MakerBots, RepRaps and UP!s: It's 3D printers!

The Something Awful Forums > Discussion > Hobbies, Crafts, & Houses > MakerBots, RepRaps and UP!s: It's 3D printers!

«‹›727 »

MickRaider: Aug 27, 2004; Now I smell like lemonade!

I saw someone post about how they're confused why there's not more large scale printers. I mean all you have to do is increase the bar length? Right?

RIGHT?!

No you fools, The longer the span the larger the moment.

Always makes me laugh when I see people with no mechanical aptitude saying they're going to design and build some 1m x 1m x 1m printer

# ? Jul 29, 2015 19:41

Adbot: ADBOT LOVES YOU

# ? May 8, 2024 20:55

Harvey Baldman: Jan 11, 2011; ATTORNEY AT LAW
Justice is bald, like an eagle, or Lady Liberty's docket.

MickRaider posted:

I saw someone post about how they're confused why there's not more large scale printers. I mean all you have to do is increase the bar length? Right?

RIGHT?!

No you fools, The longer the span the larger the moment.

Always makes me laugh when I see people with no mechanical aptitude saying they're going to design and build some 1m x 1m x 1m printer

I'll be honest and say I am one of those people who doesn't get why things can't be made bigger easily.

# ? Jul 30, 2015 01:15

Fayez Butts: Aug 24, 2006

poo poo gets hosed up more easierly

# ? Jul 30, 2015 01:17

MickRaider: Aug 27, 2004; Now I smell like lemonade!

https://plus.google.com/u/0/101170277470139301307/posts/JZqRoWvkp2N?cfem=1

Brook Drumm knows his stuff

MickRaider fucked around with this message at 01:21 on Jul 30, 2015

# ? Jul 30, 2015 01:19

Bad Munki: Nov 4, 2008; We're all mad here.

I seem to recall some guy converting his joe's 4x4 to be a 3d printer, but honestly, WHY.

# ? Jul 30, 2015 01:20

mewse: May 2, 2006

What's the term for settings in slicer software to deal with extruder velocity screwing things up? I've been googling recoil, whiplash :smith:

# ? Jul 30, 2015 02:18

peepsalot: Apr 24, 2007; ��PEEP THIS...
��BITCH!

mewse posted:

What's the term for settings in slicer software to deal with extruder velocity screwing things up? I've been googling recoil, whiplash

You mean retraction speed?

# ? Jul 30, 2015 03:09

Linux Assassin: Aug 28, 2004; I'm ready for the zombie invasion, are you?

Harvey Baldman posted:

I'll be honest and say I am one of those people who doesn't get why things can't be made bigger easily.

Alright- so here is the issue- you want to move something very small, very discreetly, over a distance.

You use stepper motors, because servomotors are expensive- but stepper motors don't really 'know' their position, they do 'know' how many clicks they have tried to move. So you need to make sure that the stepper is not going to 'miss' a step. this is achieved by putting more voltage to the stepper, which makes it want to 'click and shake'- you handle the shake by putting the movement under tension (belt, or screw, or whatever). Now if you start making the movement longer then that 'click and shake' gets amplified by the length of the bar, and the tension that you need to put it under wants to full on warp the movement rod, and there will be more irregularities over the entire movement that may cause the stepper motor to miss a step.

Then you have your print head- and the plastic has to get there somehow, various ways this is done, but the important part is that the longer the plastic has to go the heavier the weight on the print head, the heavier the weight the stronger the gear/feed mechanism needs to be- which puts even more weight, all that weight amplifies the problem above.

Finally, the plate itself needs to move up and down (or the print mechanism needs to move up and down)- now you have a major problem because every extra bit of length increases the static weight of moving the plate (or every bit from problem 1 and 2 if your moving the print mechanism rather than the plate for Z). The reason why this is a problem is that now you have a force acting in only one direction on that stepper motor; it WANTS to move down thanks to gravity. You can counter it by friction (which means more strain on the stepper with each step it takes), or with a stronger stepper (which means more 'click and shake'), or by counterbalancing the Z access (not sure if anyone does this, but it would be very mechanically complex).

Now that's not to say these are insurmountable problems, in fact they are really surmountable (but cost $$$), use ball screws and servomotors. BUT that's not proven technology for 3d printing, and it may introduce all kinds of weird eccentricities to the print that won't be evident until someone builds it and then starts countering those problems much like how stepper motors and drive belts have slowly improved in performance (Ball screws and servomotors will likely not experience eccentricities related to size).

