Chain & Sprocket Triangular Kinematics

Continuing the discussion from Throwing my hat in the sled modification ring:

I’m, splitting this idea from @awlinc out of the massive ring sled discussion, because it’s my new favorite and I think it’s very promising. A lot of the other triangular kinematics designs require custom hardware like those offered by @bar (laser cut steel ring & v-groove bearings upcoming), @dlang (steel top ‘pantograph’), and @pillageTHENburn (laser cut birch 45 degree ‘pantograph’). All of these solutions look awesome, but the parts may be hard to come by in remote locations.

I’ll summarize what was discussed so far, as I interpreted it. Most of these ideas are someone else’s, my apologies if I didn’t give appropriate credit, please feel free to mention if these are some of your ideas. The basic design would look like this, but with the grooved ring elevated off of the base sled for proper COG offset.

The ring could be made with a maslow, or with a router, a board, and a nail, to a high degree of accuracy. To build with a router & board, the nail would be hammered through the board into plywood, and attached to the router on the opposite end, with the nail moving positions in the board (but at the same center point in the plywood) for successive cuts.

Construction would involve one 1cm deep dado indentation on the outside of a ring, followed by a full depth cut a little further out, leaving a circle of ply with an indent on the outer edge. A succesive cut would remove the center ~2" in from the circle, leaving a ring. This would be done twice, then a 3rd thinner ring with no dado indent would be stacked on top. This would create a circle 3 layers of ply thick when stacked, with 2 grooves to hold a #25 chain. The groves could be sanded to give them a slight taper to help with chain alignment.

Once the circle is built, it would be installed a few inches off of the base of the sled, near CG, with standoffs. Then, approx 1.5 meters of a chain loop would be rapped around each indent in the circle. An idler sprocket (these can come with built in bearings!) would be placed inside the loop of chain on each, then an intermediate linkage, similar to the one @bar uses on his ring, would attach the sprocket to the chains.

This approach shares a part with the standard Maslow, #25 chain, and the idler sprockets are available just about anywhere the chain is sold. The overall price could be as low as $30, and no special tools or skills would be required to build. Here are some parts I’ve found online that would work for a total of 22.78 + some shipping:

Long enough chain for both loops - $16.28 incl shipping

Idler sprocket with 8mm / 5/16" opening - $3.25 each + shipping
https://www.phidgets.com/?tier=3&catid=43&pcid=36&prodid=809

One thought that was mentioned was that the chain wouldn’t need to go all of the way around, and less chain could be used if screwed in to a section of chain that wouldn’t have to lift from the ring, but there’s a chance this could introduce some error and inconsistency.

I hope this is helpful, and I’d love to get everyone’s thoughts :slight_smile:

7 Likes

Thanks for breaking this out @ImpetuousWombat, I think it’s time for some testing.

Stopped by the local bike shop and dug around in the junk parts box. I’ll try to get some testing done this weekend with stretched chain and junk spare parts :grin: (free!!)

That is about half the price of any idler sprockets for #25 chain that I could find. I was planning on trying with a #43 roller chain and use bicycle derailleur sprockets, both of which can be found pretty cheap. #43 is .5" pitch and .125" roller width, which I think will work with bicycle sprockets.

It’s about the same price as #25, about $15 with shipping,
https://www.ebay.com/itm/TRITAN-43-1R-X-10FT-Precision-Roller-Chain-1-2-Pitch-10ft-Box/112409396964?hash=item1a2c1f46e4:g:6tMAAOSwR29ZGy64

Amazon has the rear derailleur sprockets aka “jockey wheels” 2 for $10 ,

or you can get them for $3 a pop straight from China on ebay (if you feel like waiting :slight_smile: )
https://www.ebay.com/itm/New-Mountain-Bike-Jockey-Wheel-Rear-Derailleur-Pulley-11T-13T-For-SHIMANO-SRAM/332360567857

2 Likes

All of these solutions look awesome, but the parts may be hard to come by in
remote locations.

The wood links can be made from a maslow (even an inaccurate maslow)

The ring could be made with a maslow, or with a router, a board, and a nail,
to a high degree of accuracy.

not from a maslow, until you have it accurate.

Construction would involve one 1cm deep dado indentation on the outside of a
ring, followed by a full depth cut a little further out, leaving a circle of
ply with an indent on the outer edge. A succesive cut would remove the center
~2" in from the circle, leaving a ring. This would be done twice, then a 3rd
thinner ring with no dado indent would be stacked on top. This would create a
circle 3 layers of ply thick when stacked, with 2 grooves to hold a #25 chain.
The groves could be sanded to give them a slight taper to help with chain
alignment.

Once the circle is built, it would be installed a few inches off of the base
of the sled, near CG

This seems to be ~3" or so from the surface of the sled, so mount the rings to a
hunk of 2x3 on edge and you should be very close.

