Are we talking the same one?
I was talking:
By @ImpetousWombat up above… To me if that works, it seems like it kind of naturally is “45deg” but…
It looks like it would negate issues with router clearance.
1 Like
@pillageTHENburn that sounds great!!
what happens if the sled rotates enough that one set of arms collapse together? (or if you were to only have one set of arms and had the other line fixed?
That is a vertically-truncated version of the original, but the mounting holes are still vertical. The way I drew it makes the horizontal arms sort of look 45ish but the arms that the chain will attach to are still vertical (along with the mounting points). Yes, that idea also clears room for the router. It’s basically a hybrid between the 3-bar linkage and my original pantograph idea (which used 6 bars per assembly, 12 total). The pantograph mounted above the router as well; I think mounting the 3-bar above the router should work, I just haven’t tested it in our real world physics yet 
When I’m referring to mounting it at 45˚ I am referring to the mounting holes and the center bar (that the chain attaches to). Those would be 45˚ from the centerline of the sled. I’ll draw some pictures…because apparently that’s what I do. Ha!
-Logan
Fully working machine.
All of the arms are 10 inches in length. Making it 9 inches on center for the holes. I have complete range of motion to cover the entire work surface now.
I don’t think it moves at all. The only movement I see I believe to be generated by the motor arms flexing in a bowing fashion. At some point I plan to stiffen up the arms.
I really like this design, it seems to work very smoothly. I will try to make a pen holder this week to map out a grid, but I believe any error that exists is much less than the original design.
I am happy to answer any questions anyone has. I am sure others will improve upon the design so I didn’t spend the time to draw it up in CAD.
6 Likes
It looks like you removed the handles from the router base. Were any other modifications to the router needed?
Correct. Handles removed, as a warning to others, I could not get the protective caps off the handles without damaging then. I drove a screw through then and pulled them out, they are in there tight. Without the handles it is much easier to affix the zaxis strap.
No other router modifications were needed.
1 Like
Thinking about what is needed for accuracy here and how to get accuracy when you don’t have precision tools to start with.
I’m going to start with the @ImpetuousWombat design (I think that it is going to be the better one, it avoids any interaction with the router
you need to make two matched sets of “horizontal” arms and two “vertical” arms that exactly match the holes in the sled. Since the vertical arms need to resist the bending force caused by the chain tension (up to 60 pounds) the shorter they can be, the better.
Tools Required:
-
A saw that can cut boards in half lengthwise. A table saw or bandsaw is ideal for this, but you can do this with a jigsaw, circular saw, or even hand saw if needed.
-
Some way to drill vertical holes. ideally this is a drill press, but hand drill in a jig, or just careful work with a hand-held drill can work.
for the horizontal arms, take two pieces of wood (say 1x2") drill holes through the 2" thickness on each end, then rip the boards in half, making 4 1x1 boards. you now have two pairs of arms that match exactly.
for the vertical arms, take a 1x2 board and drill one vertical hole in it. Drill a hole of the same size through the center of your sled (you need to do this before you cut the large center hole for the router).Bolt the board to the sled. Now drill two holes through the board and the sled. just make sure the first hole out from the center is far enough to clear the router base, but it probably will work better the closer the assembly is to the router.
notes:
None of the spacing of the holes is critical, because everything that matters is drilled at the same time
You can flip the boards so the cut edges don’t rub against each other in the final assembly, so it doesn’t matter if the cut edge is uneven.
Instead of drilling holes in a solid piece and then cutting it in half, you can take two pieces and stick them together (double-sided tape, glue with a layer of paper between them, or even solidly clamping them together) and then drill them.
If you can position a chunk of 2x4 edge up under the sled when you drill the holes for the vertical pieces, you can drill down into it, and then drill it out as well. You can then use that 2x4 to space the entire assembly out from the sled so that you are close to the balance point .
When bolting things together, try to get bolts that have a smooth shoulder and are long enough that that shoulder is hat contacts the moving parts.that will greatly reduce the wear compared to having the material pivoting on threads.
Use nylock nuts on all the joints so that they won’t work their way loose.
1 Like
For the rectangular arrangement, the lengths of the horizontal arms again need to match on each side, but their length doesn’t really matter and they don’t have to match, so they can be drilled/cut in pairs.
The vertical arms need to match the spacing of the mounting holes, so again you connect them to the sled and drill them all at once. Since the spacing for the mounting holes to the bit need to match, drill the center holes (chain mounts) first, bolt that to a center hole in the sled, and then drill the end holes.
