Ultimaker Extruder Bearing Assembly
Bertho Boman

Original Pressure Plate Assembly

A goal was set to be able to upgrade the Ultimaker Extruder Driver without having to make a complete new unit and a second goal was to make it easy for other people to duplicate. The result is a design that is interchangeable with the original driver and no modifications to the driver itself are necessary. All parts can be printed on the Ultimaker except a spring. I printed the parts from PLA and so far they are OK but I have not done any material strength calculations and ABS would be a better choice.

Original Design

Above is the standard pressure plate assembly used in the Ultimaker 3-D printer. It is laser cut from plywood and from Delrin. It functions reasonably well but has two issues that can be improved.

First, the pressure against the filament is through a piece of Delrin instead of a much lower friction ball bearing.

Second, a much bigger problem is that the pressure is controlled by the fixed position of the pressure plate. This causes serious problems when feeding the filament. A very slight reduction of the filament size and the feeding pressure will be much too small and the driving wheel (knurled bolt) slips. If the filament is slightly larger, the force will be much too large and the driving wheel deforms the filament and it jams in the Bowden tube.

Original Inside
Original Delrin Pressure Plate

New Design

New Side Plate

Two identical side plates were printed (or they could be laser cut). Note, the hole spacing has been changed and the part is slightly squared up to be exactly matching with other CAD designed parts. There are very small errors in the original extruder drawings which prevents a real CAD program to properly snap parts together for alignment.

Replacement for the old Delrin Piece

This part matches the original size so the new assembly will fit without modifications. It is printed in PLA or ABS. There is no need for the low friction properties of Delrin.
Also in this view the position of the bearing can be seen.

Screw and Bearing Hub added

The bearing center hub is slightly wider than the bearing to make sure the bearing outside is not touching the walls.
A long 3mm screw has also been added. Both sides of the head were filed off to form a "T-Handle".

Yoke, Spring and Nut added

This is now the finished assembly except missing the regular 3mm screws. A yoke has been added that applies force to the ball bearing shaft. The force is controlled by the spring and adjustment knob.

The recess in the knob is visible

From this view the recess for a 3mm nut can be seen. The recess was intentionally made to an exact dimension expecting that as usual the hole to be slightly too small. That allowed the nut to after heating be securely pressed into the recess.

First printed assembly

A washer was added between the spring and the knob. Also an "acorn" nut was added to avoid the sharp end of the screw.

End View

The new assembly inserted into the Ultimaker Extruder Driver

In place and functional! The quick release functions as usual except pressure needs to be applied to the assembly while sliding out the little handle.

Also visible in this picture is my replacement of the original Bowden tube holder. This design locks the tube rigidly to the extruder driver frame with absolutely no movement.


Low-cost digital luggage scale

A very useful device to be able to make good measurements is a force scale. The simplest and lowest cost are the luggage scales that can be found at many locations, often for less then $10. I have checked several brands with precision standards and they have been VERY accurate. It is a recommended accessory since it can be used to measure various variables on the 3D printer.

I measured my initial setting that I guessed at and it is 3.4 kgf. I have not had time to evaluate and find the ideal setting but that is working fine so far. It is on my to-do-list....

In the original design, a 0.2mm change in filament size would typically cause printing failure and possible jams because of the drastic changes in the force against the driving wheel with the rigid pressure plate.
With the new spring loaded design, a 0.2mm filament change is only creating a 1% change in force in a 20mm long spring; obviously insignificant.

I have only been using this new design for a few days and I have not seen any problems or surprises.

Now I have to learn how to properly put up a project on Thingiverse so other users will get a chance to evaluate and try the design.

Bertho Boman
Vinland Corporation
11600 NW 20th Street
Fort Lauderdale
FL 33323
(954) 475-9093
Email: boman33 at Vinland dot com