Best scan selection

Told you all I would be back. I created 6 variations of the same scan. Relearned with Nominal, Master, Find Nominals and Defined with Nominal, Master, Find Nominals.

I specified 500 hits for each of these 6 different (but same) scans. Why then after the runs do I have a spread of 139-194 hits per scan? I can verywell understand Relearn adjusting my counts, however I struggle with Defined. Even when 'generating' the scan features over those same start and end points I was recieving somewhere around 240 hits (offline).

What settings need to be adjusted for me in order to achieve a denser point collection?

Additionally, after the first test run through each of the 6 different scan groupings, I re-ran them all in a group 1 right after the other and the point collection changed in 4 of the 6 after their first run. (Relearned/Defined with Master were the only ones to consistantly report the same alotment of points on both runs)

MASTER sets the XYZIJK points and they STAY THE SAME, they do not (and never will) update to new XYZIJK.

To get more points, decrease the increment, which is the spacing between points. Keep in mind, I only have touch trigger, so I don't know what additional options there may be for scanning heads.

MASTER means it is MASTER, not changing (like a master gage).

FIND NOMS does just that..... and even with MASTER, find noms WILL have an effect. ALL machines have drift, all of them. When the probe drifts away from the nominal XYZ, it will re-pierce the cad to find new nominals for that point. MASTER will not let it add or subtract points, but each point will now have new XYZIJK nominals (even if they only change by a few microns). Eventually the scan will drift away from the original generated scan, but the number of points will not change.

RELEARN will generate new XYZIJK every time as well as new numbers of points (either more points or less) since it is re-learning everything, XYZIJK and # of points.

So, it has been a while and I completed writing an entire program, now going through the trouble-shooting phase. I must say, THANK YOU for the help I have recieved here! Hopefully there is more to come:

I've started to mess around with the OPT probe settings moreso lately along with the "max" increment. This option is really puzzling me. When I had first established a scan along a R0,40mm it showed me about 20 hits it was willing to recoup the data for. Since the initial settings, I have upped the point alotment in the opt probe settings and brought the "max" increment down to less than 50% of what it had originally been set to, you would likely expect (atleast I did) that this would allow for MORE points to be taken by the scan. However the MOST points I get now are 2-6 in most of these. I get higher number of points showing up in the scan when I use a max increment of say .002 and am ranging from 2-4 using .0008, anyone have any idea what could make something like this happen?

Do I need to reset my 'D' in my scan dialog box? Thats the only thing I can see that continues to shift itself. My start and end points remain the same as they had been when I first created the scan and have stayed the same durring any edits or running of the part program.

Matthew,

I'm creating linear open scan sections on surfaces with complex ("irregular") geometry. These surfaces do not deviate from the CAD model, usually, more that .020 of an inch. I am measuring a complex set of peramiters with these scans (combined geometric relations to a datum, contour, profile form & location, etc.). I'm doing all of this with touch probes (with points set at .100" intervals) as the surfaces have not been polished to a fine finish when they come to my lab. From what I've read from your responses in this thread, I'm assuming that you believe I should be scanning using the MASTER options for the nominals? If I'm programming offline, when I "GENERATE" the new scan, PCDMIS is piercing the CAD to find the nominals based on my current alignment when using MASTER and these nominals are then locked in? Are you saying that, even when using the MASTER method, the nominal will self adjust due to "drift"?
I'm just looking for some clarification on this, as it's always been a little fuzzy for me.

Matthew,

I'm creating linear open scan sections on surfaces with complex ("irregular") geometry. These surfaces do not deviate from the CAD model, usually, more that .020 of an inch. I am measuring a complex set of peramiters with these scans (combined geometric relations to a datum, contour, profile form & location, etc.). I'm doing all of this with touch probes (with points set at .100" intervals) as the surfaces have not been polished to a fine finish when they come to my lab. From what I've read from your responses in this thread, I'm assuming that you believe I should be scanning using the MASTER options for the nominals? If I'm programming offline, when I "GENERATE" the new scan, PCDMIS is piercing the CAD to find the nominals based on my current alignment when using MASTER and these nominals are then locked in? Are you saying that, even when using the MASTER method, the nominal will self adjust due to "drift"?
I'm just looking for some clarification on this, as it's always been a little fuzzy for me.

The 2 options, NOMINALS and EXECUTE, when set to MASTER and NORMAL when you generate and create (without measuring) the scan, lock the XYZIJK values in, they will not change.

You DO need to have FIND NOMS DURING EXECUTION un-marked in the F5 setup screen. I would also un-mark FIND NOMS in the same setting menu.

IF you have EXECUTE set to RELEARN... that is for a continuous contact probe (either laser or touch). This allows the machine to 'flow' with the part, no matter how much it changes. If it isn't set to re-learn, then the probe will not maintain the correct contact pressure (coutinuous contact) or correct read distance (laser) when scanning. This SHOULD only be used for continuous contact or laser scanning.

IF you have NOMINALS mode set to FIND NOMINALS then each and every touch made will find new XYZIJK value for each touch. This can be used for any type of scanning, touch, continuous contact or laser.

So, since you are using a touch trigger probe and you want the XYZIJK to stay the same, you want them set to MASTER and NORMAL. This will keep the scan point's XYZIJK the same forever and ever (unless something "gleeks" the program).

Now it's CRYSTAL clear. Just the kind of detailed answere I was hoping for, Thank you.