hexagon logo

Use Results for mating components.

Hello,
At work they gave me the task of finding the way of proove that the measured part will mate another part even before trying this.
We have a whole team working with PC DMIS and the give us the results as reports, but to present to the customer we need to show graphically that the part will mate the mating part.
I use Solidworks and I was thinking on the possibility of importing the results, but maybe there is even the possibility of doing this directly in PC DMIS(?)
So the tasks are:
-to overlay the results form the CMM on top of the STP file or Solidworks model.
-Add the mating part and proove that there are no interferences.

Do you have any ideas?
Thank you very much

Franco Attolini
  • Unless you have a point cloud (full part scan) you won't be able to say yes or no about any interferance.

    Also, I assume that each of these parts has tolerance. Tolerance is there for a reason, meaning that if the part is within those tolerances, then they SHOULD build correctly, each part has tolerance. Comparing a point cloud of PART-A to the cad of PART-B will probably NEVER show that there is no interference since part A will be + here and - there. In a typical assembly design, where parts are to mate, there is no 'design gap'. So a part with even +/-0.005" tolerance on a mating surface will pretty much show that points from PART-A will interfer with the design of PART-B.

    Sure, you can over-lay the data, even in Pcdmis, but in Pcdmis, it will not do you any good since the 'data' for PART-B is CAD data and the data for PART-A will also be CAD data, Pcdmis is not a cad software, you can't do distance between cad entities. You would have to do this in a cad package. But, I can already tell you, even if it isn't point cloud data, you WILL have 'interference' between the data for PART-A and PART-B
  • I agree with Matt. Unless there are very specific mating requirements, ie interference fits, specific minimum gaps, CAD is usually drawn assuming all parts are exactly nominal and mate as such.

    You would likely need full scan data of both the part and the mate, which would still only be helpful for that 1 part and mate. It couldn't guarantee that it will mate with any other part, or that any other part will also mate exactly the same.

    It sounds like they have no confidence in their tolerancing abilities.
  • Yep, pretty much what Matt said.

    Sounds to me like their design engineer has little to no confidence in his ability to design a component.
  • Thank you for your answers.
    Well sorry I avoid saying that this is for parts that are outside tolerance, we manufacture parts that takes 200hr of machining and several tons of steel, so we is out of tolerance it cost alot to re-machine or ultimatetly to scrap it. Tolerances are well set but so are tight to avoid the worst combination and sometimes is just one dimension out of tolerance and could not be so critical. But as I said the customer wants to see an analisys and graphical is the easiest way of convincing the customer, especially for the thermal expansion conditions.
    I was thinking to lay the points from the CMM into solid works to then make a surface and mate the mating part in nominal max condition and see the clearances.

    Thanks

    Franco
  • In all seriousness, a laser or white light scanner is *far* better suited to this type of project (although the Solidworks idea is a good idea)

    I'm guessing these parts cost $$$. If this happens somewhat frequently maybe consider pitching it as an idea for them?
  • Option 1, is very best solution:

    In all seriousness, a laser or white light scanner is *far* better suited to this type of project

    This will take far less time and deliver far more data than option 2...

    Option 2, requires many many hours for CMM:
    Make scans across all mating surfaces in two directions.
    If you have an analog scanning probe it will be not too bad, but if you only have a touch-trigger probe you are looking at millions of beeps.

    For BOTH options, the measured data should be exported in IGES format for someone talented with a CAD software to perform the actual fit comparison work.

    PC-DMIS is a metrology software, and metrology is the science of measurement to traceable standards.

    3D fit between mating objects is a more complex task than simple dimensions.
  • Thank you for you answers.

    Well the idea is to use the data taht already exist to make the "report" that comes our of PC-DMIS, and just try to adapt it to make it more visual. I will try to check the "scaners" options. Bu as I said, we use to just assess the information without a visual guidance but the customer likes that.
  • Thank you for you answers.

    Well the idea is to use the data taht already exist to make the "report" that comes our of PC-DMIS, and just try to adapt it to make it more visual. I will try to check the "scaners" options. Bu as I said, we use to just assess the information without a visual guidance but the customer likes that.