AnsweredAssumed Answered

Part Axes are Slightly Rotated

Question asked by Johny Abarca on Feb 20, 2017
Latest reply on Feb 20, 2017 by Johny Abarca

Like • Show 0 Likes0
Comment • 13

Hi all,

I am making an assembly and changed some underlying pieces to the whole thing and have been chasing down the errors ever since. I am almost finished expunging the errors, but one thing that is bugging me is that one of my blocks has its axes rotated 0.44 degrees and its ruining all my mates. The part that it is referencing is still square to Front, Right, and Top planes of the assembly but now this block has a 0.44 degree tilt in it and I can't figure out what would have caused it. So now if I sketch a rectangle on that block the rectangle itself is not square to the axes of the assembly.

I suppose one fix would be to just manually rotate the block back 0.44 degrees but I am just curious what would cause this in the first place. Is there a way to check to make sure the axes on the part are square and parallel to the axes on the assembly?

Thanks!

Johny

Attachments

Bottom Die Block.sldprt.zip542.3 KB

No one else has this question

Outcomes

13 Replies

John Stoltzfus Feb 20, 2017 11:17 AM
Mark Correct
Correct Answer
It is impossible to diagnose your issue without pictures or models, however to diagnose any issues is to open the assembly file and suppress every item in the feature tree and then start at the top first inserted part and unsuppress, do this one by one till the issue pops up (like a metal detector), then start digging from there..
Like • Show 1 Like1
Actions
- Johny Abarca @ John Stoltzfus on Feb 20, 2017 11:44 AM
  Mark Correct
  Correct Answer
  
  My apologies, I added the offending part to my original post. The issue here is that for whatever reason the global coordinates are rotated 0.44 degrees to the rest of the assembly. If I draw a rectangle with the edges "vertical" and "horizontal" it makes those edges at that angle to the rest of my assembly. This is easy enough to work around with parallel and perpendicular relations to the rest of the assembly components, but the real problem lies when I go to dimension everything. Because of the angle in the coordinates in the part I am not measuring the "true" distances but rather the distances in some funky rotated space. So a 1.00 in dimension becomes something like 1.001 and change in my assembly and this throws all the mates way out of whack.
  
  This is the offending part in the assembly. With it suppressed there are no issues because everything is referenced to my strip, but I can't mate my holder properly because everything on it is slightly skewed.
  
  This is my first real "top down" assembly and I will agree the order in which things are in the feature tree is not ideal (I moved parts around to better 'organize' them, though, now they are not in the order in which they are referenced i.e. the holders should be last and the forms first with the strip at the very top) but such is life.
  Attachments
  - Top Planes.png314.0 KB
  Like • Show 0 Likes0
  Actions
  - John Stoltzfus @ Johny Abarca on Feb 20, 2017 12:13 PM
    Mark Correct
    Correct Answer
    I did open the above file and my first concern would be that the first feature plane is listed as a face off of another component. Secondly - You mentioned that you moved parts in the feature tree, that could be your answer, you might have moved a pc that was mated with a distance mate from point to point which would allow it to move if any other relation has been moved...
    
    See attached for a Top Down or Inside Out design
    Attachments
    The Skeleton Sketch Part Introduction.pdf655.4 KB
    Like • Show 1 Like1
    Actions
    - Johny Abarca @ John Stoltzfus on Feb 20, 2017 12:39 PM
      Mark Correct
      Correct Answer
      I appreciate the feed back and the pdf, it has a lot of useful advice in there that I unfortunately didn't follow this time around.
      
      As for the feature plane, the reason it is like that is because it is being extruded down from the bottom of my strip. I created this part in my lower die assembly relative to the strip (like everything else is/should be). I figured this would keep the axes locked in the same as my assembly coordinate system, but maybe as you said it probably got swept around by some point to point mate or something. I was not aware that mates would do that to the underlying coordinate system.
      
      Either way it was my mistake moving things around in the assembly feature tree and even creating some parts in context then removing them and reintroducing them and mating them back into the assembly. In theory I shouldn't ever have to mate the parts that I add in context correct? They should be created as fixed as I define the underlying geometry relative to something already in the assembly?
      Like • Show 0 Likes0
      Actions
      - John Stoltzfus @ Johny Abarca on Feb 20, 2017 1:00 PM
        Mark Correct
        Correct Answer
        Johny Abarca wrote:
        
        Either way it was my mistake moving things around in the assembly feature tree and even creating some parts in context then removing them and reintroducing them and mating them back into the assembly. In theory I shouldn't ever have to mate the parts that I add in context correct? They should be created as fixed as I define the underlying geometry relative to something already in the assembly?
        
