the normal maslow and the m4 is amazing, a truly innovative and creative approach to affordable good cncs, but the normal Maslow had its limitations, mostly speed and needing to be vertical but the m4 improved on those issue greatly and is continuing to improve at a very impressive rate which i love to see, but no matter what the m4 has its limitations, this style always will, it rides on the material and that means you have to deal with that as a problem, but what about a m8? 4 more motors, way more headache but more precision and speed potential and with enough tension and a round frame for all 8 belts it may be able to hover over work pieces and just do its thing, rigidity headaches aside of course, and im sure yall have already thought of this spiderweb of a design but just wanted to mention the potential of this idea if you hadnt, even know i know its a hell on earth of a project to even consider while still advancing the m4 tech.
I really like the compact portability of the sled design even with it’s limitations. Something in three d would require a lot more frame infrastructure. But for fun let’s play. So the idea is a frame with anchors up and down with a router suspended in space? Mathematically you could make a stable point in three d space with just 4 defined distances on tense ropes. So the current four belts and almost the same kinematics might already work. I could see it working with two high in one diagonal and two low on the other diagonal at the corners of a rectangle. The main issues I can see adapting the current design would be twisting of the arms and spools up and down. This could be limited if the belts passed through stable defined holes fixed to a sled instead of rotating around the central cylinder. This would probably limit the ratio of functioning cutting area to frame size. I would also redesign the belt end anchors to rotate freely in three space by having a horizontal pin as well like a CV joint. Also to make it work I would have much longer z screws to reach from that floating sled down to the work piece. I’ll sketch something up later today. I think it could be done with the existing four motors and math. Fun design puzzle. Not sure I really want to make one.
Renaming this: project sky spider.
are you saying it’d have 4 belts above it onto like a dome shape suspending it? i was thinking a cross shape of belts filling in the gaps of the normal belts on the same plane as them(just more stacks of arms in the center), maybe with some reinforcements to the 4 cross belts that would hold the machines weight and hold it taught, i look forward to seeing what you sketch up, best of luck
Caleb Yager wrote:
the normal maslow and the m4 is amazing, a truly innovative and creative
approach to affordable good cncs, but the normal Maslow had its limitations,
mostly speed and needing to be vertical but the m4 improved on those issue
greatly and is continuing to improve at a very impressive rate which i love to
see, but no matter what the m4 has its limitations, this style always will, it
rides on the material and that means you have to deal with that as a problem,
but what about a m8? 4 more motors, way more headache but more precision and
speed potential and with enough tension and a round frame for all 8 belts it
may be able to hover over work pieces and just do its thing, rigidity
headaches aside of course, and im sure yall have already thought of this
spiderweb of a design but just wanted to mention the potential of this idea if
you hadnt, even know i know its a hell on earth of a project to even consider
while still advancing the m4 tech.
take a look at the hangprinter that tried to do that. It ends up being MUCH
harder to position things in 3d than in 2d
David Lang
Caleb Yager wrote:
are you saying it’d have 4 belts above it onto like a dome shape suspending
it? i was thinking a cross shape of belts filling in the gaps of the normal
belts on the same plane as them(just more stacks of arms in the center), maybe
with some reinforcements to the 4 cross belts that would hold the machines
weight and hold it taught, i look forward to seeing what you sketch up, best
of luck
more belts will not increase accuracy, it will make it easier to end up in a
situation where the belts are pulling against each other.
David Lang
PROJECT SKY SPIDER!
This is not a good idea…. but I think it might work.
You would need:
A new sled base with four holes for ropes and probably little table legs
Two small tripods
8 rope anchor loops
The general plan is to flip the maslow upside down with the active end of the router hanging out the bottom.
Take the arms off of the maslow and tie them to a new sled base using rope.
Tie the belt ends to ropes.
Attach two belt end ropes to tripods above the moving plane of the maslow,
attach two belt end ropes below the plane on the floor.
Enter the z offset positive and negative into the kinematics.
Enter the belt extension values from the center of the arm hole to the center of the new sled plus the little loops of rope at the anchor ends.
Calibrate the maslow in a plane using four little table legs attached to your new sled.
I think you would end up with a very small useable area in the middle where the maslow could move in a virtual plane above the cutting surface.
