Throwing my hat in the sled modification ring

having posted that, I’ll point out that the example here has l and s = 0, but not h3 :slight_smile:

also for people building this, remember that you only need one chain on a flex mount, the other can be fixed and let the sled rotate (as long as the cord and dust hose don’t cause problems)

you can use wheels like these

different CNC supply places have them in different sizes.

remember that there are vitually no side loads on these wheels, so even a slight
V like this may be enough to keep them in place.

you may also be able to fund stock already made into rings for various purposes.

1 Like

The V bearings is where I started with the Uber sled but never could find rings or discs with the corresponding chamfer. Easy enough to make (or have made) but I was looking for 100% off-the-shelf.

1 Like

another (slightly larger) metal v-wheel

and searching ebay for v grove bearings comes up with a LOT of options.

The V bearings is where I started with the Uber sled but never could find
rings or discs with the corresponding chamfer. Easy enough to make (or have
made) but I was looking for 100% off-the-shelf.

they should be able to work well with a round surface instead of chamfered
surface (if you get one large enough)

Just off the top of my head. the pin in the middle out front the chain would slide over and be held in ( Just like the Pulley wheels ) with a nut. There would be bearings between shafts and the pulley wheels.

It could be way more simple. Just wanted to bang the concept out quickly. The hoop would ride pinched between the three pulley wheels.

I think nylon pulley wheels could be gotten for way less than stainless V wheels.

you are horribly over-engineering this

forces are only in one direction (tension on the chain), and so if you go with multiple points of contact with the ring, you just add uncertainly and error (exactly what is the distance from the end of the chain to the center of the ring?)

something like what @chauhuh posted, but with only one wheel and a tab on the outside for the chain to attach to (a 1/4" thick tab with a hole in it, then use a master link to connect the chain)

This all seems like it is quickly being over engineered… Bar’s solution is 1 extra bracket with some bolts, 1 steel dowel, self bent, and 2 blocks. the only necessary upgrade is the blocks, and we found those for ~$20… everything else seems like overkill.

Interesting for sure, but not necessary imho. lets see some comparisons to the ‘default maslowCNC’ and dial it in from there.

No offense intended, these are all interesting, but the idea is to keep cost and availability as low as possible.

Well yeah it’s over engineered, but three points of contact eliminate ALL slop. and it could easily be done with plate and angle or tube. trust me the first one wouldn’t look like that. It would be junkyard fabulous.

Though I’ve learned from many projects in that past that: it’s way easier to over shoot your design and work back from it. Plan for the moon reach the stars.

So, leaving aside the issue with the pulley/block/whatever - it seems to me that if the ring isn’t perfectly circular, you are introducing more error than you are solving.

My math understanding wrong on that? (probably)

If not, then how to make an accurate ring would be my issue (a metal worker I’m not).

like this… Hoop roller

Takes a littel work and you have to do the math for spacing to get your desired radius, but can be done with just two wheels. Three would be easiest though…

Maybe a simple plate.

for the wheels, look at

18mm wide, 40mm diameter, 15mm shaft

@iRoc999, the problem is that if you introduce too much friction, the block won’t slide on the rod, any inperfections on the rod will bind your three-wheel setup, while one wheel will just conform.

@bdillahu, you are correct that imperfections on the ring will affect the position, but that’s actually much easier to detect and fix than the various error sources from the stock design.

I’m not even sure if we need the third bracket, if the hoop is stiff enough to avoid bending (which you need anyway), it shouldn’t matter if it moves up and down a smidge in relation to the workpiece (just like it doesn’t matter much if the chains aren’t perfectly parallel to the workpiece)

you could probably get away with a flat 1/4" thick plate with two holes in it, one for the chain, one for the bolt. you don’t even need something on the other side of the bearing (which makes getting the sled out of the way really easy)

combined with the idea of letting the sled still rotate, you could have one chain attached to the traditional mount, and a quarter circle with a single bearing for the other chain.

1 Like

Everyone seems to be instinctively adding at least two points of contact, and I think that is the right idea. I’ve noticed that all of the similar things that I can find out there use at least two wheels like these zip line trolleys:

I think it’s because with just one point of contact there is a tenancy for the block to pivot in place and until a bit of strain builds up, then suddenly catch up with where it should be. A second point of contact would prevent that pivoting in place. That’s just a hunch, but I certainly am seeing behavior where the block waits until some strain has built up, then catches up suddenly.

That leads me to point #2. I know nobody has proposed using a block without bearings, but we need real bearings. All the blocks available at the local hardware stores rely on the bronze wheel sliding on a steel pin, which has almost exactly the same coefficient of friction as the bronze wheel sliding on the steel rod so sometimes the block won’t even rotate, just slide which isn’t any good.

I’m stuck here at the moment waiting for the USPS guy to come, but as soon as he does I’ll drive out to the sailing supply store and see if they can get me some better blocks.

I’m not super worried about the perfection of the circle. If this idea pans out, we can have bent rods made professionally and sell them for a couple bucks for anyone who wants a perfect circle and doesn’t want to bend it.


Exactly! Which is why roller coaster wheel assemblies are build like they are. The first tries didn;t hold the track from enough points and build up tension that suddenly lashed out and caused lose of contact.

Not the same thing, but technically the same principles. And mind you that roller coasters work on inertia and friction is the enemy. Trust me three points of contact for eacj contact point should be minimum. Or you are asking for sudden and violent backlash.

Honestly I’d the hoop would need more internal bracing to really work fluidly.

Really the Zipline trolly is basicly what I’m proposing, but I’d just add a thrid wheel to help keep the hoop tracking better, since its more of a circle than a stretched zip line cable. Which can’t be flat, but comparatively is way close to level thatn a small metall hoop.

@bar for this application any small sailing block will work, as long as the sheave can ride on the metal bar well.

These blocks have no bearings, nor do i believe they are needed as long as the tolerances are good on the block. Most will have well below 1mm of play w/o any loading and once the sled is hung, they’ll stay put and roll well. I’m familiar with the effects you’re reporting with the hardware store pulleys, and you’ll see they’re night and day different from even the cheapest sailing blocks. It’ll roll well, with at most for maintenance, some dry lubricant spray every now and then.


I think your idea of using a sailing block is a really good one. I’m planning to stop and pick up one on my way home this evening and test it first thing tomorrow morning. Some of the deck mount blocks look like they could give us a good way to test a two or three point of contact system so I’ll get some of those too :smile:

1 Like

I think this is a great idea. We’d probably want some sort of safety so that the sled couldn’t ever fall, but the simplicity is really nice!