High Speed Printing
Posted by Bill Clark
|
High Speed Printing February 14, 2014 05:58PM |
Registered: 10 years ago Posts: 239 |
I run my CNC mills at 500 "/min and thought that was fast. Being new to 3DP one of my first observations was how painfully slow most run. Then I found this [www.youtube.com] Is this typical for advanced users? What kind of hardware/software is required to do this? Any other examples out there and any input would be appreciated. This is surely the direction I am headed
|
Re: High Speed Printing February 14, 2014 07:08PM |
Registered: 10 years ago Posts: 57 |
If you notice, the print looks terrible at that speed. Also, the entire frame is wobbling back and forth. You can certainly make your printer move that fast by jacking up the current and increasing the acceleration, but you will not achieve acceptable print quality at those speeds. The plastic simply won't melt fast enough. You'll need a material with much lower thermal mass, or a completely different technology to build that quickly...
|
Re: High Speed Printing February 14, 2014 07:16PM |
Registered: 10 years ago Posts: 38 |
I've ran my prusa i3 that fast before and it was unsettling. The thing was flopping all over and threatened to thrash itself to bits. Im sure I took a year off its life. The print quality was better than I expected for how much it was wobbling, but layer adhesion sucked. I need strong parts, and parts printed at that speed aren't. Maybe if you made a 3d printer out of a rigid cnc machine, you could make shelf-bound knick knacks really really fast (relative) but don't hope to make anything functional.
|
Re: High Speed Printing February 14, 2014 07:20PM |
Registered: 10 years ago Posts: 73 |
You'll want to look at this, too: [www.youtube.com]
The developer, Alden Hart, spoke at the local hackerspace recently and said that the Ultimaker was sitting on a hard tabletop and didn't move or wobble at all.
It was all in the Tiny G controller, and how it handled jerk and acceleration.
Way cool.
The developer, Alden Hart, spoke at the local hackerspace recently and said that the Ultimaker was sitting on a hard tabletop and didn't move or wobble at all.
It was all in the Tiny G controller, and how it handled jerk and acceleration.
Way cool.
|
Re: High Speed Printing February 14, 2014 11:20PM |
Registered: 10 years ago Posts: 239 |
|
Re: High Speed Printing February 15, 2014 12:46AM |
Registered: 13 years ago Posts: 1,797 |
|
Anonymous User
Re: High Speed Printing February 15, 2014 12:53AM |
|
Re: High Speed Printing February 15, 2014 02:39AM |
Registered: 10 years ago Posts: 474 |
Bill I don't think you'd be happy with the results of anything printing like that I know you come from machinist background and the prints at that rate of speed are terrible at the moment plastic filament can not be laid down that fast with good results. Plus the threaded rod machine you can see all of its shortcomings flexing all over the place terrible. 3-D printers were originally designed for one up prototypes the guy working with tiny G is probably working with the accelerations the problem with that being from what I've found is so far is you need to have good fast acceleration and deceleration for prints if you reduce this you can achieve that speed but the print is not going to be good notice he's not printing
Edited 1 time(s). Last edit at 02/15/2014 02:47AM by cnc dick.
Edited 1 time(s). Last edit at 02/15/2014 02:47AM by cnc dick.
|
Re: High Speed Printing February 15, 2014 08:12AM |
Registered: 10 years ago Posts: 67 |
But, high speed is achievable:
http://www.youtube.com/watch?v=zhV2jwhb9n8
"Never argue with stupid people, they will drag you down to their level and then beat you with experience."
http://www.youtube.com/watch?v=zhV2jwhb9n8
"Never argue with stupid people, they will drag you down to their level and then beat you with experience."
|
Re: High Speed Printing February 15, 2014 09:19AM |
Registered: 12 years ago Posts: 1,236 |
TinyG uses a slightly faster CPU than typical AVR Arduinos, but the clever bit is in the firmware which implements constant jerk motion, all other firmwares do constant acceleration with some hacks for jerk. In principle the same algorithms from TinyG could be applied elsewhere, it's open source. The TinyG uses 2.5A TI DRV8818 similar to DRV8825.
The printer in the video of the OP is using LinuxCNC with Gecko 540 drivers, otherwise a fairly standard Prusa Mendel, Wade extruder, unknown hot end.
So all the ingredients for fast printing are there, it's just a question of optimizing performance and integrating them all into one platform. You wouldn't need to spend a lot of money on a mammoth printer to do it either.
What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license
The printer in the video of the OP is using LinuxCNC with Gecko 540 drivers, otherwise a fairly standard Prusa Mendel, Wade extruder, unknown hot end.
So all the ingredients for fast printing are there, it's just a question of optimizing performance and integrating them all into one platform. You wouldn't need to spend a lot of money on a mammoth printer to do it either.
