Vertical H-bot with x and z axis?

Hey everyone

I've been designing a compact reprap, and have found that the simplest and easiest way for me to keep things compact is to use a H-bot vertically. This means that instead of having the H part on top of the bot, it would have it sideways, driving the x and z axis instead of x and y. So sort of like a prusa, except energy from the Z axis is not used to move the heavier y carriage.

My main concerns are...
Will this require a lot of torque and will it even stay without the extruder dropping down?
Will it be less accurate because of the lack of lead screw?
Has this been done before?

Any help you could give would be greatly appreciated

Edited 1 time(s). Last edit at 02/27/2013 04:22PM by j4ysen.

[www.buildlog.net]

Ahh thanks, I had stumbled on the ord bot before, but not this one!!

Thank you so much for your help

For what it's worth, I think coupling an axis that is mostly static (Z) with an axis that moves a lot (X) is a horrible idea. But to each their own.

crispy1 Wrote:
-------------------------------------------------------
> For what it's worth, I think coupling an axis that
> is mostly static (Z) with an axis that moves a lot
> (X) is a horrible idea. But to each their own.

I agree, mechanically it just seems a little off.

However I suppose the rostock couples z X and y and that seems to work well.

Quote
konwiddak
However I suppose the rostock couples z X and y and that seems to work well.

No, Rostock has A, B, and C rails with stepper motors.
Rostock does NOT HAVE any X, Y or Z stepper motors.

---

Bob Morrison
Wörth am Rhein, Germany
"Luke, use the source!"
BLOG - PHOTOS - Thingiverse

I did suggest this configuration in another post, but on reflection, a H-Bot might be better for XY axes rather than XZ.

Mechanically a H-Bot is probably less accurate than one with separate belts for each axes as the belt is effectively twice as long (or at least twice as long). The belt stretch and slack will make the positioning less accurate. OK you gain a bit by the doubling of the belt runs, but I can't imagine this to be much benefit. The errors probably wouldn't cause much of a problem in XY, but when printing with fine layer heights in Z this could be a killer.

There is an industrial T-Bot in the video below which is similar to what you are doing, but I think its only for pick and place operations where accuracy is less important.

T-Bot

Having said all that, it would be interesting if you did build this to see if it was feasible. Even if it wasn't useful for FDM I am sure the design would be of interest as a lightweight milling machine or pick and place device.

rhmorrison Wrote:
-------------------------------------------------------
> > However I suppose the rostock couples z X and y
> and that seems to work well.
>
>
> No, Rostock has A, B, and C rails with stepper
> motors.
> Rostock does NOT HAVE any X, Y or Z stepper
> motors.

The head moves in XY through a coupled movement of three motors. These three motors are orientated for linear movement in the z direction. Circa you have coupled your XY motion to a motion in a z direction, which at the head you want to keep constant.

Therefore on paper it looks even more balmy than a vertical h bot, but actually works very well.

I still don't think a vertical h bot is a good idea though!

I'm not as expert as the other people here on movements and coupling, but I would worry that printing an object on a vertical base would allow the object to have some sag due to gravity as it cools during the printing process. Not sure how much, or how big of a deal this would be, but it is a concern, and more so as the printed object gets bigger and farther from the base plate. A "tall" enough object might extend out far enough so that the torque due to gravity (at the plate-object junction) might be large enough to have the object fall of the base during printing.

Regards,
aeronaut

I know this is a dead thread but I feel that it is important that I have implemented a bot that uses the CoreXY arrangement for the XZ plane. I call it CoreXZ.

CoreXZ Video

Some of the above comments suggest that coupling a slow moving axis with a fast moving one would be a bad idea. I avoid that by wanting the z axis to be fast moving. I want to start doing non-planar slicing and this along with GUS can be a good test bed.

My main motivation was to get rid of a stepper and to try out my pulley idea. I use 2 608s sandwiched together to simulate a grooved pulley.

I wasn't expecting it to be a great printer but it out performs my expectations. There is some ever so slight banding because a layer equates to a 1/2 step instead of a full step. The effector also drops when the power goes off. I like the design so much that I am redesigning it with smaller pulleys to reduce the banding and a passive ratchet mechanism keep the gantry from falling. I will make 4 of them for my school. The frame is laser cut and assembled with zip ties. I use very little hardware. I was able to get a print volume of 200x200x200 out of 290mm rods. I use no limit switches. I just set the effector on the bed where I want to print and then print. If I am trying to print something big, I put the effector in the bottom left corner and make sure I slice appropriately.

Blah, blah, blah. I just though you guys/gals would find this interesting.

---

ConceptFORGE
Wally, GUS Simpson, LISA Simpson, THOR Simpson, Sextupteron, CoreXZ

very interesting nicholas . Thank you for this proto. I had this idea and was wondering how it would work. I see it does the job

that makes me win lot of time . cool .

---

the Smartrap project

[smartfriendz.com]
[www.thingiverse.com]
[reprap.org]
[github.com]
doc assemblage: [reprap.org]
NEW: Forum smartfriendz: [smartraptalk.smartfriendz.com]