g54.2 question

[Jeremy Herron@DBS Solutions™]

I have 2 variaxis 730's with matrix controllors, and I use g54.2 all the time. We always try to have A and C at home during our setup, but sometimes this is not possible. I had an operator setup a tube like parallel with the Z axis and we used A and C to make it vertical. He had entered all the numbers into his g10 line correctly to set his dynamic offset. The program indexes to A-90. and C0. calls g54.2 and drills six radial holes in the part that are perpindicular to the centerline of the part. The initial hole is correct, but as the C indexes to the next hole only X, Y, and Z are adjusted to maintain position. A stays at the same angle and it has to adjust to stay perpendicular. I cannot count how many time we have done this and it worked properly. Am I missing something here. Anyone else seen anything like this.

 

thanks in advance.

[MIL-TFP-41]

The limitation I have come up against with using G54.2 in a 5 axis machine with Fanuc controllers is that it only compensates for an offset of the program origin from the rotation center of the C-axis, it doesn’t compensate for an angular displacement for the table surface- IE if a part doesn’t sit square on a fixture.Lets say your guy had to put the A-Axis out minus 2 degrees at C0 for the part to dial in. Rotate to C180, and now your A would have to be positive 2 degrees for the part to dial in. I am not sure about Mazak's flavor of G54.2...tho I can say that Fanuc will not do this. They do have yet another option, G54.4, which does

[Jeremy Herron@DBS Solutions™]

Mazak has confirmed exactly what you have described and oddly enough they just released their own version of g54.4 this week.