Hard time figuring this out!!!

[andyb]

Ok Guys,
I need help on this. My brain is fried. I am cutting a 6"x6" piece of .25 acrylic with a .1875 pocket that is .125 deep for a customer. See the attached pic. I'm using an Onsrud 63-510 bit for acrylic. Here is the problem. The 6"x6" cuts out dead on the pocket cuts .125 deep but the pocket does not cut the content width of .1875. The pocket on side A down the X cuts .164 wide, the side D down the Y cuts .169 wide, the side B cuts .168 wide and side C side cuts .175 wide.

Now I can see if the piece cut out the wrong size that I may have a machine problem but like I said it cuts dead on. So why is the pocket cutting out 4 different widths. One other thing is that I move the material around the table and zeroed the X and Y there then cut a sample. I have cut six sample and the problem is content.

If you have Aspire I have attached the file.

Andy B.

[bob_s]

Andy
I'm guessing that what you are seeing is bit deflection. I know that it should be the same for the whole path, but I have seen it with wood fiber doing this for a 1/8 bit. I don't know if the plastic has any internal grain?
Possibly try doing the cut in 2 passes, or offsetting in .010 and then cutting a profile inside and outside the vectors that make your pocket. If you find a way that works, combine all the cuts into one tool path. I usually try cutting the problem in multiple passes as my first fix. Getting really precise can get frustrating.

Bob

[jerry_stanek]

I agree with Bob and I think it has to do with witch way the bit is traveling going one way the bit will deflect in two directions

[letterman7]

If you insist on using an 1/8" bit, slow it down and use a climb cut. Like the others say, deflection seems to be the issue. Personally, I'd just mill it with a 3/16" and not worry about having to hog the excess.

[andyb]

Bob and Jerry,
I don't think it is bit deflection. I can run an inside profile cut after I pocket it and I still have the same measurements. If it was bit deflection this I would think the outside would be off also.

Someone on the Vectric forum suggested that the gantry may not be parallel. Not sure about that but will look into any suggestions.

Letterman,
I am using a climb cut at 1 ips. I shouldn't have a problem using a 1/8" bit. I cut acrylic all the time without any problems. It is just this pocket that is giving me a headache.

Andy B.

[bob_s]

Andy
Instead of running a pocket try it a a profile inside of your outer vector and another outside of your inner and check the size on those. That way the bit will have the same material on both sides as it cuts. I think that is what may be causing the deflection. Open on one side of the bit and cutting on the other. maybe?
Bob

[andyb]

Bob,
That is sort of what I tried and get the same results. I ran the profile after I ran the pocket cut.

Andy B.

[myxpykalix]

I've read the suggestions others have offered and if they don't apply, the only thing that hasn't been mentioned is, have you put a set of calipers to your material to make sure all 4 sides of it is the same thickness? Because if it's not the bit, and not the machine then it might be the material. Let us know that you measured the material and what it was so we can cross that off the list as a part of the problem.:confused:

[jerry_stanek]

Have you measured your bit and if it were that your gantry is off it would cut off the same for the profile and pocket unless you turn turn the material.

[gerryv]

If you run the file with no changes on a piece of wood what happens?

[Brady Watson]

Your material could be warping after you pocket out .125" deep and .1875" wide in only .250" thick material...Is the part still flat after it is cut or does it have a convex back face?

Try a piece of scrap 3/4 MDF, plywood or hardwood with the SAME bit, toolpath and other settings.

-B

[gene]

are all of the vectors closed ? could it be cutting in the line on one side and on the inside of the others?

[andyb]

Jerry,
I'm not sure if I can get a good measurement off the bit. The Onsrud 63-5xx series is a basically spiral Super-O and the Super-Os are offset a little.

Gene,
I'll try that.

Brady,
Yes it is still flat. Will try the MDF.

Gene,
They are closed. The pocket makes several passes to cut out.

Andy B.

[Brady Watson]

Are your calipers reliable? :eek:

-B

[jerry_stanek]

one other thing I would do is run the file a second time after a few minutes to let the material relax from the heat you are generating.

[andyb]

Brady,
Yes. Two calipers are giving the same reading, the customers and mine.

Jerry,
Will do.

Andy B.

[tomwillis]

Andy - if my step over distance is set too large - some funny things happen when I do pockets. Have you tried adjusting those settings?
Tom

[andyb]

Tom,
That is another thing I am going to try. I have had that problem before. I'm also beginning to wonder if it is a problem with the file. the original file was a DXF file the customer created that I imported into Aspire.

Andy B.

[paul_z]

Andy,

Use preview toolpaths to create a preview in the 3D view. At the lower right of the screen the current position of the cursor is displayed in X,Y and Z. A quick calculation and you have the widths of any cut.

Paul Z

[gene]

I would redraw it in aspire and give it a try

[Brady Watson]

Will try the MDF.


Did you try machining another material or??? This should not be that tough to resolve.

-B

[Rick W]

Andy,

Your cut doesn't look very clean. Is it possible that some acrylic has stuck to your bit and made areas wider than others???

Rik

[Rick W]

Andy,

It looks like you are melting the acrylic, not cutting it.

Rik

[andyb]

Sorry Guys... I haven't had a chance to work on this anymore. Have 115 sheets of Melamine to cut and get out the door.

Gene,
I already have it redrawn in Aspire to try.

Brady,
I totally agree with you. It shouldn't be. I just lost my employee yesterday so I have no time to work on it right now. It will be Saturday before I can cut more test pieces in other material.

Rick,
The cut looks good. The rough edge you see is the protective plastic film.

Andy B.