Here are diagrams. Maybe I try this when I actually retire. This is for mostly entertainment.
huh. pretty. https://www.youtube.com/watch?v=S9vbJFpzeTQ
just do this and drop your router in it. Hope vibration doesn’t cause problems.
looks like someone was still working on it in 2022 Torbjorn Ludvigsen
your cut area would be 1/2 the size of your sled, otherwise the belts going up
will hit your workpiece (and how are you mounting your workpiece??)
David Lang
The belts would be on a long angle the workpiece would be on the floor. The dotted line is the virtual plane that the sled moves in the cutting would happen on the ground.
Nope, terrible idea. too much rotation. Will think on it some more. Thank you for the fun puzzle though. I really don’t think it would ever be a good idea. because the flexing and wobble would just keep adding up no matter how you suspended it. I might play some more but Maslow4 is cool as a mostly portable machine. At some point you would be engineering a giant box which is back to a gantry based machine. I am continually impressed how much accuracy is possible with the maslow 4 setup and very clever mathematical programming.
Five thoughts in the spirit of play and anaylisis. I want to keep putting my own energy generally into making the current maslow4 better. Personally I can help with documentation and frame design. I have little meaningful experience programming and haven’t learned anything about maslow’s programming so other than vibe coding experiments I’m not doing anything there. This is a fun thing to think about for me but I personally don’t want to get too distracted from making the good thing we have work more reliably and better. That said, by playing with the idea of a hanging maslow I am learning a bit more about the current design.
-
Suspension is really hard because as you tighten something up it gets infinitely more force as you try to get a rope with a load pulled horizontal. The maslow is a suspension machine that acts in a plane. That’s going to make it really hard to use the clever controls in maslow to work in three d. If we just tried to lift the maslow off the ground and pull the belts really tight to get it in a plane it would require huge tensions in the belts. It looks like the hangprinter people solved this by making a very very long vertical line to act in the zdirection. I really respect the people who are doing the programming side of maslow and I don’t know much about it but when we type in a different number for work thickness maslow might be thinking about its belts more as a pyramid so maybe one could type in a very very large number for work thickness to make a hanging pyramid with reasonable tension that won’t break things and still get some meaningful movement?
-
When we hang things with four belts we have enough information in the system to define a single point in three D space. So if you had a router spindle light enough and four high points you could define that point and then suspend a router from it which is what I was trying to do above but then any rotation in the sled would be magnified as you move away from that point of suspension. In my design above the router is a 2 foot lever so even a small tip in the sled would make a very large movement in the working end. The Maslow design is pretty brilliant in 2d because it is all centered around the router spindle so it doesn’t matter very much how the sled is rotating or jiggling around that point when it is working in a 2d plane. In three D to use this same logic I would ideally make the controlled point as close to the cutting end of the router as possible and as small as possible. Perhaps a small ring around the router bit. The current planar Maslow has the motors stacked up pulling from different heights so there is some rotational noise introduced which is counteracted by the large flat sled, in trying to control something in three d this would be a lot harder.
-
So then to control a stick (router) in three D we would need two controlled points spaced out along it’s length. This would requires 8 ropes under tension or with some bodging and using gravity as the fourth tension force at least 6 controllable lines under tension. To keep things as simple as posssible (ha) I could imagine suspending a maslow from two sets of 4 identical belts with one set directly above and parallel to the first. Think two upside down pyramids nested inside each other. If you made the second set of belts exactly mirror the first with movements and control then you could skip a lot of hard math and still get two controlled points moving in a vertical line together. so a new improved Project Sky Spider design might be two maslows stacked on top of each other hanging in a slack pyramid with one just mirroring the movement commands of the other but still needing to do the subtle encoder monitoring and tiny adjustments to keep the belts at the same length.
-
The most useful thing I have learned so far from playing with this is that the maslow arms could be taken off the spindle and as long as they are still free to rotate around that central point they could be moved out from the center and the machine wouldn’t notice much (other than the clever math that corrects for the different heights of the arms) . Looking back at the original maslow designs this was accomplished by using a steel ring with pulleys riding along it. In Maslow4 the spools rotate around the router body. For playing with my own machine then I could see putting a plastic or metal ring in place of the arm on the stack and then moving the motor out from the center along a rope. I don’t know why I would want to do this but it would work.
