What's new
What's new

No 4th axis: Will Helical Interpolation do this?

gmoushon

Cast Iron
Joined
May 18, 2006
Location
Illinois
Have a job opportunity to mill a series of cosmetic windows into the side of 1.25" aluminum tubing and I'm trying to devise a way to deburr the edges without a 4th axis. There will a total of about 1000 windows so I would prefer not to do them by hand.

The windows are purely cosmetic and the tolerance is very broad. The customer says the main hole can be cut with an end mill through-cut. But to deburr the edge, I need to follow the edge of the finished profile with a chamfer or ball nose. That is where I'm having difficulty.

Machine is a Mycenter0 with a Yasnac MX3 control.
It does have helical interpolation but I have very little experience with it other than tapping.
Also, I've wondered if changing the plane selection, (G18 or G19), would help.

I've attached a representative image that shows the window in two different shapes. The tilted parallelogram is the shape I would prefer but thought I may have to go with the rectangle if my only option is to use G18 or G19.

It seems to me there has to be a way to do this since I'm not trying to hold tight tolerances. In fact, the chamfer doesn't even have to be uniform...it just has to knock off the sharp edges.

Any help would be appreciated!
gm.
 

Attachments

  • Tube.JPG
    Tube.JPG
    91 KB · Views: 517
I think I'd just buzz around the inside and outside edges with a lollipop mill.
Delete the radius off your model and use the sharp edge to generate your code. You can use either the rectangle or trapezoid shape but the rectangle could be forced to convert to G18/G19 to save on code.

Either way, you're going to have a lot of code to make the edge breaks look halfway descent, so you may want to consider making it a sub-program written in incremental.
 
Usually for weird shaped chamfers like this I will just surface it with a ballnose endmill.
 
hy :)

... finish toolpath is planar
... you compute for each point the height on the cilinder
... maybe aproximate the 90degrees at corners with 4*[90/4]degres arches
 
You don't need a 4th axis for that part.

I didn't think that it did, but I can't figure out how to simultaneously circularly interpolate the Z-axis while linearly interpolating the X and Y.

I'm comfortable doing subroutines with looping etc. and could do this with very small linear Z-movements. I just thought there must be an easier way.
 
Not sure what you mean.
Chamfer the radius on the inside corner of your pocket?
G3 XYZIJ Simple 90 degree helix move

Or chamfer OD of the tube?
G18/G19 (depending on orientation of the tube)
G03/G02 (depending on which side of pocket) X/Y (depending on part orientation) I/J, K
G17 (back to X-Y plane)
 
I don't think it will be a "helical" move, but a 3d toolpath with alot of XYZ moves is what it looks like to me. I think you could get a ballnose to trace the edge and look good.
 
I don't think it will be a "helical" move, but a 3d toolpath with alot of XYZ moves is what it looks like to me. I think you could get a ballnose to trace the edge and look good.

Probably, but I think you could fake it with a helix and still look descent.
Especially if he's hand coding it.
 
That's a piece of cake in Mastercam with the 3D Chamfer tool path. I think it would be exceptionally difficult to program by hand or with conversational programming due to the simultaneous Z moves. Additionally, the program must take into account the changing diameter of the cutter as it either heels in or heels out as the Z plane changes. Here is a sample program of a .020" 3D chamfer on a .5"x.5" pocket on a 1" diameter tube using a 1/4" 45 degree spot drill with a .050" tip clearance. (XYZ Zero in the center of the pocket, top of the tube).
Part1.jpg

