Update 08Feb2026 - Side filament sensor is replaced with new version with switch for TPU. Uses Prusa's design idea.

I had rattling sounds at resonant speeds on my Core One. I narrowed down the Y Bearings as a root cause and decided to exchange the Y Rods with Linear Guide rails.

The result was very positive - no more rattling. Resonance can still be heard slightly next to the printer, but is more a low frequency vibration from the enclosure. And in general the printer became quieter during printing. Another benefit for future is that the carriage has greasing nozzle and must not be disassembled for re-greasing.

I did this end of October, have more than 100 hours since then and no issues, so I decided to share it.

Not everything was perfect as I took some dimensions too rough. Now when Prusa shared the complete CAD I made small changes to get to nominal.

Unfortunately not everything can be printed and it requires one milled aluminum part.

Further, the XY Frame must be removed from the printer to drill the holes and cut threads for the new rails.

1. Aluminum T-Holder

To attach the X holders to the linear guides I made an aluminum holder. As raw material I used a T-shaped 35x35x3mm rail from the local hardware store, which I shaped on a small CNC which I have. A non-CNC milling machine will work as well. The external dimensions are not that important, but the position of the holes should be within +/-0,1. 

Here I have to point that with this holder I intentionally shift the X axis 0,3mm higher. The reason is that I use M3x6 screws to fix the rails and they protrude 0,5mm above the XY frame. Having in mind that screws have tolerances, some may be even more. But some can be also lower and this is the reason why I did not try using washers. If the screw sticks out than I am sure that I use the whole thread in the frame, which is 2mm thick.

This of course will move the pulleys and the rollers higher. Even before that the belts were always on one side and not centered in the pulley. I don't see any signs of wear and tear.

And last but not least - I had no 3D models at that time and took the dimensions directly from my printer. The XY-Carriage was 0,7mm above the frame. Now I see that nominal distance in Prusa's model is 1mm, but due to the bigger holes on the linear holders, all tolerances go down. Now, with the new design I have 0,95mm gap, measured with filler gauge.

I upload a step file of the holder. It is symmetrical and same for left and right. (in case someone notice, the holes are no symmetrical on the picture I uploaded, but this is because of my imperfection - I mounted the rails too much to the back and the carriage was sticking out 1mm from the rail at front position. So I had to go for shifted left and right version)

2. XY Frame rework

M3 threads must be tapped on the frame to screw the linear guides in. For this purpose the XY frame must be removed. Follow the Core One assembly instruction how to do that.

For drilling the holes I made a gauge which can be screwed to it. From one side it uses the most inner M3 thread of the tensioner, and on the other side the hole for the motor holder with screw and nut.

Because the gauge is still flexible, it is better to clamp it in the middle as in the attached photo.

The gauge I used is with 4,5mm holes (must be tapped with 4,5 drill after printing) and press-fit 0306DU bushings. Then I used a centering drill with 3,15mm diameter to mark the positions. At the end I drilled with 2,5mm drill and tapped the threads.

But I think that it can be done with smaller holes and directly 2,5mm drill. Or use the gauge only to mark or punch the positions and drill without it. Therefore I add an STLs with 2,5 and 4,5mm holes.

Can be printed in one piece on Core One, rotated at 49° (or 131°).

Material PLA. As visual quality is not important I used 0,4nozzle and 0,2 Speed profile with gyroid infill.

3. Side Filament Sensor

Update 08Feb2026 - Filament sensor is replaced with new, TPU version, which has a switch to open the lever as in Core One +.

The new carriages are wider and will hit the original filament sensor. I had to make a new one which is lower.

Cable connector must be plugged before you insert and screw the sensor in the base. A separate thin sheet is used as a flag over the switch. When moved to open position, it changed the background of the "X" sign. The rest of the assembly is similar to the original sensor.

Refer to the attached picture and to the list of needed screws and nuts at the bottom.

Material PETG, Orange and Black.

Printed on Core one, non-HF  0,4Nozzle, 0,2mm Structural, 25% Gyroid Infill. The lever needs supports everywhere. The rest of the parts are without support.

4. Tipps

The T-Holder must lay without gap to the side wall of the carriage.

Mount one rail parallel to the edge of the XY frame. For example, use caliper to align it. Keep the other rail loosen. Mount all other parts including the X-Axis bar. Then move the bar to the most front and rear positions and tighten slightly the screws. Ensure that it moves without friction all the way.

5. Hardware needed

For Guiding rails:

• 2x T-Holders from the Step file

• 2x 300mm HIWIN MGNR12R - guide rails. !!! The edge must be 12,5mm away from the first hole. (in the middle between the holes). In this way the rails are symmetrical. Accuracy of +/-2mm is still OK.

• 2x HIWIN MGN12H, Z1 - Carriage with light preload

• 32x M3x6 DIN912 (ISO4762) - 24x to attach both guiding rails to the frame and 8x to attach the T-holders to the carriages. 

• 8x M3x4 DIN912 (ISO4762) - to attach the X axis holders to the new T-Holder

I recommend to use thread locker (Loctite 2400/243 or similar)

For Filament Sensor:

• 1x M3x12 DIN12 (ISO4762) - for filament sensor lever (original was M3x10)

• 1x M2x6 DIN912 (ISO4762) - for filament sensor PCB (original was M2x8)

• 2x M3 Square nuts (same as the original if you can remove them)

I used the original Ball, Magnets, Collet and cover Screws (M3x8) and of course the sensor itself.