Welcome to the Onshape forum! Ask questions and join in the discussions about everything Onshape.

First time visiting? Here are some places to start:
  1. Looking for a certain topic? Check out the categories filter or use Search (upper right).
  2. Need support? Ask a question to our Community Support category.
  3. Please submit support tickets for bugs but you can request improvements in the Product Feedback category.
  4. Be respectful, on topic and if you see a problem, Flag it.

If you would like to contact our Community Manager personally, feel free to send a private message or an email.

Mate Connector Directions NOT Intuitive

The axial (Z) behavior of mates, except maybe Fastened, doesn't seem to have any logic, which can't be true because its software.  This has been frustrating the crap out of me for the past 3 years I've been using OnShape (the first 2 I thought it was just my ignorance).  Its a total crap shoot what the directional behavior of any mate with motion is going to be.  Its pretty irritating to have to create air cylinders with negative strokes (most of the time, but not always).  A scan of posts shows that its not just me.  You can't look at two mate connectors on your screen and reliably know how they'll orient themselves.  Venting... having to correct the orientation of virtually every mate gets tedious.

Comments

  • mahirmahir Member, Developers Posts: 1,307 ✭✭✭✭✭
    Yeah, I feel your pain. You're right. There is logic, but it's hard to see intuitively. Most people don't think in vectors. I submitted an enhancement request years ago to at least be able to flip dimension direction easily (a la SW), but it never got any traction.
  • marten_hutchisonmarten_hutchison Member Posts: 29 EDU
    The crew I work with also uses SW.  One of the consistent bits of pushback I get from them using OnShape is "mates are easier in SW".  I don't use SW so I can't compare.
  • billy2billy2 Member, OS Professional, Mentor, Developers, User Group Leader Posts: 2,068 PRO
    edited February 2022
    I find it confusing also.



    I can't give you any wisdom.

    I will add though, once you get into assemblies and the fact that they can inherit mates correctly from sub-assemblies, I stopped complaining. Try pulling sub-assemblies into higher level assemblies and watch how everything works. I've never seen this before and it really helps creating a library of working sub-assemblies. Hopefully soon, we'll be able to download robots, and stages that actually articulate and we don't have to create sub-assemblies that work. When I download a pneumatic cylinder from mcmaster carr, why do I have to make it work? What's wrong with this world?

    A word of warning, limit mates are confusing also. Just remember that the lower limit mate always has to be a lower value than the max limit mate. When limit mating a lot, you'll find some go from negative infinity to zero and others go from zero to infinity. I can't tell you why and I've tried to figure it out.


  • shawn_crockershawn_crocker Member, OS Professional Posts: 865 PRO
    billy2 said:
    When I download a pneumatic cylinder from mcmaster carr, why do I have to make it work? What's wrong with this world?
    Yes.  I'm amazed there is not some standard that can be implemented into .step and the like.  Something that any software can look at and translate into its own mates "language"
  • billy2billy2 Member, OS Professional, Mentor, Developers, User Group Leader Posts: 2,068 PRO
    edited February 2022
    @marten_hutchison It's iterates me when I hear people stuck in SW over small issues involving interface.

    OS assemblies solve reverse kinematics:

    You can mate an effector and OS will tell you the robot's angles. If you're designing automated equipment, this is a huge benefit. The last time I used SW, it wouldn't do this.



    And then there's assembly inheritance where you can take a switch assembly that toggles, add it to a panel, add it to a chassis, add it to a rack, add it to room, add it to a building and then the switch still toggles in the building assembly. The explode state (whatever its called)  in SW doesn't work and forces you to build crazy assembly structures that are meaningless.

    I don't understand why not understanding the mate orientation would stop someone from switching when the OS assemblies are superior.

    I'm thinking online sales doesn't work, this is a simple objection that should be handled easily.


  • billy2billy2 Member, OS Professional, Mentor, Developers, User Group Leader Posts: 2,068 PRO
    edited February 2022
    @shawn_crocker I agree, why can't step be upgraded? We don't even have properties that get transferred.

