I am having some trouble with measuring accurate Diameter size, I have read some threads on this type of thing but the solutions provided pertain to mostly Bridge Type CMMs and I am using a Romer Arm. So I'm interested in any tips you guys may know of to get more accurate results. ... oh and it seems I have more trouble when working on sheet metal parts and I can imagine its due to the thickness of the material. Thanks in Advance
Lots to learn. Been doing it for about a year now and still feel like I'm just scratching the surface. I troll the romer forum frequently... daily... and ask questions from time to time myself. some of them are probably pretty dumb but there ya go... only way to learn unfortunately is to be dumb for a while.
I have been using a Romer Arm for more than six years now, and they just don't do diameters better than +/-.001". I've done many side by side comparisons with a Bridge type CMM, and while the arm will measure within a +/-.0002 when measuring position or profile on a small volume part, diameters are always inconsistent. This is not unique to PC-DMIS as I have had the same results with other software. I think it may be due to how it calculates diameters with six or seven axes in play, versus two with a standard CMM, but I may be wrong.
One thing I have found is that when calibrating to a sphere, I get varied measurements. The single point calibration always gets me the best results. With single point calibration, you are using the full rotation of the six and seventh axis, instead of holding the arm straight up with a sphere calibration.
I've done a few hundred thousand holes like you describe. Here's what you want to do:
1) Use an Auto Circle, not a Measured Circle. Because this allows you to...
2) Use 3 sample hits to create a local plane that the circle will be flat to. Because this allows you to...
3) Use Plane AutoTrigger. Because this allows you to probe hits NOT where you trigger the arm, but when it crosses the cut-plane.
4) Use the appropriate number of hits and a depth = 1/2 the thickness of the metal
In this way, you can probe 3 hits around the circle, the arm will emit a tone, and you can 'rub' the probe on the thin cylindrical surface that defines the bore. Every time you do, the arm will capture a hit exactly where you want it to be, at the Depth you've set.
Review your training if you don't remember how to set up the Plane AutoTrigger. Do not use the Point Autotrigger in this instance, it won't work...and I can't get into why, today.
Now your hits are placed correctly. If you have enough hits, then your circle will be repeatable. Good luck.
PS.
Most of the time with the arm, we really don't need to precisely probe points. Whenever you do, and your body's precision is the real problem, look towards one of the two Autotriggers for help.
Not as extensively as I have with probing. The accuracy will vary depending on a multitude of factors (part color and finish, operator, ect.). If you have a +/-.010 call out, scan away. Auto Circle extraction works well once it is setup, it may require some tweaking at first, but is well worth it if you are establishing a program to measure a large number of parts. My only gripe with scanning in PC-DMIS is the lack of filtering features found with other types of software, there tends to be a large amount of artifact which does alter results.