# ? Jul 30, 2015 03:29

mewse: May 2, 2006

peepsalot posted:

You mean retraction speed?

No, like there's a minor amount of sway with my printer but the software assumes a perfect world where it can whip the extruder 20mm to the left at fairly high speed and immediately start a line on the y axis without any kind of pause. It was causing these wiggles in the print that were obvious when they happened on every single layer. Extruder movement speed I guess?

# ? Jul 30, 2015 04:07

Bad Munki: Nov 4, 2008; We're all mad here.

mewse posted:

No, like there's a minor amount of sway with my printer but the software assumes a perfect world where it can whip the extruder 20mm to the left at fairly high speed and immediately start a line on the y axis without any kind of pause. It was causing these wiggles in the print that were obvious when they happened on every single layer. Extruder movement speed I guess?

I don't know about your software but most cnc config stuff I've seen just calls it acceleration.

# ? Jul 30, 2015 04:15

mewse: May 2, 2006

Bad Munki posted:

I don't know about your software but most cnc config stuff I've seen just calls it acceleration.

YES I think this is what I was looking for, thank you so much, stupid high school physics

# ? Jul 30, 2015 04:17

Some Pinko Commie: Jun 9, 2009; CNC! Easy as 1️⃣2️⃣3️⃣!

mewse posted:

YES I think this is what I was looking for, thank you so much, stupid high school physics

You can usually lower that in the firmware on the machine itself. Look for Default Acceleration and Max Acceleration values in Arduino, adjust and upload to the machine, then try a calibration cube or something to see if it made things better.

The slicer has nothing to do with that part. I've set that around 500mm/s/s for X, Y and E on my Robo because 3000mm/s/s default and 10000mm/s/s max was loving insane when pushing a lot of short, rapid direction changes.

# ? Jul 30, 2015 20:03

Some Pinko Commie: Jun 9, 2009; CNC! Easy as 1️⃣2️⃣3️⃣!

Did anyone see the homemade ceramic 3D printing delta?

https://www.youtube.com/watch?v=OOqAxePyJBg

# ? Jul 30, 2015 20:04

mewse: May 2, 2006

Wade Wilson posted:

You can usually lower that in the firmware on the machine itself. Look for Default Acceleration and Max Acceleration values in Arduino, adjust and upload to the machine, then try a calibration cube or something to see if it made things better.

The slicer has nothing to do with that part. I've set that around 500mm/s/s for X, Y and E on my Robo because 3000mm/s/s default and 10000mm/s/s max was loving insane when pushing a lot of short, rapid direction changes.

I am kinda hosed because I built my printer from an aliexpress prusa kit and the controller board is a generic chinese thing running Marlin firmware. I can control it with repetier-host but I'm very wary of flashing firmware to it.

I did see in repetier that the default values for cura slicer has travel set to 150mm/s which might be high.

Is there a way to download firmware from the board and maybe yanking the marlin values out of it?

# ? Jul 30, 2015 20:28

cakesmith handyman: Jul 22, 2007; Pip-Pip old chap! Last one in is a rotten egg what what.

Linux Assassin posted:

Alright- so here is the issue- you want to move something very small, very discreetly, over a distance.

You use stepper motors, because servomotors are expensive- but stepper motors don't really 'know' their position, they do 'know' how many clicks they have tried to move. So you need to make sure that the stepper is not going to 'miss' a step. this is achieved by putting more voltage to the stepper, which makes it want to 'click and shake'- you handle the shake by putting the movement under tension (belt, or screw, or whatever). Now if you start making the movement longer then that 'click and shake' gets amplified by the length of the bar, and the tension that you need to put it under wants to full on warp the movement rod, and there will be more irregularities over the entire movement that may cause the stepper motor to miss a step.

Then you have your print head- and the plastic has to get there somehow, various ways this is done, but the important part is that the longer the plastic has to go the heavier the weight on the print head, the heavier the weight the stronger the gear/feed mechanism needs to be- which puts even more weight, all that weight amplifies the problem above.

Finally, the plate itself needs to move up and down (or the print mechanism needs to move up and down)- now you have a major problem because every extra bit of length increases the static weight of moving the plate (or every bit from problem 1 and 2 if your moving the print mechanism rather than the plate for Z). The reason why this is a problem is that now you have a force acting in only one direction on that stepper motor; it WANTS to move down thanks to gravity. You can counter it by friction (which means more strain on the stepper with each step it takes), or with a stronger stepper (which means more 'click and shake'), or by counterbalancing the Z access (not sure if anyone does this, but it would be very mechanically complex).

