 56 ‘base-mount’ frame type that has an integral mounting frame. This is the type of motor you’ll find driving drill presses, bench grinders, band saws, air compressors and a wide variety of shop power tools. The other type of NEMA 56 motor is called the 56-C ‘face mount’ and these go all the way down to one-forth horsepower and are generally much smaller in external dimensions. They are designed to bolt onto a fabricated mount or to another piece of 38
Shaker Tables equipment like a fan cage or a direct drive for a bench mounted belt sander. There is also the NEMA 48 base mount that has a slightly smaller footprint. Any of these motors with various mount types are available from almost any industrial supplier such as Grainger’s or McMaster-Carr but they aren’t inexpensive and having to buy a new motor is where most home-built Shaker Table projects get relegated to the sideline. This is where ‘scrounging’ and ‘cannibalism’ really become important. You can usually find a huge variety of used motors at swap meets, yard sales, junk stores, salvage yards and used appliance stores. Sometimes you can buy an entire piece of equipment with a good motor in it for next to nothing and then reuse the motor and throw all of the other parts in the trash. I’m constantly on the look out for motors and pumps at garage sales so don’t buy a new one unless you absolutely have to. One of the ideal equipment candidates for being cannibalized is any old diaphragm water pump as these usually have very durable but small motors and some kind of eccentric drive mechanism all in one nice housing. Regardless of what you end up with as a power plant there is usually little doubt that it’ll be running at either 1140 or 1725-40 rpm’s. Of course the Shaker table only needs between 200 and 300 cycles per minute so you’ll end up with some kind of pulley and vbelt system to reduce the rpm’s between the motor and drive shaft. If you luck out you might end up with a variable speed motor and then you could go with some kind of direct drive but this is unusual. You can buy a speed controller for most motors but then again they are relatively expensive. During the mockup stage you can use a regular old variable speed electric drill to drive your table as they can be set to run in the rpm range you’ll need. In practice you use some electrical tape or a hose clamp to keep the trigger depressed at the speed range position you want. Large hose clamps work really well because you can adjust the speed range using the clamp screw. Assuming that you’re on a budget and find a good electric motor that typically will be running between 1725 and 1740 rpm we need to find a way to reduce this speed to drive our eccentric shaft to provide movement to the table deck support frame at around 290 to 300 rpm. The easiest and cheapest way to do this is with a system of pillow block arbors and different sized pulleys. There are all kinds of Internet pulley ‘calculators’ that you can use to find the various combinations of pulley sizes that you might need for your specific situation but in general the rule of thumb is one of ratios. For instance a motor fitted with a 2-inch pulley will drive a shaft fitted with a 4-inch pulley at one-half the motor speed. It’s a simple 2.0:1 ratio. The same motor driving a 5-inch pulley would have a 2.5:1 ratio and driving a 639
Shaker Tables inch pulley it would have a 3.0:1 ratio or speed reduction. It doesn’t take long to realize that a motor running at 1740rpm needs to be turning a shaft fitted with a 12-inch pulley in order to get the speed down to 290rpm. For a small table there probably won’t be enough room to use a 12-inch pulley so we need to look at multiple or staged speed reduction by using more than just two pulleys and pillow block bearing assemblies. If you’re lucky enough to latch on to a motor that only turns at 1140 rpm you can get by with just a 2-inch primary drive pulley and an 8-inch final drive pulley which simplifies things tremendously. All of this gear takes up a lot of space and in some cases the drive assembly will actually be larger in physical dimensions than your table. Unfortunately I have no idea what kind of motive power system you’ll be using so I can’t give any hard dimensional drawings for this part of your particular table and this is one reason that I recommend that you build the motive system and it’s support frame as a separate element. You can bolt it to your table frame or incorporate it inside your carriage or stand when the table is finished. I wish I could be of more help in this area but the variables are simply to great for me to illustrate specific design alternatives. As an example I set up two motors on the bench and both are rated as 1/4 horsepower. One is 120 volt and the other is 12 volt. Both motors have almost the same torque curves but as you can see in the snapshot the physical dimensions, shaft size and mounting points are radically different.
