what is the strength of the spring you recommend? I thought you were using mini blind springs in your frame build? did you switch the out for more powerfull constant force springs?
To roll up the slack chain not in use, this springs should be sufficient. But I have not tested it. To keep my slack chain horizontal across the top beam I need 2kg (~ 4.4 lbs)
1 Like
So I drew up a standard maslow motor mount configuration and the math on the standard everyone is using seems just as complicated. as the sled goes from the top
to the middle the distance between the pivot points on the chain sprocket changes by about 1/8" on each end or 1/4" total. does the software compensate for this. or is it just considered close enough for horseshoes and hand grenades?
2nd drawing shows the entire scale of measurements.
I do know it compensates for how much chain wraps around the sprocket based upon where the sled is. I haven’t figured out the math to it though from just reading the code. A diagram of how it all works would be helpful to have.
4 Likes
I was wondering if you could 3-D print a chain guide that always has the chain coming out at a single point… something like this…
bottom right corner is where the chain feeds out and right side is where the chain feeds in (from the takeup).
I guess one could, but you would have to tension the chain so it wouldnt’ jam. PERSON WHO tried this before specifically said chain jamming was an issue.
since maslow software already accounts for chain wrap it would be easiest to just add an extra 10 tooth sprocket up where the motors are at and in essence it would work the same way mechanically as the exsisting setup so no new math would need to be figured out
I was envisioning the motors and chain takeup on the sled (say mounted on top of the bearing rollers of a ring kit) and so you needed the chain to feed off the motor directly in line with the router bit.
I don’t think there will be any real solution for storing the excess chain on the sled without using a separate motor for it.
a spring loaded cansister like those used in retractable power cords would work, but the above drawing is probably cheaper, only requires two extra chain sprockets bolted to the top two corners which is $9 more. About what the L brackets would of cost anways.
I still think having the chains collect on the sled somehow is cooler.
How does the chain wrap around the sprockets/motors on the sled? I’m having a hard time picturing what is happening there.
sorry updated drawing with the missing fixed chain end point
I don’t think the diagram correctly captures the force on the sprocket - i think it is divided between the chain to the fixed point and the chain to the slack turning point, though not equally. I think the chain to the fixed point will have the greater effect, and more during upward travel that during downward. I think this varying effect would make calculate even messier.
1 Like
I was thinking about replacing the chain with something like this:
The “no stretch” characteristic would need to be tested, but it could solve the issue of how to store the extra cable
2 Likes
I’m curious, what is the perceived advantage(s) to having the motors on the sled? Is it just so that you can take the sled and electronics off? If so, wouldn’t it be kind of difficult to make sure they are put back on in the correct position on the chains to not have to do calibration every time?
@bar I find that interesting, what would your thoughts be for something like this? Instead of a counterweight system just have the motors wind and unwind this cord? I’m not really familiar with how things are calculated since I haven’t looked at the software, or even gotten around to assembling my Maslow yet, but seems like it would probably through a lot of the current calculations off since I’m sure the cord sag would be different than chain sag? I might order a spool of this stuff to mess around with it. Would make things a lot simpler not having to deal with counterweights or bungee material to keep chains taut. But, I suppose doing that the question would become, could the motors shafts hold up to the full weight of the sled on them? Currently, some of the weight is transferred off to the fixed point end of chain / counterweights that is keeping the chain taut. I don’t know the physics behind it to know exactly how much is transferred off in this manner though. At the same time, I suppose if that was an issue, the motors could simply turn a gear / belt that is attached to the load bearing spool fixed to the frame instead.
One challenge is measuring how much cord is unspooled… I don’t think you can reliably do so from measuring the rotation of the spool since how much spools out per rotation is affected by how much is on the spool and how it’s wound around the spool (neatly or not). But, if you had something like a couple of sprockets that the cord passes through that will turn as it does, then you could put the encoder on one of those… might need some gearing as well to up the number of rotations.
1 Like
One reason is if you can do this, then all you need to do is nail one end of the cord to the end of your top beam and you are done. Would make for a much more consumer friendly design. You could even make a bracket that attaches to the top beam such that the cord can be terminated at different distances away from the beam to allow you to handle different thickness workpieces (but then again, with no chain skip, how important is it to have it perfectly aligned?)
And if you can get it to work out well, you could add two more motors to the bottom of the sled and tie their cords off in the bottom corners of the frame. With four motors, you could eliminate rotation of the sled and accuracy issues in the corner. you may even be able to make the frame smaller.
1 Like
I’m imagining a version with four cables to replace the bricks we use now. I think that there would be two big advantages. First, the whole machine would be contained in a small part and could be stored or even shipped easily. Second, calibration has always been a thorn in our side and with four cables I think we could automate it completely. If each two cables are pulled tight in sequence we would have enough information to determine the locations of the four points the cables are attached to the frame and could compute all the information we need.
8 Likes
I smell a Quad Maslow kickstarter, 
2 Likes
I’m not sure I’m following how the four cables would be used in this scenario.
I’ve been following the threads on trying to add tension to other points on the sled (mostly on the bottom / sides) to improve accuracy and it has seemed from there that most of them result in added tension in directions that aren’t very beneficial. So not sure if you are talking about something along these lines or not.
If you use 4 motors then it is much “easier “ to control everything in theory
1 Like
Let’s say you want to spray paint “Romanes eunt domus” in huge letters on a wall in the middle of the night. If you had a sled with the motors already mounted, then all you would need to do is mount the ends of the chains to points at the top of the wall (a known distance apart, of course).
1 Like