3D scanning and printing

The amazing parametric slew bearing!

In my final project I will need the ability to have magnets spinning very fast along with a vertical wind mill. This requires some sort of bearings that allow the magnets and mill to spin freely while otherwise keeping them firmly secured in their place.

At first, I researched ball bearings, specifically the type that can be printed in place, thinking that they would be ideal for this purpose. However, during my research I discovered incredibly good instructions on how to design parametric slew bearings.

Created by Christoph Laimer, the instructions take you step by step through creating parameters, drawing the sketch and extruding the design. This process helped me gain a much deeper understanding of how to use parameters in 3D design.

Incorporating the slew bearing in final project

I expanded the sketch to include an area for the magnets, a coil holder and a second bearing at the top, tying it all together. It should be noted that this design is flawed and does not create any electricity. In order for it to work, the coil must be arranged perpendicular to the axis along which the magnets spin, not parallel. This will be rectified in future designs.

I measured my magnets and added parameters to the design to incorporate them perfectly.

The reason why it should be made additively

Additive creation methods are methods by which material is "added" to create the design the creator envision. Subtractive creations methods in stead "subtract" material from a chunk to achieve similar results. One of the key benefits of additive manufacturing is that the tool you use to print with, in this case the "nozzle" can move freely until the layer has been added. There is nothing in the way, which is why it can create complex shapes with holes and odd shapes on all sides.

The primary reasons why this design should be made additively and not subtractively are the bearings and the magnet holder. The bearings include a v shaped indentation along the interior of the whole race. This would be prohibitively difficult to create in subtractive processes.

The magnet holding element is circular, with races along the top and bottom and holders that enter it from six different angles. This would be a nightmare to create subtractively, but is done with ease in an additive process.

Printing with Ultimaker

I used fairly standard print settings for all these prints. I picked the pre-set "fine" profile, changed from the raft to the brim (since the raft is unnecessarily cumbersome for flat items like these), changed the initial layer height to be 3mm for added adhesion and chose the triangular infill type, as it is slightly faster without loss of integrity compared to the hexagonal. I did not find that these prints required any support, although some required light sanding afterward. I kept infill density at 20% although I am certain now that 10% would be plenty.

I like to place my prints as far back as possible, because the heat is more constant there, which minimizes warping. I also like to put my designs in the corners rather than the middle, so that we don't wear the middle of the rails down faster than the rest of them.

Easy to use, difficult to master

Our printer is an Ultimaker 2 Extended, and when I originally printed this, it was not properly calibrated and we were not good at setting it up, so we had to use a brim, gluestick, door and thicker base layer in order to have acceptable adhesion. Since then, Carl came to finish Fab Academy with us, and he is some sort of 3D printing wizard, so now we don't need anything more than good printer settings to get perfect adhesion. Since he came and calibrated our printer, print times went down by hours, and the prints come out better than ever. (Hire him)

He noticed that our printer needs to be set to higher temperatures than the ones in Dubai, and he deduced that this was most likely due to our much lower ambient temperature. Since then, we printed at higher temperatures, and everything has been working more smoothly

Rough and imperfect

One of the limitations of 3D printing is that the prints generally require some work before they are usable. For example, these 20 rolls all needed to be individually sanded and worked before they were smooth enough to use in the mechanism. For the bearings, it is critical that they are wider than they are tall, as only the rotating sides should touch the races, the ends spinning without friction. This is one of the things which would benefit from being molded and cast, as I needed 40 of these, and they took 6 hours to print, followed by 2 hours of sanding and cleaning.

Some assembly required

To assemble the slew bearing you place the outer race on the table, with the inner race inside it. Then you arrange the bearings inside tilted alternately towards the inner and the outer race.

Some assembly required2

Once all the bearings have been arranged, you place the second inner race, in this case, the magnet casing on top, securing the rings together.

It spins!

And the moment of satisfaction comes in. The design works as far as the spinning is concerned. It began rough and loud, but after some grinding it back and forth, and adding lubrication, it spins much better.

On the whole, this was a very satisfying prototype creation. The parameters are fine, the spinning works, and the print was of a high enough quality that it required only minor sanding and post processing. There was some minor warping of non-critical areas, such as the outside of the races, but this has no impact on the functionability of the model. In cases where the warping is unacceptable, I will use my new knowledge of fan settings to combat it.