Figure 34
40
Shaker Tables My recommendation is to design for accommodating an 8-inch pulley somewhere in your power transmission system, as this will be the largest pulley that you’ll probably ever use. Allowing room for this size then makes it easier to work things out if you developed a drive system that uses smaller pulleys. A rudimentary schematic diagram of a simple power transmission system is shown in Figure 35.
Figure 35
This system would be appropriate if you have a variable speed motor or a conventional motor equipped with a speed controller or even a motor that ran at a fixed speed of around 300 rpm’s. In action this system spins shaft ‘A’ that is mounted between a pair of pillow block bearing assemblies and in turn rotates ‘cam-3’ that actuates the eccentric control arm or hits the tables bumping block. Figure 36 illustrates a more complicated system that relies on the use of various pulleys to reduce the speed of the final drive shaft. Shaft ‘A’ is mounted between a pair of pillow blocks on the lower portion of the table support frame and shaft ‘B’ is mounted in a similar manner but onto the upper portion of the support frame. The entire assembly may be completely contained within the confines of the table frame or mounted on a platform at one end of the table frame. In action the motor drives shaft ‘A’ at a fraction of the motor speed by use of a larger pulley. This reduced speed is then further reduced between shaft ‘A’ and shaft ‘B’ by an even larger pulley. 41
Shaker Tables
Figure 36
Shaft ‘B’ then drives the bumper cam or eccentric at around 300 rpm. This may look complicated but in application it is extremely simple and cost effective to implement compared to the cost of a motor speed controller; Note than I haven’t shown the belt tension idlers or other minor details in these simple illustrations. Once you have bought or scrounged all of your drive gear you’ll be wondering how to cram 5 pounds of stuff into a 2-pound bag but it can be done with some creative design work. Figure 37 depicts a portion of an engineering drawing for a small table equipped with a Gemini table type drive system and a rather large 115-volt motor. All of the gear fits into a space that is 24-inches long by 12-inches high and only 18-inches wide.
42
Shaker Tables
Figure 37
As compact as this particular installation is it can get even smaller if we leave behind the old 19th and 20th century design concepts on which it is based. Most of the older table designs use what I call a ‘horizontal’ design scheme since they are easy and economical to build but newer tables, having very small footprints, are using a ‘vertical’ design scheme where the shafts are oriented vertically instead of horizontally. A good illustrative example of this scheme that most of us can relate to is the average modern drill press which can be adapted to table use quite easily.
Table Carriage Assembly and Stand So far on our prototypes we’ve only been two parts to illustrate the general operational nature of Shaker Tables. One was the deck support structure and the other was the lower part of a typical carriage. Now we need to look at building a complete carriage and stand. The reason this section may seem out of sequence is because the design of the carriage and stand will depend to a large extent on what type of motive power and drive system you end up using. That’s why the ‘power’ section of the article was placed ahead of this one. Once we have an idea of how much space will be required for the power system and whether it will be mounted internally or externally we can finish up the carriage frame and stand, if we’re using one. Figure 38 is a snapshot of our link prototype with the upper part of the carriage finally installed.
43
Shaker Tables
Figure 38
Now we’re to the point of having something that actually looks like the basis of a small Shaker table. Always keep in mind that the prototype doesn’t have to be a work of art so long as it’s reasonably accurate and strong enough not to get beat to pieces on it’s first test run. The internal components shown in this picture are just shimmed up on temporary blocks so I could verify spacing and make sure that I could actually get access to the parts with a wrench and screwdriver. This scheme requires that the motor be mounted upside down below the bearing block supports in a ‘table saw’ fashion down inside a stand that we’ll build later. This is an extremely small carriage, being only 12x18x6-inches in dimensions but it uses full sized components. By moving one of the shaft assembly mounting points to the upper carriage frame I could actually get the motor installed inside even this small frame. This is why prototyping is so important. I’ve found through experience that you can develop a design far faster by using a combination of drafting and building the bits and pieces of prototypes, than by simply drafting the conceptual ideas on paper first. Most designs based exclusively on paper design engineering tend to end up being larger in dimension than they really need to be. Remember that what we’re illustrating in this article so far, represents the most basic concepts that are cheap and easy to build without recourse to any fancy tools or equipment so they illustrate fundamental concepts more than hard and fast building instructions or design ideas.