    I'm hoping with cloud cad, vendors get wise and build onshape models to distribute. A vendor could push current pricing and availability to your design.  You could easily look at a BOM and see if the next build has any issues. The vendor could derive a usage count and supply demand. 


  • marten_hutchisonmarten_hutchison Member Posts: 29 EDU
    Sounds like we maybe getting off track... I'm not disputing OS vs SW assemblies.  I'm not discussing .STEP imports.

    I'd simply like more predictable orientation behavior when I mate parts in assemblies.  I've realized that I unconsciously cringe when I have to hash through a large assembly because I have to contend with guessing the mate orientation for virtually every mate and then correct them.  For small assemblies its not a big deal, but larger ones where I have to place 20+ components, it gets old.  I can understand its a difficult challenge for developers to create software to guess the local coordinates of a part.  I'd be happy if just the Z's of the mated parts always did the same thing.
  • shawn_crockershawn_crocker Member, OS Professional Posts: 865 PRO
    Sounds like we maybe getting off track... I'm not disputing OS vs SW assemblies.  I'm not discussing .STEP imports.

    I'd simply like more predictable orientation behavior when I mate parts in assemblies.
    Over time, and with my ever increasing love for onshapes mating system, I have begun to not really notice the issue you are talking about.  I do think it is an issue though.  Anytime I have to configure, for example, a revolute mate, I struggle briefly with rotating the mate this way and that to try and capture the appropriate positioning so that I can have a nice clean, 30 deg in the configuration means 30 deg in the mate.  Not 90 minus the configuration value or some other off hand massaging of the data.  And sometimes, weather I am feeling particularly dull that day or what, I just have to resort to an unclean setup to get what I am trying to accomplish in a timely manner.
  • marten_hutchisonmarten_hutchison Member Posts: 29 EDU
    @shawn_crocker your Revolute example is what I'm referring to.  I can make ANY mate work, but as you say, I prefer "clean" logical mates, not "unclean" do whatever, to make it work.  Nor do I want to waste the extra time.

    I'll refine my statement... OS's mating concept overall is intuitive, but the initial behavior of the two part's local coordinate axes, when the mates are selected is not.
  • eric_pestyeric_pesty Member Posts: 1,885 PRO


    I'll refine my statement... OS's mating concept overall is intuitive, but the initial behavior of the two part's local coordinate axes, when the mates are selected is not.
    Yeah, I can see what you mean better now...

    Adding the mate connector editing launcher (whatever it's called) in the mate dialogue would go a long way as it would make it a lot quicker to fix the MC themselves rather than adding offsets/rotation in the mate!

    Something like this:



    I already raised it as an IR but from the "contact support" menu so there is no forum post to vote for... feel free to make one as well!
  • nick_papageorge073nick_papageorge073 Member, csevp Posts: 823 PRO
    I agree with the OP. I've been using OS professionally for just over one year now. I still move parts in the wrong direction quite often. I also did do all the training on my own at the very beginning.
  • billy2billy2 Member, OS Professional, Mentor, Developers, User Group Leader Posts: 2,068 PRO
    Sorry, I didn't mean to derail the thread. I too see past the confusion and just goto to town with assembling. Honestly, I don't know if it's a bug or I just don't understand it.

    @eric_pesty I've submitted this enhancement 3 times and each time support tells me I've already submitted it. I've been waiting for this for a while now and asking multiple times doesn't seem to help.

  • eric_pestyeric_pesty Member Posts: 1,885 PRO
    @billy2
    Right, I don't think an individual submitting the same IR multiple times help, but they do keep track when multiple people ask for it so that's what I meant!
  • mahirmahir Member, Developers Posts: 1,307 ✭✭✭✭✭
    Is there an enhancement associated with this issue that has more steam than the others? That way we can lump our votes onto the lead horse. I have my enhancement below, but it only has 4 votes. I guess I could have worded it better to get more attention.

