question about axis for cartesian 3d printer

[DdeWaard]

So I have been designing a cartesian 3d printer for a few days now, one that can handle a build platform of (width x depth x height) 500x300x300 in an as tight as possible package, thus have a paper scanner style x and y axis and the z axis for the bed.
So far i've designed it like the ultimaker, (with crossing axis) with the motors on the carriage and the extruder aswell, weighing it down(of course, made stronger with double smooth rods on the axis and 20mm alu extrusions for the frame and it will get cover panels for extra strength and the looks...
I don't really see it coming together since the cariages are so heavy now, it might get inaccurate, especially on such large distances. and i really dont want the bowden tube anymore, i have it on my rostock max and i think a direct drive printer will work better, also for flexy filament.

I have been thinking of either an H-bot or a makerbot style configuration.

With the H-bot, i have heard that the long belt gives the printer a great amount of stress on the belt, especially with such a big printer. and is it harder to control one of these printers, because the x and y axis motors do the same?

here you can see the shorter X-axis, over which the longer Y axis with the head is moving

Now i have really liked the idea of the makerbot, which uses seperate axis as where the y axis rides on the x axis and the y axis is moves the head left to right.
and i have a question about that, why is the head travelling over the longer Y axis and not the shorter X axis, so that the axis the head travels on is shorter, thus stiffer? Are there any benefits to this setup?

What would be your choice? And do you have any tips for me?

Thanks in advance,

Demy

[Polygonhell]

The HBot design has issues, I've built 4 prototypes, if you draw the belt out, and think about the forces involved it becomes obvious that there is a significant racking force across the carriage, and everything needs to be stiff enough to withstand that. The issue becomes the cheap LM8UU bearing people use, they simply have too much slop.

I built a CoreXY prototype, it removes the racking force, but it's very picky about belt tension, basically both belts have to be tensioned the same, or you have racking issues.

I have a couple of ultimaker (etcha-sketch) style prototypes and IMO it's by far the best design for an X/Y stage, the parallel belts means bearing slop is largely irrelevant, they are sensitive to getting the rods exactly parallel and the X/Y stages exactly aligned.

There is nothing wrong with having one axis with moving motors, you'll lose some acceleration, and you'll need to ensure the device can support the weight.

Even at 300mm 8mm rods are barely sufficient, much longer and there is measurable distortion over the length with no load, you'd want to go at least 12mm, and possibly bigger. 20mm extrusions might also be on the small side, I know with 15mm extrusions, I couldn't get rid of the lack of torsional stiffness with ~280mm envelope.

Now having said all of that, I think the biggest engineering challenge you'll have on a printer that size of the moving Z axis, you'll need to either support it from both sides, or get enough vertical separation on the fulcrum to remove the bearing slop.

[Eaglezsoar]

So I have been designing a cartesian 3d printer for a few days now, one that can handle a build platform of (width x depth x height) 500x300x300 in an as tight as possible package, thus have a paper scanner style x and y axis and the z axis for the bed.
So far i've designed it like the ultimaker, (with crossing axis) with the motors on the carriage and the extruder aswell, weighing it down(of course, made stronger with double smooth rods on the axis and 20mm alu extrusions for the frame and it will get cover panels for extra strength and the looks...
I don't really see it coming together since the cariages are so heavy now, it might get inaccurate, especially on such large distances. and i really dont want the bowden tube anymore, i have it on my rostock max and i think a direct drive printer will work better, also for flexy filament.

I have been thinking of either an H-bot or a makerbot style configuration.

With the H-bot, i have heard that the long belt gives the printer a great amount of stress on the belt, especially with such a big printer. and is it harder to control one of these printers, because the x and y axis motors do the same?

makerbot rep2axis.jpg

Now i have really liked the idea of the makerbot, which uses seperate axis as where the y axis rides on the x axis and the y axis is moves the head left to right.
and i have a question about that, why is the head travelling over the longer Y axis and not the shorter X axis, so that the axis the head travels on is shorter, thus stiffer? Are there any benefits to this setup?

What would be your choice? And do you have any tips for me?

Thanks in advance,

Demy

My best tip for you would be to take that Makerbot to the Range and see if you can put a .45 caliber slug through the o in Makerbot.