44
Shaker Tables Once all of the power transmission parts have been collected and we’ve decided on a couple of different ways to arrange and mount them it’s time to look at how to couple that power system into something that actually provides the ‘Shake’ to a Shaker Table and I call this the ‘Motive’ or ‘Motion’ system.
Motion Systems This is a good time to go back and re-read the first parts of this article that talked about the various types of tables built in the past and what types of ‘Motion’ systems they used. Perhaps the simplest of these early designs was the Rittinger Table that used a cam to drive a bumping bar that was held in tension by a ‘spring’ made from a long piece of 1x4 pine. Even today this design is more than adequate for a Shaker Table and actually forms the basis for almost all modern table motion systems except the Wilfleys. The Rittinger system consists of a ‘cam’ that moves the table deck in one direction and then a ‘spring’ that pushes the table in the other direction where it hits a ‘bumper’ that provides the jolt or impulse that causes the particles on the deck to move. This same system, in various guises, is still used today on the popular Gemini and Keene tables. To incorporate a similar system into one of the prototypes the first thing we have to do is install a ‘spring’ at one or both ends of the table. The easiest and cheapest way to do this is to adopt the ‘flexi-strip’ concept developed by Rittinger. This is the same ‘flexi-strip’ system we’ve already described elsewhere and well tested over time by numerous table manufacturers but in this instance we’re using it in a slightly different way. Figure 39 shows our link table prototype fitted with a flexible strip spring at one end. The strip shown here is just a piece of 1/8-inch plywood about 1.5-inches wide by 6inches high and even though it is relatively small it provides a significant amount of resistance and could be used on tables having decks as large as 24x48-inches. You can spend a lot of time trying to develop coil spring alternatives to this antique design concept and come up with next to nothing but you can pretty much try anything that meets the requirements of your particular table concept. This system works and it works extremely efficiently so I suggest that it be adopted without a huge amount of time spent in experimentation. On large tables you might need to build up a ‘leaf’ type spring from several laminations of thin materials.
45
Shaker Tables
Figure 39
Figure 40 is another snapshot of the strip taken from a different vantage point.
Figure 40
There is 3/4 of an inch space between the strip and the carriage frame that is more than enough to provide clearance when the deck cycles through it’s throw range.
46
Shaker Tables As the table is cycled through a motion period we need to provide some way to control, limit and/or dampen its movement and for this function coil springs do work well where they function basically as ‘shock absorbers’ to dampen the reaction from the tension of the plywood leaf spring. On most tables the secondary tension adjuster also serves as an adjustment to shorten or lengthen the throw of the table and to strengthen or lessen the ‘action’ of the motion and the ‘bump’. For such s simple little device these adjusters are a real pain in the butt to tune properly. Finding the parts that are ‘just right’ for your particular table can also be a real pain and five or six trips to the hardware store can be expected.
Figure 41
Figure 41 shows the secondary tension adjuster in place on the table. It’s that little object in the middle of the block of wood held to the motion frame with a C-clamp. On this table it’s at the end opposite the leaf spring. Some tables will have a tension adjuster at both ends. The glue was still drying when I took these snapshots. The wooden block is doweled into the frame. Figure 42 is the adjuster taken from another angle. You can more easily see that its made up from two coil springs some nylon bushings, washers and a bolt that runs all the way through the motion frame and the carriage frame. The hole for the bolts needs to be fairly well oversized so the adjuster doesn’t bind as the table cycles through its motion. When the table is idle and not in motion all the adjuster does is keep the table deck centered over the carriage frame.
47
Shaker Tables
Figure 42
In figure 42 the table is stationary and you can see that the spring between the two frames is in full extension while the spring on the outside of the motion frame is in full compression. This particular spring is a little too tight but I was too lazy to run down and buy a screw that was 1/4-inch longer. In Figure 42 I’ve but some pressure on the motion frame and pushed it back against the tension of the leaf spring and you can see now that the inner spring is in compression. As the table deck is cycled back and forth under power these springs constantly change functional roles from compression to extension and thereby control and regulate the throw and ‘impulse’ strength. You can spend a lot of time making these tension adjusters fairly fancy and by using allthread rod and an adjuster nut at both ends you can change the rate of each spring independently. This is one area where individual experimentation can pay big dividends. These are very small adjusters on this mockup table and you’ll have a much wider variety of springs and other parts once you begin on the full sized table project.
