Milling chatter in slow motion and audio spectrum analysis video

Milland · Feb 27, 2021

Endmill chatter at 20,000 Frames-per-Second - YouTube

Very interesting video, although for full disclosure the guy is using a CNC router rather than a proper VMC. But the principles are there, and if anyone here has the capacity to do similar on a real mill it would be great to see the results.

See? Home Shop Harrys do have their place (especially when they know a bit of science and have access to HS-video cameras...).

Rickyb · Feb 27, 2021

Interesting vid. Does make you think.

Milland · Feb 27, 2021

Rickyb said:
Interesting vid. Does make you think.

Sure does. I'm trying to figure out if I can borrow a HS camera to do my own tests, but it won't be soon, unfortunately...

boslab · Mar 1, 2021

From my limited knowledge would it be fair to say chatter is in fact resonance?
You hit the natural resonant frequency of the system, to overcome somthing needs to change, speed feed clamping etc?, when I get chatter I tend to change speed (the worst chattery thing I ever turned was about a 12’ ring on a craven lathe, awful, it rung like a bell, we did it on a vertical no problem
Just thinking
Mark

Marty Feldman · Mar 1, 2021

An excellent vid, not your typical YouTuber. Thanks for this one.

-Marty-

Milland · Mar 1, 2021

Marty Feldman said:
An excellent vid, not your typical YouTuber. Thanks for this one.

-Marty-

Hi Marty,
I thought the same, and it's worth checking some of the others the guy has done.

I don't know if we keep a listing of good Ytubers anywhere (like this guy and David Wilks), but we should...

Milland · Mar 1, 2021

boslab said:
From my limited knowledge would it be fair to say chatter is in fact resonance?
You hit the natural resonant frequency of the system, to overcome somthing needs to change, speed feed clamping etc?, when I get chatter I tend to change speed (the worst chattery thing I ever turned was about a 12’ ring on a craven lathe, awful, it rung like a bell, we did it on a vertical no problem
Just thinking
Mark

Hi Mark,
NRF certainly plays a role, but I think it's more complicated than that, due to the variations in force of the cutter digging in or bouncing away.

In our machines it's not as dramatic (at least, I don't think so) because of their inherently greater stiffness, but that's why I'd love to try to film my own HS video of cutters at different parameters on a real VMC (well, just a Haas VF-series), to see how much motion can be seen under abusive cutting.

Shawnrs · Mar 1, 2021

A couple of things, I think I work with people that move that slow and it is real interesting seeing the endmills start to break up and shatter.

DavidScott · Mar 1, 2021

Too bad he doesn't have a real machine. The work piece moves .009" and the tool .02"?

Joe Gwinn · Mar 1, 2021

boslab said:
From my limited knowledge would it be fair to say chatter is in fact resonance?
You hit the natural resonant frequency of the system, to overcome somthing needs to change, speed feed clamping etc?, when I get chatter I tend to change speed (the worst chattery thing I ever turned was about a 12’ ring on a craven lathe, awful, it rung like a bell, we did it on a vertical no problem
Just thinking
Mark

Yes, chatter is a kind of self-sustaining resonance. The resonance is made to oscillate due to power amplification, where intermittent contact between tool bit and work causes the tool and/or work to ring in unison.

CarbideBob · Mar 1, 2021

Would have been interesting to see the shallow stepover done with lower spindle speed or higher feedrate to maintain effective chip load the same as the other tests.
Bob

boslab · Mar 1, 2021

What I was thinking , would it be possible to use audio to actually find out what frequency to avoid, like ringing a grinding wheel, bash it record the audio to give you a predictable chatter zone?, also is this resonance implicated in say fatigue in aircrafts or example, or cracks in pipes
So many thoughts pop in excuse me, do electrical systems have resonance
Mark

CAMasochism · Mar 1, 2021

boslab said:
What I was thinking , would it be possible to use audio to actually find out what frequency to avoid, like ringing a grinding wheel, bash it record the audio to give you a predictable chatter zone?, also is this resonance implicated in say fatigue in aircrafts or example, or cracks in pipes
So many thoughts pop in excuse me, do electrical systems have resonance
Mark

This is actually the whole point of MSC's MillMax program. We've done a test with it and are looking at another. Took a 3/4" endmill doing a full width cut about 5/8" deep and you couldn't even hear it running using the settings they recommended.

