what is the performance? Where's the limit
Posted by TomNagel
|
what is the performance? Where's the limit May 09, 2009 05:02PM |
Registered: 14 years ago Posts: 7 |
I've been reading on the Reprap for a few hours now, and I have not found anything yet on the limits of the Reprap.
* What is the resolution with which the parts are printed?
* How well do the printed parts follow the 3D design?
* What limits the above?
* how fast can the Reprap print?
* and what limits this?
I have the impression that the focus of improvement by the contributors of this project is on simplification and decreasing the costprice, and less on speed and precision. Is that true?
* What is the resolution with which the parts are printed?
* How well do the printed parts follow the 3D design?
* What limits the above?
* how fast can the Reprap print?
* and what limits this?
I have the impression that the focus of improvement by the contributors of this project is on simplification and decreasing the costprice, and less on speed and precision. Is that true?
|
Re: what is the performance? Where's the limit May 09, 2009 06:33PM |
Admin Registered: 17 years ago Posts: 1,915 |
TomNagel Wrote:
-------------------------------------------------------
> I've been reading on the Reprap for a few hours
> now, and I have not found anything yet on the
> limits of the Reprap.
> * What is the resolution with which the parts are
> printed?
~0.1 mm for most machines
> * How well do the printed parts follow the 3D
> design?
depends on how well you design them.
> * What limits the above?
practice
> * how fast can the Reprap print?
depends on the machine. most of the darwins, I hear, manage about 8 mm/sec per axis. Adrian has a darwin going around 30 mm/sec. I mill with tommelise 2 reprap at 8.3 mm/sec and can print at about 12, when I get finished milling a extruder.
> * and what limits this?
>
> I have the impression that the focus of
> improvement by the contributors of this project is
> on simplification and decreasing the costprice,
> and less on speed and precision. Is that true?
yes and no. some people like myself and vik are working on simplification and cost. other people are aiming at higher speed. precision is pretty much constant for the past year or so.
-------------------------------------------------------
Hell, there are no rules here - we're trying to accomplish something.
Opportunity is missed by most people because it is dressed in overalls and looks like work.
Thomas A. Edison
-------------------------------------------------------
> I've been reading on the Reprap for a few hours
> now, and I have not found anything yet on the
> limits of the Reprap.
> * What is the resolution with which the parts are
> printed?
~0.1 mm for most machines
> * How well do the printed parts follow the 3D
> design?
depends on how well you design them.
> * What limits the above?
practice
> * how fast can the Reprap print?
depends on the machine. most of the darwins, I hear, manage about 8 mm/sec per axis. Adrian has a darwin going around 30 mm/sec. I mill with tommelise 2 reprap at 8.3 mm/sec and can print at about 12, when I get finished milling a extruder.
> * and what limits this?
>
> I have the impression that the focus of
> improvement by the contributors of this project is
> on simplification and decreasing the costprice,
> and less on speed and precision. Is that true?
yes and no. some people like myself and vik are working on simplification and cost. other people are aiming at higher speed. precision is pretty much constant for the past year or so.
-------------------------------------------------------
Hell, there are no rules here - we're trying to accomplish something.
Opportunity is missed by most people because it is dressed in overalls and looks like work.
Thomas A. Edison
|
Re: what is the performance? Where's the limit May 10, 2009 05:34AM |
Registered: 14 years ago Posts: 7 |
Forrest Higgs Wrote:
-------------------------------------------------------
> TomNagel Wrote:
> --------------------------------------------------
> -----
> > * How well do the printed parts follow the 3D
> > design?
>
> depends on how well you design them.
>
Can you explain this a little further? How does the design influence the precision with which it is made?
And how about warpage? Is that a problem, or is it negligable?
And another question about the abilities of the printer (I think FDM printers in general). What if the object that I am trying to print has overhanging parts, no matter in what direction i try to print it. Is that simply not possible? Or do I have to design a support structure that is printed with the object, and cut off later?
-------------------------------------------------------
> TomNagel Wrote:
> --------------------------------------------------
> -----
> > * How well do the printed parts follow the 3D
> > design?
>
> depends on how well you design them.
>
Can you explain this a little further? How does the design influence the precision with which it is made?
And how about warpage? Is that a problem, or is it negligable?
And another question about the abilities of the printer (I think FDM printers in general). What if the object that I am trying to print has overhanging parts, no matter in what direction i try to print it. Is that simply not possible? Or do I have to design a support structure that is printed with the object, and cut off later?
|
Re: what is the performance? Where's the limit May 10, 2009 10:14AM |
Admin Registered: 17 years ago Posts: 1,915 |
TomNagel Wrote:
-------------------------------------------------------
> Forrest Higgs Wrote:
> --------------------------------------------------
> -----
> > TomNagel Wrote:
> >
> --------------------------------------------------
>
> > -----
> > > * How well do the printed parts follow the 3D
> > > design?