48
Shaker Tables
Figure 43
Figure 44 shows the opposite side of the motion frame so you can see that the outer spring is now in full extension.
Figure 44
49
Shaker Tables Once we have the primary and secondary tension systems in place it’s time to look at the various ‘motion systems’ and the most elementary system is what’s normally called the ‘counter weighted flywheel’ system.
Flywheel Motion Systems This is not an especially old motion system for tables since it first appears around the first quarter of the nineteenth century when gasoline and electric power became more readily available in processing plants and drive shaft speeds began to increase as a result. In application motion is transmitted to the movable portion of the table by the transference of kinetic energy generated by the used of a ‘weighted’ flywheel as seen in this old patent drawing of an original plywood Gemini table from 1922.
Figure 45
I’ve highlighted the weight on the belt pulley (flywheel) in red. If you’re printer doesn’t support color the weight is object number 85 in the illustration. This is probably the most fundamental motion system, even more primitive in many ways than a regular cam and it does work but you have to spend some time in fine-tuning the system for your particular table. I have repaired several old antique tables that used this system but I’ve never had a chance to run any of them with wet slurry so I don’t really know what the motion looks like except during dry runs. 50
Shaker Tables
Building one of these systems is incredibly easy. You start out with a good-sized final drive pulley, the larger the better and drill a hole through one of the spokes. Then you make some ‘slugs’ from pretty much any material you have on hand using a standard bimetal hole saw.
Figure 46
You can see some of the ‘slugs’ I have on hand drilled from material as thin as 1/4-inch to as thick as 1-inch. I typically like to make slugs 1 to 1.25-inches in diameter, as they’re easier to fit on the smaller pulleys that are more common on modern machines. You can ‘stack’ together slugs with different thickness to arrive at very precise amounts of weight. The slugs are placed on each side of the pulley and secured with a thru-bolt, washers and a nut as seen in the mockup shown in Figure 47. You can balance the slugs with some judicious sanding and grinding so the wheel doesn’t have a tendency to wobble but they are usually within 2 or 3 grains of being identical in weight right after they are drilled out. You can also change the ‘action’ of the wheel by having various holes in the spokes so the weights can be located further out or closer in to the axle hub.
51
Shaker Tables
Figure 47
The larger slugs shown here weigh 3 ounces each so on this wheel we have a total of 6ounces of weight attached 5.5-inches out from the center of the pulley. The exact amount of weight needed will vary depending on the weight of the movable portions of the table and the size of the pulley used as the flywheel. Even a few ounces on a big 10 or 12-inch wheel has a tremendous effect. If you’ve ever driven a car or cycle with a unbalanced tire you can appreciate how much energy is generated through this system of motion transfer. One of the problems with this system is that the final drive shaft, or the motor itself, depending on many speed reduction pulleys you use, needs to be directly below the flywheel so the changes in belt length caused by the deck movement aren’t to extreme requiring a secondary belt tension device. Tables using this type of motion system also need a slightly different carriage design since we need to fabricate a structure that actually houses the flywheel since it moves coupled to the deck frame as seen in the following diagrammatic sketch. On other types of systems the final drive pulley is attached to the stationary frame so they can be slightly more compact.
52
Shaker Tables
Figure 48
Figure 49 gives you some idea of what this type of arrangement might look like.
Figure 49
53
Shaker Tables As you can see all we did was to take our small prototype link table and turn it upside down so that the carriage became the motion frame and the original motion frame now becomes part of the table stand. The motor of course would be placed below the pulley inside the stand once we added legs and bracing to it. I think you can get the thrust of the concept from these two images.
Cams and Eccentric Motion Systems The next system we’ll talk about is the very old cam method of transmitting movement to the table deck. I won’t spend a lot of time on this system since there are much better methods we can use but in some instances a cam drive might be easier for some people to fabricate. Figure 50 illustrates a typical old-fashioned 3-lobed cam drive in the position where the table deck is furthermost to the front (towards the drive end) after the bumping timber has slipped off the end of the cam and is forced against the bumping stop by the tension spring, shown in full extension.