What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license
|
Re: High Speed Printing February 15, 2014 11:50AM |
Registered: 10 years ago Posts: 474 |
Bill I guess it all depends on the quality that you want or need like that YouTube video at 750 mm second he doesn't give you a close-up of the part he printed at that speed. We all know moving things around at that speed is possible printing nicely is another story I will will keep an eye out on that tiny G and see how that turns out
Edited 1 time(s). Last edit at 02/15/2014 12:21PM by cnc dick.
Edited 1 time(s). Last edit at 02/15/2014 12:21PM by cnc dick.
|
Re: High Speed Printing February 15, 2014 03:55PM |
Registered: 10 years ago Posts: 73 |
I don't care that much about crazy-fast speed; it was the smoothness of direction changes that got me.
With the TinyG, it looks like infills of narrow parts would be very smooth and create less vibration and stress to the machine.
That should contribute to better quality parts, even without uber-speed, no?
With the TinyG, it looks like infills of narrow parts would be very smooth and create less vibration and stress to the machine.
That should contribute to better quality parts, even without uber-speed, no?
|
Re: High Speed Printing February 15, 2014 04:48PM |
Registered: 13 years ago Posts: 2,947 |
Quote
DonaldJ
I don't care that much about crazy-fast speed; it was the smoothness of direction changes that got me.
With the TinyG, it looks like infills of narrow parts would be very smooth and create less vibration and stress to the machine.
That should contribute to better quality parts, even without uber-speed, no?
That is all about the Ultimakers mechanism. Here is a video of Tantillus (same type of mechanism) doing rapid infill running marlin. [youtu.be]
Although it looks like it is doing slow straight infill it is actually doing tiny little zigzags from 80 percent hex infill.
9000mm/s^2 max acceleration 3000mm/s^2 default acceleration 20mm/s jerk 45mm/s perimeter 90mm/s infill. 300mm/s travel.
| FFF Settings Calculator | Gcode post processors | Geometric Object Deposition Tool Blog |
| Tantillus.org | Mini Printable Lathe | How NOT to install a Pololu driver |
|
Re: High Speed Printing February 17, 2014 07:23AM |
Registered: 10 years ago Posts: 67 |
That link where he is printing 750mm/s his print got trashed due to the controller not keeping up. But here is a link to another print he did at 500mm/s with settings and finished product:
http://ichibey.exblog.jp/18401224/
You will have to translate as it is in Japanese. But his blog is worth it. This guy has done some interesting experiments in printing including designing his own HBOT and delta.
"Never argue with stupid people, they will drag you down to their level and then beat you with experience."
http://ichibey.exblog.jp/18401224/
You will have to translate as it is in Japanese. But his blog is worth it. This guy has done some interesting experiments in printing including designing his own HBOT and delta.
"Never argue with stupid people, they will drag you down to their level and then beat you with experience."
|
Re: High Speed Printing February 17, 2014 11:39AM |
Registered: 10 years ago Posts: 474 |
Bill I guess it's all in what you want or need but for me personally that print that he showed printed at 500 mm a second is not up to my standardsQuote
cnc dick
Bill I guess it all depends on the quality that you want or need like that YouTube video at 750 mm second he doesn't give you a close-up of the part he printed at that speed. We all know moving things around at that speed is possible printing nicely is another story I will will keep an eye out on that tiny G and see how that turns out
|
Re: High Speed Printing February 21, 2014 11:12PM |
Registered: 10 years ago Posts: 239 |
Yes, I'm starting to get my head around this. I do believe we will migrate towards faster and faster print speeds. Seems like a natural progression. I have absolutely been swimming in all this stuff trying to get up to speed. Very excitingQuote
cnc dick
Bill I don't think you'd be happy with the results of anything printing like that I know you come from machinist background and the prints at that rate of speed are terrible at the moment plastic filament can not be laid down that fast with good results. Plus the threaded rod machine you can see all of its shortcomings flexing all over the place terrible. 3-D printers were originally designed for one up prototypes the guy working with tiny G is probably working with the accelerations the problem with that being from what I've found is so far is you need to have good fast acceleration and deceleration for prints if you reduce this you can achieve that speed but the print is not going to be good notice he's not printing
|
Re: High Speed Printing February 21, 2014 11:17PM |
Registered: 10 years ago Posts: 239 |
Synthetos (TinyG) are very hard at work pushing the envelope (mainly with milling/routers) but have a new version coming out (the V9) which is going to be very powerful/versatile. Riley didn't give me specifics but when I presented my desire to build a machine that prints and mills with more than just XYZ he said this would be the ticket. Thanks again for the link. It put me on the path I wanted to go onQuote
DonaldJ
You'll want to look at this, too: [www.youtube.com]
The developer, Alden Hart, spoke at the local hackerspace recently and said that the Ultimaker was sitting on a hard tabletop and didn't move or wobble at all.
