From RepRap
Revision as of 15:45, 24 November 2011 by Glenn (talk | contribs) (+Category:RepStrap)
Jump to: navigation, search


Scanz is a RepStrap, essentially from old flat bed scanners. It is also an experiment to see if a cartesian robot with a plastic extrusion end effector (proper term for extruder) can be made without access to a RepRap, 3D printer etc, but a simple drill press.

Break down of the project

Extruder, is also made from junk, though it has so many iterations, its more machined than was originally intended. Thermal break is BakeLite, essentially the same plastic that is used for pot and frying pan handles. Scanz thermal break is a pan handle, bough specifically to try out the plastic, cost $15, to get the same length in peek $100 easily. The nozzle and fibreglass coated nichrome wire was bought from MakerBot. The barrel my friend made for me on his work lathe.

Unlike Teflon (or PEEK for that matter) BakeLite gives off no fumes when over heated, which is a huge bonus, as heating PTFE above 250 degrees celcius, gives of highly toxic fumes, PEEK is better, BakeLite is very cheap.

The extruder is a very basic pinch wheel on a stepper NEMA 17 pushing up against an idler bearing under lateral force by a spring tensioner.

Z Axis is quite literally the drawe rails you can buy from any well stocked hardware store, very cheap. They are mounted in parallel to the sled as opposed to how they are in drawers. This simple method saved my sanity when I figured it out. The sled plates are fibreboard that I found lying around at work, cut down and drilled where necessary. Its so well done I don't actually need lateral bearings to keep good contraint.

The X axis is a flatbed scanner with a plate that will get glued on or screwed on, not sure yet.

Y axis is above the bed and holds the Z axis, and is currrently a flatbed scanner, though this is about to change due to under constrainment on it by the twisting that occurss during movement and stopping due to the pendulum effect. this method is ok for anyone wanting to build a cartesian robot, but forget it if you want to do anything precise, like 3D prining sub 1mm resolution.

The Electronics are Gen 3 MakerBot running RepRap firmware, thus I running RepRap host software. Reason for this is to support the open source movement. At some stage I will make a RepRapped shield for the Arduino Mega for the Mendel that this machine will make..

Communication will be done via Bluetooth, because unlike most people I cannot have my PC adjacent to the rpinter, and there is a walk way between them, thus Bluetooth has proven to work for data transfer.

Bluetooth communication is done with a BlueSMiRF, which had to have its baud rate reduced to 9600, and found that the tool on to be the best tootl for doing this. I already created a page on how to setup your module for this [here]


Initial Conclusions have found that it is theroretically possible to make a sub 1mm resolution 3D printer to make parts for a RepRap from scavenged parts, though I have yet to try if extrusion hot ends can be made by drill press alone, though others have tried and succeeded.

My project is just not showing these reults to a reliable standard.

Lessons Learned

Indeed if I were to try this again, having an X Y moving bed is the better optin, using just the drawer rails for constrained movement, and use the parts from inside a scanner to drive those sleds. The sleds would be designed in the same way as my Z axis is now, but instead with belts.

Using just sleds like I have mentioned would make the device look like the McWire desin.