Figure 50
Figure 51 depicts the same table after the cam has rotated further and now the bumping timber is pushed all the way to the end of its travel range by the lobe on the cam. The tension spring is now seen to be in full compression and there is a gap between the bumping stop on the frame and stop block on the timber.
54
Shaker Tables
Figure 51
The motion cycle seen in the two diagrams is the result of a typical cam driven mechanical system. There are countless variations as to the arrangement of the various individual parts but they all work in an identical manner. Back in the old days when tables were driven by water wheels most cams were four or five lobed since that was the only way to get the number of ‘bumps’ up to a respectable two to three hundred per minute. As time progressed and steam and later gasoline and electrical power came into play the number of lobes was reduced to 3 then 2 and finally 1 as the better power sources could run at much higher rpm’s. Today almost all cams are of the single lobe variety except on tables driven by manual cranks. Plate cams, that is cams made from pieces of flat steel (or wood) stock can be designed so as to provide movement in three different ways. The most rudimentary is called the ‘constant velocity’ cam where the cam follower (the bumping timber) moves a steady and constant amount in direct proportion to the movement of the cam. This type of cam is basically half of a circle. The second most common movement is called ‘accelerated motion’ where the cam follower is moved at an ever-increasing acceleration. This is the type of cam lobe seen in almost all modern high performance automobiles. This profile when adapted for use on a Shaker Table might look like the arrangement seen in the diagrammatic sketch shown in Figure 52. In this diagram the cam is on its ‘short’ side and the table is pushed back to the lobe by the force of the tension spring.
55
Shaker Tables
Figure 52
In Figure 53 the cam has rotated a full 180-degrees and now the cam is pushing the table against the tension spring with its ‘long’ side or ‘nose’.
Figure 53
Note that unlike the old antique tables we’ve eliminated the bumping timber and replaced it with a modern roller wheel that serves as the ‘cam follower’. Almost all modern cam driven tables will use a roller cam follower since it reduces friction and noise but also doesn’t require lubrication. It’s also much easier on the cam itself and reduces wear to a significant extent. The third type of cam movement is called ‘harmonic or sinusoidal motion’. A cam having this type of profile is often referred to as an eccentric. Figure 54 illustrates some of the more common cam profiles. Some may look very similar but all have mathematically different characteristics that will generate slightly different table motion.
56
Shaker Tables
Figure 54
There are several good books on Cam design that I urge the reader to study if they want to build a cam driven table but one of the best, even though it is old, is ‘Cams, Elementary and Advanced’ by Furman, published by Wiley in 1921. You can often find it in used bookstores. The cam designs we’ll provide patterns for here are of the most elementary types.
Eccentrics Eccentric type drives can take many forms and at least one of these types is a close cousin to the conventional plate cam but easier to build and implement. This system was first adopted for Shaker Tables around 1896 but became very popular around 1922 in the second and third generation Gemini tables. In its earliest form it consisted of nothing more than a circular wood lug with the hole for the axle shaft drilled off-center. This lug rotated eccentrically on the driven shaft between two rubbing blocks attached to the tables motion frame and as a result caused the table to oscillate back and forth between the blocks as it was restrained by some wood leaf springs. Later tables used coil springs and a steel lug with steel rubbing blocks. It takes a little while to get your head wrapped around the theory of operation but once you grasp the concept its one of those things that is so elegantly simple that you’ll wonder why you didn’t think of it yourself. Figure 55 is a simplified schematic diagram showing in separate pictures how the rotation of the eccentric wheel creates motion to the table deck. 57
Shaker Tables
Figure 55
58
Shaker Tables Beginning with sketch ‘A’ the eccentric is at the top of it’s rotation cycle and not touching either rubbing block. The table deck is at it’s center or ‘neutral’ position. As we rotate the eccentric counterclockwise in sketch ‘B’ it begins to come into contact with the rubbing block on the left side and as the rotation continues the table deck is pushed to the left against resistance provided by a tension and compression spring. As the eccentric continues to rotate the table can now come back to its neutral position as seen in sketch ‘C’. As the eccentric begins to rotate back upwards to the right it begins to come into contact with the right-hand rubbing block and the table deck is forced to the right as seen in sketch ‘D”. Continued rotation brings us back again to the position shown in sketch ‘A’. The system described here is actually very ancient with respect to machines in general and was one of the first means used to convert rotary into linear motion back in the 17th century but it is perhaps one of the very best systems available for small economical Shaker Tables. As mentioned earlier this system was incorporated in the second generation of Gemini tables and for good reason. It is one of the few drive systems that permits a table operator to change between a ‘bumping’ action to a ‘jerking’ action or even to an oscillation action by making a few simple adjustments. We’ll provide links to some high quality engineering drawings for a modern adaptation of this drive system at the end of the article.
