Want to build new Corexy and so many concerns
Posted by n9jcv
|
Want to build new Corexy and so many concerns February 12, 2017 12:41AM |
So a little background, I have been 3d printing for about 3 years. I first built an I3 aluminum frame from a kit. Then I build a cherry pi delta. When I built the cherry pi delta, I designed a new effector, arms, carriages, partially due to my want and partially due to materials I already had vs what I could not obtain cheaply, etc. The cherry pi turned out great. Next I built a 100% custom I3 from 2020 extrusion. I desiged every part on this machine. I have since sold my delta and aluminum i3, and my custom 2020 i3 is the best so far. I have built a few others, like a small i3, build 6x6x6 for my god son, that was a good project for he and all, all custom.
Now I want to build a corexy. Top contenders for me are Hypercube and FB2020. Don't take the following wrong. The people that designed these machines did an excellent job and I commend them for all their efforts, and sharing with the community. But there are again things I don't like or things, that I would rather change. Some small some large.
I should mention, I do fairly well in Sketchup. But that is about the extent of my 3d modeling skills and I don't have $ for solidworks, autocad stuff like that. Also, I have hand tools, but no metal working tools or experience. So I know some use aluminum exclusively, but for me it is extrusions and plastic parts.
One example, FB2020 seems to use M4x8 screws, well I have about 1000 M5x10 and would rather go that route, but that means I would be modifying all those parts to be a bit thicker and bigger holes, not a biggie right.
Then both the hypercube and fb2020 use cantilevered beds. Now I agree most are dual 12mm rods with a lead screw and at the size and weights of the bed, those 12mm rods will not flex. But to me the design just is bad. I know many others have different opinions. It is like this bike fork, I see this and I say to myself, why would you do that, you are just asking for trouble, yes it may work, but do you really want to do that? You could make it some much stronger by doubling up with the legs of the fork.
So what I want to do with the bed is use 3 8mm rods, 2 in rear and one center front, and leadscrew between 2 in rear for movement. To me just seems better, like a fork with 2 legs seems better to me
Has anyone else actually used this Z design successfully in a configuration like this?
Then some have vertical x carriage and others horizontal. For now I am leaning with a horizontal to maximize Z height. Some have motors in and some out. For me I want the motors in, I would like everything in the frame.
One question I have is can the corexy movement function properly using only 2/3 of the width of the cube. Reason being, use 2/3 for bed and the other 1/3 for eletronics,power supply etc, perhap that electronics section being enclosed to look nice. I think as long as the x and y belts have the same length and route, then it should function properly, it should not matter what size the actual frame is, correct?
So how many others have had these same dilema? At first I thought maybe I could compromise and just go with a published design, but my brain refuses to cooperate, it says I MUST modify certain things lol
Now I want to build a corexy. Top contenders for me are Hypercube and FB2020. Don't take the following wrong. The people that designed these machines did an excellent job and I commend them for all their efforts, and sharing with the community. But there are again things I don't like or things, that I would rather change. Some small some large.
I should mention, I do fairly well in Sketchup. But that is about the extent of my 3d modeling skills and I don't have $ for solidworks, autocad stuff like that. Also, I have hand tools, but no metal working tools or experience. So I know some use aluminum exclusively, but for me it is extrusions and plastic parts.
One example, FB2020 seems to use M4x8 screws, well I have about 1000 M5x10 and would rather go that route, but that means I would be modifying all those parts to be a bit thicker and bigger holes, not a biggie right.
Then both the hypercube and fb2020 use cantilevered beds. Now I agree most are dual 12mm rods with a lead screw and at the size and weights of the bed, those 12mm rods will not flex. But to me the design just is bad. I know many others have different opinions. It is like this bike fork, I see this and I say to myself, why would you do that, you are just asking for trouble, yes it may work, but do you really want to do that? You could make it some much stronger by doubling up with the legs of the fork.
