Would this work?
Posted by Dave_robinson_022
|
Re: Would this work? April 24, 2015 10:22AM |
First off, I don't understand the red and blue lines in the lower left corner. It looks like you want the X-Y belts to also run in the Z axis - But at the same time it is my understanding that you want two Z axis rods to double as linear guides and transfer rotation to the X-Y gantry?
Assuming you want to make a normal system but with the steppers transferring through these linear rods -
I don't think it will work.
Look at your rectangular rod which is supposed to turn and transfer rotational movement to the outer cape. What is preventing your rectangular rod from just rotating? The friction force from the rod turning and pushing the steel ball to one side and thus creating a lock will maybe be enough but I am not so sure. It will anyway require some steep tolerances to avoid that you over time wear down the balls - Especially if you then afterwards have movement in the z direction and the ball does not start to turn - That will make a flat point on the ball.
The idea certainly has merits. You could use a standard spline and grease that heavily and let a bronze bushing run on the outer diameter. It will create some more friction for the Z-axis to overcome and you will maybe end up needing two spindles drive to avoid skewing and thus locking - But should work.
I made a quick sketch:
A spline intermeshing looks like this:
It is important to understand that when the Z-axis moves up and down this connection will have to slide by the grease / lubrication that would need to be applied. For this reason it could be necessary to encapsulate the entire thing to avoid plastic, hairspray and what not to stick to the spline.
Edited 1 time(s). Last edit at 04/24/2015 10:30AM by LarsK.
Assuming you want to make a normal system but with the steppers transferring through these linear rods -
I don't think it will work.
Look at your rectangular rod which is supposed to turn and transfer rotational movement to the outer cape. What is preventing your rectangular rod from just rotating? The friction force from the rod turning and pushing the steel ball to one side and thus creating a lock will maybe be enough but I am not so sure. It will anyway require some steep tolerances to avoid that you over time wear down the balls - Especially if you then afterwards have movement in the z direction and the ball does not start to turn - That will make a flat point on the ball.
The idea certainly has merits. You could use a standard spline and grease that heavily and let a bronze bushing run on the outer diameter. It will create some more friction for the Z-axis to overcome and you will maybe end up needing two spindles drive to avoid skewing and thus locking - But should work.
I made a quick sketch:
A spline intermeshing looks like this:
It is important to understand that when the Z-axis moves up and down this connection will have to slide by the grease / lubrication that would need to be applied. For this reason it could be necessary to encapsulate the entire thing to avoid plastic, hairspray and what not to stick to the spline.
Edited 1 time(s). Last edit at 04/24/2015 10:30AM by LarsK.
|
Re: Would this work? April 25, 2015 06:23PM |
Apologies for the lack of clarity on my drawings. You are correct that I want to transfer the rotary motion through the rear supporting rods and your spline idea is what I was trying to get at. The second diagram is a little confusing; basically I was trying to implement a smaller idea:
to the Z-axis. This may be over engineering the solution, but I notice that most Z-Axis ate fixed and moved along the rear plane... I thought that if I could balance the control by trying to apply the above diagram to either side then this may help with levelling? The other option i considered was using the exact plan above mired on the left and right side, driven by an axel to operate both simultaneously?
What I am trying to achieve overall is to keep the gantry as light as possible as in a delta style by removing the stepper motors from them but keeping the control as direct as possible and removing the complications of delta math from the situation. Is there anything else like this?
to the Z-axis. This may be over engineering the solution, but I notice that most Z-Axis ate fixed and moved along the rear plane... I thought that if I could balance the control by trying to apply the above diagram to either side then this may help with levelling? The other option i considered was using the exact plan above mired on the left and right side, driven by an axel to operate both simultaneously?
What I am trying to achieve overall is to keep the gantry as light as possible as in a delta style by removing the stepper motors from them but keeping the control as direct as possible and removing the complications of delta math from the situation. Is there anything else like this?
|
Re: Would this work? April 28, 2015 10:42AM |
Hello Dave,
First off, I agree with the others. The concept with two motors will complicated and the gains rather limited. If you want to move the gantry then sticking to the leadscrews moved by one motor is better. But for the gantry movement motors; You should remember that the motors are not really that heavy and the Z axis do not move that often. Mass reduction versus loss of stability is not black and white.
I wanted to show you something that I have been working on which is sort of related. Here I am moving just the nozzle on the carriage. This means I have to move this stepper motor all the time and thus I will lose speed. But it is more compact then having a moving bed (and in some aspects simpler). In this setup it allows me to print to a max 50mm height but I have found the vast majority of my parts to be in that range. But it is still in early design phase. My current main issue is resolution on the z axis. When driving without gears (directly on the stepper) it will move the nozzle too much.
This was particularly bad on my first design and in this design I would also have the problem that the movement would not be entirely linear and I would maybe need software for that.
This is the current design:
First off, I agree with the others. The concept with two motors will complicated and the gains rather limited. If you want to move the gantry then sticking to the leadscrews moved by one motor is better. But for the gantry movement motors; You should remember that the motors are not really that heavy and the Z axis do not move that often. Mass reduction versus loss of stability is not black and white.
I wanted to show you something that I have been working on which is sort of related. Here I am moving just the nozzle on the carriage. This means I have to move this stepper motor all the time and thus I will lose speed. But it is more compact then having a moving bed (and in some aspects simpler). In this setup it allows me to print to a max 50mm height but I have found the vast majority of my parts to be in that range. But it is still in early design phase. My current main issue is resolution on the z axis. When driving without gears (directly on the stepper) it will move the nozzle too much.
This was particularly bad on my first design and in this design I would also have the problem that the movement would not be entirely linear and I would maybe need software for that.
This is the current design:
Sorry, only registered users may post in this forum.