New fixed motor movement system?

So, I was derping around looking at Core XY machines to build (Specifically HyperCube EVO), and I started to think "Hey, crossing belt CoreXY is nice and all, but the belts getting misaligned is a bit of a pain". Then I also looked at the HEVO's approach but noticed that not crossing the belts induces a torque onto the extruder assembly. So, I thought, "Hell, why not try to design a new system?".

And thus after a bit of tinkering I made this:


The main idea here was to balance the forces as much as possible. Since the gantry is connected to opposite symmetrical points, as well as at the same level of the center of mass, the rotation alongside the X axis should be non-existent.
Then there's the X carriage, which is connected right at the center to reduce all rotations along the Y axis. This in turn would allow the use of rods without having to worry much about lubrication or binding.
I'd like to receive any and all feedback, as this would really help me decide weather it's worth trying to build something like it or not.

Edited 1 time(s). Last edit at 09/05/2018 12:20AM by Heracles421.

There's an error in your drawing.
I marked the belt in question with a red arrow. It is fixed in length and hinder the gantry to move 'down' or it'll go slack, when moving 'up'

... the other belt too - there's no compensation for displacement ...

---

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

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

Woops, I screwed up there, hehe. I guess I'd need to have the belt fixed on a single point, thanks for noticing

Hmm, after giving it a second look I now notice this is totally flawed... Back to the drawing board

Edited 1 time(s). Last edit at 09/05/2018 07:51AM by Heracles421.

Ok, new design... Any thoughts?






Edited 1 time(s). Last edit at 09/05/2018 02:00PM by Heracles421.

Now it seems right, but the kinematic is a mix of H-Bot and CoreXY.

dX = dA ( red belt )
dY = -dA + dB ( blue )

I don't know if it's worth adding extra length to the blue belt and get more stretch/ringing there, just to fight extruder tilt. ( which is neglectible when using linear rails and carriers)

edit: I just saw that both belts are equal length, so forget my above comment. It's even worse: both belts are extra long which is a big trade off IMHO.

Edited 2 time(s). Last edit at 09/06/2018 12:56AM by o_lampe.

What problem is being solved here?

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Quote
o_lampe
Now it seems right, but the kinematic is a mix of H-Bot and CoreXY.

dX = dA ( red belt )
dY = -dA + dB ( blue )

I don't know if it's worth adding extra length to the blue belt and get more stretch/ringing there, just to fight extruder tilt. ( which is neglectible when using linear rails and carriers)

edit: I just saw that both belts are equal length, so forget my above comment. It's even worse: both belts are extra long which is a big trade off IMHO.

Yea, that's the one downside I can see to it, belt stretching and adding inconsistency. I wanted to use something like this for linear bearings more so than rails.
I can't think of a way to make the belt any smaller while keeping them in a cartesian-esque configuration...

Edit: Actually, now that I look at it, the belt lengths are exactly the same as with either an H-Bot or a CoreXY, the only difference being you have to sets of belts (one for each axis) rather than a single belt for both, but it shouldn't really be that much different from either configuration. If anything the belts need to stretch less as they endure half the force. However, I just noticed that having that H-Bot esque config beats the whole purpose of this system, as it applies a torque on the frame...

Quote
the_digital_dentist
What problem is being solved here?

Trying to figure out a belt system that has a cartesian configuration (unlike the h-bot and coreXY) so that there's no unbalanced forces being applied where you don't want them to (for example, rotational forces), but without having the motors move like in a traditional cartesian configuration

Edited 2 time(s). Last edit at 09/06/2018 09:42PM by Heracles421.

Do these "unbalanced forces" cause problems?

With careful positioning of pulleys, a single-level corexy mechanism could be made free of "unbalanced forces", I think.

OTOH< if you want to try to come up with a new belt layout...

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Quote
the_digital_dentist
Do these "unbalanced forces" cause problems?

With careful positioning of pulleys, a single-level corexy mechanism could be made free of "unbalanced forces", I think.

OTOH< if you want to try to come up with a new belt layout...

Depends on what motion system you use, but for rods and extrusion rails they can cause problems. Not so much with linear rails due to the tolerances they're made at.

And yes, a crossing belt CoreXY can be made balanced, but the crossing of belts and the fact you use a single motor to move in two axis is what I don't like about CoreXY

Edited 1 time(s). Last edit at 09/07/2018 09:14AM by Heracles421.

Have a look at the Sl3dr by richrap, ultimaker/quadrap or the XY stage of an etch a sketch. All interesting mechanisms that (i think?) fit your goals. I've looked at them extensively for similar reasons as you.