During second alignment in same program,recall use_part_setup and recall: A1 , What is the meaning of use_part_ setup .Pl find below image for your reference.
For some reasons ( Mostly about reduce checking time), we do not use ITERATE alignment. What we do : First, use read point to set up initial alignment. Second, measure datum A Plane_A, datum B and Datum C cycle or round slot datum B and cycle or round slot Datum C. Last, recall startup and build final alignment with Datum A ,B,C. From time to time, I can tell the variation about our way was less than 0.2mm through measurement, usually around 0.10 to 0.15 mm. My question is if we use ITERATE alignment, what kind of accurate we could have? Under 0.05 mm? I am not sure whether there is an answer, just curious
Impossible to say without seeing exactly what you are doing in every situation. For example, with your read point method, after you measure your "PLANE_A", do you add an alignment which levels to it BEFORE proceeding to measure your circles / round slot? If not, then you are introducing error because the workplane the 2D features (circles lines and slots) are projected into will not have been updated to take into account the part's orientation on the table / fixture (because it will still be based on the CMM's home co-ordinate system - STARTUP). For iterative alignments, it depends on the types of features you are using, where they are on the part, the number of iterations performed, the iterative alignment tolerance and, if you are using more than 3 features for the LEVEL, the fixture tolerance.
Already we aligned holding fixture that is A1 iterative.In same.part with pattern holes.(4 holes) individual reference from that 4 holes.we need separate alignment..So again we need separate iterative alignment bcoz that face consist of compound angle.
Thank you so much. Yes, we start with datum B or Datum C first, usually it should be vector 1 or -1 in one of 3 directions, after B and C , then we measure the plane A. You are right the cosine error would affect the accuracy of result depends on the angle definitely since we only measure all those features one time.
