My kit will not arrive until February, however I am trying to solve some of the major issues with my design and setup. Chain sag appears to be a major problem maintaining accuracy. The sag is obviously due to gravity acting on the weight of the chain.
Has anyone considered using a lighter weight chain? As a cyclist the first thing that comes to mind is a road bike chain. The motor drive sprocket would need to be changed or perhaps combine the current motor/sprocket with an idler shaft - #24 sprocket on one end and bike sprocket on the other. Any ideas??
Chain sag adds up to a couple mm accuracy loss, max, and that’s only at the corners. Several talented people are working on software to account for the sag right now. If I were you I’d stick with the stock chain so that the new fix doesn’t require tweaking for your non-standard chain.
Sled weight has a direct effect on chain sag as well as the weight of the chain itself. If a software fix is found wouldn’t we need a specific weight and weight distribution for the sled as well? It seems everyone has their own sled variation with different types & location of weight. Even bricks do not weigh the same. That’s what leads me to suggest removing as much chain sag at its source: the chain, because there are so many other variables involved.
I’m also considering extending the placement of the motors. It appears many designs locate them about 12 feet apart. Perhaps 14 feet apart and another foot higher would remove some of the chain sag as well.
Based on what I’ve read in the development thread, this is something that would be accounted for, although maybe someone more involved in the process could confirm that.
I have not tested this theory, but wouldn’t a longer chain lead to more chain sag, not less? The reason for extending the top beam is to reduce the angle that the chain is at when it is at the far left and right sides of the bed. The more vertical the chain is, the less force the other chain is exerting on the sled. I actually have a shorter top beam on my machine and it helps illustrate the point. My top beam is 112". I have noticed some distortion >2’ from the center of the bed, and unacceptable distortion another foot beyond that.
That being said, I’m in the early stages of closing on a house that has a 2 bay garage! So long as everything goes well, the Maslow is going into said garage. One of the first upgrades then will be adding a longer top beam so I will be able to test length to see what effects that has on machine accuracy and chain sag
Yes, you make a good point, adding more chain would add to the sag. The effect of a shallower chain angle with extended motor spacing might more than compensate for that increase. Trial and error would tell - or is there a mechanical engineer in the forum? Has anyone tried this?
It might be useful to note that while extra weight can be good there is a maximum. The absolute max weight of your sled shouldn’t be (can’t be) more than about 39.5 pounds. With a 39.5 pound sled at the top center of the work piece that puts 65.68 pounds of tension on each chain. The motors stall at 66 pounds, so a 39.5 pound sled is about all it can handle (I personally would aim for something well under that to allow a margin of error).
going from ~10’ apart to ~12’ apart doubles the angle of the chain in the bottom
corners, you would have to go to 16’ apart to get another doubling.
going from ~11’ of chain out (stock) to ~12’ of chian out (12’ apart) makes
little difference to the sag, but adding another 2’ of chain would start to add
more sag.
There has been talk of testing this, but as far as I know there hasn’t been anything done yet. I believe @mrfugu has plans to do this with his unistrut frame. As far as mechanical engineers, I know that quite a few people on these forums have quite extensive knowledge into the mechanical workings of the machine. See @dlang’s post above mine, he is very good with this sort of thing.
My sculpture degree doesn’t really give me much formal background in engineering, but I do have a bit of relevant work experience. Still doesn’t make me an ME though. If I could tell high school age me what to major in, I would have guided myself in that direction.
This is a good point. My sled is somewhere around the 40 lb mark and I haven’t run into any issues with the motors stalling. I need to more accurately measure its weight at some point, my current numbers come from adding together the weights of all the components. I have been wondering if I’m putting too much stress on the motors, though.
This could be because your motors are closer together right? The closer your motors are to each other the more acute the angle is between the chains when the sled is at top center (given the same motor height). The more acute that chain angle is the less tension there is on the chains.
That 39.5 pound number is based on “standard” machine dimensions (I don’t recall off the top of my head what numbers I used but I could look it up if you’re curious). The equation to find the force on two chains under load at a given angle is: F = (m x 0.5) ÷ cos(a x 0.5)
where: F is the calculated force on each chain m is the mass or in our case the weight of the sled a is the internal angle between the two chains
So if you measure your chain angle when the sled is top center you could plug in your numbers and figure out your max sled weight. (The max is 66 pounds per chain).
dlang - thanks for your input on motor spacing. This helps with my design. I’m planning a rigid unistrut type of top beam. The longest I’ve found locally is 120". Piecing something together is a possibility for wider spacing although as you suggest the benefits diminish.
pillageTHENburn - so essentially what you are saying is that wider and higher motor spacing reduces the load on the chain at the edges of the material.
Chain alternatives might include:
300 grams at 116 links - might require 2 lengths per side of Maslow, about $13 each at Planet Cyclery.
Better yet - made for single speed use:
Can’t find a weight but surely lighter for about $16 each at Planet Cyclery.
These alternative chains are more expensive than a generic #25 however the accuracy gained might be a savings in the long run with efficient & reliable use of materials.
The pitch of the standard road bike chain is 0.5". Finding a sprocket to fit would be required.
raising the motors make it so that the angle at the top center is steeper, which
helps. But it means that the angles in the bottom corners are steeper as well,
which hurts.
right now, the problems we have are:
bottom corners due to steep angles resulting in very low chain tension (chain
sag and just lack of force to move the sled)
top center due to the shallow angles having high tension, resulting in the frame
flexing in the stock design.
going to a top beam style frame solves most of the problems with the flexing, at
which point you only need to go higher if you have a heavy sled or different
sprockets (larger sprockets will move the sled faster, but reduce the force that
the motors can put on the sled)
so the single best thing to do is to move the motors out a little bit.
Unfortunantly, 10’ material is common, 12’ material a little less so.
The current software fix for chain sag (for triangular kinematics) that is undergoing testing accounts for sled weight, chain weight, and the angle the Maslow is leaning.
I found a local plumbing supply with 20’ lengths of unistrut !
I know…that’s overkill however it does allow 12’ plus and extra piece to use in the structure.
Still looking for a sprocket to handle 0.5" pitch bike chain. I don’t really want to use a larger one than the one in the current kit considering what you said about the reduction of motor force.
by comparison, it’s just over 40g for 48 links (12") of #25 roller chain, so the
equivalent of 116 bicycle links would weigh just under 200g (assuming I didn’t
mess up the calculations, 0.089 pounds/ft)
Thanks for pointing this out. The first thing I should have done is looked up the weight of the #25 chain. Obviously even the lightest bike chain at 250g/116 links is at best equal weight. Sorry for the diversion, I’ll do better.
I wouldn’t worry too much about it. This is how we figure out solutions to the problems we’re currently faced with. Sometimes it takes a couple of duds to get to a good fix.
