The Dale C. Maley Family Web Site

I initially thought that wood links would not be precise enough so I went with a brass marble lifting mechanism.

Although the brass is accurate enough, the lead solder joints keep breaking.   I may have to attempt to silver solder that model.

I decided to investigate building the model with wood links.

I will use maple for the round end links and 1/4" oak dowels.

I was going to make some end links 1" diameter and some others 3/4".............until I found out the smallest hole saw I have is 1"................so I will try to do it with all 1" diameter links that are 3/4" thick.   

I have to build 8 links.

Below is the start of my wood link design......

1. Hole saw using 1" OD bit from 3/4" hard maple (center drill makes 3/16" dia hole in center)

2. Kiss both sides on horizontal belt sander to remove black marks

3. Drill center hole larger to 17/64", a hair bigger than the 1/4" dowel that will be used

4. Put on 1/4" bolt on lathe, tighten with nut, make chamfer on 2 ends, sand

1. I made dowel to connect 1 end piece and center piece, then put in fixture

2. On other end piece, drill 1/4" hole clear through 1" round link

3. Place end on fixture, get long 1/4" bit, use end link for guide, start drilling hole in center 1" OD piece at the right angle

4. Take end piece, put in vise, finish drilling 2nd  hole in center piece

5. Glue 3 rotating pieces and 2 mating dowels together on fixture

This process worked great and gave me a perfect fit between the 3 rotating points.

I found a piece of scrap plywood and will use it as my test base............because I will probably have to adjust the design as I build it.

On the base test assembly, I turned the crank and got interference with the rear vertical link, and the lower base. I sort of knew there would be interference, but I decided to measure and adjust the model, then adjust Sketchup accordingly.

I was really suprised how well the right hand side of the model worked in the prototype stage !!!!!!!!

1. It was smooth motion

2. very repeatable in terms of where the lifting link was in its motion

3. took little force to activate it

You can use this link to watch a short video of the prototype working great.

I expected some interference between arms, and the worst was the curved arm just to the left of center.  I am going to have to redesign it to a different style.

I chose to scroll saw the curved link from 1/4" thick red oak, then glue 1/4" spacers on each side to get the necessary 3/4" thickness.  I used this method on another project and it worked well.

Once I changed to a curved link, both sides of the mechanism worked well when I hand cranked it !!!!!!!!!

You can use this link to watch a short video of testing the mechanism at this point.

I noticed the right hand lifter is not on center with the LH side lifter...........because I did not put the shaft bearings 1/4" away from the lower base.  That piece of plywood was so dark, I could not see my carbon paper lines very well.

I am going to go ahead and make the final base piece, and transfer the mechanism to the final base.  I want to make some changes later, including a loop for the marble and a round sink, but I am going to assume I can fit them on the original base board..........worst case I add onto the base board, or make a bigger one.

the crank location with respect to the link base is critical, so I need to switch to the final base now.

I made a pine final base.  I also made the holding piece for the marble prior to the mechanism picking it up..........I copied the design from the brass model.........and located it on the base board in the same spot as in the actual brass model.

Next challenge was converting brass design to wood for:

1. Pick-up on left hand side of the model that picks up the marble and transfers it to right hand mechanism

2. Left hand pickup that accepts transfer from right hand mechanism, then final groove removes it

3. Final groove that dumps marble into wood track

My concept is take a piece of wood 3/4" wide with a rounded front so it guides into the LH lifter, then put a groove in the piece so the marble runs downhill to the regular track.  I made a test piece shown below, but it is hard to see the groove I cut in with the Dremel 1/2" drum sander.

My drive axle is not yet pinned to the arm that drives the mechanism, so sometimes I have to push the mechanism when it slips............but this short video shows the pick-up mechanisms working ok.

You can view the short video using this link.

Build the little tower that holds the regular chute and attach to the base, so I can fine tune the drop-off design.  I will make the tower solid since I am not dropping the marble down the inside like on the brass model.

It took a lot of trial and error, maybe 5 attempts, to get the RH side pick-up to work with the grooved unloader chute!!

Once I got that done, I added the loop and the round bowl.

Now I  have to do some fine-tuning to make sure the marble always makes its way around the loop.  I am probably marginal in terms of design of having enough fall to make it around the loop.  Early testing I did said I needed a minimum of 4.5 times the loop radius for the minimum and I am at 4.0...........3.5" loop diameter is 1.75 loop radius and about 7 inches of fall gives 4.0 ratio.  Some options.........

1. raise drop hole from vertical chute about 1 inch.

2. move bowl to left and decrease angle of fall into the loop

3. sand and remove any friction spots in the loop (there is one at top)

4. make a new loop with something smaller than 3.5 inch diameter

You can use this link to watch the model in action at this point.

Since the 4.0 inch diameter loop would not work 100% of the time, I decided to make a 3.0 diameter loop and try that.  Sketchup design is not perfect, but good enough to make the parts.

A compromise between 4 and 3 inch would be a 3.5 inch diameter loop.  I designed one in sketchup but did not build one.

I decided to smooth the inner track as I built it, because it is tough to smooth if after the guard or guide outer pieces are in place.

I did not pay attention to the approximate entrance angle I would have from the vertical tower down to the loop.  I kind of guessed and made that angle on the loop.  When I final assembled everything, I found I made the entrance angle too steep on the loop, so I removed some material using the Dremel.  With a small diameter, like 3.0 inches, it is more important to try to match the entrance angle on the loop with the angled track !!

I put a piece of blue masking tape on the entrance joint to the loop, and bingo, she started working.  I will try to fill that area, then sand it nice.

