Magically changing nominals

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

  • Magically changing nominals

    We are running PC-DMIS 3.7mr2. We have this chronic problem with nominals "automagically" changing without warning. We have the "find nominal" turned off, but it still does it. I have been tempted to put a comment in front of every dimension command with the correct nominals in place, just so the engineers (who never seem to believe a CMM report) can see that yes...we know what the correct nominal is.

    Seems odd that the program can even change them at all. I should think that once you put in a nominal, it should be embedded in stone.

    I have talked to B&S about it, but they haven't been particularly helpful, except for mentioning turning of the find nominal thing, which doesn't work.

  • #2
    Update to mr3 solves a few bugs from what I hear
    sigpic
    if you had soap on a rope it would be tied to yer ankle

    Comment


    • #3
      When you are dimensioning the feature(s) do you have to "tweak" the number to be a whole number? e.g. 9.9995 change it to 10.000? If you "tweak" the value at the dimension you need to answer "yes" to the question, "send new nominals to feature?". Otherwise everytime you run it it is going to change. Don't worry it should adjust for diferent alignments. Just be careful because I don't rememmber 3.7 mr2"s "quirks".
      sigpic

      James Mannes

      Comment


      • #4
        Make sure find nom durning exe is not checked

        Comment


        • #5
          Originally posted by JamesMannes View Post
          When you are dimensioning the feature(s) do you have to "tweak" the number to be a whole number? e.g. 9.9995 change it to 10.000? If you "tweak" the value at the dimension you need to answer "yes" to the question, "send new nominals to feature?". Otherwise everytime you run it it is going to change. Don't worry it should adjust for diferent alignments. Just be careful because I don't rememmber 3.7 mr2"s "quirks".


          I agree with James. Sometimes when dimensioning I will find that PCDMIS inserts its own nominals (based on what I don't know). Just make sure you do the above and click "yes" to the question of sending new nominals to feature. It's all part of the SMA conspiracy I say
          I used to be high on life but I built up a tolerance.

          Brown & Sharpe Global Advantage
          PCDMIS CAD++ v2011mr2
          PH10MQ/SP600M


          sigpic

          Comment


          • #6
            I think it is an MR2 thing. I have nominals set to the mean in my dimensions but not in my features (for monkeys reading the report). I can be fine for weeks then bam I find out we are running bad parts because the dimension nominal changed from the mean to the feature's programmed nominal. Say we are running an ID at 1.500 ±0.005. I might program it at 1.495 (this is only an example) but put 1.500 in the dimension nominal. Everything is fine for a while then one day the dimension says 1.495 ±0.005 and all night long the monkey ran his parts at 1.490 because the CMM dimension was not in the red. Go to MR3, as long as you are running MR2 I do not believe there is a fix.

            Craig
            <internet bumper sticker goes here>

            Comment


            • #7
              I don't have 3.7 MR2, but I run into this plenty times and what I found best to do is to re-type the nominals when dimensioning - and when the question "Carry nominals back to the feature?" comes up I answer NO - most of the times it takes care of the problem.

              Joe
              PC-DMIS CAD++ 3.7 from 4.2 MR1

              Comment


              • #8
                I really wish I could help all y'all, but I have never had the nominals change or jump on me. 99.99% of the time I use regular alignments (no iterative) with CAD=PART (which I also do when doing Iterative alignments). Also, 99% of the time, my alignment is NOT part of the check program, but is it's own program, saved to external file, recalled by the check program. I do not know if this is why I don't have any problems or not, all I can do is tell you what I do and let you decide from there.

                I have been doing it this way as long as I can remember. V3.2063, V3.5, V3.7, various MR's, DOS version, etc. Never had nominals jump or change on me.
                sigpic
                Originally posted by AndersI
                I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

                Comment


                • #9
                  No cad here and it has been prevalent on 3.7 MR2. This is the only time I have had problems with (other versions included) it and it is sporadic.
                  <internet bumper sticker goes here>

                  Comment


                  • #10
                    Oh yeah, I also have cad data 95% of the time.
                    sigpic
                    Originally posted by AndersI
                    I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

                    Comment


                    • #11
                      I've complained of this a few times on this forum. I've answered "yes" and
                      "no" to updated features and haven't found a difference in either. The problem has dimenished some since switching to MR3. Right now we have all of our data going to a SPC software that has the nominals locked in and we mostly pay attention to the Plant Monitor that this SPC software is equiped with. We have had a couple of incidences where the CMM report will show nothing out of spec and the plant monitor flag an out of tolerance condition and sure enough a program that has been running for several months in Operator Mode will have changed nominals.
                      sigpic

                      Comment


                      • #12
                        Interesting. I am in the process of downloading mr3, so we will see if that helps. I am not sure about the "carry nominals back to the feature" thing. I found that rather annoying when I was programming, so I changed it so it wasn't always pestering me about it. I think I have the default set to "no", but I will have to check.

