question about this design

hello,before many years i have spotted a interesting plotter design, i made simple sketch, but i dont know whats name of this design.

it looks simple, and very light, what do you think for xy?

... this is a simple 2D-shaped parallel kinematik setup - here a 3D-type: [forums.reprap.org]

This is common for high dynamic positioning of lightweight toolheads, but you have to re-calculate the angular displacement of the head for performing XY-positioning ...

---

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

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

Welcome to the ACL (Anti Cartesian League!)

Its one type of "parallel robot". I think it would make a great robot design.

Here's my variation on a similar theme. Your concept would be a little more compact though.

HF04

How would you implement the Z axis? Rising table, moving the linear rails in Z or a Z axis on the end effector?

Do a Google image search for "parallel robot" or "parallel linear robot" there are one or two similar designs in the field which may inspire you.

well,

first, i have diy cnc router for few years, so fabricating parts isnt problem.

i look those anti cartesian machines for atleast 10 years

Z axis, well rising table looks ok to me...
in diy I always choose to fix 2 axis to frame, because if i have play in first axis, that play multiplies with play on second axis and so on...

design goal is to make it ultra light and, actually this design is best of both world, simple linear movement, no need for gear reduction (scara), and low mass, so it could be driven with weak steppers or it could be very fast

but bad side is that we have cca 1/3-1/2 of linear rail unused,..

HF, one question, how hard is to make kinematics for design like this? is there any tutorial for that?

for now, i dont have knowledge to do this

I'm building a delta. In time I will form a final opinion about parallel bots vs cartesians.

Right now I see some advantages, we will see how it pans out in the real world.

Mainly- less parts!
Modern electronics allow is the power to run a complex bot and it's math. This means we can get much simpler with machinery while still accomplishing the same tasks (at the cost of needing ever-growing software capability). Parallel arms and deltas are a great demonstration of this philosophy.

Inverse Kinematics are fairly simple for a delta and even easier for this configuration as Z is removed from the calculation.

You probably only need pythagoras rule in this instance as arm lengths (hypotenuse) and distances in X are known.

By the way you have no control of the head rotation of this robot. The extruder rotates around Z as you move in X. I would add an extra arm in a parallelogram format to one linear rail so you have a fixed plate at the head. Its not really a problem if you are only using one extruder, but if you add two or do pick and place work it could become an issue.

All of the linear rail will be used except for a portion at the far end. You can just shorten these and fix the end to the frame at that point. Even better use some openrail or open bearings so you can support the rails along its length.

Edited 1 time(s). Last edit at 02/28/2013 05:45PM by martinprice2004.

yes, for extruder...when hot end nozzle is centered in the middle of bearing, there is no problem, right?

one dumb question...in wich file are kinematics config files? for example i open repetier firmware and in configuration.h are delta settings(arm lengths etc) but i missed subroutine wich calculate that parameters to linear motion-defines delta kinematics..

sorry for stupid question, like i said, mach3 makes me very dumb

I haven't done any FDM printing off my various delta and scara robots, but I cannot imagine there will be a problem if the nozzle rotates slightly when moving across the part. There might even be some benefit.

There is a potential problem if you use a bowden extruder (which would be best for this robot) as there will be twist in the outer bowden cable. This is bad, so you would have to float either the head or extruder (or both) or use a very long cable to give rotational compliance.

I cannot answer the question about firmware as I wrote my own from scratch, but the Rostock Delta I believe is firmware modified from standard Marlin cartesian, so it should be easy to find where the equations are in that firmware. I suggest you download it and have a look (just look for some COS, ATAN and ATAN2 functions!). It would then only be a case of modifying that bit of code with your own equations.

There are many people here who would be able to help.

If fillament is 1.75, OD of PTFE tube is 4mm.
solution is to tight fit little bearing with inner diameter 4 mm to extruder, and glue ptfe to inner of bearing...or slightly bigger bearing to fit one more printed part inside that will tight fit to tube and extruder..one of solutions

i have ordered arduino+ramps, stepper motors, when components arrives then i play with it

also have i think 12 pcs lm20uu from previus failed project and now arriving 10 psc lm10uu