3D Printing

Kossel Carbon Fiber Arms

Leave a comment

Continuing the saga of converting a Rostock Max to a Kossel, the arms which attach the carriages to the effector platform from the Rostock were not long enough for the new build. The appropriate length for the Rostock Max was around 11.5″, while the new Kossel build needs ~16-17″ in order to reach the outside of the bed.

I sourced carbon fiber tubes from Goodwinds.com and inserts from Amazon. I divided up each of the 3 rods into two 15″ pieces using a dremel to form the 6 arms that attach the carriages to the effector platform. I used super glue to attach the inserts to the carbon fiber tubes. When fixing the inserts it is very important that all 6 rods are exactly the same length. This measurement plays an important role in the accuracy and resolution the printer can get. In order to achieve this I set up a simple jig to hold each rod in the same place while the glue dried:

IMG_20150205_222307

A lot of information I read talked about the length of the carbon fiber rods themselves needing to be exact. Taking this approach allows you to simply slip the ends in and call it a day. I found that using this jig allowed me to “fudge” the lengths of the rods. Since the jig is holding the position of the ends exactly where they need to be, a slightly shorter rod doesn’t compromise the total length, the rod end simply sticks out a little farther from the rod. This allowed me to have small differences in my rod lengths (I was using a dremel free hand) without sacrificing precision in the total length.

IMG_20150205_222319

I rotated the rod ends in the rods towards me so they touched the nut, ensuring each of the ends were completely flat as well. After trying out one rod I could see the printer would be able to print to the edge of the heated bed with ease, and hopefully better accuracy.

IMG_20150206_070106

On an unrelated note I finalized the work on the 24v power supply by attached the 12v stepdown unit and the SSR to the PSU to keep things neat:

IMG_20150206_070129

In the future I may make the PSU modular to ease the work required to move the printer.

3D Printing

Kossel Redesigned

Leave a comment

After assembling the main portion of the printer I noticed too much flex was present in the frame. I overestimated the required side of the frame to fit the heated bed comfortably. If I was going to keep the same size frame i would have needed to move up to 30mm extrusion or larger, as well as bulking up the design of the corner pieces. After researching online, as long as the bed was within the triangle, even if by only an inch, I wouldn’t lose precision.

IMG_20150127_211209

With a base that large I would have needed ~26″ arms for the effector as well which can get costly. I cut the frame base pieces from 25″ to 15″, which still leaves room for the edges of the bed, but reduces the arm length to 18″ and raises the print height to around 12″. There was also significantly less flex when moving the frame:

IMG_20150131_005658

A size comparison with the Rostock Max v2 bottom frame laid over. In the end each tower was moved away from the center of the bed about 2.5″. While this seems like a small amount, that’s the beauty of equilateral triangles, the area of the triangle was greatly increased to encompass the heated bed.

I noticed both the top of the old Rostock Max and the snow flake the bed used to be mounted to weren’t level anymore. It wobbled when sitting on the frame. I decided to use extrusion to attach the bed to the frame. I clipped the old tapered screws shorter and bent the nuts to slide them into the extrusion. Two washers were placed between the frame and the bed to make sure the bed would be level and to help keep heat from the bed being absorbed into the frame causing cold spots on the top side of the bed.

IMG_20150131_111736

IMG_20150131_111746

Hopefully using extrusion instead of melamine will keep the bed from warping while heating up. Moisture doesn’t affect the shape of the extrusion, it should stay level no matter the climate.

IMG_20150131_111823

The printable area of the bed is within the triangle formed by the towers.

After the extrusion was attached to the heated bed and laid onto the frame I didn’t observe any wobbling.

I am reusing the stepper motor dampers from the Rostock to keep the noise to a minimum. I suspect without the wooden frame of the Rostock acting as a sounding board I should be able to get the printer even quieter.

I’ve ordered some carbon tubes and swivel arms for the cheapskates. I need to find out how I can attach the bed to the frame in the mean time. I will also need to isolate the bed wires from the frame to prevent short circuits.