> >
> > depends on how well you design them.
> >
>
> Can you explain this a little further? How does
> the design influence the precision with which it
> is made?
You have to remember that a printed design consists of both perimeter and infill roads and that the perimeter and infill are done as two separate operations. You also have to remember that when you print a road the extrusion has both a thickness and, more importantly, a width.
Most of the printers, iirc, still have a bit of problem turning the extruder on and off instantaneously at the end of a print road. What that means is that you can wind up with too much or too little plastic where a print road meets the perimeter. As well, if you put a bolt hole, for example, too close to the perimeter of an object you are designing you can have perimeters for the hole and outside of the object overlap. You also have infill roads that are very short between the hole and perimeters. That gets messy. Go through the blogs and look at closeups of printed objects and you will see this problem fairly frequently.
With milled objects you have to remember that your end mill has a diameter. Milling objects with features smaller or in the same range of size as that diameter can get quite tricky. You have to be careful how you design stuff.
> And how about warpage? Is that a problem, or is it
> negligable?
It's definitely not negligible, especially for objects of any size. You need to think about printing in an environment heated to about 70 C to minimize warpage. Metalab and Adrian have come up with ways of doing this without unnecessarily complexifying your printer.
>
> And another question about the abilities of the
> printer (I think FDM printers in general). What if
> the object that I am trying to print has
> overhanging parts, no matter in what direction i
> try to print it. Is that simply not possible? Or
> do I have to design a support structure that is
> printed with the object, and cut off later?
You either design the object in pieces which don't need overhangs {my strategy with milling} or you include support material.
-------------------------------------------------------
Hell, there are no rules here - we're trying to accomplish something.
Opportunity is missed by most people because it is dressed in overalls and looks like work.
Thomas A. Edison
-------------------------------------------------------
> Forrest Higgs Wrote:
> --------------------------------------------------
> -----
> > TomNagel Wrote:
> >
> --------------------------------------------------
>
> > -----
> > > * How well do the printed parts follow the 3D
> > > design?
> >
> > depends on how well you design them.
> >
>
> Can you explain this a little further? How does
> the design influence the precision with which it
> is made?
You have to remember that a printed design consists of both perimeter and infill roads and that the perimeter and infill are done as two separate operations. You also have to remember that when you print a road the extrusion has both a thickness and, more importantly, a width.
Most of the printers, iirc, still have a bit of problem turning the extruder on and off instantaneously at the end of a print road. What that means is that you can wind up with too much or too little plastic where a print road meets the perimeter. As well, if you put a bolt hole, for example, too close to the perimeter of an object you are designing you can have perimeters for the hole and outside of the object overlap. You also have infill roads that are very short between the hole and perimeters. That gets messy. Go through the blogs and look at closeups of printed objects and you will see this problem fairly frequently.
With milled objects you have to remember that your end mill has a diameter. Milling objects with features smaller or in the same range of size as that diameter can get quite tricky. You have to be careful how you design stuff.
> And how about warpage? Is that a problem, or is it
> negligable?
It's definitely not negligible, especially for objects of any size. You need to think about printing in an environment heated to about 70 C to minimize warpage. Metalab and Adrian have come up with ways of doing this without unnecessarily complexifying your printer.
>
> And another question about the abilities of the
> printer (I think FDM printers in general). What if
> the object that I am trying to print has
> overhanging parts, no matter in what direction i
> try to print it. Is that simply not possible? Or
> do I have to design a support structure that is
> printed with the object, and cut off later?
You either design the object in pieces which don't need overhangs {my strategy with milling} or you include support material.
-------------------------------------------------------
Hell, there are no rules here - we're trying to accomplish something.
Opportunity is missed by most people because it is dressed in overalls and looks like work.
Thomas A. Edison
|
Re: what is the performance? Where's the limit May 22, 2009 06:59PM |
> * What is the resolution with which the parts are
> printed?
~0.1 mm for most machines
I've seen this number quoted in a few places, but I'm not sure what it means. I assume it means that the extruder can be positioned to 0.1 mm, not that the plastic laid down is positioned to 0.1 mm. Is that right?
Imagine printing a set of 5 parallel lines.
Can each line be 0.1 mm think, with 0.1 mm of space in between them?
If not, what's the minimum width of a line you can print, and what's the minimum spacing between them?
Thanks.
> printed?
~0.1 mm for most machines
I've seen this number quoted in a few places, but I'm not sure what it means. I assume it means that the extruder can be positioned to 0.1 mm, not that the plastic laid down is positioned to 0.1 mm. Is that right?
Imagine printing a set of 5 parallel lines.
Can each line be 0.1 mm think, with 0.1 mm of space in between them?
