3D Printing Building an FDM printer, maybe...

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

vcp

Well-Known Member
Joined
Jan 27, 2009
Messages
1,473
Reaction score
577
Location
Meridian, ID
My old printers are pretty... old, limited, not printing well anymore, or are in a zillion pieces. I've had intentions for years to build a new one; maybe now I should get around to it. Not my first rodeo; my original 2012 Replicator has been modified into unrecognizability, but it's the one in pieces and it would still be limited in size.

Requirements:
  • Core-XY mechanism
  • 300 mm^3 minimum, more z is better
  • AC heating bed
  • dual z-axis screw lift; belt or dual-motor syncronized
  • dual cantilevered bed
  • 3-screw leveling
  • enclosable
  • direct-drive extruder
  • 24V
  • need to keep additional component purchases to <$500
  • 220/221, whatever it takes.*
What I have:
  • plenty of decent NEMA 17's
  • power supplies, fans
  • SSR for the heat bed
  • handfull of odd bearings
  • bunch of M3, M4, and M5 hardware
  • bunch of odd extrusion; unfortunately, none of it the common 2020 (all 3030, 2525, 1010)
  • excellent 6mm aluminum build plate
  • GT2 belt
  • the printer I have working will be adequate for printing new printer parts in ABS.
What I need to decide/acquire:
  • controller (I'm assuming for the moment, Duet2wifi)
  • extruder/hotend - open to suggestions (not familiar with current crop) [as an aside, I had my Replicator set up for hotend water cooling, for hi-temp printing; worked well so that's something I'd be thinking about.]
  • silicone AC heating pad
  • rails and/or rods
  • z-screws with anti-backlash nuts
  • assorted bearings, idlers, pulleys.
What I'm looking at:
  • Hypercube, Hypercube Evolution (HEVO) - large, active users group, many builds and modifications.
  • D-Bot/C-Bot Core-XY - huge, active users group, many builds and modifications
  • FORKlift MK1 - small users group, some mods available, CAD files available; and with beautifully engineered printed parts.
  • SecKit Core-XY - not an original build but a kit that checks most of the boxes; unfortunately while it's somewhat configurable, it would still duplicate much of the hardware I have.
  • V-King Core-XY - while I've got some v-wheels around here, I think I'd rather stick with linear bearings or rails; interesting though, and a unique z-axis lift.
  • Jubilee - don't know much about this but it appears to have a workable DIY toolchange system.
  • Anything else I should be looking at?

First thing I need to do is go dig up and inventory my extrusions. Thinking about trial printing some of the FORKlift parts (really like the adjustable motor mounts).

Thoughts/suggestions?


* That's a joke, for all you Mr. Mom's out there.
 
Last edited:
That's a subtle change in direction ;).

When faced with a tough decision, why not do both? While waiting for the Photon, I FDM printed a few test pieces from the ForkLift noted above. The guy that did this really thought out the pieces well, considering supports (none needed) and layer orientation for strength. The test motor mount that I printed came out really well, so I'm going to look closer at the Forklift, and see if I can adapt the design to the parts that I have in-house.
 
I have build one based on the Prusa Mk3S but with parts and then installed the MK3S software on another board. It came out and word very well. I have looked at the HupeCube, but will probably wait till April. Prusa is releasing a Prusa XL for 1200-1500. HyperCube Evolution list a cost of $530 for the parts. One builder reportedly build one for $680. I price the parts at 800+.

I like the magnetic build plate of the Prusa. I might wait but it is tempting.
 
Have you looked into the Voron, has by far the best support group/documentation and proven performance out there as far as corexy is concerned
. You could get away fairly cheap with your existing parts pile and save a buncha dough by not needing a Duet combo (Core xy requires more than the standard # of stepper drivers) seeing as how it runs on Klipper. Otherwise look at the Vivedino Troodon, Their claim is that it's not a Voron clone, just Voron "inspired" Whatever the he*& that means. Not cheap, but good value for the money.Seems likely these days that you can buy a complete machine for less than you'd pay for the parts. You could sell off your spares to recoup some of the initial cost..
 
Last edited:
My plan is to build one someday. But, if I happen to fall into a pile of cash I'd probably just buy the Troodon and save a ton of time and energy.
 
I have build one based on the Prusa Mk3S but with parts and then installed the MK3S software on another board. It came out and word very well. I have looked at the HupeCube, but will probably wait till April. Prusa is releasing a Prusa XL for 1200-1500. HyperCube Evolution list a cost of $530 for the parts. One builder reportedly build one for $680. I price the parts at 800+.

I like the magnetic build plate of the Prusa. I might wait but it is tempting.
Chuck: After getting tired of having to hammer and chisel parts off of glass, I put an overpriced Wham Bam PEX magnetic/flexible print surface on my cR-10 and can truthfully say that it is by FAR the best upgrade I've made. I dread having to use my other printers that still have glass. There are tons of cheaper options out there, you just need to verify the max temp the magnet can handle. Alot of 'em are only good to 80c ish...
 
