Dave Smith's ATM Grinding Machine Page


"Grind more, Worry less!"

 :o)

I owe most of the insight into the building of this monster to my fellow Aggie and ATM list member, Dave Lilly.  He was kind enough to send me images and descriptions of the working components of his machine.  He has also agreed to allow me to post images of his machine and I've tossed in some e-mail conversations we shared during my construction efforts. They may prove useful to someone who undertakes such an adventure from scratch like I did!

My grinding table is essentially built from scrap and salvaged parts. The basic table itself is built from two trashed 2" thick laminated particle board doors sawn in half (from the dumpster of a door manufacturing company just down the street ;^) ).

(For any of the images here in the description, just click on the image to see the full size image.)

I did have to purchase the 5/8" steel shafts, snap rings, flanged bearings, flanges (or hubs, as Chicago Die Cast Mfg. Co. calls them), 2" pulleys and several l-o-n-g belts (74 and 80 inches). Chicago Die Cast has no web presence of which I'm aware.  However, you can contact them the old fashioned way; I found their service to be prompt and courteous (no other affiliation with said company).  Contact info: Chicago Die Casting Mfg. Co.,

 OK, let’s walk our way through how this grinding/polishing table was built and is supposed to work: It took me a couple of weeks to figure this out from David Lilly's description and pictures (I'd never seen one in real life...and I'm slow!  Thanks Dave!).  A video of the beast in action would make the following MUCH shorter, but for now it’s the best I can do!  

I made two 18" wooden pulleys to gear down a salvaged (i.e. "free") 1750 rpm AC motor. I used a jigsaw to cut out four 18" disks with a 45-degree bevel on the outer edge. Placed two together with bevels facing in to make a "V" to hold the belt. (In hindsight I should have used my router to cut and notch the pulleys.  At the time I didn't know circles were so easy to make with a router and a simple homemade jig!  Doh!)  I used wood screws to join the bevel-edged circles together, but epoxy or wood glue would probably work fine. Then I drilled a 5/8" hole in the dead center and mounted a shaft flange or hub (5/8" bore) concentric with the hole and made a pulley (see the second image). I used another homemade 8" wooden pulley to drive the overarm. You can see the flange and pulley configuration in the above pic. I bought the flanges from Chicago Die Cast Mfg. Co. and they call them hubs. (They were tough to find!)

The original AC motor had a cheesy adjustable platform so I rebuilt it using some 1/8" thick aluminum salvaged from a machine we killed at work. I now use the original tension adjustment knob on the motor to tighten the first belt. The first belt runs from a 3-step pulley on the AC motor to an 18" wooden pulley on shaft #1 (see the schematic above). A 2" pulley on shaft #1 is connected to a second 18" wooden pulley on shaft #2 by belt #2. Now, the second shaft runs up through a 5/8" id 1 3/8" od flanged bearing in the table to the 24" (4" thick) turntable on top.  The turntable is attached to the shaft via a flange or "hub" just like the homemade pulleys. The turntable rides on six pair of 1.5" plastic caster wheels (salvaged drive rollers from a copy machine) evenly spaced to support the turntable, glass, tool and steel weights (up to 75 lbs). On shaft #2 there is another 2" pulley which is connected to an 8" wooden pulley on shaft #3 by belt #3. Dave Lilly's original design had movable shafts and Dave told me that tightening the belts was a constant battle.

So I designed and built idler pulleys for belts #2 and #3 and mounted shafts #1 and #2 permanently in place. The idlers were riginally designed in aluminum, but implemented in plywood.

 

                    design                           implementation

Each shaft is mounted through two bearings, one on top of the table and one below. In order to support the shafts I screwed two 2x4's together and drilled 5/8" holes in the desired location right down the seam in the joined 2x4's.  Then I recessed the holes on top for a bearing, mounted the bearings and then stabbed and screwed the pair of 2x4's into place as the last step in assembling the table (i.e. AFTER the shafts and pulleys were attached and in place). The shafts are held in place by hack sawing a groove near the end of the shaft and attaching a snap ring. One important thing to note that will not be visible in any of the pics is that every flange has a setscrew that holds it in place on the shaft. I drilled 1/4" deep holes into the shafts allowing the setscrews to penetrate the shaft and prevent flanges and shafts from slipping.  

With the motor at 1725 rpm and belt #1 in the 3" slot on the motor (i.e. in the middle of the 3 step pulley), shaft #1 spins just under 300 rpm and shaft #2 (i.e. the turntable) spins just over 30 rpm. I can change the speed by shifting the motor up or down so that either the 2" or 4" slots of the 3-step pulley on the motor are in line with the first 18" pulley. The 2" slot will give me 22 rpm at the turntable. With the 4" pulley, I get 43 rpm at the turntable. Currently this is the only way I can control the speed of the turntable and overarm. So far the high speed seems just fine.

Since the turntable shaft drives the 8” pulley on shaft #3, I have a constant ratio of turntable rotation to stroke of about 4:1. The shaft through the 8" pulley also passes through a bearing in the table and mounts rigidly to a stack of 4" x 4" blocks.  This block was added to elevate the point of attachment for the overarm connection.  I mounted a 4" x 12" piece of plywood to the top of this block (You can see it coming right at you in the first of the three pics.) It has a slot cut into the middle.  It connects the drive directly to the overarm (which is off and upside down in the first pic) via shaft #4, the 8" long 1/2" dia. threaded rod. The block and 4x12 unit spins, effectively making it a "disk". Don't need to waste plywood on a disk here, though, so I just used a small scrap piece of ply. I routed a groove or slot, centered lengthwise along the rectangle.  The all thread is bolted to the 4x12 (via nuts and washers on each side of the 4x12) through the groove. This allows continuous adjustment of a symmetrical start/stop position for the overarm swing. Using the traditional 1/3 diameter, center-over-center (COC) stroke, I can grind any size mirror up to just under 24" (or bigger with some modifications. :-) ).

