Replacing the threaded rod

I noticed that there is a plethora of threaded rod on the reprap. If I understand this is to provide rigidity. Has anyone considered using a much larger polymeric tube? An example of this is a square acrylic tube of 20 mm width and height with a wall thickness of 3 mm could support the same as a 8mm diameter steel rod. The tube could be assembled by either printing it. Or if its a laser cut repstrap, by cutting it and assembling it the same way the print head base is.

What do you think?

Theory
Deflection is dependent upon boundary conditions, matierial prop (E), and geometry (I). So assuming we want the same rigidity in the system then E_steel*I_rod = E_acrylic*I_tube because the second moment of area increase with a factor of h^4. Because of this if the section is hollow or not it doesn't matter all that much.

Could it work? Yes.
Would it be cost effective? I dont' think so.

Personally I would rather have to cut the steel rod twice then have to glue 4 pieces of acrylic together and then have to worry about the glue joints breaking.
Plus, acrylic (or any non thermoset plastic) tends to warp over time when under load.

Although I have considered using thick wall PVC plumbing pipe.

I agree it probably would not be cost effective. Let's face it steel threaded rod is dirt cheap and pretty easy to use. But if one of the goals of the reprap is to have a 100% self replicating machine then it provides a means to do so. Instead of gluing the acrylic together you could screw it together (as done now in the laser cut version for the print head holder) or print it in sections and assemble it together.

As far as creep goes I don't think it'll be as much of a concern as you think. Creep is a stress-time relationship. While the structure needs to be stiff the stresses are actually pretty low over time.

Wow, someone that knew I was talking about creep. :-) You must be an ME.

Anyway, it would work. For the long term replicatable parts, I was considering printing out 6 inch (15 cm) long beams with joints on the end to let the beams snap together like puzzle pieces or lego pieces - using glue, screws, pins, or whatever to hold the joint. The thing holding me back on the design is knowing the weight that it needs to support. I'll have that weight as soon as I get my tool head up and going.

Cris, et al,

IMHO, if you want to avoid creep, design in a generous safety factor, to keep the strain down. So you could over-estimate now, and get started.

Another thing to consider about plastic structure is its relatively large thermal coef. of expansion. Even heating would cause linear expansion, but uneven temperatures could cause significant bending/warping -- not a good thing for a machine using open-loop positioning.

---

Larry Pfeffer,

My blog about building repstrap Cerberus:
[repstrap-cerberus.blogspot.com]

Well my plan is to finish building the one I'm working on now and then start replacing metal parts with reprapped ones.