I could use some help figuring out where these swirl marks are coming from and hopefully be able to eliminate or greatly reduce them.
This is HDPE colorcore material, where I am engraving at a depth of .075" to get to the white color. I have a onefinity elite foreman/masso with a redline 2.2kw spindle. It is sitting on a sturdy torsion box style table that I built. I also have a vacuum hold down system to keep the material flat with tape around the border of the sheet, and it works great - it sucks the material nice and flat on top of a surfaced MDF spoilboard. I am using a brand new 2 flute upcut 1/8" endmill with corner radius, which is meant for producing a clean bottom finish. And I have tried all kinds of toolpath strategies in vcarve pro, from lower stepovers, slower/faster RPM and feedrates, multiple shallow passes, even a last final pass of just .005" depth. I even tried greatly reducing the acceleration settings on my Masso to just 25% of the default values. I also oiled the machine not too long ago, so everything is lubed up nicely. But I did notice this hapenning before I maintenanced the machine as well.
It seems no matter what I try, I keep ending up with this result of these circle swirl marks. At my day job, I work on other CNC routers, and have used this exact bit on this exact material before without the issues I’m having, all while not having to really dial in the feeds/speeds (almost any feeds/speeds in the general range cut clean). So, I already have a very good knowledge of proper feeds/speed/DOC. I have already tried other bits too, including O-flutes with corner radius, and do get some varying results, but every bit still has this exact thing occuring, that same sort of swirl pattern.
Yesterday, I made sure to check my spindle was nicely trammed and I gave the spoilboard a fresh surface mill with a flycutter, so that everything would lay nice and flat. Put a sheet on, where it layed nice and flat, and still getting the same result.
I was starting to think that maybe my setup just wasn’t rigid enough. But then, when I greatly slowed down the feeds and acceleration settings, and doing final shallow passes, I was still getting the same exact result. This has me thinking that there is something else going on.
Did you tram the Z slider?
It could be its not 90 degrees to the table in all directions
Check the finish with another bit.
It could be that new bit has one larger flute than the other.
Not the Z slider itself, but rather I used a tramming tool coming directly out of the spindle to get the spindle itself trammed nicely in all directions.
I have tried other bits. Some definitely are better than others. But they all share that same pattern going on.
It’s definitely worse when doing directional changes and curves and diagonals.
Here’s a pic of using a 1/8” Straight O-flute end mill. Definitely better. But still has that same thing going on.
Sorry, I forgot to mention. I also used a dial indicator on the toolbit to check for runout, and it was less than .001” runout. So, I don’t believe it’s the collet.
Looks odd to me that the swirls arent consistent. In the top photo the edge looks like a more consistent swirl but the middle sections look like your tram would be off and tilted down and to the right which allows the left side of the tool to be lower.
Not sure about your file but I also noticed on the far right side that one of the edges has what looks like chatter unless its possibly just an artifact of the vector you are following.
the outside countour cleanup of the section on the left, and that far right side (which is just a single line contour) both use a .060” diameter engraving v-bit, which is why the swirls look different.
and yes, that is indeed chatter that you are noticing. the vectors are clean - i made them myself in v carve pro.
So, maybe whatever is causing that chatter is causing those swirl marks?
That particular section on the left with all of those pronounced swirl marks are a section where there is a lot of directional changes in a small section.
Heres a screenshot of the toolpaths from the vectric file for reference.
Yeah its possible that the cause of the chatter might also be the cause of your swirls. If the bit is wiggling then its going to be impossible to get a completely smooth bottom. I would try a different collet/nut. Maybe its slightly off balance and is causing some vibration?
I will definitely try a different collet and nut to test that out. Will let you know how it goes. I’ll be at my workshop tomorrow again to test out some more. Thanks for the suggestion.
The chatter is not from x, y axis. It looks to be from the z axis. I had the same problem which was caused by the z slider ball screw end play problem. Check to see it you can push up or down on the ball screw on the slider. Mine had about .012-.015” end play that I corrected by drilling and tapping the end of the ball screw and adding a screw and washer to draw it up against the bearing.
Yesterday, I had to make a couple signs, so I tried using a different collet nut (a nice quality one from avidcnc), and it didn’t make any difference.
Big thing I noticed: The spindle itself vibrates pretty strongly, which I imagine must be causing the chatter, while cutting. When it’s just spinning in air, there is no vibration. But, during cutting, especially long stretches, if I touch the outside of the spindle housing, I can feel pretty strong vibrations.
Next time I am there, I will check the ball screw on the slider, as Derf suggested. Hopefully it is that. I am a bit worried that it could be my table is not as rigid as I initially thought. My table uses a kreg table frame, and I made a torsion box setup at the top using 2x4’s, and then with 3/4” plywood on top, and then 1/2” phenolic on top of that with vacuum grids engraved for a vacuum table setup that I built, and then 3/4” MDF spoilboard on top of that. My depths are quite consistent, and the vacuum table works. But perhaps the whole table is just vibrating?
Just a thought: How many nodes are along your curve? If the tool stops at each node then you might get this. If there are a lot of nodes then use the curve fitting to smooth the curve out and eliminate all the extra nodes.
Some more clue and answers to some people’s questions:
Here is some play that I noticed in the z motor coupler. It’s a video of me using an allen key putting some torque back and forth (i’m not using a lot of force here). If I do use more force, it moves the z-ball screw. Not sure if this is intended - and this is while the machine is energized. I checked all the set screws in the couple, and they are already tight. Any thoughts about this? Is this normal? Edit: after looking again closer at the video, at least to me, it seems like the z-ball screw is moving with the coupler when I move it, but not 100% sure.
Here is another pic of another angle of a new sign I made this morning. Using a single flute straight o-flute 1/8 end mill with a small radius I filed on it to make it clean the bottom better. It isn’t so terrible that I would consider it unacceptable for signage, but I’d really like to do better. I know it’s possible to do better with these bits/toolpaths/material as I have done it it my day job on other machines.
