Primo Z axis anti-backslash

My MPCNC had noticeable backslash on Z axis, this caused up/down vibrations on some materials. So I added same lead screew nut to the bottom of Primo core with ratchet wheel.

This model is created in FreeCAD and project file is available for download. Should be printed from PETG with layer height 0.2mm. Lead screw is pushed down over rubber spacer, the pushing force can be adjusted by ratched wheel. If the steps are too big, you can use thicker spacer with lower density for more squeeze steps. Squeezed material is used instead of spring and the overall height is 6mm+height of spacer.

This ratchet wheel is on the bottom side of the core. Only simple modification is needed - re-drill the bottom hole to bigger size with 11mm drill. (I enlarged it via vacuum hole in my desk, so no gantry disassembly needed).

Because I am new user here, I cannot add attachments. You can find more pictures and file download on printables, just search for: z-anti-backslash-compensation-wheel

3 Likes

I have noticed this also. There are a couple other threads that brought this up and people were told that the weight of the router would be enough. I was doing a rough mill on a piece in hard maple yesterday and noticed some ridges on some cuts like there was some variation in the Z. I rested my hand on the stepper and it would cut slightly lower than with my hand off. I didn’t notice this with poplar or cherry.
How well did this work for you?

Interesting. I am one who says often that the router weight should eliminate backlash i ln the Z screw.

Adding weight to the Z axis would cause some things to flex. So I would not be surprised the bit would sink.

The nut has to rise up from the acme threaded rod to necessitate backlash compensation. Or the coupler needs to be flexing.

I can see movement between the screw and nut. Maybe my nut is worn. I have only done 4 small 3D pieces so I did not expect it to be worn this fast.

I mostly do HGW2082, dibond and plexi on vacuum table top, because consistent height is important for chamfering and slotting. My lead screw and nut came from never used donored Anet 3D printer and the play was around 0.4mm. I dont have components from different manufacturer for comparsion.
Im satisfied. The squeezed material work similarly as spring in comercialy available antibackslash nut.

The nuts should have backlash. But the weight of the router and gantry should always pull down (obey gravity, it’s the law!). So the nut and leadscrew should always be at one extreme of the backlash. That’s the theory anyway. If the router/bit is pushing up with a force more than the weight of the router and gantry, then the backlash will be a problem.

I can not really imagine a scenario in which this happens, actually. We have seen the Primo mill steel and dive pretty quickly into aluminium without that happening. Maybe if you had a Dremel that is very light… but besides that it would really be hard to achieve that.