Noob seeking information to get started.

Hello folks, First I'd like to Introduce myself.

I'm a student of precision machining, and I will most likely be building a CNC router over the next year. I am also very interested in the reprap and I am a open source enthusiast. I am currently a CNC operator for a living, and will be able to make my own hardware for my project.

Frankly, I'm a busy guy and all the information I seek to get started seems to be decentralized, with some lingo and conversation that I don't quite grasp yet. (board names, controller info, 100 places to buy things I may or may not want etc.)

The rough plan at this point: To build a modified and upscaled aluminum and steel mendel, or a derivative of such.

My programming experience is limited. I am familiar with G-code, dos batch scripting, some basic shell scripting, and that is about it. I run windows xp, windows 7, and last but not least ubuntu.

I can solder, and I have a good multimeter, but I'm no electrical engineer.

My understanding of computer hardware is basic, I have built my own computers since around 1998, but I have never gotten into assembly language or fully understood chipsets.

QUESTIONS: (please feel free to point me toward more information, I don't expect you to author a whitepaper in my thread on my behalf.)

What are the length limitations, and why?

Why belt drives? I am thinking about using acme threaded screws and backlash nuts to gain accuracy that thus far I have not seen from any builds. Can someone show me snippets of code that dictate travel vs. rotation of the stepper motors? I'd need to re-code portions to achieve this. (id est: if 1/4 rotation equals 0.100" of travel, I need to be able to recode it to match my thread pitch, which may be more or less travel per R.)

How much torque can the common stepper motors used exert without issue?

Why do some repraps have heated tables? I imagine it is to keep the plastic fusion uniform. I see an issue with this as I may build a larger model, it may be more difficult or at least costly to heat. Any suggestions?

What is up with the outdated CPUs being used for firmware? Has anyone been successful in building more powerful boards? Is there even a point?

When the G-code is programmed, do feed numbers represent the xyz travel as in a standard cnc? And does the s code dictate the volume of plastic?

Basic plan at this point: I'd like my print envelope to be at least 2'x2' if not larger. I have no issue making the machine rigid enough for this. I'd like to use acme threaded screw drives rather than belt drives. I would like to be able to swap out the plastic printing head of a small router for finish work. I'd like my tombstone(table) to be capable of fixturing work, including but not limited to the plastic parts generated.

Thanks for putting up with me!

Welcome! Sounds like you have all the relevant skills. You'll be fine.

Sluark Wrote:
-------------------------------------------------------
> What are the length limitations, and why?

None, just a compromise between cost/time. There are a couple of repraps around that have larger build areas, but build time for really large models becomes an issue.

> Why belt drives?
Fast, cheap, accurate enough when using microstepping stepper motors. Leadscrews are certainly useable, but are more designed to push a tool through a resistive material, which belts would struggle to do. But reprap doesn't have much weight and is depositing a material, not pushing through it.

> How much torque can the common stepper motors used exert without issue?
Depends on the stepper motor. Here's the specs for the NEMA17 I use, which are the top end of torque needed for Mendel class repraps.
[www.slidesandballscrews.com]
Here's the reprap wiki page on stepper motors: [reprap.org]

> Why do some repraps have heated tables?
For keeping prints flat and allowing them to cool uniformly. The larger the print, the more likely it will warp. ABS is worse than PLA. I print PLA on blue painter's tape without a heated bed at the moment.

> What is up with the outdated CPUs being used for firmware? Has anyone been successful in building
> more powerful boards? Is there even a point?
Arduino is cheap, does the job and is well developed. The community will always welcome more powerful CPUs, but it will increase the complexity of software solutions, and there has to be a significant advantage to such a move to make the development worthwhile. Most CNC machines run on Centronics printer interfaces - and you call reprap backward?!

> When the G-code is programmed, do feed numbers
> represent the xyz travel as in a standard cnc? And
> does the s code dictate the volume of plastic?
Yes. But s code is used for command parameters such as heater temperature, e code is for extruder stepper motor control (mm of extruded plastic). Here's the reprap gcode wiki dictionary page. [reprap.org]

> Basic plan at this point: I'd like my print
> envelope to be at least 2'x2' if not larger. I
> have no issue making the machine rigid enough for
> this. I'd like to use acme threaded screw drives
> rather than belt drives. I would like to be able
> to swap out the plastic printing head of a small
> router for finish work. I'd like my
> tombstone(table) to be capable of fixturing work,
> including but not limited to the plastic parts
> generated.
Sounds great. One of the principles behind reprap is self-replication, hence the use of plastic parts, but it is just one branch of fused deposition modeling (FDM) class printing, which is one branch of 3D printing (http://en.wikipedia.org/wiki/3D_printing). But there's no reason that they can't be built out of T-slot or converted router tables etc. Leadscrew will work fine, and are suited to multi-role machines. They may not have the speed of belt drive, which is approaching 150mm/s printing and 300mm/s travel speed. [www.mendel-parts.com] Though it would shake my printer to pieces!

Hope that helps!

Edited 1 time(s). Last edit at 07/16/2011 06:12PM by droftarts.

belts simply allow faster travel than threaded rods. If you can get a threaded rod and stepper setup to go fast enough for you go for it. Everyone knows it's doable but more expensive.

heated beds prevent warping of plastic.

if you can keep a 2' x 2' stable you'll be fine. Parts that large would take a long time to print, even with the fastest speeds people report. At some point you'd have to trade off detail for speed, if you extrude larger you can print bigger parts faster.

Arduinos are a natural choice for translating gcode commands into stepper movement, but not necessary, there are quite a few people who have done other things.

My suggestion is to build a standard reprap for as cheap as you can do learn what's going on before doing the larger project.

Thank you for the replies, I believe this has satisfied most of my curiosity at this point. I'll bump this thread later when I find myself asking more questions.

I know of screw machines which can run quite fast, but the stepper motors they use would be double the cost of a mendel kit (each!) I may just follow that advice and build a standard mendel, or 1.5 scale to get started, and then sell it off to a friend for cost when I decide to upgrade.