1 Like
took 1 minute and drew this up in emachineshop.com
how can I share this file with the community?
it’s basically 1"x10" bar of steel with 1/4" holes spaced 1/2" from ends
Thinking about accuracy here.
If you use the box arrangement where the hole is in the center of the vertical arms, I think the critical items are:
- the mounting holes for the sled must be either
A. exactly in line with the bit (example shown here)
or
B. exactly perpendicular to a line that goes to the bit (what I think @pillageTHENburn is talking about) Think of a T here the ends of the top are the mounting holes, and the bit is at the bottom of the T, this can be rotated to any convenient angle on the sled
-
the horizontal arms must be exactly the same length (hole to hole) on each size (one side can be longer than the other.
-
the vertical arms must be exactly the same length (hole to hole) as the mounting points on the sled.
-
the distance from the top hole in the vertical arm to the chain mounting must be exactly the same as the distance between the top mounting hole in the sled and the bit (or the vertical bar of the T)
great to see the fully working setup
If I am seeing things correctly, a line between the mounting holes for each side
is centered on the router bit, correct?
could you test a version where these holes are offset? (like the brick holder
holes are, in a T arrangement with the bit as I described above)
@dlang, I agree the primary issue with this design is consistency, not accuracy. Your suggestions for ways to accomplish this without advanced tools are creative. I think it is critical to stack the partnered pieces together so that each operation is done to both of them at the same time.
My issue with this design, is that it will be difficult to get the arms positioned properly to ensure that the center of rotation is around the bit. As I talk about below, I used a simple method to achieve this in my design.
Yup, watched this happen in seconds today. The other option is to use two bolts tightened against each other.
Yes everything in mine is centered. One nifty trick to accomplish this. As you can see in the first picture here. I drilled a shallow 1/4 inch hole on the side of the chain arm. I then placed this over my 1/4 router bit while the router was mounted in the sled and used the arm to mark the holes to be drilled through the sled. This ensured that both the spacing of the holes was correct and that the bit was at the center of the invisible arm.
If I change the mounting points as you describe, it will certainly cause errors, are you just trying to get an idea of how much error it causes?
1 Like
Onshape seems to have good sharing. I think there is something in Fusion360 as well.
However, I stress that the design probably can’t be this simple. I have three contact points on all the rotational bearings. This is crucial for the wood design, but is probably still a good idea for any design, otherwise whatever we use as a axle will almost certainly bend.
1 Like
Nope, just changed to the triangular setting and inputted the chain connection distance setting. I don’t think this setting is super critical to get right, I believe any errors with this only result in shift of the sled from the expected position, but not any distortion.
Since the total rotation of each arm is only ~70 degrees, I suspect you could design an irregularly shaped 4 bar linkage that would get you a reasonably accurate rotation about the router bit inside of that range.
I have zero training as an engineer, all of my designs are based off of intuition, so I wouldn’t even begin to know how to figure out that geometry.
1 Like
I have the arms cut and drilled (did one, then used a transfer punch so they’re all the same), using 1/2" 1x11 birch with 10" hole spacing, also marked and drilled the sled with the same method. I’m using two pieces for the outside uprights with a shaped (fancy name for whittled away on the disk sander) spacer that shouldn’t hit the arms at any reasonable angle. Will be interesting to see if they have any hope of working. The glue will dry overnight, we’ll see. More plywood for phase 2 and up. Pics taken, will start a new topic if it works, otherwise we can change the subject
The floor breaking workers are supposed to show up tomorrow (their past promptness is why I’ve been hammering and hauling), so there might be a delay in implementation.
Just for clarification:
I feel like if the linkage parts don’t interfere with the router then there is no advantage of the 3-Bar Top Mount over the 3-Bar Balanced. The 3-Bar Balanced is probably easier to layout mounting points on the sled but I don’t think that’s a big deal with either one. 3-Bar Top Mount is also currently untested, but I think it’ll work the same.
I will try to build physical tests tonight or tomorrow. I have to balance work and family too!
Thanks for making this brainstorm so productive!
-Logan
3 Likes
I think this distance is pretty important, if it’s incorrect it makes the
calculated chain length incorrect, which can cause major errors (especially top
center)
In reference to the rotation axis for triangular kinematics setting.
Huh, is that right? I thought it would just cause a shift.
Ok, then I need an explanation of what we consider to be the “end of the chain.” Is this the very tip of the chain? I entered this as the distance from the bit to the surface that the chain pin sits on, but I wasn’t really sure what I should be measuring to.