Why is this program making my mill chatter?

[TXFred]

I could use a little help here. I'm using my Tormach 1100 on steel for the first time, and I'm getting some serious chatter issues.

I'm running a brand new, 1/2", four flute Meling end mill in a set screw holder. I'm using trochoids, with a stepover of 0.0625" and a cut depth of 0.08" Feedrate was determined by GWizard (see screenshot).

The workpiece is a slab of 1" steel plate from my local metal supplier.

The initial plunge takes place off of the edge of the workpiece, and then the trochoids create the initial slot.

Given my depth of cut and stepover, it seems like I am babying the mill. But the chatter was so bad that I had to hit the emergency stop.

Also, I've tried this same code with the feed reduced to 8.6 IPM, and it still chatters. I've also had some issues with the TTS holder pulling out of the collet.

Am I asking too much of the mill, or is my code just really bad? I welcome your advice.

Sincerely,
Frederic

[TXFred]

With one call to my metal supplier, I may have answered my own question. I programmed for 1020, but what I'm cutting is A36.

GWizard doesn't list A36, only 1020, 8620, 1045, 4130, 1090, 52100 and tool steel. Which of these is the closest to A36?

[pete from TN]

But here is a link...

http://www.eaglesteel.com/download/t...eel_Grades.pdf

A36 is the most common grade of hot rolled steel and it actually kinda difficult to machine moreso than 1018/1020 cold rolled steels... You can read up on it there...peace

Pete

[photomankc]

Try hitting the more button next to the material selection in GWizard. It should have a huge array of materials you can set. I did A36 in a 7" round on the lathe and machined pretty darn well, but I never milled it.

[Scott_M]

Hi Fred

Your speeds and feeds seem pretty reasonable but I think the problem is with the trochoids. Looking at your code the feed rate dosen't change for that ity bity radius it is probably making those little arcs pretty fast.
I have never had any luck with the trochids toolpaths in Sprut. Have you tried just a regular slotting path at those speeds and feeds ? I'll bet it works out fine. Try a test cut on the steel with the jog wheel.

Scott

[TXFred]

Good advice from one and all. I got it to cut without chatter. Although it still cannot seem to handle anything more than about 0.1" DOC. Is this normal for a Tormach? What's the deepest cut you can do on your 1100?

Frederic

[BobWarfield]

Actually, the regular slot will want to be slower.

I downloaded the g-code and loaded it into G-Wizard Editor, then ran the Revision to convert from IJK to R mode for arcs. I wanted to see how small the little buggers are--0.012 or so.

Makes me wonder a couple of things:

- Do Trochoids do what you think if the loop is smaller than the tool? I bet not, but I don't know what the threshold relationship is. Sure enough, this paper suggests Trochoids lose their effectiveness when the slot is less than 1.2 x tool diameter:

http://mmc.me.kyoto-u.ac.jp/pubs/pap...m03_yamaji.pdf

So, try setting up your trochoid to be 0.6 or even 0.75" and that may improve the results. Your part looks like there's plenty of room to do this.

- How well does the Tormach track those little tiny loops and is the impression there is chatter just a lot of vibration from banging around on the loops?

Cheers,

BW

[bobeson]

Frederic,

Are you using the Tormach TTS power drawbar? If so, I'd recommend swapping out your belleville stack for an equivalent-length solid bushing. In my experience, this helps when cutting steel with a width of cut less than the tool radius. I have cut lots of A36 with my pcnc, with generally satisfactory results. I usually remove all the "skin" (mill scale) with a face mill first, during blanking. Lately I've taken to buying a lot more cold-rolled 1020 to avoid this step. I frequently make full-thickness cuts in 3/8" 1020. In addition to swapping out the drawbar springs for a rigid bushing, I find that making cuts wider than the cutter radius greatly reduces chatter.

Good luck,
-Bob

[TXFred]

I hate reading a thread where the original poster doesn't follow up and say how they eventually solved the problem. So I'm going to embarass myself in this post. Feel free to laugh at my expense. It won't be the first time.

One thing I didn't mention in my posts is that I was seeing a lot of TTS pullout along with the chatter. I thought that this was due to the chatter overwhelming the grip of the collet on the toolholder. I was wrong.

First of all, my feeds and speeds were off. And as Bob stated, my trochoids were too small. I didn't feel like troubleshooting this last night, so I eliminated the trochoidal cuts. To get the part done, I cranked the DOC way down and finished the part.

But I have found and solved another problem which should make a huge difference. This problem was located between the mill and the chair. When I read the manual for the power drawbar, I read,"90 PSI Minimum" and thought it was "90 PSI Maximum." So I set the regulator to 90 PSI, and set the drawbar tension to where the PDB could just barely release it. I also failed to put anti-seize on the threads or outside of the collet. I did, however, have a good amount of grease and dirt on the inside of the collet.

From Tormach's document on preventing collet slip,"Improper surface preparation of the inside of the collet can reduce tool holding force by 66%. Improper maintenance of the drawbar and lack of lubrication can reduce tool holding force by 70%." Now that I'm running 125 PSI on the PDB, I also have more clamping force since I can tighten the drawbar farther. My previous pressure was 72% of what it is now. Let's do the math on this one. My clamping force can be expressed as a percentage of the maximum 100%.

100 * .33 * .30 * .72 = 7.128

In a worst case scenario, I had only 7% of the full clamping force available at the collet. Is it any wonder that I was having problems?

I'm going to take some leftover A36 and do more testing. I'll post my results. I have a feeling that with a good trochoid, the right feed and speed, and having a tool that is properly coupled to the mill, it's going to cut a whole lot better.

Sincerely,
An Idiot

[BobWarfield]

Fred, looking forward to hearing how well the revised Trochoid works!

Best,

BW

[flyinchips]

Fred

No laughing here. Thank you for following up. Personally, I find the problem is between the Mill and the Chair 90+ % of the time, but that's just me.

[Scott_M]

This syndrome is also know as a "Pebkac" error.

"Problem exists between keyboard and chair"

We all do it !! Glad you got it sorted out.

Scott