Interim Notes I’ve sent draft copies of this article out to several people who have experience with Shaker tables and they have provided some feedback that I need to incorporate before we go much further. One point that was brought up was to make a larger distinction between the preliminary prototypes as being rough ‘mockups’ of what would later become actual ‘prototypes’. To do that I should remind the reader that so far we’ve only addressed the issue of building rough prototype ‘mockups’ of ‘conceptual’ table ideas. That’s the point of having this article titled ‘Fundamentals’. We’ll get to far more detailed and specific fabrication information in the second part of this series. A second critique was that I did not fully explain the difference between using wood as opposed to using steel in both the mockup and final phases of the fabrication process. For tables up to around 3x6-feet there is nothing wrong with using wood as the primary structural members and in fact almost all of the old tables, even very large ones, were 59
Shaker Tables built almost entirely from wood components. For long-term use however I do recommend that you build the structural framework and secondary components from steel or aluminum. Wood can be used on the decks of almost any sized table but the frames take a horrible beating in a constantly wet environment over time. For prototypes however, where the tables will only see occasional use wood makes good sense from a standpoint of economics and for the experimental mockups wood can even be used as a substitute for almost every component.
Figure 56
Figure 56 shows some wood substitute parts. The 12-inch long Maple links can be bushed with bronze bushings just like a steel link. If you know the dimensions of the pillow block bearings you’ll be buying you can make temporary blocks from wood as seen in the snapshot and in fact you can use wood dowels instead of steel rod for the shafts during the mockup phase of the work. You can even make plywood discs as substitutes for the aluminum pulleys. Sometimes however trying to get by with makeshift part substitutes can cost more than the real parts to begin with so you have to watch your building budget and make tradeoffs where they are cost-effective. Every table builder will come to their particular project with a different set of skills, different budgets, different tool and equipment resources and different ideas about what they want from a Shaker Table. Some builders will want a product for occasional weekend use while others will need something that can handle relatively large quantities of raw material. This type of article cannot address every conceivable range of possible 60
Shaker Tables table requirements or designs but I hope that we’ve hit on the high-points so a prospective builder can fill in on the fine points that meet their unique requirements.
Bits and Pieces So far we’ve only talked about the larger components of tables and the theory behind some of the mechanical parts but before we move on to Part-II we need to look at ways to actually build and/or assemble many of the bits and pieces we’ve covered in a little more detail.
Making Eccentrics One of those small components is the eccentric that provides table motion and as we mentioned earlier these can be made from wood but they are very easy to make in steel and last far longer. To do the job right however you do need access to a good drill press, as this isn’t something that can be successfully accomplished using a hand drill. To start you need to decide on the diameter of the eccentric which can range from as small as 1-inch to as large as 3-inches depending on how you decide to set up your drive system. You can usually buy a small section of scrap 3/4 or 5/8-inch steel plate from most fabrication or machine shops. In a pinch you can order the material from one of the several on-line steel supply outfits.
Figure 57
61
Shaker Tables Once you have a piece of steel you can buy a bi-metal hole-saw, in the diameter you need at almost any hardware store. I typically drill a 1/4-inch pilot hole for the hole saw pilot bit to follow so that the saw pilot doesn’t actually have to do any drilling and just follows the pre-bored hole.
