My first thought was to drill two holes in the where the rope passes through on some quality pulleys. Still allowing enough room for hoop. So bolts could be used to secure three to a top and lower plate. Which the lower mounts to the sled. I just could help but over build 
Hy everyone my name is Iroc and I’m a habitual over builder…
Take my money!! I’m ready now
2 Likes
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.
In our case, the chain will flex instead of the block pivoting in place. In
either case until the rollers start moving there will be a little distortiion,
it’s just a matter of where.
I’m thinking something like this:
add a big washer or similar if you are worried about it falling off
3 Likes
@bar ask the shop you had make the brackets what they would charge to make a quarter circle in either anodized aluminum or stainless steel with tabs that would attach to the existing brackets. The inside edge could either have a V (ideal) or be rounded if it’s easier to just bend a hunk of rod. from what you’ve said about getting the brackets done, it sounds like this would be fairly easy for them to do.
The chain angles only vary by about 60 degrees (top center, the angle is ~145, bottom corner the angle is ~90 so if the sled is balanced and one side fixed, the other only needs 55-60 degrees of motion)
Awesome! Really hope thus is part of the solution to get maslow even more accurate
I am working on a 12 inch diameter sled made out of 3/8 inch thick steel plate which is heavy enough where one doesn’t even need bricks
2 Bearings were $40
http://m.ebay.com/itm/282350157139?varId=581465390207
Sled was $10 laser cut
$50 total
What price are you trying to get below?
3 Likes
Awesome!
@aluminumwelder You did a really good job sourcing those parts, those all seem like really good deals. I don’t know if I have a target price in mind, but I just want to keep it as affordable as possible.
Here’s how far I’ve gotten.
I’ve learned that anything with a ball bearing in the center seems to work pretty well, and that making the circle that the pulleys run on smaller makes them run more smoothly because of the way the mechanical advantage works.
Larger pulleys seem to be slightly better, but it doesn’t make a huge difference.
1 Like
can you try with one chain fixed and the other movable?
that will put all the force on the one wheel to move instead of it being spread
between both of them, which may make it work better.
a couple clamps (or possibly even zip-ties) on the rail to keep one chain from
moving will do for a quick test, no need to make any actual mechanical changes
If one side is fixed to the sled the intersection point of the chain vectors will no longer be at that center point of the sled (the bit) for all points in the cutting area. Seems like that adds more math back into the system similar to what we are currently trying to eliminate.
@bar, would moving the bricks to more of a 6-o-clock position help to stabilize the sled rotation in those tests? Seems like the goal is to keep the sled vertical (for lack of a better term) at all times, am I right in that assumption?
It seems to me that you have thrown your ring into the sled modification hat, rather than the other way around!
2 Likes
I agree with @keith that if one point is anchored it will change the math, right? That doesn’t mean we shouldn’t give it a try.
@blurfl I tried 1/4 3/8 and 5/8ths and they all worked pretty well. 1/4 was a little flimsy but really easy to bend. I’d recommend finding cold rolled steel over what ever they sell at most hardware stores because it’s going to be easier to bend
I guess the “bonus” us after kickstarter orderers will have is lots of options!
1 Like
If one side is fixed to the sled the intersection point of the chain vectors
will no longer be at that center point of the sled (the bit) for all points in
the cutting area. Seems like that adds more math back into the system similar
to what we are currently trying to eliminate.
If the weight is balanced so the CG is at the bit, the entire sled will rotate
so the chain vectors will remain pointed at the center of the sled.
@bar, would moving the bricks to more of a 6-o-clock position help to
stabilize the sled rotation in those tests? Seems like the goal is to keep the
sled vertical (for lack of a better term) at all times, am I right in that
assumption?
moving the weight down makes the sled more like a pendulum, which will make it
more likely to swing when you start or stop moving.
There’s nothing special about the sled being upright.
1 Like
I agree with @keith that if one point is anchored it will change the math, right?
not if you balance the sled (put the bricks on opposite sides of the router,
same total weight, but the CG is at the bit)
if you think about it, if the sled is balanced, the torque from the motor should
gradually rotate the sled until one chain is against the stop. Everything would
still work.
I am just proposing taking advantage of this to simplify the construction.
1 Like
Ah, I see where you are going with that, and I can see how it works thanks to your explanation. It sounds like balancing will be pretty key in that setup. I would worry that getting the sled perfectly balanced would be difficult (seems like the resulting error due to an off center CG could be simulated), and that trailing items like the z-axis cables and vacuum hose might act against that balance. Of course, that may also be an issue with a two bearing setup.
I am interested to see how these set-ups perform under cutting conditions… just let me grab some popcorn 
having a uniform center of gravity is one reason why I’m using a 3/8" thick x 12" diameter sled. The steel is about as heavy as the bricks and if they were laser cut in bulk the price is under $10.
one issue with Bar’s current curved bar approach is that it might be somewhat difficult to get the semi circular bar EXACTLY centered. unlike squares it is VERY easy to be off 1/8" or more from center, not sure how that would affect cuts, but anyone that has worked a lot with circles know that unless you have a pinpoint center reference they are very difficult to center.
try making 3 discs with a 6 bolt flange pattern and getting them to line up by hand, almost impossible.
1 Like
one issue with Bar’s current curved bar approach is that it might be somewhat difficult to get the semi circular bar EXACTLY centered.
not that hard.
make sure you have the same distance from the center hole to each bracket.
or have the maslow cut the holes for the bracket mounts so they are in the right
place.
If you are making the sled manually, it’s easy to cut holes the same distance
from the center, you use a circle cutter jig (something that holds the router
and has a nail/screw in the center of the workpiece), you just rotate around
that
center to cut, or to position the router to drill mounting holes
It sounds like balancing will be pretty key in that setup.
yes, but I’ve blanced enough things that I don’t think it’s that hard.
As you say, we can simulate how much the CG being off a bit causes errors, and
I share your concerns about the vaccuum hose (I don’t think the cables will
matter).
But the only way to know for sure is to try it.