Yet another silly idea

that's when you don't have your password in mind... you have to use another stupid name

Okay, nevermind.

I just ripped apart some window lifter I got hold off,
or was it for a rooftop? I don't care...

In fact it lifts the window with two bowden cables,
that seems to be a nice idea for a limited movement.
Bowden cables are cheap (even compared to leadscrews) and if it can moves up a window some 30cm it should also move an Axis of the cartesian, no?

Just some pulley for winding the core wire and two "feeding arms"
(I guess that's all printable the one I do have right now is made of PP)
Here's a picture of another cabledriven windowlifter just for illustration
(until I've shot a picture of mine )


Why do I mention this at all?
Well I think if we are able to place our motors wherever we want, we could reduce the moved mass resulting in smaller motors and cheaper cartesian bots.
To be honest, I don't know if thats viable at all,
but it looks nice from my point of view

What do you think?

'sid

Hi sid,

look here:


and the atached images in this post: [forums.reprap.org]

In this system i fixed both motors (7) at the base and moved the laser-head with steel-wires in a pulley-manor in X and Y ...

Viktor

Edited 1 time(s). Last edit at 08/31/2008 03:42PM by Viktor.

Seen that before

But what I thought of is even simpler than that.

The motors don't have to be connected to the cartesian at all (except by the bowden cables of course)

I'll make a quick sketch...

(ten min later)


Sorry quick and dirty
Just think of how your bicyclebrakes work.
'sid

[edit2] edit picture

Edited 2 time(s). Last edit at 08/31/2008 07:46PM by sid.

Hi sid,

... main problems here would be first: - the friction in the hoses and second: - the displacement of the winding-path on the winding-axis.

For relative short ranges [second] wouldn't be a problem, but when you want to move longer distances, then you have to wind some ten to houndred turns and the wire-coils would 'travel' along the roll, so it's not linear in tension/length and can 'overlap' at the outer positions ...

But nevermind, in my laser-plotter i solved this with thicker rolls, a bigger distance to the first 90

Well,
The movement of the original Mountingplate that was moved by the motor
was about 34cm so just a bit more than needed than needed for a standard darvin for example.
The wheel has a diameter of about 40mm and is capable of holding six windings
and because it's about (Pi*40) 125mm per winding it'll hold up to 750mm of wire.
I think that is more than enough

The wheel itself is about 14mm wide So the maximum displacement of one wire is
~14mm for 75cm movement
~7mm for 37cm movement
~6mm for 30cm movement

The wire guides of the wheel housing are 70mm long (without the throughbolt)
that's an error of 0.08mm for 6mm displacement
The only thing I would consider "bad" is the resolution. because it's 0,3125mm/step (@400s/rev)
So I think we need some microstepping here (maybe one day we can )

Friction.
Good point, The friction of the original wire/hose was pretty small but due to a lot of grease.
Unfortunately I had to cut all the original wires, so I can't tell exactly.
But, I had ptfe hoses on my bike for the bowden cables of the clutch; not the cheapest you can get, but from my point of view they do help a lot!
(at least on a motorbike) maybe we can deal with the friction that way.

I will have to give it a try (as soon as I'm back from vacation that is)

'sid

[edit]made a mistake..

Edited 1 time(s). Last edit at 09/01/2008 06:38AM by sid.