Idea: A RepRap Hanging From the Ceiling

It makes cirkles now. Kinda...
Vimeo

---

torbjornludvigsen.com

I did some openscadding (see attached).

I'm wondering about the roller-fairleads, if I would be better to just use fishing rod leads since I want to aim for cheap and simple. The roller leads adds a lot of complexity to the build...

---

torbjornludvigsen.com

check this out, this was my diploma project form a couple of years ago, now i am taking on the project again. but it was a similar concept.

[galomoncayo.net]

Wow, cool. Does it do precise movements?

---

torbjornludvigsen.com

... instead of three motors you can use only two and a spring-loaded rewinder or a weight at the free end of the wire.

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Definitely, swapping one motor with a spring would be possible. Saving a motor is a huge win but there are drawbacks.

The main one (I think) is that the printer wouldn't be able to accelerate quickly along the action line of the spring. The printer will be heavy because of all the motors onboard, so it will carry more momentum than for example the Sky Delta.

A smaller issue with a spring is that it would make imprecise firmware harder to detect (lines would never go slack if accelerations are always low). Another smaller issue is that tension would be different depending on position. To get closer to constant-force springs, we would probably use a very long rubber band and wind it many times around a pulley-system, like on the Sky Delta. Not a big issue, but maybe eating up some og the complexity gain from dropping a motor.

With all that said, I don't really know since I haven't tried. I will have to sit down and calculate how large force the spring must have to enable reasonable accelerations along its action line...

---

torbjornludvigsen.com

Sorry, I read answered as if you had written just "spring", and not "spring loaded rewinder". A spring loaded rewinder would of course be much smarter than just a spring,
since we could use spiral torsion (constant force) springs for winding. Let's limit discussion to springs are made for one full rotation (360 degrees) or less (don't know about realism in this assumption, not my area of expertise). Since the spools on the hangprinter are limited to a ca 80 mm diameter, and we want the lines to be at least 2 m long, we need 8 full rotations. That is, we need to gear up the spring by a factor of 8.

The constant force we need can be approximated like this: If we want acceleration of 500 mm/s^2 when spring is at a 45 degree angle, and the printer weighs 2 kg, we get

F = sqrt(2)*2*0.5

Say F = 1.5 N for simplicity. The spring will need to exert a minimum torque of 8*1.5*0.04=0.48 Nm. Even doubling that to take reality into account sounds totally possible to me. Avoiding the assumptions about the springs and gearings, I imagine retractable dog leash strong enough to lift a mass of ca 300 g in a 2 m span to be entirely realistic.

Avoiding springs all together, weighs at the end of lines would be dead simple, dirt cheap and easy for experimenting with different weights...

To be honest, I'm surprised by these numbers and I'm a litte unsure what to do with it. I'm quite close to actually testing the first prototype now, with 5 motors. If the spring-solution still looks promising when I see the 5-motor protoype move (and test needed forces empirically), I might just try to build the 4-motor 1-spring version! Computational and economical gain seems quite obvious. If you want to try buliding the 4-motor 1-spring before I get the time, and want to fork/branch the repo, please do =) [github.com]

All the best!

Edited 1 time(s). Last edit at 06/21/2015 04:11PM by tobben.

---

torbjornludvigsen.com

Still collecting my thoughts about the weight-at-end-of-line solution, I just wanted to mention a couple of developments in the project:

I'm doing my masters thesis about Hangprinter! So we're rich in potential work hours =D

E3D have kindly sponsored the project by sending me a free Volcano power pack, a hobbed bolt and 2 kg of filament =D Big thank you to E3D!

I have put up some pics an comments over at [vitana.se]

---

torbjornludvigsen.com

... looks really interesting, but too a bit complex for me ... if I'll start with a hangprinter, then it would be a plane 2motor+rewinder stringy at the ceiling (hold magnetically) and three rollos/blinds releasing the printbed

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Hello again. The printer is ready to be mounted in somebodys ceiling... Video

Looking forward to seeing what the prints are going to look like

---

torbjornludvigsen.com

Has anyone done a design where a gyroscope is placed in the effector?

That would reduce or maybe eliminate the need to maintain perfect tension.

@Ezrec: I don't know anyone who have made a hanging RepRap, even without a gyroscope, but it's definitely a good idea!

It moves now. Lines tight and all (for a little while before it crashes into the floor). Check it out: Vimeo.

---

torbjornludvigsen.com

Did a calibration 2d-print today:

---

torbjornludvigsen.com

Hey guys, it works!

video of print on vimeo
See blog for more explanations and stuff.

Does anyone have recommendations regarding driver chips?

I'm soldering on the drv8825s to put them in fast decay mode, like Quentin Harley does here. I would like to get rid of the soldering routine, but keep the smooth running and high current. Would this be achievable with for example SilentStepsticks?

---

torbjornludvigsen.com

... I'm using a RADDS board with RASP128 drivers (1/128 microstepping), which are good for 1.7A continuous or 2.2A with aggressive cooling ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Thanks, hadn't heard about them before. RAPS 128 looks very interesting. I'll order some of them and check it out

---

torbjornludvigsen.com

Congrats! Are you planning on doing scaling tests next?

Yes, I'm thinking about something that's hard to build with other methods, like a really tall auger.

---

torbjornludvigsen.com

Wow! That's excellent, @tobben!
While some people are working at getting printers small enough for their desktop, you're pioneering in the other direction!

Maybe someday we'll have folks building Clercks in smaller frames on their desks. For those who can't dedicate a whole room to a 3D printer

Thanks @maso!

I've done some rough calculations on resulting forces and momentum if lighter Nema14 motors were used. My qualified guess from this is that hanging RepRaps with smaller motors are feasible, so I will try to go for smaller, cheaper more easily frameable in my next design.

I have a really patient wife who lets Clerck take over the kitchen for now, but in a months time I've got to dismount it.

---

torbjornludvigsen.com

Clerck is printing reliably and prints come out good enough for self replication (even with 1.2 mm nozzle). Check out Vimeo and blog.

I see this concept as proven now, maybe I'll put on a smaller nozzle just to be able to comparable level of detail directly to other RepRaps. Final exams are getting close, so development will slow down again.

---

torbjornludvigsen.com

That's excellent, very impressed.

Thanks!

---

torbjornludvigsen.com

Splendid work!
There is much potential in this.

Awesome work !

Thx =)

---

torbjornludvigsen.com

Quote
tobben
Thx =)

By the way, if you ever want to implement an arm solution for this machine-type into Smoothieware, I can send you a broken board ( with like a mosfet or ethernet not functional ) for free.

I have no ARM plans myself right now. If you would like to develop it with me I'd be happy to co-develop using your fixable Smoothieboard =)

---

torbjornludvigsen.com

If you want the board, I can definitely help you with the coding. If you've already done the coding for Marlin, it should be trivial to add to Smoothie. Just tell me.

Hm... I think the Smoothieboard version of Clerck would be best developed by someone "scratching his own itch" (someone who really wants a Clerck with Smoothie and understands the Smoothie benefits well). That's not me, so I won't do a physical build with Smoothie myself.

The moment you decide to do a Clerck+Smoothie build, send be the board, and I'll help you as much as I can. Since you have the coding skills, chances of success are high

---

torbjornludvigsen.com