 

An aside: Originally, I built the overarm to not only be continuously adjustable with respect to the distance of the "throw", but also fully adjustable in positioning the start and stop of the stroke.  (Unfortunately, in all of these images the apparatus that controls this motion has been removed.)  In other words, I could start and stop the stroke anywhere I want and vary the length of the stroke independently. David Lilly showed me how to do this...by not mounting the overarm directly to the rectangle, but rather mounting it to another continuously adjustable rod. This rod needs a pivot point out near the end. I mounted a scrap piece of plywood onto a salvaged bicycle front wheel axle/bearing using wood screws. Cut a dado in the top of the ply and "presto" a swiveling pivot point. It works real well--cheap fix. Cheap is good.

 

The overarm has a fork in it and the shaft has to be able to move freely in the fork. Here I have the overarm off and upside down next to the shaft with the pair of bearings exposed. I position the bearings on the 1/2" all-thread with pairs of nuts tightened against each other at the desired height on the rod. I cut a 1" groove into a 4x4 and lined the inside and bottom edges of the groove with some thin scrap galvanized steel angle. The pair of bearings mounted on the 1/2" all-thread takes the abuse of pushing the loaded overarm back and forth. The overarm also rides on a big flat washer mounted below the bearings on the 1/2" all-thread shaft. I started out with a baby food jar lid, but wore completely through it! (It has been the ONLY part on the machine to fail!)

At the other end of the overarm I drilled a 1/2" hole for the pivot point. I mounted a 12" or so length of 3/4" threaded plumbing pipe via a threaded plumber’s flange to the table and it serves as the pivot point for the overarm. Originally I just used wood screws to mount the flange permanently to the table. Later I learned that getting mirror and tool COC is a piece of cake if I clamp the pivot point in place--that makes it "continually adjustable". I use 2 C-clamps to hold it in place. In 'tbltop.jpg' you can just make out one of the C-clamps at the far end of the overarm coming up from behind the turntable. You can also see the 3/4" pipe coming up through the overarm at the far end. I'm using a pair of vice grips to hold the vertical position of the overarm on the 3/4" pipe. No bearings here... not much need for one I don't think. There's only 15 to 30 degrees of motion at the pivot point and the force and weight are mostly at the tool/mirror and at the bearing contact on the overarm slot.

A word or two about mounting bearings and shafts: I bought 5/8" steel shafts and 1 1/4" od, 5/8" id flanged bearings from a local hardware store (B&B in Culver City, CA). I used a 1 1/4" wood-boring bit to recess the bearings down to the flange. Initially I had problems with the bearings popping out of the recesses, so I used a couple of large button-head wood screws to fasten the bearins permanently in place. They haven't come loose since. A cross-section of the table would reveal a table top of 2 sheets of 3/4" ply, supported around the perimeter by 2" thick door halves (more or less).  Two pair of 2x4s run the length of the table. A pair of boards are placed vertically and screwed together.  One pair is butted directly to the underside of the table top; the other is mounted below all of the pulleys. The shafts run right down through the seams between the two boards in both sets so that the shafts do NOT flex or come loose.  The 1 1/4" recesses all face 'up'.  Shaft #2 doesn't fall through the table because I hack sawed a groove around the circumference of the shaft about 1/4" from the top of the shaft and inserted a snap ring into the groove.  I used this same technique to hold the shafts in place for the idler pulleys, as well as to elevate the ply rectangle that drives the overarm.  The idler pulleys required a snap ring on both outside faces of the wooden shells.

 

While polishing the 13", 1/2" thick plate glass mirror, I noticed the motor started getting pretty warm so I mounted a fan to constantly blow on the motor while the machine is on.

 

Now a comment on the design of tool/mirror attachment to the overarm:  (images coming soon) Regardless of whether you grind/polish MOT or TOT (mirror on top or tool on top), you'll need a way to hold the disks (mirror or tool) loosely in place. Let's assume TOT for this description. The mirror is placed in the center of the turntable and four 1.5" squares of 3/8" ply are screwed into the turntable spaced evenly around the mirror. A small gap (1/8") is left between the squares and the mirror to allow the mirror to "slip" during grinding/polishing to avoid astigmatism. To hold and drive the tool an upper disk of multiple plywood circles are stacked and screwed together and a 5/8" hole is drilled in the top about 1" deep into the dead center. Another set of four 1.5" squares of ply are mounted vertically to the bottom outside edge of the stacked disks and extend only far enough down beyond the edge to contain the tool or mirror. The overarm has a vertical 5/8" hole positioned directly over the center of the turntable. I ran a section of 5/8" steel shaft through the hole in the overarm down into the "button" or hole in the top of the disk. This is a loose connection. I'm hoping the "slop", so to speak, will help avoid exact repetitive strokes and astigmatism. I stack one or two 35 lb. barbells in between the overarm and the multi-layer tool disk for weight.  The short shaft passes right through the middle of all of it down into the 'button'.