I have an overly crowded garage (to say the least) and I was thinking about what could be done to eliminate the extra length of the bottom beam and the chain supports there.
I though of a way that we could make a chain takeup that would gather the slack chain. The idea is that the chain attaches to the front disk, and a spring/bungee cord attaches to the smallest disk so that as the chain is gathered, the bungee cord pays out.
These would be hung under the motors (it’s shown with a 5" long 1/2" bolt, which could easily go through a 2x4), and would also serve to keep the chain aligned with the sprockets (inspired in part by the two level chain guide)
I did these with 16" disks for gathering the chain so that there are only going to be 2 layers of chain stacked (to minimize the chance of the chain binding against itself, but it would work with smaller disks. This is designed to be built out of 3/8" or 1/2" plywood (it may be able to be made out of 1/4" plywood, but you may want to add an extra layer of the smallest disk)
They have these rotational springs that they use for garage doors; I wonder how expensive they are, and if they could be a cleaner replacement for the stretchy rope.
There is a Maslow forum topic about using the stretchy rope to provide counter force on the sled, to help reduce chain sag. I think this idea could align very nicely with that idea. In the original implementation, a 3/2 motion ratio was created by using 3 lengths of stretchy rope to pull on 2 lengths of the chain. Only discrete motion ratios can be created with that implementation, 1/2, 2/3, 3/4, etc. With this implementation, the motion ratio can take on an infinite number of values, because it comes from the ratio of the two wheel diameters. This idea, combined with that idea, could be very cool.
They have these rotational springs that they use for garage doors; I wonder
how expensive they are, and if they could be a cleaner replacement for the
stretchy rope.
they tend to be long and narrow, for this we want something short.
We don’t need a lot of force, so most springs that are designed to lift weights
are going to be drastic overkill. I think that using the stretchy rope can work
well (adjust the smaller spindle to the best diameter for the amount of pull we
need), the stretchy rope would go horizontally across the bottom of the top beam
(getting tighter as the chains get shorter, so it wouldn’t interfere with the
sled)
There is a Maslow forum topic about using the stretchy rope to provide counter
force on the sled, to help reduce chain sag. I think this idea could align
very nicely with that idea. In the original implementation, a 3/2 motion
ratio was created by using 3 lengths of stretchy rope to pull on 2 lengths of
the chain. Only discrete motion ratios can be created with that
implementation, 1/2, 2/3, 3/4, etc. With this implementation, the motion ratio
can take on an infinite number of values, because it comes from the ratio of
the two wheel diameters. This idea, combined with that idea, could be very
cool.
There was also some talk about using counterweights channeled inside PVC pipe to keep it clean and contained. I like the idea of counterweights as it provides a constant force on the chain, but at the moment, the bungee cords are working for me, so it is low on the priority list at the moment.
In the HUGE new frame design thread @bar and some other folks have seemed to be playing with a way of having the chain wrapped around the sprockets the opposite direction with extra chain routed along the top beam… seems like a great plan to me.
From appearances, @bar has this working, but it would seem to require some code changes. The physical aspect I can do… the code… well, I do code, but haven’t touched Maslow and that would be weeks to figure out
So… is this a direction GroundControl is moving in? Anybody now?
Initial patches have been merged for the firmware and Ground Control to allow the chain to be wrapped either direction around the sprocket. No comprehensive testing has been performed yet as it is pending frame designs, but it does exist in the current master code!
