Aluminum Extrusion Frame For The win

With the exception of 3d printer kits I’ve bought, all of my DIY projects have employed aluminum extrusions. Specifically, t-slot aluminum extrusions sourced from Misumi USA.

Misumi HFS-15 For Past Projects

In previous projects, I used their HFS15, or their 15mm profile aluminum extrusion. Their 15mm profile aluminum extrusion is a great choice for small projects for a number of reasons: M3 bolt pairs well with it, M3 low profile square and hex nuts work well as t-slot nuts. The extrusion is VERY inexpensive. Misumi USA will cut the aluminum extrusion to length and will even handle drilling various holes as required.

Misumi Series-5 20mm Profile Aluminum Extrusion

It’s amazing what 5mm in each direction will do for the rigidity of the aluminum extrusion. In addition, with the larger profile, the t-slot channels for the nuts are likewise wider and deeper, the clearance for the bolt is wider, and the amount of aluminum contributing to the overall rigidity and strength is quite nice.

For this project, while the amount of motor induced torquing is minimal, since there is no X/Y quick acceleration and direction change, there is a good deal of force being applied during the immersion of the build plate into the resin toward the FEP film, and when the plate is lifted away from the FEP film, due to the viscosity of the liquid resin and the suction force generated by the surface area.

On a cantilevered arm, like in my original design, this would have applied a forced on the arm and the z-axis pillar, pulling the whole structure towards the center of the build space. To counter this, I went with 20mm profile aluminum extrusion in the 40mm by 40mm configuration, which greatly increases the strength of the structure.

This is further reinforced by the use of 20mm profile 20mm by 80mm frame side panels, to which the z axis tower is attached to with 4 40mm by 40mm corner brackets. That z-axis column isn’t going anywhere.

With the adjustment to the tower design to add a second z axis tower to provide a balanced arm, any inward force would be countered by the common arm.

All Metal Connectors, slides, and Brackets

The second decision I made with regards to the construction of this printer was that all or most mechanical connections would be done with metal and metal only, where ever possible.

All of the corner joints are handled by 20mm profile 40mm x 40mm corner brackets with 4 M5 bolts per face.

The linear motion is handled by recirculating ball linear slides on a linear rail of hardened steel. This is mounted to the Z axis column with 8-12 M3 bolts along the length of each rail. Each Z axis column has 2 linear rails on opposite sides of the Z axis column. For a total of 4 linear slides between the two towers.

The corner brackets are also sourced from Misumi USA. While Misumi USA also sells linear slides and screw drives, I chose Amazon to source those parts.

Other Frame Material Options?

You can go with other options for the frame. For instance, 80/20, Inc. also sells aluminum extrusions. However, I’ve sourced from Misumi USA over the years, so decided to continue making use of their extremely professional services.

You can also opt to go with other materials like HDPE, Acrylic, Steel, or even Wood. However, if you opt for another building material, make sure it is sufficiently rigid, constrained, and/or protected from environmental factors like humidity.

  • Wood is subject to expansion and contraction with humidity in the air. You can make use of resin impregnated wood, which is significantly denser, stiffer, and resistant to humidity.
  • HDPE/plastics in general require significantly thicker parts to achieve similar strength. They are also subject to potential softening due to heat and exposure to certain chemicals.
  • Steel is a good option for the frame, but for the slides and accurate positioning, you would still need something to provide the 10’s of micron accurate motion. Steel is also fairly heavy, expensive, and if you need to add holes, requires more effort.

Using Aluminum Extrusion For Linear Motion (V-slot, milled surfaces, etc)

You can get milled surface aluminum extrusions. Misumi USA offers this product/service as well. However, while it is a nice low cost and convenient option, I personally decided against it.

Open Builds and other popular maker suppliers employ V-slot. That is… a T-slot aluminum extrusion where the channel leading into the slot forms a V bevel inward instead of a normally rounded corner. This forms a split V-groove and allows for a v-shaped roller to align in the groove and use it as a linear guide. This creates a nice contact surface and provides smooth linear motion.

But I opted to not choose this option for a number of reasons(in general, not specific to v-slots):

  • Aluminum extrusions are made with 6061 aluminum formulation and while strong, if exposed to a strong force, will easily dent/scratch/nick, which would lead to small deviations over time. This would happen with metal, plastic, rubber, or other kinds of rollers.
  • Rollers can build up crude and cause motion deviations. Much more so with rubber rollers.
  • Roller bearings can also fail leading to stuck/dragging rollers.

I also opted to not use round linear rods and circular linear bearings for the following reasons:

  • Out of the box, the linear bearings have a good deal of give. The frame and carriages would need to be designed with some pre-load in mind.
  • For a 4 bearing setup on two linear rods, you _can_ get a good bit of stability. This setup is used with CNC machines and various FDM 3d printers. However, without getting the linear bearings with pillow block mounts, I would need to design and build a rig to hold the linear bearings and keep them aligned. This means a custom milled HDPE/Acetal part or aluminum part, which would be expensive.
  • IGUS makes a 100% polymer slide, which comes in a pre-compressed and uncompressed form. They work well and are nearly silent. However, they, like the other linear rod solutions, requires some pre-loading in the design to ensure the amount of give/slop is minimized.
  • Use of a linear rod would require additional build space to facilitate the clearing of the rods, mounts, linear bearings, and carriage systems, all of which end up being more complex and bulky.

I went with linear rails and circulating ball carriages for the following reasons:

  • Rigid and aligned along axis for mount points.
  • Low profile for a compact build.
  • Readily available through various online sources.
  • Their installation increases the rigidity of the structure.
  • Spacing of the mount holes made them a good match for the 20mm profile 40x40mm aluminum extrusion I would be using.
  • Fairly quiet.

To be honest, for this build, I went with the option which resulted in the most off the shelf parts usage vs customized, milled, or otherwise produced parts. This time around, my goal was to build a workhorse and focus on the printing… not the tinkering and tweaking of the printer.

Speaking of producing parts…

Why I Avoided Using 3D Printed Parts

I have built a number of printers using 3D printed parts ranging from 30% infill to 90% infill. I have printed them using PLA. For the most part, the prints functionally worked. That is, they fulfilled the role of holding the parts of the frame in place. They accepted the bolts and nuts.

However, this is what they did not do and could not do:

  • Hold the frame to the same level of rigidity as metal connection parts would have.
  • Allow for stronger tightening down of bolts… excessive force causes sinking in, cracking, and print failure.
  • Withstand heat or room fluctuations without resulting in bolts and nuts working their way loose. Imagine how many bolts are on a printer and how much hunting and seeking that required to re-tighten down?
  • Handle extended periods of stress without suffering some measurable degree of deformation. Almost all parts would eventually deflect/deform by some critical amount… enough to cause issues with prints.
  • Humidity can cause swelling and/or dimensional change.

As they say, your mileage may vary. Mine was filled with quite a bit of having to compensate for problems with personally 3D printed parts. And outsourcing prints to others was expensive with no means to guarantee quality with just one printing. -_-;

As noted above, I want this build to be a work horse. Easy to maintain, little maintenance. And little to no impact to the frame/components due to humidity and temperature changes to the surrounding environment. Think, between Winter and Summer.

Resources