Repeat manual points in DCC

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

  • Repeat manual points in DCC

    I have an issue with my CMM where any manual points taken in Z will vary greatly. Even taking 3 points on a calibrated block i will find each point can differ from the last by up to .02". I have noticed that in DCC mode it does not do this so i have been adding a full dcc alignment to all of my part programs. I have an issue with a few parts that are setup on blocks though. I need a way to have the manual points taken on blocks placed randomly on the table to be repeated in DCC. Any way to have it basically use a variable point in DCC mode that calls out an existing manual point?

  • #2
    DCC alignments are prudent. You are doing the right thing. 0.02 is a lot though, almost sounds like probe comp. What size ruby?
    <internet bumper sticker goes here>

    Comment


    • #3
      For the program im currently working on im using 3 2BY20mm in a star probe. I have the same issue though with a 2BY30 and 5BY50. I have also recently switched to a TP200b from a TP200 hoping the problem would be solved and it has not.

      Comment


      • #4
        Are you saying the location of the blocks can vary so you can't get them in DCC mode? You can switch to manual mode in your program and use a comment to instruct the operator to take a hit manually. Align to it, go DCC and then take the hit in DCC mode. That 0.02 is a little concerning (metric or English?). You should be getting better than that I think.
        <internet bumper sticker goes here>

        Comment


        • #5
          Yeah because the setup can vary depending on where the operator places the blocks i cant just have the points set in DCC and have the progam repetable. I thought of doing it that way as well but it is extremely time consuming and because of the fact that some of our operators are less skilled than others it enters more operator error factor. If all else fails i will try that but if i have any alternatives i will definetly try them first. .02" is well beyond any tolerances we have in the shop and make measuring heights on many parts impossible. I would love to get to the root of this issue if possible, but it seems that its been an issue since day 1 with our SHeffield RS-670. By the way I am running 3.7 MR3 now. I have run evererything from 3.6 MR1 till this.

          Comment


          • #6
            Try throwin a message box (comment) up that says "Place the ruby over the block" or something to that effect. Take a read point then align to it and take a DCC point. That will at least keep you from having to ask the operator to take a manual hit.
            <internet bumper sticker goes here>

            Comment


            • #7
              jacyEd,

              You do need to have some sort of a repeatable set up.

              Once you establish a particular set up, take a picture.
              Next do a manual alignment, then a dcc alignment, then measure your part.

              That is why we write programs, for the "less skilled" operators, and that is why they are called "operators", they only run the programs that we write.

              Don't take any of this the wrong way. Also, stick to 3.7mr3, it is a fairly stable version of pcdmis.

              Blessings,
              ZydecoPete
              sigpic

              Comment


              • #8
                I definetly like 3.7 MR3 by far the best. MR2 crashed on me at the end of nearly every program i ran. And the verisons before seemed to have easily twice as many bugs requring all sorts of work-arounds.
                I have been adding dcc alignments to all of my programs as i run them (600+ programs) but i have issues like this that really mess things up for me. The rs-670 has a steel ribbed table as well, making it not possible for a static DCC alignment on the table surface for parts with datums on the face it rests on.

                Comment


                • #9
                  What Pete said is the best advise if you have 3.7 you have hypereporting. I like it a lot for set up instructions. If set up instructions or consitent set ups are out of the question then a readpoint would probably be best.
                  <internet bumper sticker goes here>

                  Comment


                  • #10
                    RELATIVE MEASURE is the way to go.

                    If I get what you are saying:
                    1) Place the blocks on the table
                    2) Measure them in manual mode, create a manual alignment
                    3) Now you want to measure those same locations in DCC mode, no matter WHERE they were.

                    If this is the case, set up the block the first time, take the manual hits, make the alignment. Then go into DCC. At this point in time, the DCC you are going to program will be at the XYZ of the manual hits you did in your manual alignment (they MUST be manual points, not planes or lines!). So, you go into Auto Features, and go to SURFACE POINT not VECTOR POINT. In Surface point, there is a box for RELATIVE measure. After you type in the XYZ IJK nominals for the point, click on RELATIVE measure. Then set the feature, all axis, to the manual point you took on the block. Then, when you set them up again, no matter where the blocks are, after you run the manual alignment, the DCC points will hit the block at the CURRENT location, not the original programmed location.
                    sigpic
                    Originally posted by AndersI
                    I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

                    Comment


                    • #11
                      That sounds like exactly what i need! Ill give it a shot in just a few mins and ill let you know how it goes. Thanks alot!

                      Comment


                      • #12
                        I have not been able to get this to work. Is this type of auto feature only possible while using a cad file or am i just doing something wrong? When i create the plane manually out of idivisual points then copy the XYZ/IJK in the auto point it seems to randomly recal either pre or post alignment coordinates for the feature. I can run it once and have it move to 50" over the part and seek, then i can run it again without changing anything and have it find the feature correctly. Theoreticals for the auto points are also changing at random to the pre alignment coordinates and back.

                        Comment


                        • #13
                          I have heard about issues with relative measure but I don't know enough about it to fill you in. If you can't get it you could code it so that your feature had the same nominal locations as your manually measured feature's actual locations. Say your manually measured feature is a point named PNT1then your DCC feature would have an X, Y, Z of PNT1.X, PNT1.Y, PNT1.Z. If you can't pull off relative measure then you aren't still dead in the water.
                          <internet bumper sticker goes here>

                          Comment


                          • #14
                            If you are using a global machine I have had the same thing happen.What I found the problem to be is even in manual mode the machine moves to fast when you take hits.I learned when you take manual hits I had to slow the machine down or the same thing would happen.Take a hit straight down like you normaly would then slow the machine down take another hit then look at the differance.When the machine was new it took some time to get used to.Using a tp20 with 3.7 mr3.

                            Comment

                            Related Topics

                            Collapse

                            Working...
                            X