Flat LISA variation, wall mount...

Flat LISA variation, wall mount...

Postby FMPrime » Sun Jun 29, 2014 12:36 am

It looks to me that the Lisa Simpson design could be built with all three lead screws on one plane, say mounted against a wall. The pyramid would be skewed down to the extruder point. I think the motions still work out, but the math adds up differently.

Here is a quick paint sketch to illustrate my idea.
3d printer idea.png
Black: lead screws, Red: arms, Blue: hinges, Yellow: build platform.
3d printer idea.png (6.94 KiB) Viewed 45653 times


Being mostly flat, I think it would be handy as a wall mounted printer or possibly in a briefcase build.

Please let me know what you think. Is the concept even feasible? I may eventually try to build one, but I don't have any 3d printer hardware or experience. I do like twisting geometric concepts in my head though, so I should probably get a 3d printer.

Cheers,
~Dan
FMPrime
 
Posts: 4
Joined: Sat Jun 28, 2014 11:58 pm

Re: Flat LISA variation, wall mount...

Postby Nicholas Seward » Sun Jun 29, 2014 1:01 am

Great idea. This could fold up very easily. Let me know of you need any design help.
Nicholas Seward
 
Posts: 738
Joined: Mon Nov 25, 2013 10:41 pm

Re: Flat LISA variation, wall mount...

Postby Dejay » Fri Jul 25, 2014 6:13 am

Wow nice idea! Would this work?
Why has nobody tried this before?

Shouldn't this even work with a rostock/kossel style wheel on extrusion carriage?
Dejay
 
Posts: 11
Joined: Tue May 27, 2014 1:34 am

Re: Flat LISA variation, wall mount...

Postby Nicholas Seward » Fri Jul 25, 2014 7:06 am

I tried to dig up a few videos of deltas with this arrangement but I couldn't track them down. It is most definitely possible. Extra care needs be taken with the singularities, work volume, and resolution.
Nicholas Seward
 
Posts: 738
Joined: Mon Nov 25, 2013 10:41 pm

Re: Flat LISA variation, wall mount...

Postby sane » Fri Aug 01, 2014 11:27 pm

Correct me if I'm wrong, but couldn't you mesh the two arms of Wally for the XY action and two lead screws for the height? And since the z speed would no longer be an issue you could go for a tighter thread.

Furthermore, couldn't you have a calibration laser/sensor to characterise any deviations in the lead screws with Z to allow for any old threaded screws? Maybe even just one screw? Sounds like that would be cheaper.

Wally+Lisa = Lilly ;)
sane
 
Posts: 12
Joined: Tue Jan 14, 2014 9:47 pm

Re: Flat LISA variation, wall mount...

Postby Dejay » Sun Aug 03, 2014 6:09 am

I wonder if a "flat gus" variation would work, so you have a sort of robot arm instead of a enclosed space.

Not sure how the accuracy is with delta arms so long relative to the "delta radius". And to avoid a lot of torque on the lead screws, you might also have to have one arm longer than the others, or avoid positions where all three arms are flat horizontally.

Computer vision to track and measure the position / orientation of the effector could be great for calibrating away all sorts of errors. But it would require more CPU load on the controller even after CV calibration for the inverse kinematic. You'd either have to come up with a machine specific kinematic that accounts for wobbly lead screws and solve it, or use some kind of grid or interpolation scheme to adjust the errors. This could even adjust for something like robot arm droop.
This might be awesome if we ever come up with a way to 3D print a complete printer fully assembled including motors. The result might not be very precise, but you could calibrate the errors theoretically. Well except backlash maybe. But even backlash could be measured and maybe compensated for.

The STEM system (a motion capture system / 6DOF game controller) could be easier to use for this than CV (http://sixense.com/wireless). Probably won't work because it uses an electromagnetic field to measure the position.
Dejay
 
Posts: 11
Joined: Tue May 27, 2014 1:34 am


Return to LISA Simpson

Who is online

Users browsing this forum: No registered users and 1 guest

cron