        Yes in theory you don't mate parts that are built in context. Just a heads up - if and when you do design in context of another part, just be careful how you stack or tie in geometries or planes to the newly inserted part. For a trouble free assembly you can delete any part in the assembly and only have mate errors..
        Like • Show 1 Like1
        Actions
        Johny Abarca @ John Stoltzfus on Feb 20, 2017 1:12 PM
        Mark Correct
        Correct Answer
        Yeah I definitely do not have that. I made a change to my strip in order to separate the punches and get more thickness in the die block and everything exploded. I read somewhere that deleting sketch relations and moving the geometry around rather than redrawing the geometry itself was better and it definitely has cut down on the number of reference issues I have been having, but I don't know how to realign the axes on this one particular component to coincide with the skeleton part.
        
        EDIT - Now the part's coordinate axes are at an angle of 8.36 to the assembly's coordinate system.
        Like • Show 0 Likes0
        Actions
        Rick McDonald @ Johny Abarca on Feb 20, 2017 1:51 PM
        Mark Correct
        Correct Answer
        Are you planes parallel?
        You are comparing the top plane of the Helper Die part and the Bottom Die part but Assuming these parts are aligned to the planes, the planes of each part should be parallel.
        See if mating these planes changes the alignment.
        If each part was made based on the alignment of the Top, Right and Front planes then these planes should each be parallel.
        I can't tell from your drawing but give that a try.
        When you import a part the part is floating until it is fully mated (if you have turned off "Fixed") I see one part is fixed and the others are not.
        Like • Show 0 Likes0
        Actions
        Johny Abarca @ Rick McDonald on Feb 20, 2017 2:03 PM
        Mark Correct
        Correct Answer
        I imported the strip and fixed it. It came in parallel to the assembly planes. I then added every component in context to that strip, but when changing the strip at one point it caused the planes and the axes on the die block to rotate. I could try to mate the planes and see if it unrotates the axes, but now the assembly planes are rotated with respect to the assembly axes so I do not know if that would even be helpful at this point seeing as how the whole assembly sits at about 8 degrees with respect to the x axis. I am not sure how this all happened...
        Like • Show 0 Likes0
        Actions
Johny Abarca Feb 20, 2017 1:06 PM
Mark Correct
Correct Answer
The more I mess with the block the more I am causing the underlying coordinates to twist, I didn't even know that is possible. What is causing them to do that? I was under the impression that when you mate or dimension something it does it with respect to the body not the coordinate system itself? What is the relationship between the two and which takes precedence?
Like • Show 0 Likes0
Actions
- John Stoltzfus @ Johny Abarca on Feb 20, 2017 1:36 PM
  Mark Correct
  Correct Answer
  With out looking at the entire file and pulling everything apart it is difficult to give you the answer you need - seriously based on everything you're saying I would look at every detail from start to finish and my primary focus would be distant mates or any in context from one line to another line, relating them with a dimension. Some where with in the first few components there is a part that has a relation to another part with a distance relation and it's flipping from plus to negative etc..
  Like • Show 1 Like1
  Actions
  - Johny Abarca @ John Stoltzfus on Feb 20, 2017 1:41 PM
    Mark Correct
    Correct Answer
    That wouldn't be faster at this point to just delete the twisted blocks and rebuild them from scratch? Or do I still run the risk of the same issues affecting a brand new block built in context?
    
    EDIT - Looking at it now it also looks like the assembly's planes are now rotated with respect to the x,y,z axes.
    Like • Show 1 Like1
    Actions
    - John Stoltzfus @ Johny Abarca on Feb 20, 2017 1:43 PM
      Mark Correct
      Correct Answer
      In the long run - It's always faster to trash a Troubling Design then to keep messing with it every time you open it. - but your there and I'm here, so it really is your call.
      Like • Show 1 Like1
      Actions
      - Johny Abarca @ John Stoltzfus on Feb 20, 2017 1:48 PM
        Mark Correct
        Correct Answer
        This is where I am at.
        
        I had my strip originally mated up to these planes to be square but now its twisted. Everything is suppressed and the original part isn't at an angle.
        Like • Show 1 Like1
        Actions

Part Axes are Slightly Rotated

Attachments

Outcomes

Attachments

Attachments

Related Content

Recommended Content