Fanuc 18 i-M/IJK missing moves

[Kelly Burns]

Area clearance toolpath, isnt picking up a small ijk move. After it blew thru the part and eventually into a 3" wall of the part stopping the cutter and forcing me to get a spindle regrind this morning, the points have been plotted and it is posting the correct code. The move that it is missing is coming from having a 6 in radius with only .0072 of arc. J is still set at zero. My control is not picking this up and creates a giant arc around into the part.

 

Can anyone help me with what parameter needs changed at the control? Ive pawed thru the book and contacted the seller of the machine (Daewoo/Doosan) but I am having little success in trying to find the setting. I keep looking for a type of corner rounding radius or something along those lines, but I am unsure of what needs changed. Any help on this would be greatly appreciated.

 

Thanks in advance.

[Kelly Burns]

[Kelly Burns]

N218 X-7.4959 Y-13.079 I2.8299 J.3719

N220 X-7.496 Y-13.0718 I2.8542 J0. (This line)

N222 G1 X-7.4956 Y-7.5375

[FP1]

We had a problem just like that a while back. It was a 99.000 radius very short move on a Mazak Vari-Axis. Luckily it was spotted before it could do any damage. Mazak upgraded the Matrix software but it still would run off like yours. They never did figure out why it happens.

[Kelly Burns]

Looking at these parameters:

 

1801 0011

1826 xyz all set to 20

1827 xyz all set to 10

1820 xyz sll set to 2

3410 circle tolerance 1000

 

Im hoping to jog someones memory.

[heavychevy2155]

Can you break the arc and make a few straight lines. You may have to sand to blend but it might be close.

[Kelly Burns]

If it was only a contour. Problems is its 15k. high speed path.

[Dave Ball]

Have you tried changing any of the system configuration tolerances and rechaining? particularly the curve minimum step size.

[Kelly Burns]

I ended up stripping all arc movement out and making linears to be safe. Fanuc didnt bother to help to much and never really did give me any answers.(I sent them over pictures of the part, program line that it happened on, and even a plot verification of the arc points) I did get great support from Alex at Fastech who wrote in an algorithm on my existing post to look for small arc movement and convert only those sections to linear. I still had some trouble with the next part gouging, and just to be safe went ahead and converted it all to linear. I still dont think that I have the answer to my original question, but have found a temporary work around. If anyone has anything else to offer on this it would be appreciated.

 

Thanks again!