A 3D printer… Part 1


Custom 3d Printer 

This is the start of a series of pages on the design and construction of an experimental 3D printer.


The goal ultimately is to build a large printer capable of printing entire large (approx 100cm*40cm*40cm) items in multiple materials.  The first printer however, is much smaller and really an exercise in acquiring the knowledge and experience necessary to build a much bigger printer with a minimum of silly (and/or expensive) mistakes.

A somewhat extensive survey was performed on off the shelf and kit 3d printers at this point, and some conclusions drawn.

  1. Though an important step and enabler for the amateur 3d printing industry to take off, 3d printers made from 3d printed parts fall short of being good robust useful tools.
  2. Mechanical shortcomings of printed 3d printers such as reprap notwithstanding, the software and electronics generated by such projects is invaluable and can be reused in custom designs.
  3. Off the shelf printers of supposedly higher quality than the various kit varieties are *expensive* and sometimes not a great deal better than a well constructed kit!


There was nothing for it, I was going to have to design my own completely from scratch.   Some designs were quickly drawn up and thought through.  It was decided early on that the design should be as simple as possible, even if that meant incurring extra expense, simplicity was worth it.    The machine is to function as a test bed for different 3d printing heads and so much be strong and durable, so as to provide a consistent testing environment.  The electronics must be robust and securely enclosed in its own housing.


The basic design is based on a 25mm thick slab of aluminium onto which the Y axis linear bearing is mounted.  The bearing used is a crossed roller unit that was picked up cheap from ebay.  The Z axis is composed of 2 cylindrical 20mm rails and matching bearings commonly sold by numerous ebay retailers catering for the CNC hobbyist.  These rails mount on the base plate, either side of the y axis assembly and a smaller plate sits on the top, connecting them.  Two rails are used for the bearing, but a further 2 are mounted at the rear of the baseplate in the same orientation as the bearing rails, to add rigidity to the whole assembly.  The X axis bearing and actuator is mounted on a plate that runs between each of the Z axis bearings.


I had planned to use a wire drive system for all the axes but on further investigation it became apparent that that was too complicated. Multiple turns of wire on the pullys are needed to get a good grip, the turns then wind themselves up and down the shaft and change critical angles as they do so…  the mechanics of the whole assembly started to look a bit too complicated.  I gave up in the end and resorted to timing belts for Y and X axese, Z is driven by a ballscrew as it has the extra weight of the x axis to bear.

All axese are driven by stepper motors.  There didn’t seem much point using servos for a 3d printer and the cost both of the motors and drivers is prohibitive

At this stage I dived in and started building stuff.  In retrospect I should have been rather moth thorough in the design as little details neglected at that stage resulted in many requirements for complicated precision parts.  A lesson to learn for the bigger one for sure.


The next part will contain some pictures and focus on aspects of the mechanical design. 


 Posted by at 10:37 pm