Is a dual support Z on the diagonal corners of bed a bad idea?
Posted by realthor
|
Is a dual support Z on the diagonal corners of bed a bad idea? November 18, 2015 11:42AM |
Hi,
I can see more and more dual-rod supported beds that use 2 faces of the frame box, most likely the lateral ones. But there is a busy space as the Y linear rods/carriages endstop/etc would limit the available Z height that can be fitted under them. On the other hand, having dual rod supported Z by moving one rod from the back to front would limit the access to the print, another no go.
I am thinking why not placing the smooth rods for the Z on the opposing corners of the box. One can easily re-print the corner brackets with the necessary configuration as to accept the smooth rod. I have a feeling of unstable bed due to the fact that the other diagonal is somewhat cantilevered but I don't think this is so much trouble with the specific application of 3D printing. Morevover the corner brackets can be made sturdy enough to compensate for the tilting forces.
I haven't found any such design yet so I quickly sketched it to exemplify:
What do you think?
RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff
I can see more and more dual-rod supported beds that use 2 faces of the frame box, most likely the lateral ones. But there is a busy space as the Y linear rods/carriages endstop/etc would limit the available Z height that can be fitted under them. On the other hand, having dual rod supported Z by moving one rod from the back to front would limit the access to the print, another no go.
I am thinking why not placing the smooth rods for the Z on the opposing corners of the box. One can easily re-print the corner brackets with the necessary configuration as to accept the smooth rod. I have a feeling of unstable bed due to the fact that the other diagonal is somewhat cantilevered but I don't think this is so much trouble with the specific application of 3D printing. Morevover the corner brackets can be made sturdy enough to compensate for the tilting forces.
I haven't found any such design yet so I quickly sketched it to exemplify:
What do you think?
RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff
|
Re: Is a dual support Z on the diagonal corners of bed a bad idea? November 19, 2015 03:26PM |
I am not against leadscrews by all means. They are the norm with respectable 3D printers.
Regarding torque I agree, the drill bit has a very long helix (large pitch) that would require higher torque per motor step compared to the lead screw. I can't say numbers this would have to be tried or if anyone knows how to estimate that (divide the pitches maybe -the pitch of a masonry bit is 10mm or more, while the leadscrew pitch is 1.5 or 2mm I guess).
Also the sharp edges do not exist on a masonry drill bit. On the contrary they are as smooth as the leadscrew edges. The finish is the only one that has me concerned and the lack of a corresponding nut, so a delrin or PFTE or Nylon one must be DIY-ed.
Anyways, this thread is not about leadscrews vs threaded rod vs drill bits, the discussion got sidetracked
. This is about placing the bed at corners instead of front-back. Actually there is no actuation means in the sketch I provided. I was in fact going for a belt system similar to the one created by the (now famous) AndreasL . It seems that his design is discussed all over these forums.
------------------------------------------------------------------
Now this discussion is not about his design either, but about the corner arrangement and how (if) it can be designed to avoid tilt/etc.
My CoreXY concept currently looks like this:
The green is the usable printing area. I have a concern with the increased distance between the sliding corner brackets. The further they are apart the more unstable this assembly gets.
I will choose a lifting mechanism after I decide if this design is feasible.
Edited 3 time(s). Last edit at 11/19/2015 07:07PM by realthor.
RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff
Regarding torque I agree, the drill bit has a very long helix (large pitch) that would require higher torque per motor step compared to the lead screw. I can't say numbers this would have to be tried or if anyone knows how to estimate that (divide the pitches maybe -the pitch of a masonry bit is 10mm or more, while the leadscrew pitch is 1.5 or 2mm I guess).
Also the sharp edges do not exist on a masonry drill bit. On the contrary they are as smooth as the leadscrew edges. The finish is the only one that has me concerned and the lack of a corresponding nut, so a delrin or PFTE or Nylon one must be DIY-ed.
