Cutter comp Internal corner

[K2csq7]

Hi All,

One of the guys here made a toolpath for a .250 EM into a much smaller radius (see pic). Below is the beginning of the code

N230 G41 D73 X-.1761 Y-.2544 F6.
N232 G3 X-.148 Y-.2704 R.0625
N234 G2 X-.128 Y-.2757 R3.61
N236 G1 Y.0206                                             	<<<<<<<<<<<<<<<<< X moves here but shouldn't

 

We use wear comp, and I tried with I's & j"s but got the same result.

 

When the operator puts in -.002 for the cutter comp, the control is having problems when the toolpath gets into the sharp corner. As the machine moves up in Y (line N236 above) the X axis moves as well, I know the control is trying to protect both surfaces at the same time, but something is going wrong. I adjusted his geo so the tool swings a .003 arc in the corner and all is OK, but I just need to know. Is it just as simple as the control can't comp two surfaces (lines) at the same time or is it something that will bite me down the road?

 

TIA

[Buddy J]

We used to program exclusively in control comp and checking the "optimize cutter comp in control" box

used to fix stuff like this.

Have had less issues since switching over to wear comp but I make sure the box is checked anyway, just in case.

 

Checked??

[MotorCityMinion]

Try extending the linear entry and exit line some. If that is not feasible, use a smaller cutter diameter. Use linear line lengths that are at least half the rad of the cutter, plus your lead-in lead-out values, plus some for the actual cutter deviation from nominal size (your comp value and spring). IMO, lose the R's and use IJK's if possible. Even posting with IJK's and using wear comp can give you bad results with that geometry if the lead in/out and comp values are too large.

 

You can also add start and end points to the chain for force a more suitable location for comp on and off.

[K2csq7]

Thanks Guys, good advice. Buddy, that checkbox gets greyed out as soon as I switch to wear.. MCM, I will try to legthen the lines and see if it makes a difference at the control. ( linear lead in/out = .0625 now)

[Craig-B]

Extending the lead in/out should do it, You just don’t have a enough ramp on time for the machine to apply the radius comp from the control.

[K2csq7]

We are only trying to put in -.002 for the comp.... why is the .062 not enough? If someone asked me before I ran into this, I would have said .020" linear lead in would suffice...

 

Is it just as simple as the control can't comp two surfaces (lines) at the same time?

[iowajim62]

Keith,

 

Just looking at the geometry, do you need any leadin/out arcs to clear other geometry ?

I would try no arcs and just linear leadin/out if it will clear.

[K2csq7]

Iowajim,

I did try that, same results, also tried tangent lead in with no radius. Toolpath runs correctly until it gets into that corner.

BTW Control is a Fanuc 18i-MB.

[MotorCityMinion]

"BTW Control is a Fanuc 18i-MB."

 

Use IJK's not R.'s

 

 

[valus]

You ran into an interference. The vectors of those two motions have intersected each other. Even with an .003" rad, if you compensate -.004" you will have the same problem. Try control comp.

[K2csq7]

MCM, thats the first thing I tried, same results.

Valus, " The vectors of those two motions have intersected each other.".... I have run end mills into corners (no radius) a million times and never seen this, what makes this one different?

[iowajim62]

Did you try this with a different post ? And have you tried it on a different machine if possible ?

Mostly just curious now Keith as it is rather strange. It seems as if the control doesn't have look ahead.

[K2csq7]

Did you try this with a different post ?

I can write the code by hand and get the same results, the post is working correctly.

 

have you tried it on a different machine if possible ?

We have 2 of the machines with the same control and it happens the same on both...

I do have 1 other of the same machines with a 16i control, it's got a big job on it for a few more days, if I can sneak in I'll try it out.

 

Mostly just curious now Keith as it is rather strange.

Me too, stuff like this kills me.... the code is a straight line Y move, and the machine cuts a taper. WTF...

[within a thou]

Just looking at that pic and the path shown, where is the G2 move in N234 come into play? Looking at that path you should only have G3's

[LaszloK]

Eliminate the radius that is too small and make it a sharp corner.

[K2csq7]

The entity that the red arrow points to is a big arc, the G02 is correct.

[K2csq7]

Laszlok, I did try that too but it did not affect the output code, so I didn't bother trying on the machine. The tool won't get near it it's only a .035R.

 

If anyone has a similar Fanuc control throw the code posted in the first post at it and see, the X moves while cutting line N236, completely unexpected...

[K2csq7]

If anyone wants to try it....

N200 G00 G91 G28 Z0.
N202 G0 G17 G20 G40 G49 G80 G90 G94 G98 H00
( .250 END MILL   )
N204 M21 (UNLOCK C)
N206 M11 (UNLOCK 
N208 T73 M06
N210 M00
( CONFIRM T73 H73 D73 DIA = 0.2500 )
N216 G0 G17 G20 G40 G49 G80 G90 G94 G98 H00
N218 G0 G17 G90 G54 X-.2205 Y-.2104 C0. B0. S1900 M3
N220 M20 (LOCK C)
N222 M10 (LOCK 
N224 G43 H73 Z1. M8
N226 G0 Z.1
N228 G1 Z0. F30.
N230 G41 D73 X-.1761 Y-.2544 F6.
N232 G3 X-.148 Y-.2704 R.0625
N234 G2 X-.128 Y-.2757 R3.61
N236 G1 Y.0206
N238 G3 X-.1455 Y.0286 R3.39
N240 X-.1715 Y.0343 R.0625
N242 X-.1774 Y.034 R.0625
N244 G1 G40 X-.2396 Y.0281
N246 Z.2 F50.
N248 G0 Z28.
N250 M9
N252 M5
N254 G0 G28 G91 Z0.
N256 M19
N258 M30
%

 

Here is the IJK version I tried also....

N200 G00 G91 G28 Z0.
N202 G0 G17 G20 G40 G49 G80 G90 G94 G98 H00
( .250 END MILL   )
N208 ( LOAD T73 )
N210 M00
( CONFIRM T73 H73 D73 DIA = 0.2500 )
N216 G0 G17 G20 G40 G49 G80 G90 G94 G98 H00
N218 G0 G17 G90 G54 X-.2205 Y-.2104 B0. S1900 M3
N224 G43 H73 Z1. M8
N226 G0 Z.1
N228 G1 Z0. F30.
N230 G41 D73 X-.1761 Y-.2544 F6.
N232 G3 X-.148 Y-.2704 I.0439 J.0445
N234 G2 X-.128 Y-.2757 I-.917 J-3.4916
N236 G1 Y.0206
N240 G3 X-.1455 Y.0286 I-1.428 J-3.0746
N242 X-.1774 Y.034 I-.026 J-.0568
N244 G1 G40 X-.2396 Y.0281
N246 Z.2 F50.
N248 G0 Z28.
N250 M9
N252 M5
N254 G0 G28 G91 Z0.
N256 M19
N258 M30
%