I timed it and it is taking about 10 seconds for the marble to drop from the drop-off point to exit the round table.   Sometimes I pick up an empty load using 2 marbles because of the long return time.

When I made the round table on the lathe, I was unsure how much of a fall angle to put on the inside.   I guess I should take the piece back to the lathe and cut a much deeper fall angle.  

The right hand pick-up is loose on its 1/4 inch dowel and slips down sometimes, so I have to move it back up once in a while.

You can use this link to watch a video of the model.

The model is basically working now..........so now begins the process of upgrading/fixing each issue in the model so it becomes very reliable.

Put bowl back in lathe and increased drop angle.  Reduced time from 10 sec to 8 sec, not enough.  I laid a toothpick on the bowl incline, and bingo, time dropped down to about 4 seconds, so will glue it in.

made a fence around the top of the bowl to deflect the high speed marbles that sail off the edge, plus when a marble hits a marble already in the bowl.

I added another quarter round spacer on top to prevent this.

The model is working pretty well.  On the LEFT hand pick-up, I think I need to rotate the pickup arm about 10 degrees CW viewed from the front, so it wants to pickup versus push the marble away.  You can use this link to watch the video at this point in the project.

On my marble loops, the worst spots are where the pieces on the track come together.

I have been struggling with figuring out another way to built a more friction free track and loop.

I thought about making a tube to hold the segments, which I chuck on the OD on the lathe chuck........then I place the pieces inside.......then turn the ID of the pieces to get a smooth track.  I don't know how to secure the pieces while turning, maybe 2-sided tape, but how do I remove the pieces?

Another idea:   As I build the loop and get the inner track done, and before I add the guides........glue down a piece of 3/4" wide wood veneer all the way around the loop.  The veneer hides any rough spots at the junction of the track pieces !!!!!!!!!

Process for making spur gear:

1. I want a 20 tooth gear with a 6 inch pitch diameter and a 20 degree pressure angle mounted on a 3/8" shaft

2. I inputted these parameters into GearDXF program and it drew the gear

3. I exported the gear dwg as a DXF file

4. Imported the DXF file into Sketchup

5. Scaled the drawing using the 3/8" bore diameter

6. Print out full scale paper patter and glued it onto the gear blank

7. Bore the 3/8" hole and scroll saw cut the gear

8. Mount gear on 3/8" dowel on a scrap piece of plywood. Clamp plywood onto drill press table. Rotate the gear by hand and knock off any high spots using drum sander on drill press.

Note: You can print out a pattern directly from GearDXF program if you want to.

It is critical the OD of the gear blank has no run-out to the shaft bore, or it will not work if the run-out is excessive.  I decided to make a 3x3 inch blank about 5 inches long by gluing up 4 pieces of 3/4" inch thick  red  oak.  I took the 2 center pieces and routed a 3/8" wide groove 3/16" deep in each piece.  I then aligned the grooves when I glued up the stack.  After gluing and clamping, I will drill a 3/8" hole through the square hole, which will give acceptable run-out.

The tooth spacing on the spur gear, which has to match the worm gear, is pitch diameter circumerference divided by the number of teeth.........or Pi * D/Teeth = 3.1415*6/20 = .94245 inches.

I made a template in sketchup to use to mark out the teeth on the worm gear.

It took about 2 hours to hog out the worm gear using the Dremel.  My process was:

1. Using dovetail hand saw, sawed edge of each tooth

2. Removed material using Dremel and round ball carver attachment

3. Check depth using depth gage

4. Sand on drill press using 1/2" diameter dowel with sandpaper inserted into sawcut in dowel

It could be prettier looking, but I am quitting for now. I may come back later and do some more finish carving/sanding.

Once I got everything working, about the only tweak I had to make was shortening the LH side of the LH pick-up, I shaved about 1/4" off the left hand two fingers and it worked ok.

I had to move the location of the drop-off channel slightly so it was centered with the RH pick-up.  Fortunately, I made it adjustable with 2 screws on the bottom of the chute!!

I uploaded my finished design to the Sketchup warehouse. You can use this link to download the Sketchup file of my model.

You can download a Version 7 copy of my file, which is what I drew it in, using this link.

You can use this link to watch a video of this project in action.

I initially made this model using brass linkages because I thought wood would not be accurate enough for the precise motions needed to transfer the marbles.  But, the solder joints are very weak on the brass model and easily break after limited use.

I decided to try it in wood, and Voila.............it works great !!   I anticipate this wood linkage model will hold up better to grandkids and other children playing with this model :)

I have now taken this to about 3 public events, where both children and adults played with it.  On the 3rd event, it stopped working.

Investigation found 2 issues.  The dowel for the upper pick-up arm had come unglued from its round base.

Same issue with a dowel on the link that connects to the lower lift arm.

I noticed with a 1" diameter round member joining the dowel, there is not enough length of the dowel inserted, to prevent movement and eventually ungluing.

I'm going to try to make the dowel extension length longer and see what happens.

The new links and dowels are all red oak...........versus pine links and birch dowels.

This increased dowel engagement should keep the links from coming loose, I hope :)

I added dowel engagement length on the pickup piece, by adding a tapered piece of oak turned on the lathe.

I also changed the round joints to beefier pieces with more dowel engagement.

I have learned that if you change one thing on this model, it usually impacts another piece and you have to change it also.

Once I got the 2 new links working ok, then the LH pickup arm would not extend far enough into the eared holder...........so I slide the green eared holder about 1/4" to the left, which solved the pickup problem, but then the 2nd ball would get hung up on the RH side of the eared holder !!!!!!!

So, I made a new eared holder, extended the LH side by about a 1/4" and made the top taper steeper.

You can use this link to watch this model in action.