If not, what's the minimum width of a line you can print, and what's the minimum spacing between them?
Thanks.
|
Re: what is the performance? Where's the limit May 22, 2009 07:35PM |
Registered: 17 years ago Posts: 161 |
kitep Wrote:
-------------------------------------------------------
> > * What is the resolution with which the parts
> are
> > printed?
>
> ~0.1 mm for most machines
>
> I've seen this number quoted in a few places, but
> I'm not sure what it means. I assume it means
> that the extruder can be positioned to 0.1 mm, not
> that the plastic laid down is positioned to 0.1
> mm. Is that right?
>
> Imagine printing a set of 5 parallel lines.
> Can each line be 0.1 mm think, with 0.1 mm of
> space in between them?
> If not, what's the minimum width of a line you can
> print, and what's the minimum spacing between
> them?
>
> Thanks.
The 0,1mm is the resolution with which the print head can be moved in the X and Y axes. It comes from the circumference of the wheel driving the timing belt divided by the number of steps in a full stepper motor resolution. This gives the distance the print head moves for each step of the motor. For machines using 400 step per revolution motors it is in the region of 0,1mm. For machines using 200 step per rev motors (like mine) you only get ~ 0.2mm. It will vary a bit from the wheel driving the belt too.
I think most people are using 0.5mm nozzles on their extruders, The actual thickness of the line will vary a little depending on how fast it is extruded, and how fast the extruder is moving (X, Y).
When the machines print, they draw an outline first, then fill in the inside, so the outline, which is the important bit will be accurate to 0.1mm assuming the software takes into account filament thickness, etc. A more or less continual 'thread' of filament is used to make objects.
So in theory if you are printing a cube, you should be able to make a 50mm cube, and a 50.1mm cube and be able to measure the difference, assuming the mechanics of the particular machine are up to scratch. In reality, warping will throw it out by more than this.
Part are not printed in dots like on an inkjet printer so there is not really a compatible 'resolution' in that sense. The quality and accuracy of printed parts comes down to how the sotware constructs them, so as the software progresses, so will the quality.
Your example about 0.1mm lines: I doubt very much it would be possible it would be possible to extrude a 0,1mm filament to be able to lay down 0.1mm parallel lines.
Hope this helps.
Please anyone point out and mistakes in my thinking
-------------------------------------------------------
> > * What is the resolution with which the parts
> are
> > printed?
>
> ~0.1 mm for most machines
>
> I've seen this number quoted in a few places, but
> I'm not sure what it means. I assume it means
> that the extruder can be positioned to 0.1 mm, not
> that the plastic laid down is positioned to 0.1
> mm. Is that right?
>
> Imagine printing a set of 5 parallel lines.
> Can each line be 0.1 mm think, with 0.1 mm of
> space in between them?
> If not, what's the minimum width of a line you can
> print, and what's the minimum spacing between
> them?
>
> Thanks.
The 0,1mm is the resolution with which the print head can be moved in the X and Y axes. It comes from the circumference of the wheel driving the timing belt divided by the number of steps in a full stepper motor resolution. This gives the distance the print head moves for each step of the motor. For machines using 400 step per revolution motors it is in the region of 0,1mm. For machines using 200 step per rev motors (like mine) you only get ~ 0.2mm. It will vary a bit from the wheel driving the belt too.
I think most people are using 0.5mm nozzles on their extruders, The actual thickness of the line will vary a little depending on how fast it is extruded, and how fast the extruder is moving (X, Y).
When the machines print, they draw an outline first, then fill in the inside, so the outline, which is the important bit will be accurate to 0.1mm assuming the software takes into account filament thickness, etc. A more or less continual 'thread' of filament is used to make objects.
So in theory if you are printing a cube, you should be able to make a 50mm cube, and a 50.1mm cube and be able to measure the difference, assuming the mechanics of the particular machine are up to scratch. In reality, warping will throw it out by more than this.
Part are not printed in dots like on an inkjet printer so there is not really a compatible 'resolution' in that sense. The quality and accuracy of printed parts comes down to how the sotware constructs them, so as the software progresses, so will the quality.
Your example about 0.1mm lines: I doubt very much it would be possible it would be possible to extrude a 0,1mm filament to be able to lay down 0.1mm parallel lines.
Hope this helps.
Please anyone point out and mistakes in my thinking
|
Re: what is the performance? Where's the limit May 22, 2009 07:56PM |
|
Re: what is the performance? Where's the limit May 24, 2009 05:44AM |
Registered: 14 years ago Posts: 7 |
|
Re: what is the performance? Where's the limit May 24, 2009 07:44AM |
Registered: 15 years ago Posts: 242 |
It depends on hardware as well as software.
