I need to measure a dimension to the centerline of a radius, but the radius is 2" and only covers an arc of about 25 degrees. Measured on the optical comparator, the radius is nearly nominal, but on the CMM it measures 2.117". This throws my centerline over .100" out of position. How can I create a centerline from points on the radius using the nominal radius? Can I force PC-DMIS to best fit a 2" radius to points on the surface?
Use a fixed Rad algorythem to find center of the rad. then origin on that circle. Now measure points on the radius and report the PR. It gives more of a profile but it will do your trick.
Could you type in the theoritical location of the first radius (ie create a point) or measure it, and in polar coordinates, take a few profile hits on the smaller radius and tolerance out the profile relative to the first rad?
Keep in mind that I've not tried this but sounds like it should work.
You might have one shot at making this work... the problem is, that as prev. stated, the measurement algorithm breaks down on small arcs, but also due to that breakdown should give you nonrepeatable results. As the error is a "normal" error, it is subject to statistical analysis. So, if you do get such "nonrepeatable" measurements, run the part twenty or thirty times and look for the x-bars on size and position. I have successfully used this method on other systems.
Use a fixed Rad algorythem to find center of the rad. then origin on that circle. Now measure points on the radius and report the PR. It gives more of a profile but it will do your trick.
Sean is on the right track. We just discussed this in the user group. And there is a way to get an accurate reading for a small arc segment. I have to find my notes but I think Sean is correct.
When asked about it the "experts" quickly rattled of the sequence to achieve an accurate reading. They should put a step by step set of instructions together for everyone to go by. Every time someone says that the CMM doesn't measure worth a **** they bring up the segment of an arc. I frickin hate that! That's why I wanted know how to do it. So I could stick it right up their azz!!
Pull the nominal CIRCLE from the radius in CAD, I'm assuming that you have CAD, set your origin at the center of the circle and then take polar radius hits on the arc. This is the only way I've ever gotten PCDmis to work for small cicrle segments.
Sean has my agreement also. It is NOT a Pc-dmis thing, or even a cmm thing, its a mathematical issue. Polar coordinates are the best way, from the known radus center. I wish ppl would say its not possible, if I don't know for sure, I just say that, not, "It can't be done with Pc-dmis".
I AGREE W/ THE ABOVE POST. WE GO WITH A KNOWN CENTER LOCATION THEN
USE POINTS. THIS METHOD IS SIMPLE AND VERY ACCURATE AND QUICK. YOU WILL
GET AN ACCURATE RADIUS.
Do you just take one individual point or many? If many, do you construct a circle out of all the points? If using only one individual point how can you do a True position call out?