I have one for my ender. I just can't get the thing to print reliably. I have a pinda probe and auto-leveling and the think keeps scraping the bed despite having been set up properly.
 
Chuck: I suspect auto leveling is over rated. Although for quite a while I've actually owned an original BL Touch, I've yet to use it. Haven't found the need. I rarely (maybe once a month?) have to level my bed. I'm currently building a cobbled Ender 3/A10m mess that's going in an enclosure since winter hit and I'm struggling with ambient temps (everytime someone opens a door), I'll probably try to implement it there just due to lack of accessibility. The Wham Bam (what they call PEX) build plate has been virtually flawless for hundreds of prints. Simply give it a quick scuff with 000 steel wool and wipe it down with IPA and it just works. Only gripe I have is that there are no alignment tabs, makes it tough to get on straight. McMaster Carr also sells high temp rated magnetic sheets if you care to DIY a setup.
 
Last edited:
Auto-leveling on the PrussIs not overrated. It is near perfect and almost never fails. I would be afraid to use any expendable surface with my Ender-3. I Level The printer and it seems like it’s the right level, But on the next print it dives into the PRINT surface and destroys it.

It does this with or without the Pinda probe. I am sure I got a lemon.
 
To each his own then. Everyone has their preferred method, mine works for me. Also; We are comparing different types of machines. Even attempting to compare similar models from the same manufacturer may very well give differing results?
 
Automatic leveling is not a priority for me either. Like Wallace, I seldom need to touch mine. I can understand how it would be useful for someone starting out though.
 
Auto-leveling on the PrussIs not overrated. It is near perfect and almost never fails. I would be afraid to use any expendable surface with my Ender-3. I Level The printer and it seems like it’s the right level, But on the next print it dives into the PRINT surface and destroys it.

It does this with or without the Pinda probe. I am sure I got a lemon.
Sounds like Ya got a bad one. I personally prefer to set Z 0 with a feeler gauge myself (takes 30 seconds or so), that way I'm not relying on something that "may" crash my junk.
 
I'm looking at the Voron/Troodon, but this I don't understand...
Since Klipper offloads all the grunt work onto a Pi, you can get away with a simple/cheap board. Save a C-Note at minimum. Just look into it, you'll see.
 
Corexy requires, by design, more than the "standard" number of stepsticks that boards are designed for..
 
Just now realized you may have been referring to the Troodon boast about the Duet "style" board. While it is very nice, it's simply overkill/added expense when building a Voron.
 
Corexy requires, by design, more than the "standard" number of stepsticks that boards are designed for..

A pair for the combined XY, one for each extruder, plus 1-3 depending on how you do Z.
 
Last edited:
Since Klipper offloads all the grunt work onto a Pi, you can get away with a simple/cheap board. Save a C-Note at minimum. Just look into it, you'll see.

While I have an unused Pi here somewhere, and a smoothieboard, I just googled Klipper and UDOO and it looks like people are also using that. I've also got a UDOO and if that works all I'd need would be a driver shield for it. While the Pi/smoothieboard would be possible, the smoothieboard was intended to replace the board in my K40 laser.
 
That's what I'm talking about. Duet setup would cost a whole lot more, and other than the UI, wouldn't do anything better..
 
Have you looked at Scara systems yet? Intriguing, yet probably too limited to be truly practical.
 
Have you looked here yet. It's obviously pretty thorough and they must've spent a ton of time on it. Voron w/single extruder requires 7 steppers. I'm not aware of such a single board for such? https://docs.google.com/spreadsheet...7sIsVr5nARIzHsFrJnI4P4npgs/edit#gid=798204101
OIC, they need four steppers for that massive Z lift, with eight belts and eight independent tension adjustments; all aside from any of the XY mech.

MIC 6 and ALCA 5 sold for scrap prices
I've got a really excellent plate from Misumi, when they had a deal giving away a bunch of stuff for free.

Could try your luck with something like this, but you still need to breakout the extra steppers. https://www.ebay.com/i/184027236886...vjn4W3XOOJysfZwl7JdLGYNZ7qoiSKXRoCCUcQAvD_BwE
Also wondering about Bigtreetech stuff. I understand it's a crapshoot with the clones, but they are sooo cheap.

The UDOO combines an ARM and Arduino Mega on a single board. When I got it from the original Kickstarter, I was thinking it would be a useful division of the high-level motion planning and low-level step control, but nobody had done the code at the time (and way beyond me). But it looks like Klipper has now done exactly that.
 
Yup. Klipper looks like THE thing to run either core xy or Delta format. Seems to be continuously updated also which is extremely important.
 
And you do have a valid point about the complexity on the Voron. I'm just absolutely intrigued by the stationary bed. Slinging or raising/lowering a relatively massive print surface in my mind is not good engineering. Move the light(er) stuff and leave the heavy crap alone.As an added bonus, you're not constantly moving mains current wiring for the bed heater. Proper motion wire is not cheap and using cheap wire in a moving configuration is never a good idea.
 
Last edited:
Back
Top