[labuda]

Hi,
When i run my G-code that I generate with Catia V5 (using the Fanuc16_Mill3 post). I get this arc differs error which i don't understand. I've machined many things so far with this combination of programs, and no problems.

What am i doing wrong?


Mach 3 stops me at Line 26:


( OPERATION NAME : Tool Change.1)
( T3 5/64" End Mill)
( Tool 3 machining time is 309.4 seconds)
N10 T3 M6
N12 M8
N14 S7500 M3
( OPERATION NAME : Profile Contouring.1)
N16 G54
N18 G00 X-7.492 Y18.788
N20 G43 Z2. H3
N22 G01 Z.45 F300.
N24 X-7.292 Y20.246 F75.4
N26 G17 G02 X-6. Y20.992 I1.292 J-.746
N28 G01 X6.
N30 G02 X7.492 Y19.5 J-1.492
N32 G01 Y-19.5
N34 G02 X6. Y-20.992 I-1.492
N36 G01 X-6.
N38 G02 X-7.492 Y-19.5 J1.492
N40 G01 Y-18.788
N42 G00 Z2.
N44 Y18.788
N46 Z1.275
N48 G01 Z-.1 F300.
N50 X-7.292 Y20.246 F75.4
N52 G02 X-6. Y20.992 I1.292 J-.746
N54 G01 X6.
N56 G02 X7.492 Y19.5 J-1.492
N58 G01 Y-19.5
N60 G02 X6. Y-20.992 I-1.492
N62 G01 X-6.
N64 G02 X-7.492 Y-19.5 J1.492
N66 G01 Y-18.788
N68 Z2. F1000.

[ger21]

Try changing your IJ mode.

[9lrac9]

I have had this problem when I first got Mach3 in 2006 and it was a math problem one or the other has to change. I use the ( I-J ) Mach3CPST. Mach2BPTS is where I had the problem at first, then I upgraded to M3 and still had the problem. That's when I put Mach3CPST and I have not had any problems since.

9lrac9

[guy2b1]

I encounter the same problem in Mach3, what should I do? What is IJ where can I find this IJ and how to change it?

[Superman]

Some controls don't like two G-codes on the same line
or a plane shift (G17) on an arc line, try putting the G17 on the line before the G2/G3 line and see what happens

Forget about the IJK they are more accurate than the R


N18 G00 X-7.492 Y18.788
N20 G43 Z2. H3
N22 G01 Z.45 F300.
N24 X-7.292 Y20.246 F75.4
G17
N26 G02 X-6. Y20.992 I1.292 J-.746
N28 G01 X6.

[ger21]

I encounter the same problem in Mach3, what should I do? What is IJ where can I find this IJ and how to change it?

In general config, try changing from absolute to incremental.

[HuFlungDung]

If this is only an occasional error, then it is probably due to incorrect rounding off of the arc center coordinates. The fix is better accuracy in rounding off, and both your CAM and Mach have to agree on the same set of rounding rules.

[guy2b1]

Thank you all I got mine working now by changing from absolute to incremental

[neilw20]

NCPlot is a great way to check for these problems before putting the code in the big machine.

I always run my stuff on the screen first with this. Saves heaps of work.
Great tool at a reasonable price. Picking up mode problems is really easy.
www.ncplot.com