Figure 58
Mount the hole saw in the drill press and start drilling. I used both stick and liquid cutting lubricant but the work goes pretty easily in steel stock less than 1-inch thick but it is slow going and it’ll take around 15-minutes to saw completely through a 3/4-inch slab of steel. The larger the diameter of the piece the more difficult it is to drill and I’ve found that 2.25-inch diameter is about the largest that can be drilled with the average home-shop type drill press. Smaller slugs are actually amazingly easy to do even in stock up to oneinch thick. The secret is in getting a good quality hole-saw and I personally think the ‘Milwaukee’ brand is the best and longest lasting. You can usually get two or three pieces cut before the blade gets to dull. Good cutting lubricant can extend blade life significantly. Once I get an initial ‘score’ with the hole saw I typically change bits and drill out the hole for the drive shaft since it’s easier to hold the stock while it’s still a large slab of steel that can be easily clamped down. Once the shaft bore is complete you can finish up cutting the disc as seen in Figure 59.
62
Shaker Tables
Figure 59
This particular eccentric was made from 5/8-inch hot rolled steel and is 2.25-inches in diameter. The shaft bore is 5/8-inch.
Figure 60
63
Shaker Tables Figure 60 shows how the eccentric fits on the shaft. You can weld the shaft and disc together of weld a shaft collar to the disc so the unit can be removed. Another way to make an eccentric is to simply weld a small section of tubing to the drive shaft.
Figure 61
Figure 61 shows what this arrangement would look like. This particular tubing is a little on the thin side being 1/8th wall and in reality you’ll need something with a wall thickness of around 1/4-inch to hold up without eventually becoming out of round. You can get a little fancier and make some ‘caps’ for the ends of the tubing with holes drilled for the shaft to slide through. Weld the caps to the tube and then weld a shaft collar on each end and you’ll have a removable eccentric.
Wheeled Eccentrics Another way of creating the effect of an eccentric is to make some small tabs that enclose a regular old rubber caster wheel. This is similar to what you’ll find on a Gemini table and it works well and is relatively easy to build. The example shown in Figure 62 shows a 2-inch wheel mounted but I built the tabs to accept up to a 2.5-inch wheel.
64
Shaker Tables
Figure 62
To finish this assembly I need to weld a short piece of tubing between the tabs and then weld the shaft collars in place so the unit is removable. For a permanent arrange the tabs are just welded to the drive shaft.
Cutting Cams Making cams is not much more difficult than making eccentrics but you do need a good reciprocal saw or ideally a metal cutting band saw. All I have is an old Saws-All and it works just fine on steel up to 1-inch thick. The trick is to use a relative course toothed bi-metal blade with only 12 to 14 teeth per inch. You make short straight cuts approximating the shape of the cam. Trying to make the blade follow the curves will just bind the saw. In practice you make a paper template of the cam profile that you decide to use and trace its outline on a piece of steel. Center punch the location of the shaft hole and drill this out first. The piece we’re using in this photograph is 3/4-inch thick hot rolled and the cam pattern is called a ‘Heart’ profile which is by far the most commonly used for Shaker Tables. This particular cam will allow for a maximum of 1-inch throw though it’s doubtful that you’ll ever need that amount.
65
Shaker Tables
Figure 63
If you look closely you can see the black felt-tipped pen outline on the steel plate. The eccentric we made earlier provides a good reference of size.
Figure 64
In figure 64 you can see how I’m making straight cuts with the saw progressively working my way around the pattern. 66
Shaker Tables
Figure 65
Figure 65 shows the roughed out cam. It looks pretty horrible now but once we smooth it out with a belt sander it’ll be almost as good as something made in a machine shop.
Figure 66
Figure 66 shows the final cam after smoothing alongside the eccentric, which hasn’t been smoothed and polished yet. 67
Shaker Tables It’s these small parts of the drive system that take the most time to build and unfortunately they are critical elements that you can’t just go out and buy at the hardware store. In fact I couldn’t find anyplace online that sold anything I could adapt or modify which is why I had to build these in the first place.
Part-I will be continued………………. Part-II will cover building a full sized working table………….
If you want to contact me my email address is
[email protected].
Happy Prospecting, Gary
68