Now that's not to say these are insurmountable problems, in fact they are really surmountable (but cost $$$), use ball screws and servomotors. BUT that's not proven technology for 3d printing, and it may introduce all kinds of weird eccentricities to the print that won't be evident until someone builds it and then starts countering those problems much like how stepper motors and drive belts have slowly improved in performance (Ball screws and servomotors will likely not experience eccentricities related to size).

I can't edit your post down because all of its relevant and correct, but here goes:

A servomotor is a motor with an encoder on, nothing more. If someone comes up with the control hardware, servo motors are possible. Hell you could add encoders to the NEMA steppers you're using now if the controls could handle the feedback.

To minimise the weight on the print head you have the reel supported above the printer, the only weight then is the length of filament between the reel and extruder, not perfect because of the greater distances meaning you might pull a metre of filament out then move half that distance back. So we need a controlled feed from the reel with recoil/retract settings to keep an appropriate amount of slack before the extruder. Distance between the reel mouth and extruder inlet is easily calculable, you know how much filament you're using so you can control this with another axis.

Z-axis counterbalance: either counterweight the plate with belt fixed at each corner, idle sprockets and weights(weight is chosen based on fixed weight of the plate etc and variable weight i.e. anticipated build weight) or a pneumatic cylinder underneath, electronically control the pressure to vary the counterbalance as the print progresses. You'll need a counterbalance tank otherwise you just compress the air in the cylinder giving you a very non-linear response.

So basically I'm agreeing, everything is surmountable but untested in this field. The challenge is fascinating to me, slightly more than the printing itself.

Personally for a massive printer I'd build a climbing gantry, four fixed legs, a frame between them that climbs by geared servo motors on each corner, with the x & y axis between them, keep the platform stationary and only have to counterweight the climbing gantry which would remain a constant weight.

# ? Jul 30, 2015 20:48

Some Pinko Commie: Jun 9, 2009; CNC! Easy as 1️⃣2️⃣3️⃣!

mewse posted:

I am kinda hosed because I built my printer from an aliexpress prusa kit and the controller board is a generic chinese thing running Marlin firmware. I can control it with repetier-host but I'm very wary of flashing firmware to it.

I did see in repetier that the default values for cura slicer has travel set to 150mm/s which might be high.

Is there a way to download firmware from the board and maybe yanking the marlin values out of it?

Marlin boards are surprisingly durable for firmware flashing. You have to really gently caress up to ruin an Arduino RAMPS board (and I'm talking physical over-voltage).

Use the actual software found here:

https://github.com/open3dengineering/Prusa-i3/tree/master/Firmware/Marlin

And you should be good to go. Instructions are on the GitHub post.

On a note about designing bigger printers, someone pointed this monstrosity out to me this afternoon.

http://www.cosineadditive.com/am1/

# ? Jul 30, 2015 21:11

ImplicitAssembler: Jan 24, 2013

Wade Wilson posted:

Did anyone see the homemade ceramic 3D printing delta?

https://www.youtube.com/watch?v=OOqAxePyJBg

The extruders are finally getting there. My wife is a potter and is keen on getting into this, so I've been following the development..

https://www.youtube.com/watch?v=1DjVC5MxJr4

https://plus.google.com/u/0/communities/109375785524125994679?cfem=1

(The only page I actually get anything useful from on G+!)

# ? Jul 30, 2015 21:36

CapnBry: Jul 15, 2002; I got this goin'; Grimey Drawer

mewse posted:

I am kinda hosed because I built my printer from an aliexpress prusa kit and the controller board is a generic chinese thing running Marlin firmware. I can control it with repetier-host but I'm very wary of flashing firmware to it.

There's no way to practically extract the values from the firmware but you can do an M503 command in a terminal and Marlin will print all the speed and acceleration settings. All of these can be overridden so just put your overrides in the start GCode. No need to compile firmware or take the rather high chance that you'll never be able to get the configuration back the way it was.

I've hacked on Marlin a bunch and I still don't edit configuration.h to change the values, I just put them in my start gcode.

code:

M301 P15.98 I1.07 D100 ; extruder PID
M92 X80 Y80 Z4000 E660; steps per unit
M201 X1000 Y1000 Z30 E750 ; max accel
M203 X150 Y150 Z2.5 E30 ; max feedrate
M204 S3000 T750; Accel/stationary extrude accel
M205 X20 Y20 Z0.5 E5 ; Jerk

# ? Jul 30, 2015 23:06

Aurium: Oct 10, 2010

Wade Wilson posted:

Marlin boards are surprisingly durable for firmware flashing. You have to really gently caress up to ruin an Arduino RAMPS board (and I'm talking physical over-voltage).