    Mate option to flip offset/limit direction
  • S1monS1mon Member Posts: 2,982 PRO
    @mahir
    I would also argue that any of the Limit options should have the current value in the dialog. I know it shows up on screen, but why not in the dialog too? (this is what Solidworks does)

  • NeilCookeNeilCooke Moderator, Onshape Employees Posts: 5,681
    S1mon said:
    @mahir
    I would also argue that any of the Limit options should have the current value in the dialog. I know it shows up on screen, but why not in the dialog too? (this is what Solidworks does)

    Saying "this is what SolidWorks does" is like red rag to a bull  :# - IR please!
    Senior Director, Technical Services, EMEAI
  • S1monS1mon Member Posts: 2,982 PRO
    @NeilCooke
     Saying "this is what SolidWorks does" is like red rag to a bull   - IR please!

    Haha.

    How about, as a user, it seems weird to have the limits but not the value itself all in the same place. When I'm first creating one of these types of mates, I often want to play with all three values.


  • marten_hutchisonmarten_hutchison Member Posts: 29 EDU
    If I wanted software that worked like SW, then I'd use SW, not OS.  The main reason I switched from Fusion to OS was the learning curve was faster and Assemblies were more intuitive.  I like that OS continually iterates and they roll out seamlessly.  I'd rather see a good tool I like get better, hence my comments on Mates behavior.

    I'd be happy to read a short description of the programmer's design intent with respect to the mate's coordinate axis behavior when they're selected.  Maybe that would clear up my question.

    tks MH
  • M_HutchisonM_Hutchison Member Posts: 27 PRO
    Here's an example...  You might think that based on the coordinate axis shown, that if you want Offset the part to the right, you would enter a negative Y value.  Wrong, its positive.  No idea why but its typical.

  • eric_pestyeric_pesty Member Posts: 1,885 PRO
    @M_Hutchison
    I can't tell without seeing the document but I believe you are trying to offset the moveable part which is on the right relative to the fixed part which is on the left. If you switch the order of the mate connectors it should give you the expected result.
    Remember the offsets are trying to move the first selected MC in the axes direction of the first MC.
  • nick_papageorge073nick_papageorge073 Member, csevp Posts: 823 PRO
    @M_Hutchison
    I can't tell without seeing the document but I believe you are trying to offset the moveable part which is on the right relative to the fixed part which is on the left. If you switch the order of the mate connectors it should give you the expected result.
    Remember the offsets are trying to move the first selected MC in the axes direction of the first MC.

    That could be the usability improvement. Whichever coordinate system the movement is based on could be made much bigger or highlighted, etc, so its easy to figure out what will happen. If you switch the selection order, the bigger/highlight goes to the other coord system.
  • shawn_crockershawn_crocker Member, OS Professional Posts: 865 PRO

    @M_Hutchison
    I can't tell without seeing the document but I believe you are trying to offset the moveable part which is on the right relative to the fixed part which is on the left. If you switch the order of the mate connectors it should give you the expected result.
    Remember the offsets are trying to move the first selected MC in the axes direction of the first MC.
    Well I just learned something new!
  • eric_pestyeric_pesty Member Posts: 1,885 PRO
    shawn_crocker said:

    Well I just learned something new!

    I got a bit frustrated by the apparent "randomness" one day so I created a basic test assembly and played with mates to try to figure out how it worked but there are probably still some more subtle points I am missing! That's how I figured how to apply offsets to fasten mates in a predictable manner, but it gets more complicated as well when you throw in rotations, or have more degrees of freedom, etc

    I started another thread about the need for a "tips/webinar/tech brief" type of material as this really isn't explained properly anywhere in the training, but that hasn't gotten any traction! 
    If the devs are reading this (@NeilCooke is pretty active around here!) and looking for topic ideas for tech tips or webinars, I would be really interested!
  • Lucas_KuhnsLucas_Kuhns Member, csevp Posts: 99 PRO
    I have been using OS Pro for more than 3 years now and I am still confused by which way the offset goes! It's just a running joke in the office that regardless of whether you use a + or - number, it will go the opposite direction of what you want. A tech tip or slight tweak to the UI could fix this.
Sign In or Register to comment.