Annoying (and possibly risky) things Calypso does?

[[Da...]]

This is probably not a "bug", but it is an issue which I have had happen several times in the last couple of days. When probing a PEM that is not terribly close to it's 'theoretical' location (don't worry; we get 1.5mm position AND MMC for them...), the machine detects an air scan, finishes  measuring, completes everything in the 'strategy' tab, the re-centers and measures again. No problem, although it was surprising the first time it happened to me.

So, here's the issue; I have a part with a plethora of PEMs all parallel to the A-datum. The first 8 are measured in the Y+ probing direction, with the clearance plane being Y-. Moving to PEM#9, it is measured in the X- direction, with a full swooping rotation (rotating in 2 axes simultaneously, so I need some Z+ movement first..), so I changed the clearance plane in PEM#8 to being 'before only', and add 3 machine moves; 1 away from the PEM, 1 up, then 1 to move to the location to get PEM#9, with the change in stylus direction done there. Complicated, but it gets the job done without crashing anything...

So, on those times when PEM#8 is NOT well-located, the stylus finishes, moves away, moves up, changes and moves over, and then.........moves BACK in the Y- direction, rotates, and goes back to PEM#8.....

 

 

I'd like to hear what other annoying things Calypso has done for you.

[[Mi...]]

So that's a reaction to the air scan as you already determined... it's better than an incorrect result.

You can turn it off (Bad idea)

Or you can program to try and avoid it, by helping the CMM find the hole/post ahead of time.... there's lots of ways to do this. The simplest is to use a Meas. Ref.

- Open the Feature in Question

- Click the "None" button under Projection

- Click the Meas. Ref tab

- there is 3 options, use the one that makes the most sense, in this case I will assume "Own Geometry" is what you want. If it's a circle, selection Own Geometry will make the CMM probe a circle at the expected location with a simple point strategy, then immediately come back and scan with your chosen strategy but it will use the circle it just probed as the reference for finding your stud.

Not sure how this works on an RDS, but it shouldn't do any weird moves or rotations.

[[Ke...]]

Would setting the Rotate/Swivel position in +Z avoid all of the complicated move positions?

[[To...]]

In the circle, there is an a setting under the Projection button at the top. Click the Meas.Ref tab and select "Own Geometry"  This will tell the circle to measure it using 4 points first, then use the location information to adjust the location of the scanning path, thus eliminating the need for Scanning Optimization.

[[Da...]]

Please sign in to view this quote.

Now THAT sounds useful! We measure a LOT of inserted hardware, and they are all over the place!

[[Je...]]

Hey David.  I agree that Mike's solution about using "own geometry" versus the nominal to actual comparison is a useful solution.  I've been frustrated in the past as well with the air scanning situation, although I realize from an accuracy standpoint how it is necessary in some situations.  Active sensors like VAST eliminate this problem altogether.

[[To...]]

Please sign in to view this quote.

LOL  I totally missed that Michael mentioned that.

[[Je...]]

Please sign in to view this quote.

Sometimes it's nice to hear the same thing from multiple people.  It helps to confirm that the info is plausible (or that we've all been led astray 😂).

[[To...]]

Please sign in to view this quote.

I shouldn't say this but it annoys the crap outta me when I see posts that indicate the original post was not read or was not read correctly correctly.  I've done it a bunch of times.   Sometimes I even post things that I end up regretting.  Unfortunately, the new format won't allow us to edit or delete these posts.

Edited December 9, 2024

[[Mi...]]

haha I'm not offended.

Although I am a little surprised nobody has chimed in one of the other half dozen or so ways you could accomplish helping the CMM find a small feature that moves around a lot prior to scanning it for real. Meas. Ref seems to be a fairly unknown option.

[[To...]]

Please sign in to view this quote.

For circles, it's probably the easiest...lol

[[Da...]]

Please sign in to view this quote.

Well, I tried that today while revising a program that I made last week. My first element to measure was a Ø3mm PEM stud that is (predictably) in 'variable locations' from 1 part to the next. Often when I measure it, the first pass you can see the probe shaft deflecting on one side (sometimes even going 'TWANG' as you round the high point...), then air-scanning on the other side. So, today I tried "own geometry".....

What happened was, first, it decided to start measuring from the opposite side; it takes one hit, then rams the PEM stud and alarms out. Gee, I could've done THAT......

**sigh**............

Yeah, in the future I think I'll just take points, give a good retract distance, and just take a dozen hits. Probably quicker than scanning, and really, it's just a PEM being held to a positional tolerance of 1.5mm @MMC...

[[To...]]

Please sign in to view this quote.

Did you think about reducing the retract distance on this circle?  Honestly, if the locations of the PEMS are that erratic, you may have trouble with any probing strategy.

Edited December 10, 2024

[[Da...]]

Check what the circle is set for, is it O.D. or I.D.? 

[[To...]]

David might be onto something.  This whole time, I assumed you were measuring female pem inserts but maybe you're doing pem studs. I don't believe "own geometry" works on OD's.   However, you can create a new circle, measuring the stud using 4 touch points.  Then, in the scanning circle, instead of using "Own Geometry", reference touch point circle.

Edited December 10, 2024

[[Da...]]

I've experimented, and found a 2.00mm retract works OK with these, with a clearance of 5.00mm (they are fairly close together). For the future, I'm not going to even bother with scanning on PEMS unless it's for a customer like this one. Besides, I'm really using this program as an educational exercise. I've already revised it 5 times as I've learned more about Calypso.