[Li...] Posted April 13 Share Posted April 13 Good morning, all! I have some questions and some venting to do here. LOL! We are having a company machine 3 impellers for us currently. They use PC-DMIS at this machine shop. Their programmer was able to export all the points throughout the CMM program. He has created a program that has 934 points. The one impeller has 3 blades that curve from top all the way to the bottom base of the impeller and 3 "splitter" blades. He is probing the top and bottom of each blade along with points along the edge. This is per the note section of the drawing. The exported file from his program is an XYZ file that I have put into excel. My boss and our lead SQE are adamant that he and I are "apples to apples" when it comes to the points he uses and I use, which is fine. I am just creating a point and copy/pasting XYZ, IJK. Just a lot of ctrl c, ctrl v action....lol! 5604 times! If I understood him correctly, he won't report each point individually, rather collect them all as a group and it would "in the background" get high/low deviation and he said that it will add those two results together. So it wont do max deviation x2 "profile". I will ask him about this just to re-verify. -Question #1 What can I do in Calypso to match something like this? I have 19 points pulled into a plane (recall feature points) then selected it in a Profile characteristic and did two results min/max. This result won't be "apples to apples" on a report like my team wants and expects to see. I may need to have the other programmer change what and how his program reports. -Question #2 (for those who have PC-DMIS knowledge) With myself and the PC-DMIS programmer are using the same inspection fixture, the same alignment points per CAD model points, will this result in these two different CMM software's give the same data? Is it possible for Calypso and PC-DMIS to report the same data, at least within a few tenths? When the most important items needed are present, i.e. fixture, datum scheme, and the same exact points. Link to comment Share on other sites More sharing options...
[Cl...] Posted April 14 Share Posted April 14 You could pull all Those points into a Maximum and or Minimum feature construction and that would give you the max or min point. Link to comment Share on other sites More sharing options...
[Li...] Posted April 14 Author Share Posted April 14 When I got to a certain point in the program after copy/pasting the nominal XYZ, IJK the PC-DMIS programmer was able to supply me I decided to create a few characteristics and run the program. This is being ran on my offline seat so what should have happened was to basically populate nominal data. Well, when I did this, I got big deviations on the 3 profile char. I made in the program. I have no idea what is causing this. Then I added a point using Calypso and I got what I would expect, zero deviation. There must be some sort of disconnect from PC-DMIS nominals/vectors and Calypso. Link to comment Share on other sites More sharing options...
[DW...] Posted April 14 Share Posted April 14 Please sign in to view this quote. See attached. Influences on Measuring Results.pdf Link to comment Share on other sites More sharing options...
[Aa...] Posted April 14 Share Posted April 14 Please sign in to view this quote. That's a nice graphic. In my humble opinion, by far the biggest influence is the lack of proper design tolerancing. Almost everything I see is more or less incomplete, ambiguous, illogical, or impossible. Yet all vested parties insist on moving forward with dimensional inspections so that they can see "where we're at". And from then on, it's garbage in garbage out. Improper design tolerancing forces the inspection programmer to fill in the missing information, or patch over impossible tasks. Each programmer will interpret and perform this effort differently, leading to a variety of unstable possible outcomes. Link to comment Share on other sites More sharing options...
[na...] Posted April 15 Share Posted April 15 Please sign in to view this quote. Very well put. Agrees, thanks. Link to comment Share on other sites More sharing options...
[Li...] Posted April 15 Author Share Posted April 15 Please sign in to view this quote. Yes, but our report/data won't match because PC-DMIS will take min/max deviation and add those together. This is how PC-DMIS calculates profile according to the programmer at our supplier making these impellers. Link to comment Share on other sites More sharing options...
[Ke...] Posted April 15 Share Posted April 15 Please sign in to view this quote. I wonder if they are using bilateral with two results, instead of with one result (which is max deviationx2)... or if they are using an old version Link to comment Share on other sites More sharing options...
[Li...] Posted April 15 Author Share Posted April 15 Please sign in to view this quote. I will ask him about this. According to your screenshot he should be able to do max dev x2 for profile. However, he stated that it uses all the points taken on a surface to get these results. I would be more prone to using space point distances, essentially on the impeller, there are no planes on this part, at least on the 6 blades that are on this impeller. Link to comment Share on other sites More sharing options...