Anyways, this thread is not about leadscrews vs threaded rod vs drill bits, the discussion got sidetracked
. This is about placing the bed at corners instead of front-back. Actually there is no actuation means in the sketch I provided. I was in fact going for a belt system similar to the one created by the (now famous) AndreasL . It seems that his design is discussed all over these forums.------------------------------------------------------------------
Now this discussion is not about his design either, but about the corner arrangement and how (if) it can be designed to avoid tilt/etc.
My CoreXY concept currently looks like this:
The green is the usable printing area. I have a concern with the increased distance between the sliding corner brackets. The further they are apart the more unstable this assembly gets.
I will choose a lifting mechanism after I decide if this design is feasible.
Edited 3 time(s). Last edit at 11/19/2015 07:07PM by realthor.
RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff
|
Re: Is a dual support Z on the diagonal corners of bed a bad idea? November 27, 2015 07:06AM |
Hi guys,
Two guides on the same plan is relatively easy.
Two guides on two parallel plans is more difficult to manage.
Mecanicaly speaking, it's still two guides wherever they are. Guidance cannot be better.
I'm not skilled enough to argue deeply about that configuration in general,
but on simpler systems all is usualy about the 1X 2X rule.
Distance of the force applied should not be above 2 times the bearings distance.
If you've got a diagonal distance of 300mm between rails, you should have at last 150mm between the two bearings on your guide.
But, what if the force (vibrations) is applied at the unguided corners ? This should increase drasticaly the distance.
If a plate is guided on oposite/diagonal sides, you create a virtualy a pivot point and maximise overhang on the perpendicular diagonal ends.
You probably need to install longer supports for these far corners.
On a single plan system (two guides on one side), You don't need to put the guides on the corners
To guide a 200mm plate, you can install the guides at 120 mm distance only.
Because most of the time, to deal with forces/vibrations on the overangs the two guides are working in push-pull.
The closer one on push and the distant one on pull. If your support on the oposite parallel edge is stiff enough,
you get a very stable system from one single plan. And it's the simplest to build.
On repraps we often have 1X 4X (or more), because we want compact parts,
so there's a temptation to get a more stable system guiding by two sides.
Because it reduced the overhang at 0 on the face to face line.
But at the same time you still have to deal with overhangs on the unsuported sides.
You just displace the triangle rule. And the bearings won't work in push-pull anymore.
You still need support at last 1X 2X from each guide to the edges.
Otherhand, as your guide is perpendicular to the centre of your plate you divide overhang by two but in both sides.
That's still a good point I gess. For a 200mm square plate you only need to distance your bearing of 50mm,
that's not bad if you want to get a compact system, or maximize the stroke along your gide.
Guiding from oposite sides is not a bad idea, even if it's more difficult to set-up perfeclty parallel plans.
But guiding for oposite corners increase the 2X distance, so you will need to increase your 1X distance too
and also use longer supports to hold the overangs.
If you install your corner guides on two parallel plans it's pretty the same for setup, but with a longer distance.
Precision upon longer distance is always an issue.
Thinking about it, to me guiding from 2 oposite sides is better. You don't want to install your guides on corners.
++JM
Edited 1 time(s). Last edit at 11/27/2015 07:07AM by J-Max.
Two guides on the same plan is relatively easy.
Two guides on two parallel plans is more difficult to manage.
Mecanicaly speaking, it's still two guides wherever they are. Guidance cannot be better.
I'm not skilled enough to argue deeply about that configuration in general,
but on simpler systems all is usualy about the 1X 2X rule.
Distance of the force applied should not be above 2 times the bearings distance.
If you've got a diagonal distance of 300mm between rails, you should have at last 150mm between the two bearings on your guide.
But, what if the force (vibrations) is applied at the unguided corners ? This should increase drasticaly the distance.
If a plate is guided on oposite/diagonal sides, you create a virtualy a pivot point and maximise overhang on the perpendicular diagonal ends.
You probably need to install longer supports for these far corners.
On a single plan system (two guides on one side), You don't need to put the guides on the corners
To guide a 200mm plate, you can install the guides at 120 mm distance only.
Because most of the time, to deal with forces/vibrations on the overangs the two guides are working in push-pull.
The closer one on push and the distant one on pull. If your support on the oposite parallel edge is stiff enough,
you get a very stable system from one single plan. And it's the simplest to build.