If you use an add on module with step and direction inputs, then it probably only has a common enable input, in which case there's no way to apply PWM to the coils individually for proper microstepping. I think pretty much all the PCBs using the L297 logic chip to convert from direction+step to actual stepper enable signals are fixed to a maximum of 8 steps per cycle.
The speed of the microcontroller, and any dedicated PWM hardware on the chip are also influential, as they impact on the free time (or lack of it) to do other important tasks (like motion calculations)
I chose to have more flexibility all round, so when I designed my board, I used a fast ARM chip with lots of PWM channels (4 stepper motor axis with spares !)
and I wired all the stepper control signals to the microcontroller, so that it could handle customised flexibility.
For example, I could write alternate software modules to group the coil drivers in pairs, and use 1.5 bipolar stepper motor channels to drive a 3 phase washing machine motor, use a half channel for DC bidirectional motor drive, and the other two channels for traditional bipolar stepper motor drive.
I'm only in the early stages of my software though...
If you use an add on module with step and direction inputs, then it probably only has a common enable input, in which case there's no way to apply PWM to the coils individually for proper microstepping. I think pretty much all the PCBs using the L297 logic chip to convert from direction+step to actual stepper enable signals are fixed to a maximum of 8 steps per cycle.
The speed of the microcontroller, and any dedicated PWM hardware on the chip are also influential, as they impact on the free time (or lack of it) to do other important tasks (like motion calculations)
I chose to have more flexibility all round, so when I designed my board, I used a fast ARM chip with lots of PWM channels (4 stepper motor axis with spares !)
and I wired all the stepper control signals to the microcontroller, so that it could handle customised flexibility.
For example, I could write alternate software modules to group the coil drivers in pairs, and use 1.5 bipolar stepper motor channels to drive a 3 phase washing machine motor, use a half channel for DC bidirectional motor drive, and the other two channels for traditional bipolar stepper motor drive.
I'm only in the early stages of my software though...
|
Re: what is the performance? Where's the limit June 12, 2009 10:30AM |
Registered: 14 years ago Posts: 1 |
Hi,
First let me tell you that I was happily surprised to find this site, I was googling stepper-schematics for my "poor man's" CNC-router project and... by now I've added a thermoplast extruder to my wish list ;-)
I've got no experience with 3D-printing/materials though and would like to know something about objects made. I'm interested in what to expect when it comes to strength.
For instance, I'm busy creating a panorama Cam, a micro controlled rig that points a standard camera in different directions and... takes shots. It's hard to find the right casing for this project and a reprap seems a nice tool to create it.
Suppose I do, will it be strong enough to survive normal life handling (and maybe a little more...) without being too massive ?
I know a lot will depend on materials used, size, thickness and maybe the way it was produced. As total newbie I hardly know what to expect from "DIY"-plastics, I'm (still) more into wood crafting.
Best wishes,
Bart.
Edited 3 time(s). Last edit at 06/12/2009 11:59AM by SimpsonJr.
First let me tell you that I was happily surprised to find this site, I was googling stepper-schematics for my "poor man's" CNC-router project and... by now I've added a thermoplast extruder to my wish list ;-)
I've got no experience with 3D-printing/materials though and would like to know something about objects made. I'm interested in what to expect when it comes to strength.
For instance, I'm busy creating a panorama Cam, a micro controlled rig that points a standard camera in different directions and... takes shots. It's hard to find the right casing for this project and a reprap seems a nice tool to create it.
Suppose I do, will it be strong enough to survive normal life handling (and maybe a little more...) without being too massive ?
I know a lot will depend on materials used, size, thickness and maybe the way it was produced. As total newbie I hardly know what to expect from "DIY"-plastics, I'm (still) more into wood crafting.
Best wishes,
Bart.
Edited 3 time(s). Last edit at 06/12/2009 11:59AM by SimpsonJr.
|
Re: what is the performance? Where's the limit June 12, 2009 01:51PM |
Admin Registered: 17 years ago Posts: 1,915 |
Here is a selection of things that various reprap machines have printed or milled.
[landv.net]
There have been interesting things happening since April when this selection was printed. If you want to see what a newcomer has been able to print with a BitsFromBytes Darwin, look at the bottom of this photo album.
[gallery.me.com]
He's a beginner, mind.
-------------------------------------------------------
Hell, there are no rules here - we're trying to accomplish something.
Opportunity is missed by most people because it is dressed in overalls and looks like work.
Thomas A. Edison
[landv.net]
There have been interesting things happening since April when this selection was printed. If you want to see what a newcomer has been able to print with a BitsFromBytes Darwin, look at the bottom of this photo album.
[gallery.me.com]
He's a beginner, mind.
-------------------------------------------------------
Hell, there are no rules here - we're trying to accomplish something.
Opportunity is missed by most people because it is dressed in overalls and looks like work.
Thomas A. Edison
Sorry, only registered users may post in this forum.