MW Services MillMax Pilot Program - Landing Page Video - YouTube

*added video link

mhajicek · Mar 1, 2021

boslab said:
What I was thinking , would it be possible to use audio to actually find out what frequency to avoid, like ringing a grinding wheel, bash it record the audio to give you a predictable chatter zone?, also is this resonance implicated in say fatigue in aircrafts or example, or cracks in pipes
So many thoughts pop in excuse me, do electrical systems have resonance
Mark

Yes. There was a company advertising a service to do just that several years ago. Very expensive, had to be done for every tool / workpiece combination, and since the workpiece changes as you cut it you'd have to do it for every cut to get the optimal parameters. Makes sense if you're trying to squeeze the last drops out of long run production.

I think I might have even seen something about a machine that could do it on the fly.

I don't get the viewpoint that you have to take a certain sized cut to avoid chatter and rubbing though. I regularly shave tenths to size in prototype parts on my five year old Haas, sometimes with 8" gauge length cutters, in 17-4 H900 and Ti grade 5. Works a treat every time. Maybe the cutters I'm using are better than those people had when developing that opinion in past decades?

Hardplates · Mar 1, 2021

mhajicek said:
Yes. There was a company advertising a service to do just that several years ago. Very expensive, had to be done for every tool / workpiece combination, and since the workpiece changes as you cut it you'd have to do it for every cut to get the optimal parameters. Makes sense if you're trying to squeeze the last drops out of long run production.

I think I might have even seen something about a machine that could do it on the fly.

I don't get the viewpoint that you have to take a certain sized cut to avoid chatter and rubbing though. I regularly shave tenths to size in prototype parts on my five year old Haas, sometimes with 8" gauge length cutters, in 17-4 H900 and Ti grade 5. Works a treat every time. Maybe the cutters I'm using are better than those people had when developing that opinion in past decades?

Pretty sure MSC offers a service like that. I seem to remember getting an email from them not that long ago about it.

eKretz · Mar 2, 2021

Hardplates said:
Pretty sure MSC offers a service like that. I seem to remember getting an email from them not that long ago about it.

Yes, it was already mentioned. It involves putting a vibrometer or something similar on the spindle and tapping the spindle with a glorified hammer (to simplify it) and checking each individual piece of tooling to get the most stable settings that avoid resonant frequencies. Sometimes there is only one "sweet spot" in terms of speed/feed and sometimes there are many. It apparently works pretty darn good, especially for troublesome long reach tooling or thin-walled parts - areas where the extra help can really be used.

Milland · Mar 2, 2021

And there's work going on to control and prevent chatter in "real time" using a variety of methods. Here's a video on one method, although the actual mechanism isn't explained fully:

4F Active damping for heavy-duty spindle - YouTube

Mtndew · Mar 2, 2021

DavidScott said:
Too bad he doesn't have a real machine. The work piece moves .009" and the tool .02"?

I thought the same thing when I saw this video the other day.
His machine is such a piece of shit for this type of test that it actually tells us nothing.

Milland · Mar 2, 2021

Mtndew said:
I thought the same thing when I saw this video the other day.
His machine is such a piece of shit for this type of test that it actually tells us nothing.

I wouldn't call it "nothing", but as I already said it would be nice to do the same sort of tests on a real machine. A lab I've worked with has some fancy strobe capability, but that's a "lossy" method, I'd rather use a real HS camera for capturing all the data. I need to contact them and see if they have anything I can borrow.

With Covid protocols, it may be a while...

Milling chatter in slow motion and audio spectrum analysis video

Diamond

Cast Iron

Diamond

Titanium

Titanium

Diamond

Diamond

Stainless

Diamond

Stainless

Diamond

Titanium

Stainless

Diamond

Stainless

Diamond; Mod Squad

Diamond

Diamond

Diamond

Similar threads