Delta, cartesian, corexy size question
-
I tried to do the same thing, but a bigger bed on the prusa means a lot more weight to move backwards and forwards on the Y axis, as well as the acceleration and stop problems you also have more movement in the main frame to deal with. I went for the corexy after the prusa due to the bed basically not moving apart from down!
-
A corexy is easy to make larger. The one I came up with is using makerslide with a 300x400 bed is at: http://www.thingiverse.com/thing:1470253 and with more recent updates at https://github.com/StephenRC/CXY-MSv1
I'm also working on one that uses MGN rails and aluminum u-channel from the hardware store.
-
For advice on building large deltas I suggest asking on the Google delta group at https://groups.google.com/forum/?nomobile=true#!forum/deltabot. I would be comfortable building a delta with a 400mm build diameter and 2040 extrusion, but I wouldn't want to go larger until I have built one that size.
-
Maybe dvmourik can give some tips!
-
Maybe dvmourik can give some tips!
Beautiful!!!!
I love my delta, but my concern is not being able to square the axes adequately, where something cartesian or corexy seems far more straightforward to get the physical alignment dead on, and rely less on software (not that your software doesn't work beautifully dc42). I'm just unsure which way to go. But a 400mm delta is pretty tempting!
-
Much as its not the purists viewpoint (expecting flack for this), the firmware can correct for non-orthogonal axes. Just a thought you might not need to square them that well for a cartesian printer, but I am quite certain for a corexy and especially a delta a good mechanical build will be essential.
The issue is the same in biology and the reason why no 50 foot humans exist. If you are twice as big you are 8 times heavier, so a 40x40cm cartesian printer needs to be 8 times heavier than a 20x20cm one to be as stiff. Some schematics like deltas seem to scale more easily but there is no way that the 3.5m delta shown above, beautiful and show stopping though it is, is as stiff as my 2040 kossel, or similar scaled machine. The verticals would have to be 12.25 times thicker than my 2040's or made from something much stiffer and more exotic, which in the images they are not.
-
Thanks for that DjDemonD, your printer looks really nice. I have a 2020 delta right now, it's quite good, but only around 200mm build diameter. I was considering something like an FT5 or CoreXY because of the light gantry and square build platform, but I do love how easy my delta moves (I need to do a flying extruder, my bowden is almost perfect. But not quite.
The curiosity with corexy being that i can make the frame very rigid and just have something like an e3d titan on a simple rail or pair of them and have similar advantage to the delta. Though the delta is so good looking and the movements are so smooooth. I guess the grass is always greener on the other side, right?
-
I've got both and the delta is a better machine, albeit more a thoroughbred rather than a working horse. The corexy with linear rails and now TMC's isnt far off it, delta's quicker but speed is nice rather than needed for me.
Corexy isn't that simple there's quite a complex belt path and belt geometry has to be exactly right or objects print skewed near the edges of the bed. I tend to think of them almost as a delta with two motors tied together (and a z axis), rather than 3, so simpler to tune and calibrate, but not as simple as straight cartesian machines. But they offer stationary motors, and with lightweight print heads can move very fast also. You'll always be moving more mass than with a delta though, the y slides, x rail and print head are all moving mass.
I mainly use my corexy for printing PETG and ABS with large nozzles and its enclosed (much easier to do this than with a delta).
-
Hi,
I agree with DJDemonD about stiffness. That's why our big printers got 5mm polycarbonate sidepanels mounted on the frame. (bend a polycarbonate frame cold bending and in a good way and it's really strong. only 1 panel costs around 500 euro's and you need 6 in that printer).
With this way we can make a stiffer machine than a 2020 kosel
We have 200mm bed delta's 300mm bed delta's and even 370/400mm delta's as printers too.
The most selling one of the smaller ones is the 370mm bed with a building height of 600mm.All our printers had different tests like a droptest, test to find the natural frequency of the frame, torque tests, serveral deformation tests. We won't share all information as we paid alot to get the tests done but i am sure i can help you out.
PS. If you don't want to use polycarbonate side panels (little expensive for private fun) make a strong cross in the side of the printer with steelwire or similar. If you want i can upload some pictures how to do so.
-
Much as its not the purists viewpoint (expecting flack for this), the firmware can correct for non-orthogonal axes. Just a thought you might not need to square them that well for a cartesian printer, but I am quite certain for a corexy and especially a delta a good mechanical build will be essential.
The issue is the same in biology and the reason why no 50 foot humans exist. If you are twice as big you are 8 times heavier, so a 40x40cm cartesian printer needs to be 8 times heavier than a 20x20cm one to be as stiff. Some schematics like deltas seem to scale more easily but there is no way that the 3.5m delta shown above, beautiful and show stopping though it is, is as stiff as my 2040 kossel, or similar scaled machine. The verticals would have to be 12.25 times thicker than my 2040's or made from something much stiffer and more exotic, which in the images they are not.
Correct on all parts only i am missing one thing. The printhead and mass of the arms/effector isn't same ratio as the frame. So you're mass moving around is less to get a natural frequency in the frame. Inertia is important to calculate when building a frame i think, but maybe i am wrong.
-
I'm going to transfer my parts to a bigger and more rigid kossel. This thing moves WAY too smoothly and awesomely. You guys are right.
