We're slowly replacing our old PFXs and replacing them with Global S Green & Blue CMMs with the Leitz / Hexagon scanning head. I've used spiral scanning with Zeiss CMMs with pretty good success on small holes (Ø3.25mm x 16mmL) and wanting to know how the Hexagon spiral scanning results are. Are they consistent with multiple level (circle) adaptive scanning results? or would line scanning the length of the bore be better?
The tolerances on the holes are +/-0.025mm and positional tolerances are 0.10mm.
I asked about the filtering and never received an answer, for this feature. We only have access to Gaussian and not Cylindrical ( in the auto feature). Gaussian is supposed to be for complete circles.
I tend to prefer several concentric circles over a spiral scan only so I can report a diameter at different levels (or at lease have the option to if there are issues to investigate).
I haven't done a whole lot of spiral scanning myself, but my boss has a strong opposition to spiral scanning bores. He tells me had a bad experience in the past where he was measuring engine cylinder bores using spiral scans. A while into a production run they discovered that the BF algorithm was screwing up the orientation of the constructed cylinder just enough to throw off the diameter enough that he'll never do it again. I'm a bit skeptical. I suspect that they were just being too aggressive with the scanning pitch and not doing enough revolutions, but I wasn't there so I don't know. I just know I won't be doing any spiral scans as long as he is my boss.
Benedictj1 got me thinking about filtering spiral scans. The cutoff wavelength is done in UPR (undulations per revolution). If you filter a scan of a circle that is less than 360deg or, in the case of spiral scanning, way more than 360deg, does the algorithm really take the number of revolutions into consideration? If you do a spiral scan of 5 revolutions with a 50 UPR cutoff wavelength, does it break it up into 250 total undulations (50 for each revolution) or does each revolution only get 10?
I use them all the time when doing threads, I don't use them for just holes. They seem to work pretty good for thread, as much as accuracy can give for thread.