Idées
Automatic settings for UPR and cutoff wavelength
I absolutely love the automatic scanning parameters based on size/position/form in PC-DMIS. It's a set it, forget it thing for me. However, when it comes to the gaussian filter settings, things can get complicated. The general rule is that you need at least 7 points per undulation/wavelength. A lot of the literature talks about a given UPR/wavelength, and then changing the number of points scanned accordingly. However, the point density is automatically determined based on size/position/form, and you have little influence on the scanning path. The number of points therefore is set, and you need to change the filter settings accordingly. I use a little Excel sheet where I can enter the point density per millimeter that PC-DMIS has cooked up for me (you also need to enter the diameter for round features). The sheet will spit out the filter settings based on 8 points per undulation/wavelength (1 more than the minimum). This has worked very well for me, but it's a lot of work to set up for each feature. It's also easy to forget. Especially when scanning large diameters this can cause problems. Filtering a 500mm diameter with a point density of 6 with only 15 UPR is not ideal...
In summary: I'd love to see some kind of optional setting where you can change the filter strength based on the points per undulation/wavelength, rather than some absolute value that doesn't tell you anything without knowing the scanning density and diameter of the feature. 7 would be the minimum setting, to help prevent novices from filtering wrong. People like myself could set it to 8 and make it default. If you need a more aggressive filter, just increase the points per undulations/wavelength. Or just enter the UPR or wavelength manually, if that's your jam.
Service d'assistance aux clients par UserEcho
A practical example. I'm measuring a 34 F8 (= 34 +0.064/+0.025) hole as a cilinder. There's a 0.039 diametric size tolerance. I interpret that as 0.0195 form tolerance:
Next I find out what PC-DMIS has cooked up for the scanning parameters:
Using this Point Density and the given fact I want 8 points per undulations, my UPR should be: pi × 34 × 5.525 / 8 = 74.
Ideally I'd be able to select 'Gaussian (auto)' as my filter option, and it would just do this math for me. The 8 points per undulation/wavelength variable could be a system parameter, or something you can set under setup. I really have no preference, but if you switch filter strength often it might be nice to have easy acces to it.
Here I hope a helpfull chart with general recommandations for a Gaussian filter:
If you choose,p.ex. 30, you need minimum of 30x7= 210 points on your scan.
Marek, I'm curious where you got these recommended dia to UPR filter settings? Is this from a standard somewhere. I see a variety of values being requested by some of my customers. Thanks.
Those kinds of tables is exactly what I don't want. I don't want to work backwards from a UPR. I want to work forward from the auto settings from PC-DMIS.
Of course if filter settings are specified on the drawing (or in supplementary documentation), those instructions should be followed. But 99 times out of 100 this is not the case, and in those cases I'd like a nice stable and safe strategy to fall back on that works in harmony with the workflow of PC-DMIS. UPR is the last tab in the feature definition windows, so it makes no sense to first set the UPR and then go back to point density...