PDA

View Full Version : What the ... !



alsinaj
07-16-2019, 03:50 PM
Am getting inconsistent results with something that should be dead simple, and don't understand why. I need to make 20 plastic buttons, 1.25" OD, .230" long, turned down to 15/16" over half of its length. Bought a rod of acetyl copolymer, 1 1/4" diameter, chucked it in the 3-jaw, mounted a HSS parting tool 3/32" wide, and started cutting. Set the cross-slide stop so I could plunge to the 15/16" diameter. Mounted my Noga on the front way with a dial indicator reading saddle travel.

Should be easy, right? Set the cross-slide stop, plunge to 15/16" diameter, move the saddle the width of the blade plus .115", remove the cross-slide stop, part off, rinse and repeat.

But no! I have two problems. One is that the parting tool, which I've use countless times on aluminum and steel with no problem, does not cut straight. When parting off, the resulting surface is convex, with up to .040" curvature over the .625" radius. The blade is sharp and ground straight across, with just enough stick-out to reach the center of the work. The blade is set parallel to the chuck face. The toolpost and gibs are all tight and the work is tight in the chuck. The saddle is locked while parting, and the dial indicator shows no saddle movement while cutting. I don't get it. Am I missing something?

Second problem, which may be related to the first, is that the resulting dimensions bear no relationship to the amount the dial indicator says the saddle moved between operations. For example, if I part off, then move the saddle .230" + 3/32" and part off again, the resulting disc doesn't measure anywhere near .230" -- it could be anywhere up to 50 or 60 thou off. I have six or eight of these things now, and no two are the same. I've also tried setting the offset between the side of the parting tool and the end of the work with a digital caliper. Same (completely unpredictable) results.

Seems to me the tool is deflecting and/or moving, but I can't find any looseness anywhere, and I can't believe that a 3/32" parting tool could deflect as much as half its width over a depth of 5/8", and even if it did, you would think I would notice it happening.

Any ideas?

old mart
07-16-2019, 03:56 PM
How much of the bar is projecting from the chuck?

alsinaj
07-16-2019, 04:03 PM
About 4 inches.

old mart
07-16-2019, 04:11 PM
I wonder if the bar is flexing, is there any way to have less projecting? You could try honing the blade to get it sharper than ground, and make doubly sure the tip is dead square. Don't forget to lock the saddle while parting off.

Glug
07-16-2019, 04:13 PM
Is the plastic deflecting under the pressure of the cut? My quick read suggests your symptoms seem consistent with that possibility.

I work with 7/8 acetal co-poylmer a lot. And at that OD, 4" of stick would be quite a lot. You can also get weird hamonics with plastic. Your 1.25" OD is going to be substantially more stiff. Try the parting operation up close to the chuck.

Is your parting tool biased left, right or neutral, in regard to where the nib is left? That could also be a factor.

JoeLee
07-16-2019, 05:06 PM
How this is your parting tool??
If it's like 1/6" thick try a thicker one. I have the same problem sometimes.
As far as the other inconsistencies, my thoughts are in line with what the others have mentioned.
Your work could be sliding in the chuck.

JL..........

BCRider
07-16-2019, 05:48 PM
4" stick out with a material that is as stiff as overcooked pasta.... Yep, that'll make for conical parting off fer sure. Rig up a work stop of some manner... Perhaps another dial indicator reading the carriage travel? Use only enough stickout that you can do the final parting off without the tool hitting the faces of the jaws. That should avoid the flexing of the plastic away from the parting tool and the resulting inconsistent results.

Keep in mind too that plastic is hellish and needs a razor sharp scraping edge to cut well. The slightest dullness and it'll push the plastic away and tend to cut in little lunges into the material.

Glug
07-16-2019, 05:56 PM
As you feed into the cut with the parting tool, the cut end has less and less support to resist defection from the cutting forces. Wouldn't the deflection that occurs naturally create a convex surface on the cut piece?

You should be able to visually test this. Bounce a laser pointer off the face of the stock to a distant surface. Then measure beam deflection while you take the cut.

Forestgnome
07-16-2019, 07:47 PM
The material could be moving in the chuck/collet. Acetal can be pretty hard to grip.

PStechPaul
07-16-2019, 07:59 PM
Pictures of the setup and the finished pieces would help for analysis. I agree that the likely causes are deflection of the work and possibly slipping in the chuck. It might help to wrap a piece of abrasive paper or drywall sanding screen around the piece to avoid slipping, and maybe a follow rest behind the work. Also maybe try using a live center if possible.

masheenest
07-16-2019, 08:20 PM
You didn't mention checking the center height of cut off blade. Above center is baaad.

alsinaj
07-16-2019, 08:50 PM
Thanks for all the replies. The real problem, I think, is that I didn't give enough thought to how different Acetal is from aluminum or steel. I knew that acetal is slippery, but I didn't chuck it up any harder than metal. Really cranking on the chuck key helped a lot. Using sandpaper or similar, as some suggested, would be better and easier on the chuck. All the other little hints also helped, especially reducing work and tool stick-out. Got my 20 pieces made. It cost me an afternoon, but I learned something. Thanks again. JA

Forestgnome
07-17-2019, 11:38 AM
Thanks for all the replies. The real problem, I think, is that I didn't give enough thought to how different Acetal is from aluminum or steel. I knew that acetal is slippery, but I didn't chuck it up any harder than metal. Really cranking on the chuck key helped a lot. Using sandpaper or similar, as some suggested, would be better and easier on the chuck. All the other little hints also helped, especially reducing work and tool stick-out. Got my 20 pieces made. It cost me an afternoon, but I learned something. Thanks again. JA

I forgot to mention, when I suspect slipping I put a couple of witness marks on the piece, one for axial and one for rotational slippage.

BCRider
07-17-2019, 01:16 PM
I like the idea of the sanding screen. And if the outer finish is important the idea of using finer sandpaper. Gripping the jaws tighter won't always work. At times what is needed is more of a collet like approach to get more surface area and surface friction. Tubing is an excellent example of that.

Anyway happy to read that you learned a few good lessons and got the parts made.