1. Got a question or need help troubleshooting? Post to the troubleshooting forum or Search the forums!

DIY FDM printer

Discussion in 'Off Topic' started by colton81, May 14, 2017.

  1. colton81

    colton81 Active Member

    Joined:
    Feb 21, 2017
    Messages:
    328
    Likes Received:
    71
    since i cant convince the wife to let me get a new printer (the sigma r17) i convinced her to let me by parts over time to build one. But im trying to figure what the best route would be electronics wise. The C2 uses a raspberry pi and their main board but is this the best route to go to try to design or replicate the sigma r17 with independent dual extruders? Or using the ultimaker mainboard and the A20 lime microcontroller?
     
  2. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    22,570
    Likes Received:
    7,229
    There are a lot of options out there, you need to decide which one offers you the features you want.
    There is the Arduino/RAMPS combo (which is essentially what the C2 uses and is exactly what the R1 uses). There is the smoothie board, the Rambo board, the MKS boards, etc. Do your research to find the ones that suit your needs.

    For the C2 (and any of the others) the Pi is there to add OctoPrint support. Not needed if you are direct printing or only printing via an SD card and an LCD controller. Not a bad thing to have, but not essential. All of mine now use a clone of the Robo C2 LCD / Pi combo.
     
  3. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    22,570
    Likes Received:
    7,229
    Some of those controller boards are more expensive than others so you need to sort out what features are essential :)
     
  4. colton81

    colton81 Active Member

    Joined:
    Feb 21, 2017
    Messages:
    328
    Likes Received:
    71
    ok will do been trying to research alot of this just so much info out there! might end up just going with one carriage with 2 hot ends since the independent extruders is to complex for my first build.
     
  5. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    22,570
    Likes Received:
    7,229
    So make sure you include dual extrusion in your research.
    Personally, having used just about all of the boards out there at some point I would likely go with a smoothie board.
    Next best is the Rambo (although pricier).

    For cost to performance you just about can't beat the Arduino/RAMPS but you need a Pi to get the extra functionality with that combo
     
  6. daniel871

    daniel871 Well-Known Member

    Joined:
    Apr 18, 2015
    Messages:
    1,322
    Likes Received:
    510
    There is also the ARCHIM board if you want the 32-bit equivalent to the Rambo board (though, as noted, more expensive--haven't compared too closely to Smoothieboards though).

    If you're building your own printer, spec for a single extruder to start with since it is your first printer and trying to build from scratch without ever owning a mostly-assembled 3d printer is a tremendous headache learning curve. Dual Extruding has a whole host of other things that make it an even bigger headache.

    Generally, for every extruder you add, double the amount of prep work you have to do in modeling/slicer configuration and double the printing time for a given object (unless you have an independent head setup, and good luck replicating the Sigma's design because even though it is open source, it probably wasn't written with the intent of helping someone that is an utter newbie at this kind of building to get it right the first time).

    It might be worth pointing out to your wife that building over time is going to be more expensive than getting the current popular "Baby's first 3d printer" that is sold through multiple vendors (semi-locally through Monoprice as their "Maker Select Mini v2") for about $200-250 depending on sales and the actual vendor's pricing.

    That particular printer is probably the best "get your feet wet in 3d printing for the least amount of money out of pocket" deal (built-in wifi, rave reviews just about everywhere, etc.), plus you can return it within 30 days if you find out you hate 3d printing because there are a lot of little nit-picky disclaimers about it that are very hard to articulate/commiserate with if you haven't experienced them yourself.

    And if you keep it, you have a full year's warranty for anything that goes wrong on the printer itself that is hard to beat elsewhere.
     
  7. colton81

    colton81 Active Member

    Joined:
    Feb 21, 2017
    Messages:
    328
    Likes Received:
    71
    Well i have the C2 now but im wanting a bigger build area and dual extruders. But yea im going to start with single extruder once i get it up and running how i want then i will eitger try out independent dual or single carriage dual extruder.
     
  8. colton81

    colton81 Active Member

    Joined:
    Feb 21, 2017
    Messages:
    328
    Likes Received:
    71
    Also anyone test out the .9 degree vs 1.8 degree stepper motors? Is there really any difference in qaulity? I was looking at the nema 17 im probably going to go with the smoothie board 5x from what i understand its more then capable of doing dual extruders and has built in ethernet on it as well. I know it says it needs to be 2A or under for the stepper motor
     
  9. colton81

    colton81 Active Member

    Joined:
    Feb 21, 2017
    Messages:
    328
    Likes Received:
    71
  10. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,586
    There is potential for better prints with a .9° v 1.8° steps but the reality is likely you won't be able to see it unless you are working with very small nozzle sizes and very slow prints. Save your money and get better electronics (even 32-bit)/heaters/power supplies/extruders, that is where you will see noticeable differences, either in performance and/or print quality.
     
  11. daniel871

    daniel871 Well-Known Member

    Joined:
    Apr 18, 2015
    Messages:
    1,322
    Likes Received:
    510
    I think the only real time you notice better performance from .9-degree steppers is when using them in MK-style direct-gear extruders.
     
