Machine Resonance

I have a Shapeoko Pro with an 800W spindle. I’m trying to maximize material removal, and I’m somewhat of a newbie to all of this.

I’m using a 201 cutter in Red Oak and had a good run at 1/8" DOC, 18k RPM, and 100IPM. I’m trying to maximize material removal so I then tried 3/16" DOC, 18k RPM, and 75IPM with aspirations of running faster.

It would be running fine, and then you’d hear a screaming sound in the khz range. The resonance would come and go leaving a path of destruction in the cut quality. It would occur in both X and Y movements. It did not appear that any steps were being skipped as it maintained the profile through multiple layers.

Am I trying to drive the machine too hard, or is their possibly an assembly related issue? I should also note that when it resonates, the screws for the dust boot want to start coming out, which is slightly annoying. Please help this newb push his machine to the limits!

Oak is very brittle and hard. Sounds like you are pushing it too hard. You had good results but pushed over the capability. Chip removal is important on hardwoods because you said yourself cut quality went down but the machine still cut properly just not well. You can push just so far and instead of gaining productivity you wasted time and material. Dont feel bad it happens to everyone.

Keep experimenting up to the edge and back off for a safety margin.

4 Likes

As a general rule, desktop class machines will like wider cuts more than deeper cuts. What is your WOC? Can you increase it? Is your spindle limited to 18K RPM? Can you go faster? Also, a chipload of 0.0019" (your first 1/8" DOC cut) seems low to me. I routinely cut aluminum at 0.003" chipload on my SO3 with an 800W spindle. I bet you can increase that a bit. Make sure you have strong dust collection. It matters a lot when you are trying to push your machine hard.

1 Like

I have an er11 collet so I’m limited to 1/4 end mills. I certainly can run it faster and increase the chipload. I was under the assumption that running deep and slower was better than shallow and faster so I was trying to ultimately increase up 1/4 doc for a 1/4 end mill.

The fact that the issue was intermittent (roughly 50% of the time) I think chip clearing may have been the issue. I have a dust extractor hooked to get good static pressure. In general there’s no dust left behind.

I know Winston was and to achieve 1/4 doc on a spektra bit, and wondering if that may be the solution.

I also analyzed the cut, and I was able to identify 3 resonates. There were 30hz and 15hz resonances that I’m assuming we’re machine resonances triggered by deflection. These were identified by a clear wavy pattern. There was another pattern that left diagonal line patterns so it was likely in the khz range. Some surfaces were also smooth, but off so I don’t know if the machine was just deflecting at that stage.

1 Like

There are some vendors who have 8mm (5/16") collets — that’s a sea change in terms of rigidity.

1 Like

15 and 30Hz sound like the belt system resonances to me so those are likely aspects of the machine.

Have you tried tweaking the feed override up and down when they occur? Frequently a non-resonant mode is not far away.

I’d agree with Nick, shallow depth with wide engagement and fast feed is the general way to get fast material removal on a Shapeoko type machine. There’s a pretty solid body of evidence for that now.

1 Like

I tried again today with 1/8 DOC, but unfortunately I would occasionally get the resonances again. I did not see any low frequency resonances, but there were definitely deviations from the target. I experimented with feed rate ranging from 100 IPM to 150IPM with all feed rates having occasional issues.

The more I look at it, I think my issues boil down to cutter, DOC, and material. I noticed the red oak is coming off as strings, and not as chips. I think the strings are getting wrapped around the cutter causing it to start vibrating until the strings are broken. I think my issue could be resolved by reducing DOC further, changing cutter to one with less flutes (like the spektra 1/4 upcut), or potentially using a climb cut for the initial profile followed by a finishing conventional profile.

I’m still at the prototype stage for what I’m making some I’m not too concerned. I’m going to create a fixture next and work on pocketing operations, but I don’t expect issues since the cutter will not be fully engaged like with a contour cut.

Ultimately this is a kids toy and I can sand out imperfections. I’ll probably give the prototype to my least favorite nephew.

1 Like

But tell him he’s getting the first one because he’s the best nephew :wink:

1 Like

I cut wood all day at work and can tell you that conventional tends to give a better finish with less fuzzy bits. I hardly ever climb cut. Also, a bigger chipload will help with stringing. My gut tells me that with a 1/4" cutter, your best MMR will be achieved somewhere around the 1/8" DOC but that is just a guess.

2 Likes

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.