Massive chatter and bit jumping

Greets, many thanks for all the documentation and a great design. I built my CNC and am having trouble with bit jumping. I’m currently using a 1/4" router bit, but I’ve had the same issues with 1/8" end mill and ball nose. It seems that the bit starts vibrating and due to feedback the vibrations become a massive chatter until the bit jumps. I’ve attached a link to show whats happening.

I’ve fiddled with the z-axis bold tension, leveled everything, added layers of MDF to bring the z as high up as i can go. I’ve about run out of ideas.

I’m using a very conservative feed and depth, the depth is less than 1/8" deep with a feed rate of 10 in.

Here’s a video link: https://www.youtube.com/watch?v=GIa3-fYNnwo

And to get an idea of the build, I’ve attached an image i took during the build process.

Any ideas would be great.

First off: your legs are really tall
Second off: the play appears to start largely in your tool mount. Where did you get the file for that?

Third: if you grab the bottom of the z and wrestle it around a bit with your hand, do you feel movement? Isolate where the movement is coming from.

I’d be looking for cracks in the plastic as well, that is pretty violent. Keep that to a minimum (the video is appreciated though) obviously.

I was thinking that the height of the machine may be an issue. But its not significantly taller than other machines I’ve seen being built. The tool mount is a part from Thingiverse - its mounted rock solid to the z-axis and there’s no movement at all from it. I get some movement in the z if i push pretty hard on the tool in both x and y - but it takes effort. I see no failures in the 3d printed parts.

What is your Z axis displacement? Seems to me that legs length and Z axis rail are more for 3d printing than milling…(I am a newbie as well, just thinking from the infos on the forum and V1 instructions…)

I don’t think this will help rigidity unless the vertical tubes at all four corners attached to the feet are firmly supported by the MDF you added.

Using an endmill instead of a router bit will also help.

Try a 1/8" single flute upcut bit, Make a file with CAM not manual movements. If it still happens give us all the specs and details.

You have a really tall machine, with a spindle that is known to have high run out, and a very large router bit. each adding it’s own issues. Start with the Milling basics page.

1 Like

Thanks for the replies. I’ve tried with 1/8" end mills, 1/8" ball nose, 1/4" end mill, and the router bit with the same issues. The common theme seems to be the height of the machine. I really only need 3" in z for my projects. I’ll reduce the height and give it another whack later.

The router in the picture was temporary while i was waiting on the spindle that is used in the video.

The reason I suggested this is if you open the YT video and set it to 0.25 playback, you can see the majority of the movement comes from the tool mount. The Z axis tubes stays relatively still in comparison.

1 Like

Interesting - i did not think of slowing down playback - it does seem to rock and roll - i got lazy and printed a sleeve that fits inside for my spindle where i would normally mount a laser. There’s probably not enough friction to keep things good n’ tight. I’ll attempt to lower the build and come up with a better spindle mount this evening.

Ok - I think it was the adapter sleeve I printed for my spindle. Originally I printed out a tool mount for the crappy harbor freight spindle, and rather than printing one out for the new proper spindle, I printed out a sleeve that made up the size difference.

As you can see, the sleeve was actually deformed and skewed, which tells me that during load there was not enough friction force to hold the tool completely in place, allowing slipping and causing the chatter/jumping.

I took out the spindle and sleeve and attached the HF router. Even with the 1/4" router bit the MPCNC cut through the mdf like butter without any chatter or jump.

Time to get un-lazied and print the proper mount for the spindle I s’pose. Thanks for the help!

Nice, I was wondering what that blue thing was in your video. Had a hunch it wasn’t helping!

Let us know how it works out!

I’ll let you know in 2:49 @ 16.72 meters of filament from now.

Raising the workpiece should help because at least on my (very tall) machine i measured the deflection of the legs to be less than the deflection of the z axis.

Also going to a single flute cutter will help a lot. There are some funky dynamics that can occur where the load causes deflection that changes the cutting depth and direction which in turn changes the load, and it can run away on you. A single flute cutter takes only one bite at a time so to speak and seems to me less susceptible to this.

I agree with the comments above as to your actual problem and I think you’re on track for the remedy.

1 Like