  12. colton81

    colton81 Active Member

    Joined:
    Feb 21, 2017
    Messages:
    328
    Likes Received:
    71
    Ok so if i go with a 1.9 stepper motor with a 8mm lead screw at 2mm pitch i can if need be achieve a 0.5 mm layer height with no issues? Am i understanding this correctly?
     
  13. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,586
    1.9° Stepper + TR8*2 lead screw = 0.01 mm per step, probably too little movement to actually see a difference in a printed model. Are you sure you mean 0.50 layer and not 0.05 layer?
     
  14. colton81

    colton81 Active Member

    Joined:
    Feb 21, 2017
    Messages:
    328
    Likes Received:
    71
    Yea my bad i meant 0.05 so something like this should work?

    TriGorilla NEMA 17 Stepper Motor for CNC Mill Router or RepRap 3D Printer Prusa i3 + Iintegrated 300mm TR8x8 Lead Screw https://www.amazon.com/dp/B01DVD87Q6/ref=cm_sw_r_cp_api_NvFgzbH9Q4D4N
     
  15. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,586
    TR8*8 is 8mm linear movement per rotation, so the math is pretty simple: 8 mm / 200 steps = 0.04 mm per step. This number is important because full steps are the only guaranteed movements. Then there are electrical divisions done by the stepper drivers, RAMPS board and Marlin that divide that by 16 so again the math is simple: 0.04 mm per step / 16 microsteps = 0.0025 mm per microstep.

    Please note that all of this is theoretical and the precision is simply not there. There is just no way to control any FDM printer to a precision of 2.5 microns. Also no guarantee that microsteps are actually 2.5 microns each. Also note that you are only dealing with the Z axis here since most printers drive X and Y with GT2 belts at around 80 steps per millimeter.
     
    #15 WheresWaldo, May 15, 2017
    Last edited: May 15, 2017
    mark tomlinson likes this.
  16. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    22,570
    Likes Received:
    7,229
    If you need that much precision then FDM is not your way to go*.
    SLA, DLP, CNC, etc. Those have higher accuracy (or can).
    FDM is what I call decently accurate (not really high-tolerance) and cheap to print.



    *prepare your wallet for high dollar extractions however
     
  17. colton81

    colton81 Active Member

    Joined:
    Feb 21, 2017
    Messages:
    328
    Likes Received:
    71
    ok yea ive been looking more into it. Its not that im trying to reach that accuracy im just trying to get the best possible accuracy on a fdm that is noticeable. from what im reading the TR8*8 is the standard for Lead screws. So the best possible accuracy at a full step like @WheresWaldo said is 0.04 correct? so the smoothieboard says it can do 1/32 micro stepping with its built in drivers is that even worth putting it at that level? because then im looking at 0.00125 which seems overkill for a fdm printer.
     
  18. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,586
    Smoothie is not just about 1/32 microsteps. It is much faster than an Arduino and RAMPS board combo and can handle much more complex movements than the 8-bit boards. The Smoothieboard and clones are all 32-bit processor boards.

    You have to look at the possible margins of error, if you have a 200 step Stepper, a TR8*8 lead screw and a smoothie with 1/32 microsteps enabled then each microstep can move that axis 1.25 microns. Even if there is error in the microsteps and let's say it's off by 50% you can't really control the flow of extrude that accurately. So some is lost, but what you may not gain in ultimate accuracy you get back in speed and better stepper control, built in Web interface and things like that.

    An answer to your question is that a movement of 0.04 mm is all that is guaranteed to be accurate not that the overall dimensions will be off any significant amount. To give you an example that is real world. I sell a lot of models, I try to model them in CAD with 100% accurate dimensions. So they should fit perfectly, but the printers that are used have a maximum tolerance of +- 0.2 mm overall. That is 5 times what you are talking about here. I have no issues with tolerances that far off and neither do all the people who buy my models.
     
    #18 WheresWaldo, May 15, 2017
    Last edited: May 15, 2017
    mark tomlinson likes this.
  19. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    22,570
    Likes Received:
    7,229
    That is the difference between modelling and modelling for reality

    Sent from my BNTV450 using Tapatalk
     
  20. colton81

    colton81 Active Member

    Joined:
    Feb 21, 2017
    Messages:
    328
    Likes Received:
    71
    well that was the main reason i was looking into the smoothieboard was for the 32 bit processor i was more refering to setting it as 1/32 mirco stepping instead of 1/16. which thats something i can play around with though once i get everything up and running. so as of now im looking at
    stepper motor nema 17
    https://www.amazon.com/45Ncm-Steppe...sr=8-8&keywords=nema+17+stepper+motor+bipolar

    stepper motor nema 17 with TR8*8 lead screw
    https://www.amazon.com/gp/product/B01DVD87Q6/ref=crt_ewc_title_dp_2?ie=UTF8&psc=1&smid=AVCH3LIOAKMIT

    Smoothieboard 5X
    https://shop.uberclock.com/products/smoothieboard

    12mm shaft for z axis
    https://www.amazon.com/dp/B002BBJ0CA/_encoding=UTF8?coliid=I2XWFU9IRX2V2X&colid=21M6223AFJ5CB
     

Share This Page