Use the actual software found here:

https://github.com/open3dengineering/Prusa-i3/tree/master/Firmware/Marlin

And you should be good to go. Instructions are on the GitHub post.

On a note about designing bigger printers, someone pointed this monstrosity out to me this afternoon.

http://www.cosineadditive.com/am1/

Another, somewhat smaller one: http://shop.re3d.org/collections/gigabot-3d

# ? Jul 30, 2015 23:19

mewse: May 2, 2006

Wade Wilson posted:

Marlin boards are surprisingly durable for firmware flashing. You have to really gently caress up to ruin an Arduino RAMPS board (and I'm talking physical over-voltage).

Use the actual software found here:

https://github.com/open3dengineering/Prusa-i3/tree/master/Firmware/Marlin

And you should be good to go. Instructions are on the GitHub post.

Thanks. I may try this eventually but

CapnBry posted:

There's no way to practically extract the values from the firmware but you can do an M503 command in a terminal and Marlin will print all the speed and acceleration settings. All of these can be overridden so just put your overrides in the start GCode. No need to compile firmware or take the rather high chance that you'll never be able to get the configuration back the way it was.

I've hacked on Marlin a bunch and I still don't edit configuration.h to change the values, I just put them in my start gcode.
code:
M301 P15.98 I1.07 D100 ; extruder PID
M92 X80 Y80 Z4000 E660; steps per unit
M201 X1000 Y1000 Z30 E750 ; max accel
M203 X150 Y150 Z2.5 E30 ; max feedrate
M204 S3000 T750; Accel/stationary extrude accel
M205 X20 Y20 Z0.5 E5 ; Jerk

This is wonderful, thank you. Being able to change my accel value without overwriting the firmware is exactly what I'd like to do.

# ? Jul 30, 2015 23:36

The Eyes Have It: Feb 10, 2008; Third Eye Sees All
...snookums

We might be a little past it but scaling up a 100mm printer that runs like a top to 1m is conceptually not unlike making a drat awesome and ridiculously sturdy 6" bridge out of matchsticks, and figuring that the only thing keeping you from a 60 foot bridge over the creek is some ten foot matchsticks.

# ? Jul 31, 2015 02:05

ceebee: Feb 12, 2004

Other than price and size are those larger printers worth it for somebody who wants to print out 1:1 scale stuff (props, armor, etc?) instead of getting say a Lulzbot Taz 5 or something? I was considering the Gigabot but heard some pretty iffy reviews about it.

# ? Jul 31, 2015 02:18

isndl: May 2, 2012; I WON A CONTEST IN TG AND ALL I GOT WAS THIS CUSTOM TITLE

ceebee posted:

Other than price and size are those larger printers worth it for somebody who wants to print out 1:1 scale stuff (props, armor, etc?) instead of getting say a Lulzbot Taz 5 or something? I was considering the Gigabot but heard some pretty iffy reviews about it.

Do you like spending literal days on a single print, then restarting it when something got hosed up?

# ? Jul 31, 2015 02:43

ceebee: Feb 12, 2004

Yeah I guess that'd be the worst con of a single large print. Thats why I'm considering the Taz over most because I don't mind putting multiple parts together to make one big piece.

# ? Jul 31, 2015 02:48

Rubiks Pubes: Dec 5, 2003; I wanted to be a neo deconstructivist, but Mom wouldn't let me.

What is the best way to split up a big object in to pieces to print on a smaller printer?

# ? Jul 31, 2015 03:28

insta: Jan 28, 2009

ceebee posted:

Yeah I guess that'd be the worst con of a single large print. Thats why I'm considering the Taz over most because I don't mind putting multiple parts together to make one big piece.

100% the way to do it.

Rubiks Pubes posted:

What is the best way to split up a big object in to pieces to print on a smaller printer?

Netfabb has this built in, which I think is available in the free version.

# ? Jul 31, 2015 03:48

Cannon_Fodder: Jul 17, 2007; "Hey, where did Steve go?"
Design by Kamoc

I have zero experience with 3d printing but I'm fascinated. Can someone talk me off the ledge as I'm thinking this hobbyking babby printer might be a good learning tool...

# ? Jul 31, 2015 15:38

Bad Munki: Nov 4, 2008; We're all mad here.

Cannon_Fodder posted:

I have zero experience with 3d printing but I'm fascinated. Can someone talk me off the ledge as I'm thinking this hobbyking babby printer might be a good learning tool...

