Printing pipe idea and questions.

Hi,
I am new to 3d printing but am a Fabricator so see the value and can see several ways I want to apply this. I went the punk route to get my foot in the door and bought a Solidoodle 3. I love the nature of the reprap machines and see possibilities to produce some specialty tools that are priced out of my range now. But I wanted to be printing while I am learning to build a reprap.

I want to be able to print short lengths of pipe, say a few feet to use with printed connectors to build tinker toy frames with. Basically for anything I need a frame for, green house to swamp buggy.

I think being able to print a thick outer solid fill skin while manipulating the center structures could produce pipes with good rigidity and strength compared to standard pvc pipe but stay light weight.

Here is an idea I have to adapt a small reprap to the task.

Fix the bed in the machine at the top of its z axis after cutting a hole in the center of the bed. Mount a pipe vertically bellow the hole. Cut slots in opposite sides of the pipe and mount a horizontal plate in the pipe that can come up flat with the bed and lower down into the pipe as you print.Belt and pulley system with a belt cleat on either side to use the z axis steppers to lower it decently flat. The outer pipe should help slow the cooling process but let it start to set up as it lowers so the weight doesn't deform it. Part of why I want to do it this way is square feet of shop space. Taller machine is better than a wider machine for my needs.

Questions,
1 How well do the layers bond? Would it be better to print the pipe horizontal so the layers run lengthwise?

2 Say my pipe is 3" dia and an over all fill of about say 50% with a 1/4" solid skin. Will it slump under its own weight?

Thanks and thanks for everything this community has already done.

Wow prepare for some long prints. And I LOVE the framing idea. A lot. Reminds me of when i was a kid in preschool and we had these giant 12" x12" blocks you could make 3d buildings out of. As for the idea, yes the frame will be rigid, not a chance it will slump at 50%, and you should never really go larger than 50% infill unless the wall is <2mm thick. At 5mm, you might use a 1 mm perimeter (so the 3.X mm inside would be 50% infill)

Your project is a perfect example of why I'm trying to make a new kind of printer. You are going to run into 3 things that will annoy you and you'll want to improve:
1) Print speed - Right now, the nozzles are designed for way too high resolution for such a task. Printing at 0.35mm, you'll need 8-12 hours to print such a piece at 6 inch length. I've made some.
2) Layer adhesion - You will find that a strong force in the right place can shear apart layers. The plastic people print is never hot enough. Case in point, injection molding processes are at ~50-80C hotter temps!!
3) Cost - You will find, with $30-$50 a roll being typical, this gets pricey, fast. Like $5 a pipe fast.

Please look for my work in the next few months, because I aim to make a pellet extruding fast printer than can do better layer adhesion with 300C temps. Using pellets is ~ 1/10th to 1/20th the cost, and to date, no one has succeeded as a commercial product yet. Hopefully we think the same

Edited 1 time(s). Last edit at 03/21/2013 12:14AM by Simba.

---

Measure once, Cut twice, Print 3 times.

Here is a pipe I made, about 4" long. And at 100% infill with a whopping 1.0 mm diameter nozzle, this took 4 hours.

Edited 1 time(s). Last edit at 03/21/2013 12:13AM by Simba.

Simba Wrote:
------------------------------------------------------- As for
> the idea, yes the frame will be rigid, not a
> chance it will slump at 50%, and you should never
> really go larger than 50% infill unless the wall
> is <2mm thick.

This may be true with ABS but in PLA it is not an issue.

> 1) Print speed - Right now, the nozzles are
> designed for way too high resolution for such a
> task. Printing at 0.35mm, you'll need 8-12 hours
> to print such a piece at 6 inch length. I've made
> some.

So what? Who cares how long it takes if it is high res. Its not like you stand there the entire time. You have a robot making parts and if you want 200 parts a day that take twelve hours each you build 10o robots not try and make one do 100 times the work.


> 2) Layer adhesion - You will find that a strong
> force in the right place can shear apart layers.
> The plastic people print is never hot enough.
> Case in point, injection molding processes are at
> ~50-80C hotter temps!!

Actually if you look at the data sheet for 4043D PLA it has an injection temperature almost exactly the same as we use to print. I have also had many objects break in other directions before it breaks in the layer direction.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

> So what? Who cares how long it takes if it is high
> res. Its not like you stand there the entire time.
> You have a robot making parts and if you want 200
> parts a day that take twelve hours each you build
> 10o robots not try and make one do 100 times the
> work.
>

For the above, it is critical to the OP. He wants to print part after part after part to make a frame. If it is 10 hours a print, it could take weeks to make a wheelbarrow for example, this is a huge issue. The home user can't afford 100 robots so we need to make it faster, surely you can agree with that?


I did not know that about the PLA molding temps. For traditional plastics, like Nylon, it can be in the 280C+ range! IMO PLA is awesome when done right, and quite unique because of the hysteresis on cooling compared to other plastics. That hysteresis probably is what allows the much more normal melting temps, because it can still fuse after it has started cooling, whereas traditional plastics need extra pressure to fuse by comparison.

I agree with stuff breaking in the non-Z direction depending on the forces involved. plenty of studies have shown that FDM/FFF parts are weaker in one direction than the other. Not a shocker, I'm just saying.

I actually meant to attach the link to the PLA data sheet [www.natureworksllc.com]

I do not think more at a time is the correct way to go. No matter what we will be limited to the time it takes for X volume of plastic to cool before we put more on top of it and this time takes longer the thicker the layer because of the higher thermal mass. If we print low layers at higher speeds we can still print the same volume of plastic per second and the time it takes to cool is less because of how thin it is. This means when we reach the maximum volume of plastic we can put down per second we will be better off to do it with a small nozzle at high speeds and keep the resolution then to go with large nozzles and thick layers that take forever to cool and loose the resolution.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver