given that the round line we are talking about is only 1/16" diameter, I doubt
that you will find a ribbon that would be thinner.
however, what about using a steel tape measure? cheap, strong and readily
available (idea inspired by archer maslow thread)
David Lang
My personal fear about having thin metal strips under high tension is that something might snap and the sharp edge goes flying.
that’s possible, but I think the safety margin that we will have is huge, how
string is a 1" wide tape measure?
I think the biggest problem with either the steel tape or the ribbon is that
they will be more sensitive to even small Z offsets.
www.chiefdelphi.com/t/tape-measure-tensile-strength/163921 talks about using
tape measures for climbing robots, they use an example of a 1.25" wide tape
measure and calculate that it’s good until roughly 3940 pounds
That’s about 4x the strength of the chains. So I’d be very comfortable with even
a 3/4 or 1" wide tape measure.
David Lang
another advantage of a tape measure is the idea of an optical reader to read how
much has been fed out.
another video
www.reddit.com/r/FTC/comments/3xxbpdfor_those_of_you_wondering_about_tape_measures
David Lang
I could be wrong, but I tend to think that there will be a slight twisting action on the ribbon if it isn’t perfectly in line. If true, it would be most pronounced when the ribbon is at its shortest, but even then I don’t know that it be so extreme that it will cause an issue. My current design for rope uses rollers to guide the rope where I want it to go and the ribbon will work just as well. I only constrain it in one plane which causes me problems because it’s so thin and it slides off the end stop The ribbon will actually work way better. See below (some parts are excluded from each view… top one is a side view, bottom one is a top-down view)
By going to a ribbon, the end stop will push against it with no problem. The 1/16-inch rope always found a way around the arm of the end stop when tensioned (either above or below the arm)… defeating the purpose of using one.
Regardless, I tend to think that first set of rollers will resolve any twist in the ribbon if there is a z offset issue. But whatever does happen, I have to think it will be much less of an issue than we experience currently with roller chain.
I don’t know about stretch or strength but this came to mind when thinking of tape measures.
Maybe this:
This is my pillar design for a ribbon (made out of whatever material…metal, hmwpe, etc.)
It’s based upon 3/8 inch wide ribbon. I already have 3/8-inch nylon spacers that I’m using as rollers, so 3/8-inch ribbon fits it… If a wider ribbon is used, then longer spacers, longer d-shaft, and longer spool would be needed… none of which is a problem, just rather use what I have on hand.
The first image shows a top-down view of the path the ribbon takes. The spacing is based upon a 1/16th inch diameter rope and when the ribbon arrives, I might have to adjust the spacing based upon its width.
The first two white rollers align the ribbon (fyi, their position is used in the firmware’s kinematics calculation). The second white roller is sort of a pinch roller to press the ribbon against the spool to ensure that no matter what, as ribbon is let out or in, the spool always moves. Without this, if something goes slack encoder steps can be lost. The third white roller positions the ribbon so that the end stop switch arm can be actuated when there’s tension on the ribbon. Hopefully, if the ribbon loses tension and goes slack, the arm has enough force to push the slacked ribbon outward to deactivate.
The second image shows the tentative pattern for the pillar plates. The bottom image is a side view of the assembly with a few parts missing (the first two rollers, for example). the pink bolts/nuts are there to clamp the pillar to the top plate as the top plate bolt pattern doesn’t currently match the new pillar’s design (I spread out the bolts to give more room… These bolts will go away eventually when I get to cut a new top plate that matches the pillar.
The webbing arrived today and I’ve built my first pillar. It turns fairly well and the end stop appears to work (I can hear it click when the webbing goes slack). However, the “pinch roller” really doesn’t pinch well enough because I can manage to slide the webbing through the gap without the spool turning if I do it light enough. For this to work on the bottom motors, there can be no chance for slippage when tension on the webbing is low. Maybe a rubber roller with some pressure against the webbing and spool is needed. Nevertheless, its a step forward using the webbing (vs. rope) in my opinion because it works really well with the end stop.
Any way you can give the roller some teeth? With that kind of nylon webbing, some blunt teeth should engage the texture of the strap without breaking any fibers.
I kinda did that on the spool (more soft texture rather than teeth) because it’s cut from mdf and I could readily add it. But to make it ‘sharp’ I worry about it wearing. But, if that can be overcome (maybe a different material) then that would work. But I also I found some pinch rollers on Amazon I’m going to give a try. Options are good.
I disassembled it and looked at the arrangement this morning and I think the nylon pinch roller is a little too far away from the spool. The webbing is about 0.8 mm thick, so we are talking very tolerances getting the spacing between the spool and the pinch roller “good enough”. There’s a little play in the pinch roller itself (the sex bolt that it turns on is a hair smaller than the i.d. of the roller) and the spool may not be exactly what my drawings say since it was laser cut and there’s part of the line burned away (I think laser cutting is all “follow path” vs. “outside/inside profiles”). Those two, i think cause that gap to be a little wider than designed resulting in too low of pinch force. It would be preferred to make that distance adjustable rather than re-cutting the plates until I get it right. A spring loaded arm would work, but I’m worried that everything is starting to get complex again. It needs to be simple and reliable.
Exciting! What does the stretch characteristics feel like?
I’ll check this weekend, but being a webbing I think it will have to have some level of stretch (shrink along the width as tension is applied) but I wonder that since the distance is being measured when the webbing is under tension, whether it actually matters. It would have an impact if you measure out the webbing while not under tension. For instance, when you are first trying to connect up the sled, you need to feed out webbing. If you try to measure how much is fed out, then the amount measured will be different than actual once tension is applied. I’m thinking that process will have to change to accommodate. I’m tending to look to use some automated process that once it’s hooked up, it moves around the work area and o calibrate itself. Not sure how to do that just yet.
Built a new pillar with closer spacing of the rollers and extra layers of wood. I had been using untempered hardboard I found off amazon but they’re out of stock. I needed it to be 1/8-inch thick and really sturdy. Turns out that cheap clipboards are perfect substitutes.
with the closer spacing of the rollers, there’s no slippage and any movement of the webbing through the pillar results in the spool turning. We’ll see if there’s any degradation over time, but right now I see no need for rubber pinch rollers.
I’ve been thinking about the exact same thing. I believe (but have not tested yet) that if we pull each pair of motors tight we can get enough lengths to fully define where the four corners are. I am not at a computer right now to make a diagram, but imagine a rectangle where all of the side lengths and the diagonal lengths are known. That gives us enough information to lock down the relative coordinates of the corners.
Although I guess maybe saying rectangle is assuming too much. If we have four points and we know the distances between all of them, then we can compute their relationships. We can find the lengths between each two points by pulling the cables tight in each direction.
When you mentioned that process before but I don’t really understand it. When you have time, I’d love to hear more about it.
and during operation, it will vary depending on how much tension it’s under as
it’s wound up (and I can see cases where it may 'slip) depending on the tension
as it was wound vs tension while it’s unwound)
this is why I think a steel tape would be far better.
David Lang
