Recall Startup Alignment

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Recall Startup Alignment

    When creating fully constrained subsequent alignments in a program is there any benefit to recalling the Startup alignment and building from that? I've always just built on top of the previous alignment in the past. Again, this is fully constrained with all 6 degrees of freedom accounted for (level-rotate-origin). I'm inclined to say it doesn't matter but a little voice is telling me to cut out the middle man.
    "It ain't what you don't know that gets you into trouble. It's what you know for sure that just ain't so." (Samuel Clemens)

  • #2
    EDIT: LEMME REPHRASE, there is no benefit.
    Last edited by BIGWIG7; 06-23-2015, 04:12 PM.
    Che Guevara is a communist scumbag.

    Comment


    • #3
      My rule of thumb....

      You 'recall' the alignment you measured the features in when making a new alignment. Don't jump back to a previous alignment or the startup, if you measured the features in ALIGNABC to make ALIGNDEF, stay with the default recall, which would be ALIGNABC. There is SOME reason to the madness that they put into Pcdmis at times.
      sigpic
      Originally posted by AndersI
      I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

      Comment


      • #4
        It can matter if you are using a rotation table. It can matter if you use stacked rotary tables.

        It can matter if you want to use a readpoint move in relation to machine coordinates.

        As far as alignments go. I would say as long as you never forget to update the coordinate system every time, then it would be rather irrelevant what alignment you recall.

        -----
        Yeah! You tell that little middle man to get off his high horse and #### off. Nobody likes a know it all. Almost like nobody likes vpt.se and his 5k comments.

        Comment


        • #5
          Watch the tri-thingy in your graphics.
          When in Startup and you open an alignment it is floating in all directions.
          Once there is an alignment it no longer floats.
          I have not tested to see if it really makes a difference but anytime I want to make sure I am truly leveling I recall startup.
          Its just a click away.
          I can explain it to you, but I can't understand it for you.

          Comment


          • #6
            There are programs where I need to flip the part over where I use it. If the backside of the part is not symmetrical or has different features than the top side, I will recall the STARTUP alignment and start from scratch. If the part will not move during the program, I never use it.

            Comment


            • #7
              All good points. I wasn't quite sure and don't recall reading anything about it. Just thought I'd throw it out there to get the consensus. It's usually the simple things that turn my day inside out.
              "It ain't what you don't know that gets you into trouble. It's what you know for sure that just ain't so." (Samuel Clemens)

              Comment


              • #8
                I guess you could benefit from it if you want to double check your alignment. Say you have something screwy going on and you not sure if there is a problem in one of your alignments, you could recall the startup and do a brand new onw to compare. Normally if everything is good there is no reason to recall it, just create a new one.

                Sometimes once you have restrained the align it hard to tell if its rotated property, bu recalling the startup you cna have a pice of mind to verify that its rotated or leveled correctly.
                Last edited by WolfMan; 06-23-2015, 06:29 PM.
                B&S One
                PC-DMIS CAD v2014

                Romer Infinity

                Comment


                • #9
                  One way I've compared "identical" alignments is to pull up the first alignment and go to Alignment Utilities ->Bestfit -> Transforms. Copy the "Machine to Part" values and paste into Excel. Then do the same with the "identical" alignment and look at the deltas between the two (there shouldn't be any). This (partially) is what prompted my initial question as it seems everything might be transformed back to machine coordinates when going from one alignment to another. Some of our parts are in aircraft coordinates where the origin might be 1200" or so from the physical part itself. Working in metric ends up using a little more of the available precision and I was curious about round-off error if building one alignment on top of another (on top of another...). Just wondering if anyone had ever run into any issues either way.
                  "It ain't what you don't know that gets you into trouble. It's what you know for sure that just ain't so." (Samuel Clemens)

                  Comment


                  • #10
                    Wolfman - I thought it might be easier if you can see what I'm talking about. I put together a spreadsheet that transforms an initial readpoint with a given alignment. The startup alignment shows the point as measured in machine coords (DIM LOC6). It's also output in a final part alignment (ALIGN_ABC) as DIM LOC5. The alignment comes from Alignment Utilities -> BestFit -> Transform. The workbook uses built-in Excel functions for the math. I'm more used to working with row vectors in code but wanted to provide something that could be used out of the box. Let me know if you have any questions or there's a problem with the link. You'll see some round-off error at the last digit due to the single precision. Coding would let you work in double precision.

                    The Alignment:
                    Align_ABC.png

                    The same point in both machine & part alignments:
                    Pnt1.png

                    The workbook:
                    Excel.png

                    Link to Workbook:
                    https://drive.google.com/file/d/0B-b...ew?usp=sharing
                    "It ain't what you don't know that gets you into trouble. It's what you know for sure that just ain't so." (Samuel Clemens)

                    Comment


                    • #11
                      GomoFazter, if you want to be accurate in the excel calculation, you have to export vectors with more decimal places (ASSIGN/V1=FORMAT("%3.8f,%3.8f,%3.8f",STARTUP.XAXIS) in ALIGNABC (for example) gives the first line of the matrix, with 8 decimal places.

                      Comment


                      • #12
                        You must recall Startup alignment with some assignments which use it :
                        QUALTOOLDATA("xyz", "sphere_1") gives the center of the sphere in startup alignment, not in the used alignment.
                        PROBEDATA("vector") gives the orientation of the probe in startup alignment.
                        If you want to use those assignments in auto features or move points, then you have to recall startup alignment, use them , and recall used alignment...

                        Comment


                        • #13
                          One thing you must do, if you haven't done this already is to make sure all your vectors are clean, 0,0,1. 1,0,0 etc. i am not quite sure what you trying to do there, the way I check my dim is by copying the feature and the dim in question from one align into another and look at the theo values. If its done right theo values should match accordingly to new align.

                          Personally i have not noticed any round off issue with having multiple alignments. Perhaps you right, but without seeing the whole code hard to tell what you're doing.
                          Last edited by WolfMan; 06-24-2015, 08:12 AM.
                          B&S One
                          PC-DMIS CAD v2014

                          Romer Infinity

                          Comment


                          • #14
                            Can you post the code of what you did plz. I am a little confused

                            What are you transforming? I don't understand what is the purpose of align- BF- transform function
                            Last edited by WolfMan; 06-24-2015, 08:19 AM.
                            B&S One
                            PC-DMIS CAD v2014

                            Romer Infinity

                            Comment


                            • #15
                              Jefman - Thanks for the examples. I'm not as familiar with expressions as I am with the object model and didn't know about using FORMAT as you show. I'll have to take another look at expressions and see what's exposed to use.

                              Wolfman - There is no code other than standard functions in the Excel worksheet. Download it from the link and place the cursor in one of the lower cells having a number. The math function used will show up in the Formula Bar. There are other ways to go about it but this came to mind first. It's just a method to move a feature from one alignment to another without any code. It shows the same point in different alignments using the transformation matrix given in the Alignment Utilities. You could use the sequence shown in the worksheet and code it using your preferred method.
                              "It ain't what you don't know that gets you into trouble. It's what you know for sure that just ain't so." (Samuel Clemens)

                              Comment

                              Related Topics

                              Collapse

                              Working...
                              X