It was all in the Tiny G controller, and how it handled jerk and acceleration.
Way cool.
|
Re: High Speed Printing February 22, 2014 12:36AM |
Registered: 10 years ago Posts: 1,381 |
TinyG looks like it will be good for a Delta printer.
Has Marlin firmware been loaded onto a TinyG, any programmers working on this?
Will a Smoothieboard or RAMPS-FD have this speed capability with Marlin firmware?
TinyG
(Atmel ATxmega192)
$129.99
Constant jerk acceleration planning (3rd order S curves) for smooth and fast motion transitions.
Very smooth step pulse generation using phase-optimized fractional-step DDA running at 50 Khz with very low jitter.
[synthetos.myshopify.com]
Smoothieboard
(NXP LPC 1768/9 32-bits Cortex-M3), and/or (LPC 1114 ARM Cortex-M0)
$169.97
[shop.uberclock.com]
RAMPS-FD
(Atmel SAM3X8E ARM Cortex-M3 CPU)
$49.95
RAMPS For Arduino Due
84 MHz Atmel 32-bit ARM Cortex-M3 CPU with USB OTG
[arduino.cc]
[arduino.cc]
[reprap.org]
Has Marlin firmware been loaded onto a TinyG, any programmers working on this?
Will a Smoothieboard or RAMPS-FD have this speed capability with Marlin firmware?
TinyG
(Atmel ATxmega192)
$129.99
Constant jerk acceleration planning (3rd order S curves) for smooth and fast motion transitions.
Very smooth step pulse generation using phase-optimized fractional-step DDA running at 50 Khz with very low jitter.
[synthetos.myshopify.com]
Smoothieboard
(NXP LPC 1768/9 32-bits Cortex-M3), and/or (LPC 1114 ARM Cortex-M0)
$169.97
[shop.uberclock.com]
RAMPS-FD
(Atmel SAM3X8E ARM Cortex-M3 CPU)
$49.95
RAMPS For Arduino Due
84 MHz Atmel 32-bit ARM Cortex-M3 CPU with USB OTG
[arduino.cc]
[arduino.cc]
[reprap.org]
|
Re: High Speed Printing February 22, 2014 02:57AM |
Registered: 11 years ago Posts: 661 |
Quote
Bill Clark
Yes, I'm starting to get my head around this. I do believe we will migrate towards faster and faster print speeds. Seems like a natural progression. I have absolutely been swimming in all this stuff trying to get up to speed. Very excitingQuote
cnc dick
Bill I don't think you'd be happy with the results of anything printing like that I know you come from machinist background and the prints at that rate of speed are terrible at the moment plastic filament can not be laid down that fast with good results. Plus the threaded rod machine you can see all of its shortcomings flexing all over the place terrible. 3-D printers were originally designed for one up prototypes the guy working with tiny G is probably working with the accelerations the problem with that being from what I've found is so far is you need to have good fast acceleration and deceleration for prints if you reduce this you can achieve that speed but the print is not going to be good notice he's not printing
The best way to shorten your learning curve is to build a low cost entry level machine and use that as a starting point. Regardless of any experience with subtractive machine tools additive manufacturing has elements that are best learned from building a small machine first .
|
Re: High Speed Printing February 22, 2014 08:51AM |
Registered: 10 years ago Posts: 239 |
that is great advice but just not me. I jump in the deep end and learn to swim (usually painful)Quote
vegasloki
Quote
Bill Clark
Yes, I'm starting to get my head around this. I do believe we will migrate towards faster and faster print speeds. Seems like a natural progression. I have absolutely been swimming in all this stuff trying to get up to speed. Very excitingQuote
cnc dick
Bill I don't think you'd be happy with the results of anything printing like that I know you come from machinist background and the prints at that rate of speed are terrible at the moment plastic filament can not be laid down that fast with good results. Plus the threaded rod machine you can see all of its shortcomings flexing all over the place terrible. 3-D printers were originally designed for one up prototypes the guy working with tiny G is probably working with the accelerations the problem with that being from what I've found is so far is you need to have good fast acceleration and deceleration for prints if you reduce this you can achieve that speed but the print is not going to be good notice he's not printing
The best way to shorten your learning curve is to build a low cost entry level machine and use that as a starting point. Regardless of any experience with subtractive machine tools additive manufacturing has elements that are best learned from building a small machine first .
. I did buy a assembled machine for initial learning curve however.
Sorry, only registered users may post in this forum.