This approach shares a part with the standard Maslow, #25 chain, and the idler
sprockets are available just about anywhere the chain is sold. The overall
price could be as low as $30, and no special tools or skills would be required
to build. Here are some parts I’ve found online that would work for a total
of 22.78 + some shipping:

Long enough chain for both loops - $16.28 incl shipping
https://www.ebay.com/itm/25-Roller-Chain-10-Feet-with-2-Connecting-Links-/321765763022

Idler sprocket with 8mm / 5/16" opening - $3.25 each + shipping
https://www.phidgets.com/?tier=3&catid=43&pcid=36&prodid=809

how do you attach the chain to the idler sprockets (you need to make sure that
they are flat enough to not catch on the chain links)

David Lang

1 Like

Definitely, and I think that the wood links are a great solution. The sex bolts or other bushings is where I’m thinking of the specialized hardware. Bare threads aren’t an option, and bolts with smooth surfaces along part of their length might catch a little and reduce accuracy. Easy stuff to find solutions for, but maybe not as ubiquitous as bike parts (great idea @awlinc, btw).

Good point.

Chain alignment will be important, so the motors will have to be positioned in line with the ring, but as long as the linkage attaching the sprocket to the chain is rigid I think it’ll work fine. a 2x3 with a notch cut out of the center, or anything similar to @bar’s bent metal bearing holder or @pillageTHENburn’s wood/chain linkage would work. Here’s a crappy drawing:

image

bare threads against wood are not great in that they will wear fairly quickly, but they will work

bushing material is easy to pick up at a local hardware store, a length of copper pipe cut in small chunks will work

We should put together the g-code to cut a wood linkage kit (using the strategy of cutting in the same place and moving the workpiece to get an accurate kit from an inaccurate machine)

Chain alignment will be important, so the motors will have to be positioned in line with the ring, but as long as the linkage attaching the sprocket to the chain is rigid I think it’ll work fine. a 2x3 with a notch cut out of the center, or anything similar to @bar’s bent metal bearing holder or @pillageTHENburn’s wood/chain linkage would work. Here’s a crappy drawing:
[/quote]
that doesn’t really show the important details :slight_smile:

but the key thing is to get someone to build this and test it.

Really, we need someone to test all four approaches and compare them. @bar has tested the 45 degree linkage and the ring approach, he hasn’t had time to test the top linkage kit, @pillageTHENburn hasn’t had time to test the top linkage kit either (and I haven’t had time to clear my garage enough to give me space to assemble my machine, so I’m not at all throwing stones here)

If someone is willing to commit to doing the testing of all four approaches, I’ll send them a free top mount kit.

1 Like

Chain would be a good option for this, , but wouldn’t be strictly necessary. Metal cable could probably do the job.

Making a loop and a circular ring may also be overkill, but then again maybe not. Need to think about that more.

1 Like

After further thought (and reading the other part of this discussion on the “throwing my hat…” thread, I have concluded that the loop isn’t necessary, but about 3/4 of each ring is necessary. It’s undoubtedly simpler to make whole rings, especially since the two sides would need different parts of the rings anyway.

It would help with the rigidity

And a chain is more likely to be the exact same length then a cable i think. But a chain is more likely to catch the other one where they cross.

2 Likes

If we offset the motors slightly to account for the slightly offset chains, then there shouldn’t be any issues with catching.

1 Like

What I love about this idea is the simplicity and that the main parts are available in remote areas.
The sled can made with a simple router jig and sandwich design.


Since I can’t get the idea out of my mind and as far as I know, no one else is attempting to build this, it might have to be me. Still not sure about it though.

It is clear that the design needs to be raised, to account for the tilt in the sled. I guess that the more weight you can get off the sled from above the cutter and on the lower side, the closer you can get get to cutting plane.

The little tricky part is the sprocket mounts. But since I have already figured how to align the chains on the 1 bolt mount, I might just try to recycle. If I take the 2 arms apart and get a socket on them, it could work.

3 Likes

In this design, the chain loops are fixed at the attachment point and slide around the sled? Seems feasible… what material are you making the base from?

If I was to try this, I would just make closed chain loops. I don’t see at this point why I would need to have them fixed to the sled. They will not have to slide around the sled, as the sprockets would just move from link to link.
As a prototype, I would go with 4mm and 18mm particle board, as I have both lying around. Would try to make it 3 part, grove smaller radius in the 18mm from both sides to leave a ~ 4mm in the middle at the bigger radius with 2 layers of the 4mm for the outsides.

1 Like

So the loop would have enough friction around the sled to remain in a fixed orientation while the shackle would surround a gear/bearing that would rotate as necessary? (Looking at @ImpetuousWombat’s original sketch, I think that’s what you’re describing.) The sled material could be pretty much anything in that case.

Not that it would matter if the loop rotated around the sled, or rather the sled may rotate within the loops… as long as the router is accurately centered.

1 Like

I’ll point out that the original linkage designs can be build with an inaccurate maslow, even with the temporary frame.

the key is to do all the critical drilling on the center between the motors, the movement here is very reproducable (if not the distances that were intended) and move the workpiece side to side to cut each piece.

That will let you make links that are exactly like each other. Then you can mount these links on a sled, do your calibration, and have a very accurate machine.

3 Likes

You’ve mentioned material a couple of times here.
Delrin might work really well.
And for remote areas…
Milk jugs? Or juice cartons…
A whitish almost see thru “milky” material they use in milk cartons are made out of HDPE? I think.
Anyway, if you cut them up into little pieces (I have a cool trick to cut them into one continuous strip which can be cut into small cubes) and put them in not quite boiling water they will melt and you can make REALLY strong plastic that is super easy to machine and has a slick feel to it like a polyurethane bushing…
I use it for all kinds of stuff.
I like it better than machinable wax.
Chris

3 Likes