So what I want to do with the bed is use 3 8mm rods, 2 in rear and one center front, and leadscrew between 2 in rear for movement. To me just seems better, like a fork with 2 legs seems better to me
Has anyone else actually used this Z design successfully in a configuration like this?Then some have vertical x carriage and others horizontal. For now I am leaning with a horizontal to maximize Z height. Some have motors in and some out. For me I want the motors in, I would like everything in the frame.
One question I have is can the corexy movement function properly using only 2/3 of the width of the cube. Reason being, use 2/3 for bed and the other 1/3 for eletronics,power supply etc, perhap that electronics section being enclosed to look nice. I think as long as the x and y belts have the same length and route, then it should function properly, it should not matter what size the actual frame is, correct?
So how many others have had these same dilema? At first I thought maybe I could compromise and just go with a published design, but my brain refuses to cooperate, it says I MUST modify certain things lol
|
Re: Want to build new Corexy and so many concerns February 12, 2017 09:00AM |
Three rails for the Z axis is common, but you'll find that having one on the front edge of the bed gets in the way when you want to remove a print or do any work on the extruder. It is more common to put two along the sides near the front and the third rail at the back where it is out of the way. I doesn't really matter if you use three rails, lifting with one screw is going to either bind or bounce. You need a minimum of two, and a third will help. And unless you like releveling your bed frequently, don't even think about driving them with separate motors. Use a closed loop belt to drive as many screws as you decide to use with a single motor. Look at the CoreXYNZ thread for an example of a CoreXY done right.
The size and shape of the bed don't matter, CoreXY will work fine in any square/rectangular configuration as long as the A and B belt paths are identical/mirror images.
Vertical vs horizontal extruder carriage doesn't really matter, if you go horizontal to use less vertical space, then the machine gets bigger horizontally. Most people prefer a smaller footprint and don't mind a machine getting taller. In the end you use whatever fits best with the extruder and hot-end you intend to use.
I wouldn't use anything but linear guides for all three axes. End supported round rails can flex, you have to worry about keeping them parallel (its much easier to align parallel linear guides), and the convenience of screwing things to square, flat bearing blocks can't be denied.
If you want to print ABS the printer will need to be enclosed (and warm). In that case it's a good idea to put the motors and electronics outside the enclosure. Many of the CoreXY designs you see are built without such consideration, using the frame as part of the linear motion assembly, with pulleys, belts, etc. extending out beyond the edges of the frame. If they ever want to enclose such a machine for printing ABS, they're going to have to build a big box around the machine, and then figure out what to do with the electronics. A little advance planning will make it easy to use the frame of the machine to enclose the printer. See: [www.instructables.com]
I am building a CoreXY machine of my own, constantly changing design. It will use belts to lift the Z axis and be enclosed for printing ABS. Build capacity will be 300 x 300 x700 mm. The frame is made from 40 x 40 mm slot extrusion, all linear motion is guided using linear guides. X and Y motors will be outside the enclosure as will the electronics. That one is here: [forums.reprap.org]
Whatever you do, avoid the use of plastic, especially printed plastic parts in your printer. You don't want to end up with something that looks like this (check out the upper right front corner):
And if your frame needs this level of bracing, you need to use larger cross-section frame members:
This is getting a little excessive, too (do you think those dozens of little corner braces are really adding anything to the stiffness of this thing?):
These images came from a web site that had compiled hundreds of such images of coreXY machines, but I have lost the link. Look at other designs and be critical. Then you can design your own machine avoiding other people's mistakes while copying their good ideas.
Edited 6 time(s). Last edit at 02/12/2017 12:50PM by the_digital_dentist.
Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]
The size and shape of the bed don't matter, CoreXY will work fine in any square/rectangular configuration as long as the A and B belt paths are identical/mirror images.
Vertical vs horizontal extruder carriage doesn't really matter, if you go horizontal to use less vertical space, then the machine gets bigger horizontally. Most people prefer a smaller footprint and don't mind a machine getting taller. In the end you use whatever fits best with the extruder and hot-end you intend to use.