-
So to make money try to sell kits of two maslows to people who want to try an untested and unsupported design for tying a robot with a spinning blade to their ceiling. This will end well. : )
The new and improved Project Sky Spider design:
Ahem, soooooooooo I have been quietly playing around with this idea in the background of the other stuff I’ve been doing, for like…months, I was going to test it a bit more before I talked about it, buuuuut:
The main advantages:
- With some clever geometry you can get the attachment points all the same height.
- You can get a sub-5kg assembly.
- You can fit a Makita router easily.
- I think all these mean you could run a higher speed lower torque set of motors and push feedrates higher: (hence this thread: Feedrate limits of the Maslow (theoretical).
However.
(And this is why I hadn’t posted it yet)
I’ve only tried it with 2.5m leads, and I am not sure how 5+ metre leads would work, which is what you’d need for a full sheet. Voltage drop and interference could well scupper the whole thing.
I’ll get off my backside and actually write up what I have so far later.
That is pretty cool. I guess it makes a bit more of a wire nest. Neat that you made it work!
This was / is very much my proof of concept setup - it was the minimum length of wires I could get it to work in my shed setup, but that also means suspending them. Longer wires could be routed down and around the frame, but I wasn’t sure it would even work with these lengths of wires 
you are only one more maslow away from making project sky spider a reality. You already have one maslow anchored at the corners of a room. just sayin…..
Looking forward to reading about your actual project. The vertical rails and handle and computer mount look pretty cool
@Caleb_Yager Ok, one more idea. Much more practical. It completely avoids suspension problems and extra 3d math. If you really wanted to carve something with large voids with the maslow4 If you set up the maslow as normal but on a very smooth surface and then flipped the router upside down in it’s housing the blade would be sticking up into the air above the maslow, not below it. You could then screw your workpiece into the bottom of a short table and put it above the maslow. The machine would act as normal but the blade could move up and down cutting whatever. This would be a bit more dangerous with the blade being up and exposed but it would require very little new engineering.
It would be bit tight to squeeze the top of the router and computer in upside down I would unscrew the stuff from the sled and add some layers of wood to get the extra height and add some weight at the bottom. If you look at the router be gone threads in the forum I bet you could flip a spindle upside down much more easily.
In my first idea for this it was all taped to a ceiling. Floor is much better.
Presenting PROJECT FLOOR SPIDER! aka Maslow under a table
Dave wrote:
This is a key thing in simplifying the math. In theory you can account for
offsets from this center of rotation, but if the sled then rotates, things
change. (andd we’re having enough problems with getting the math right as it is)
Ahem, soooooooooo I have been quietly playing around with this idea in the background of the other stuff I’ve been doing, for like…months, I was going to test it a bit more before I talked about it, buuuuut:
The main advantages:
- With some clever geometry you can get the attachment points all the same height.
- You can get a sub-5kg assembly.
- You can fit a Makita router easily.
go to a spindle instead of the router and you can save more wight
- I think all these mean you could run a higher speed lower torque set of motors and push feedrates higher: (hence this thread: Feedrate limits of the Maslow (theoretical).
you will be pushing the rigidity of the system, but you are working on that as
well.
However.
(And this is why I hadn’t posted it yet)
I’ve only tried it with 2.5m leads, and I am not sure how 5+ metre leads would work, which is what you’d need for a full sheet. Voltage drop and interference could well scupper the whole thing.
the electronics are light, so I don’t see a lot of advantage in moving the
controller board off of the sled, but I look forward to see your work.
David Lang
@RandomDave This is spectacular. I love how thoroughly you are taking the kit as a starting point and experimenting and building new things. It makes my day every time I see one of these posts.
Project sky spider is still not a good idea on the other hand Project Floor Spider has real possiblilites. but I am having a lot of fun with Abundance seeing how fast I can make sketches in 3d. Only sharing it because it seemed fun.
https://abundance.maslowcnc.com/run/wouldchuckit/PROJECT_SKY_SPIDER
I am really having fun with abundance. this was about 2.5 hours worth of work. Not a bad way to sketch something up .