On repraps we often have 1X 4X (or more), because we want compact parts,
so there's a temptation to get a more stable system guiding by two sides.
Because it reduced the overhang at 0 on the face to face line.
But at the same time you still have to deal with overhangs on the unsuported sides.
You just displace the triangle rule. And the bearings won't work in push-pull anymore.
You still need support at last 1X 2X from each guide to the edges.
Otherhand, as your guide is perpendicular to the centre of your plate you divide overhang by two but in both sides.
That's still a good point I gess. For a 200mm square plate you only need to distance your bearing of 50mm,
that's not bad if you want to get a compact system, or maximize the stroke along your gide.
Guiding from oposite sides is not a bad idea, even if it's more difficult to set-up perfeclty parallel plans.
But guiding for oposite corners increase the 2X distance, so you will need to increase your 1X distance too
and also use longer supports to hold the overangs.
If you install your corner guides on two parallel plans it's pretty the same for setup, but with a longer distance.
Precision upon longer distance is always an issue.
Thinking about it, to me guiding from 2 oposite sides is better. You don't want to install your guides on corners.
++JM
Edited 1 time(s). Last edit at 11/27/2015 07:07AM by J-Max.
|
Re: Is a dual support Z on the diagonal corners of bed a bad idea? November 27, 2015 08:57AM |
We are talking 2 kind of forces that act on a bed:
1) the weight of the printed part
2) the forces resulted from moving the bed, due to inertia
Forces 2) do not influence the print, as there is no printing during the Z lifts (at least not yet, not with the current firmwares, but that is something that is pioneered in wall construction, interesting for a side note).
Forces 1) what weights are talking about? hundreds of grams? Compare that with a heated bed. Is it really that serious of a problem? Printed models can be symmetrically positioned on the bed as to avoid any 30 grams of the printed component to tilt the bed on the supported corners axis.
I am thinking that a 2-corners bed lift would later on allow for a 3rd corner addition, which would make it a 3 points plane. I don't think assymetrical 3 points plane is any less effective as a symmetrical 3 points plane.
What are "F" forces in your drawing that would be so great as to affect a diagonal supported bed?
Because I don't even see it as a problem with cantilevered beds. There is no real Force there originated in the print's weight but with cantilevered beds the own bed weight is a problem and the flapping and vibrations that are amplified at fast infills towards the margins of the bed. With two sides supported the shaking of the bed is not amplified towards the supported ends but it is absorbed. The only shaking/flapping effect is limited to the unsupported corners/sides.
I see the situation like this:
The best option is of course 3 rods or even four rods, 2 an 2 on opposite sides.
Edited 3 time(s). Last edit at 11/27/2015 09:08AM by realthor.
RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff
1) the weight of the printed part
2) the forces resulted from moving the bed, due to inertia
Forces 2) do not influence the print, as there is no printing during the Z lifts (at least not yet, not with the current firmwares, but that is something that is pioneered in wall construction, interesting for a side note).
Forces 1) what weights are talking about? hundreds of grams? Compare that with a heated bed. Is it really that serious of a problem? Printed models can be symmetrically positioned on the bed as to avoid any 30 grams of the printed component to tilt the bed on the supported corners axis.
I am thinking that a 2-corners bed lift would later on allow for a 3rd corner addition, which would make it a 3 points plane. I don't think assymetrical 3 points plane is any less effective as a symmetrical 3 points plane.
What are "F" forces in your drawing that would be so great as to affect a diagonal supported bed?
Because I don't even see it as a problem with cantilevered beds. There is no real Force there originated in the print's weight but with cantilevered beds the own bed weight is a problem and the flapping and vibrations that are amplified at fast infills towards the margins of the bed. With two sides supported the shaking of the bed is not amplified towards the supported ends but it is absorbed. The only shaking/flapping effect is limited to the unsupported corners/sides.
I see the situation like this:
The best option is of course 3 rods or even four rods, 2 an 2 on opposite sides.
Edited 3 time(s). Last edit at 11/27/2015 09:08AM by realthor.
RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff
Sorry, only registered users may post in this forum.