I wouldn't use anything but linear guides for all three axes. End supported round rails can flex, you have to worry about keeping them parallel (its much easier to align parallel linear guides), and the convenience of screwing things to square, flat bearing blocks can't be denied.
If you want to print ABS the printer will need to be enclosed (and warm). In that case it's a good idea to put the motors and electronics outside the enclosure. Many of the CoreXY designs you see are built without such consideration, using the frame as part of the linear motion assembly, with pulleys, belts, etc. extending out beyond the edges of the frame. If they ever want to enclose such a machine for printing ABS, they're going to have to build a big box around the machine, and then figure out what to do with the electronics. A little advance planning will make it easy to use the frame of the machine to enclose the printer. See: [www.instructables.com]
I am building a CoreXY machine of my own, constantly changing design. It will use belts to lift the Z axis and be enclosed for printing ABS. Build capacity will be 300 x 300 x700 mm. The frame is made from 40 x 40 mm slot extrusion, all linear motion is guided using linear guides. X and Y motors will be outside the enclosure as will the electronics. That one is here: [forums.reprap.org]
Whatever you do, avoid the use of plastic, especially printed plastic parts in your printer. You don't want to end up with something that looks like this (check out the upper right front corner):
And if your frame needs this level of bracing, you need to use larger cross-section frame members:
This is getting a little excessive, too (do you think those dozens of little corner braces are really adding anything to the stiffness of this thing?):
These images came from a web site that had compiled hundreds of such images of coreXY machines, but I have lost the link. Look at other designs and be critical. Then you can design your own machine avoiding other people's mistakes while copying their good ideas.
Edited 6 time(s). Last edit at 02/12/2017 12:50PM by the_digital_dentist.
Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]
|
Re: Want to build new Corexy and so many concerns February 12, 2017 05:06PM |
Sketchup will allow you to design your printer, and even design for 3D printing but its not a parametric CAD program and you will be banging your head against the wall when you struggle to design even the simplest of models. Curves are impossible (or at least bloody annoying) and anything even remotely more precise then a mm will cause weird manifold issues in your build where sketchup tries to 'guess' what length something should really be.
Fusion360 is essentially free. Not bad for a free program either, however weird shortcuts and having to manually click tools and manually orbit makes me pull my hair out sometimes.
Has anyone tried that EAA deal!? Especially interested to see if anyone from the UK has tried to sign up. $40 is insanely cheap for Solidworks.
Also, Linear guides are not the only option. Check out V-slot. It works out about the same price as linear guides, but allows for the flexibility that extrusion brings to your design. Here is a close up of the layer alignment I get at 300 microns. I can repeat this across 200 or even 100 micron layers. Not tried 50 yet, but I rarely use it.
Also, I believe cantilevered beds can work and have the advantage of being a nice simple design, obviously this may have some compromises when you have to over engineer the design to avoid the issues Digital Dentist has already mentioned.
I would avoid plastic parts as much as possible. They work, but come with more compromises then advantages. Make your Prototype in plastic and move onto metal as soon as possible.
Fusion360 is essentially free. Not bad for a free program either, however weird shortcuts and having to manually click tools and manually orbit makes me pull my hair out sometimes.
Has anyone tried that EAA deal!? Especially interested to see if anyone from the UK has tried to sign up. $40 is insanely cheap for Solidworks.
Also, Linear guides are not the only option. Check out V-slot. It works out about the same price as linear guides, but allows for the flexibility that extrusion brings to your design. Here is a close up of the layer alignment I get at 300 microns. I can repeat this across 200 or even 100 micron layers. Not tried 50 yet, but I rarely use it.
Also, I believe cantilevered beds can work and have the advantage of being a nice simple design, obviously this may have some compromises when you have to over engineer the design to avoid the issues Digital Dentist has already mentioned.
I would avoid plastic parts as much as possible. They work, but come with more compromises then advantages. Make your Prototype in plastic and move onto metal as soon as possible.
Sorry, only registered users may post in this forum.