                        As for alignments. I try to use a readpoint to start most of the programs I write, unless a part is really weirdly shaped. However, I also run a lot of older programs written by my predecessor which use a manually taken initial alignment, followed by a more comprehensive DCC alignment. Gradually I am converting them over to my "lazier"method, but this necessitates taking new setup photos, etc.

                        Interesting about doing the alignment as a separate program. I thought PC-DMIS could do some sort of program chaining, but I haven't tried it yet. The closest I have come is having a program stop in the middle, switch to manual mode, prompt to have the part flipped over so as to run the back side, then do a new manual alignment, then do a dcc and take off in auto once again.

                        Comment


                        • #13
                          Oh...and I don't do CAD on any of our programs. Most of our stuff is so old that there isn't any.

                          Comment


                          • #14
                            Originally posted by eestle View Post
                            Interesting. I am in the process of downloading mr3, so we will see if that helps. I am not sure about the "carry nominals back to the feature" thing. I found that rather annoying when I was programming, so I changed it so it wasn't always pestering me about it. I think I have the default set to "no", but I will have to check.

                            As for alignments. I try to use a readpoint to start most of the programs I write, unless a part is really weirdly shaped. However, I also run a lot of older programs written by my predecessor which use a manually taken initial alignment, followed by a more comprehensive DCC alignment. Gradually I am converting them over to my "lazier"method, but this necessitates taking new setup photos, etc.

                            Interesting about doing the alignment as a separate program. I thought PC-DMIS could do some sort of program chaining, but I haven't tried it yet. The closest I have come is having a program stop in the middle, switch to manual mode, prompt to have the part flipped over so as to run the back side, then do a new manual alignment, then do a dcc and take off in auto once again.
                            I don't "chain" the programs together. I run the alignment program, close it, open the check program. However, if you have multiple programs for a part, this method works GREAT. For example, I will have a TRIM DEVELOPMENT program with trim points as close as a couple mm together so we can develope the blank for the part, but I have no desire or need to check that many for the "full check" program. I will also have an SPC program that checks the spc points (DUH!) and possibly others for very specific things. The point is, though, I only have to run the alignment program once and I can then take my choice of the other programs for the part. And yes, before ANYONE says it, you CAN do this same kind of thing using MARKED SETS, but doing it that way can make for a HUGE program, even if you are only running a small marked set, it WILL run slower (PC speed) due to the amount of 'code' that Pcdmis has 'open' for the program.
                            sigpic
                            Originally posted by AndersI
                            I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

                            Comment


                            • #15
                              I had the same problem and this is what WilcoxAssoc. sent me so try this if you need to it help me somewhat with 3.7MR2. Lots of less crashing and nominal changing. Good luck



                              Here are a few things you can try to prevent PC-DMIS from crashing.
                              You must be the administrator for these changes to take effect and make sure PC-DMIS is not running
                              As far as PC-DMIS crashing all the time, there is a folder in the PC-DMIS directory that contains files that can become corrupted. It's either called "administrator" or will have the pc's logon name, depending on how you log on when the pc first boots up. Check the contents of the folder for a file called gbarstate.dat., if your in the right folder, back out 1 level and rename it by right clicking on it, and selecting rename. PC-DMIS will recreate this folder the next time it's started, and will look as if PC-DMIS has just been installed. All of your toolbars and print settings will go back to their default state, along with any text color changes you've made. You might see a probe toolbox window on the screen, just close it.
                              None of your part programs will be affected.
                              You can also increase the speed PC-DMIS operates (although it may not be noticeable) by adjusting Windows XP performance settings. To do this, press the start button, then go to settings, then go to control panel, then click on the "system" icon. Next, click on the advanced tab. Under performance, click on the settings button, then select the "adjust for best performance radio button". This shuts off a lot of the fancy graphics options that Windows use's that can slow things down.

                              If you go to the www.wilcoxassoc.com website and enter the site, there is a link to a Microsoft hot fix to improve memory allocation errors. Click on the link, then, under current press releases, go down to the 5th item and click on details and follow the instructions for downloading the file.(To download, click on the white arrow with the red background). You must be logged on as the administrator to install this.

                              Other things to consider are verifying you have the latest graphics card driver installed. You can do this by going up to the card manufacturer website. Another issue can be a fragmented hard drive, you should defrag on at a regular interval
                              Also, verify that no other unneeded processes are running in the background while PC-DMIS is running.
                              Finally, Remind all operators that PC-DMIS (V3.7) takes approximately 30 -45 seconds to start, depending on system speed. Clicking on the shortcut while PC-DMIS is starting will put you "offline" and will corrupt files.
                              sigpic

                              Comment

                              Related Topics

                              Collapse

                              Working...
                              X