Do it, and post a trip report. Throw yourself upon this 3d printing grenade for the betterment of the thread.

# ? Jul 31, 2015 16:03

torpedan: Jul 17, 2003; Lets make Uncle Ben proud

Bad Munki posted:

Do it, and post a trip report. Throw yourself upon this 3d printing grenade for the betterment of the thread.

I am really curious about their other printer. I have been wanting to pick up a printer for home for some time and the price point on either of their printers is pretty tempting. Hopefully they both do not ten out to be trash.

# ? Jul 31, 2015 16:24

blugu64: Jul 17, 2006; Do you realize that fluoridation is the most monstrously conceived and dangerous communist plot we have ever had to face?

Cannon_Fodder posted:

I have zero experience with 3d printing but I'm fascinated. Can someone talk me off the ledge as I'm thinking this hobbyking babby printer might be a good learning tool...

I was thinking of doing that too but I heard the shipping was outrageous.

# ? Jul 31, 2015 16:28

torpedan: Jul 17, 2003; Lets make Uncle Ben proud

blugu64 posted:

I was thinking of doing that too but I heard the shipping was outrageous.

That is typical for them when they ship internationally. Their weight cakes are usually over estimates which makes it worse. If the stock them in a warehouse that keeps them from being shipped half way around the world it should be better.

# ? Jul 31, 2015 16:31

moron izzard: Nov 17, 2006; Grimey Drawer

They are likely on a slowboat to the rest of the warehouses now. hobbyking can be lame like that.

with shipping it will be 238 to 248 for me (depending if i do fedex express or ems 3 day)

moron izzard fucked around with this message at 19:08 on Jul 31, 2015

# ? Jul 31, 2015 17:08

Rubiks Pubes: Dec 5, 2003; I wanted to be a neo deconstructivist, but Mom wouldn't let me.

Coming from someone who just got in to the 3d printing thing about a year ago, I would say order a printer that you can put together from a kit. It helped me a LOT when it has come to troubleshooting since I know how the thing goes together and works.

# ? Jul 31, 2015 17:30

foosel: Apr 2, 2010

Rubiks Pubes posted:

I would say order a printer that you can put together from a kit

Strongly seconding this. Even though various marketing departments love to claim otherwise, those things are not plug and play, you will need to fix things, and it helps tremendously how your machine works in these kinds of situations.

# ? Jul 31, 2015 19:15

Mofabio: May 15, 2003; (y - mx)*(1/(inf))*(PV/RT)*(2.718)*(V/I)

The <$500 Mendel90's a great machine. Everybody's different, but the more kit-like (or worse, off-the-shelf-like) the printer is, the more I turn my brain off, feel married to following the assembly SOP, the less I end up knowing about the printer, and the more the assembly feels like work. There's just enough thinking to do in the Mendel90 assembly to preserve the joy of DIY.

The major argument to not kit/DIY is if you needed the printer yesterday. I figure that I went almost my entire life without a 3D printer, and yet I didn't have a burning want for it then. So what's a little more assembly time to save money, learn some skills, and feel the joy of building?

# ? Jul 31, 2015 21:34

insta: Jan 28, 2009

Seconding the Mendel90 if you have the budget for it.

# ? Jul 31, 2015 23:10

CommieGIR: Aug 22, 2006; The blue glow is a feature, not a bug; Pillbug

I am happy to report that the Robo3D R1 I had such issues with is working, the problem was the hobbed bolt.

# ? Aug 3, 2015 02:20

Portable Staplefrog: May 21, 2007

I never noticed this thread! I do not have a 3D Printer, but I constantly design things to be printed, which is done at Shapeways. I'm thinking of getting a Mini Fabrikator, which will surely print at lesser quality than what I'm used to, but it might be good for prototypes. I think that's what Cannon_Fodder was talking about a few posts up...

My designs (all printable) are at tywtb.

# ? Aug 7, 2015 16:23

insta: Jan 28, 2009

None of your designs are going to print period on anything less than a $600 machine. You made the mistake of starting with Shapeways :v:

, they've got $50k+ printers they print those parts on.

# ? Aug 7, 2015 18:02

Adbot: ADBOT LOVES YOU

# ? May 8, 2024 20:55

Portable Staplefrog: May 21, 2007

Yes, which is a good reason to use them. If I got a cheaper printer, I'd have to design around that. And would probably continue with Shapeways for things I want to come out really well.

# ? Aug 7, 2015 18:12

The Something Awful Forums > Discussion > Hobbies, Crafts, & Houses > MakerBots, RepRaps and UP!s: It's 3D printers!

«‹›727 »