(T5|1/4 SPOT DRILL - 0.25|H5|D5|TOOL DIA. - .25)
N100 G20
N110 G0 G17 G40 G49 G80 G90
(1/4 SPOT DRILL - 0.25|TOOL - 5|DIA. OFF. - 5|LEN. - 5|TOOL DIA. - .25)
N120 T5 M6
N130 G0 G90 G54 X.22 Y0. S6000 M3
N140 G43 H5 Z.3
N150 M8
N160 G1 Z-.051 F75.
N170 Y.0529 Z-.0538 F35.
N180 Y.1052 Z-.0622
N190 Y.1557 Z-.076
N200 X.2197 Y.1632 Z-.0802
N210 X.2185 Y.1711 Z-.0847
N220 X.2145 Y.1852 Z-.0943
N230 X.2081 Y.1979 Z-.1042
N240 X.1995 Y.2091 Z-.1141
N250 X.1892 Y.2183 Z-.1232
N260 X.1833 Y.2221 Z-.1273
N270 X.1772 Y.2253 Z-.1308
N280 X.1707 Y.2278 Z-.1337
N290 X.1639 Y.2297 Z-.1358
N300 X.1535 Y.2311 Z-.1375
N310 X-.1535
N320 X-.164 Y.2297 Z-.1358
N330 X-.1708 Y.2278 Z-.1336
N340 X-.1772 Y.2253 Z-.1307
N350 X-.1834 Y.222 Z-.1272
N360 X-.1946 Y.214 Z-.1188
N370 X-.2041 Y.2037 Z-.1092
N380 X-.2116 Y.1918 Z-.0992
N390 X-.2169 Y.1783 Z-.0894
N400 X-.2186 Y.1711 Z-.0847
N410 X-.2199 Y.16 Z-.0781
N420 X-.22 Y.1557 Z-.076
N430 Y.1052 Z-.0622
N440 Y.0529 Z-.0538
N450 Y0. Z-.051
N460 Y-.0529 Z-.0538
N470 Y-.1052 Z-.0622
N480 Y-.1557 Z-.076
N490 X-.2197 Y-.1632 Z-.0802
N500 X-.2185 Y-.1711 Z-.0847
N510 X-.2145 Y-.1852 Z-.0943
N520 X-.2081 Y-.1979 Z-.1042
N530 X-.1995 Y-.2091 Z-.1141
N540 X-.1892 Y-.2183 Z-.1232
N550 X-.1833 Y-.2221 Z-.1273
N560 X-.1772 Y-.2253 Z-.1308
N570 X-.1707 Y-.2278 Z-.1337
N580 X-.1639 Y-.2297 Z-.1358
N590 X-.1535 Y-.2311 Z-.1375
N600 X.1535
N610 X.164 Y-.2297 Z-.1358
N620 X.1708 Y-.2278 Z-.1336
N630 X.1772 Y-.2253 Z-.1307
N640 X.1834 Y-.222 Z-.1272
N650 X.1946 Y-.214 Z-.1188
N660 X.2041 Y-.2037 Z-.1092
N670 X.2116 Y-.1918 Z-.0992
N680 X.2169 Y-.1783 Z-.0894
N690 X.2186 Y-.1711 Z-.0847
N700 X.2199 Y-.16 Z-.0781
N710 X.22 Y-.1557 Z-.076
N720 Y-.1052 Z-.0622
N730 Y-.0529 Z-.0538
N740 Y0. Z-.051
N750 G0 Z.25
N760 M5
N770 G91 G28 Z0. M9
N780 G28 X0. Y0.
N790 M30
 
try 4+4*3+2+2=20 lines :)

4 : for the sides : 2 constant Z + 2 variable Z
4*3 : 4 corner arches * 3 segments each
2 : in & out movement in XY plane
2 : in & out movement among Z axis
 
As said prior, this is just 3D profiling (plannar) OR you could extract the edge of the pocket where you need to chamfer. Offset the shape by the 1/3 the tool radius. This is your centerline path of the tool.
By moving the entity in Z- you will increase the amount of chamfer your chamfer mill will generate.

Doug.
 
Thanks for the feedback Brian and Booze.

And yes, I'm hand coding this.

A 3D toolpath is what I thought it would need as well and I didn't know if there was a g-code that would allow me to move the X&Y and one feed rate while simultaneously moving the Z at a different feed. That would be one way I could see doing it without having to make a bunch of small discrete moves.

The second way I thought was to use a helical motion but, again, I still consider myself a novice on a lot of this and am not quite sure how this.

To answer your question Booze, the goal is to relieve the sharp edge of the window on the OD of the tube.
 
there is a real nice deburr tool made just for this.
ezburr.com, or jwdone.com both make nice products.
 
On the internal pocket corners, the Z is going to change non-linearly as you complete the arc.
Smaller Z moves per XY at the beginning, larger Z move per XY at the end. There isn't a G code to do that.

Best I could suggest is to look at your solid and explode that radius into lines and inquire the end point at the beginning, middle and end of the arc.
Program 2 helix moves.
G01 XYZ (beginning of arc)
G3 XYZ IJ (middle of arc)
G3 XYZ IJ (end of arc)

You should be able to get a reasonable looking chamfer.

Or post the size of the pocket, inside corner radius and tube OD and I'm sure someone would be willing to run it thru their CAM and post the code.
 
Thanks Booze. That makes a lot of sense.

I've ordered some scrap tubing to play with this weekend. We'll see how it looks.
gm
 








 
Back
Top