[Li...] Posted April 16 Author Share Posted April 16 (edited) "If I have one point that is .0005 as my max deviation on the plus side and -.005 as my max deviation on the minus side, then it will report the profile as .0055. The tolerance I have to give to the dimension if it’s .020 is ±.010 so that it maintains the symmetry of the profile. If I had a max deviation of 0.009 one way and -0.009 the other way it will report it as 0.018 but it will not be outside the limits of the profile even though my tolerance will say ±.010 it will stay as conforming in PC-DMIS." Above is what the PC-DMIS programmer sent me as to how it reports profile. Is there a way to do this in Calypso? Pulling all the points into a plane does not work. Probably because of the geometry of the impeller. I don't the option of Curve or Free-Form on this offline seat. We do have CMM's out on the shop floor that have Curve/Free-Form, but these are used for production I would have limited use throughout the day. I know that Curve adds the min/max of the form it has scanned. We use Curve to verify our "cam-shaped" I.D.'s on specific parts we machine here. What can I do mimic the PC-DMIS reporting of profile how he has explained it. Edited April 16 Link to comment Share on other sites More sharing options...
[Da...] Posted April 16 Share Posted April 16 Can you recall into a general surface? Link to comment Share on other sites More sharing options...
[Li...] Posted April 16 Author Share Posted April 16 I don't see how to recall points into general surface when you open it up. Link to comment Share on other sites More sharing options...
[na...] Posted April 17 Share Posted April 17 (edited) Please sign in to view this quote. Is this PC-DMIS specific? If I have one point that is .0005 as my max deviation on the plus side and -.005 as my max deviation on the minus side, then it should report the profile as .01 or is there something wrong with ASME maybe? 🙂 Edited April 17 Link to comment Share on other sites More sharing options...
[Da...] Posted April 17 Share Posted April 17 Please sign in to view this quote. You are mis-counting the zeros; it's .0005 and .005. What he is describing is a non-symmetrical or bi-lateral profile tolerance. How do you report this on Calypso? Haven't found that yet.... On simple things like lines, you could dimension it as a bilateral tolerance of +.005/-.0005, or just as a range of [.0055/-.0005]. Link to comment Share on other sites More sharing options...
[Ke...] Posted April 17 Share Posted April 17 (edited) Please sign in to view this quote. Unless I am missing something... I think that you need to report them all as Space Point Distance, and then report Max/Min Result... and then use the Result Element to add them together - abs(getActual("Maximum").actual)+abs(getActual"Minimum").actual) It's a lot of programming, and reports are cleaner if you mask off the individual Space Point Distances. If you are going to be doing a lot of work like this, then getting the Curve/Freeform software is 100% worth the investment Edited April 17 correction Link to comment Share on other sites More sharing options...
[To...] Posted April 17 Share Posted April 17 Please sign in to view this quote. Yes, I believe PC-DMIS reports the delta between max and min. What is the result if max = +.005 and the min is +.0045? That would make the answer .0005, which in most people's mind think "Wow, that is really good." One former PC-DMIS programmer, now avid Calypso program challenged me to find the calculation in ASME. I was surprised that I couldn't find it. Has anyone seen it? Link to comment Share on other sites More sharing options...
[Li...] Posted April 17 Author Share Posted April 17 Please sign in to view this quote. At this point I am just messing with one part of this program trying to get results that would mimic his report. At this one part of the program it is taking 19 points along the edge of one of the blades of this impeller. I have 19 space point distances along this edge. I kind of get what you are saying, but how would get all the points together? They way he set it up in PC-DMIS is that it finds the min of all the 19 points and the max of all the 19 points and then adds the two extremes. I have Curve on a CMM out on the floor. I was able to work this out by creating a curve feature, recalling the 19 points and putting that curve feature into Curve form characteristic. Seemed to work good. Took some trial and error, but got it. Link to comment Share on other sites More sharing options...
[na...] Posted April 17 Share Posted April 17 (edited) A pdf shared by Mark, bellow it's a good discussion too. -> Also check pdf shared by Tom -> Edited April 17 Link to comment Share on other sites More sharing options...
[Aa...] Posted April 17 Share Posted April 17 Please sign in to view this quote. I recall it being defined in detail here: Y14.5.1 - Mathematical Definition of Dimensioning and Tolerancing Principles It's also covered in the relatively new reporting standard: Y14.45 - Measurement Data Reporting These standards define single value actual measurement results for both equal bilateral and unequal bilateral profiles. I'm not sure I like the single value concept for unequal profiles. I think I would prefer a two-result min/max definition for unequal bilaterals. Link to comment Share on other sites More sharing options...
[na...] Posted April 18 Share Posted April 18 Deviations for GPS seems to be specified in ISO 17450 (could be also in some other ISO number). Link to comment Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in