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.


Assemblies, sub assemblies and parts

kevin_quigleykevin_quigley Member Posts: 306 ✭✭✭
Firstly, I am not where I can log into Onshape to check this, but this week I have had to deal with some issues in SolidWorks that I would be interested to know how OnShape could improve on.

Scenario. Medium sized assembly of glazed steel structure similar to but not this company ... http://www.clovis-canopies.co.uk/patent-glazed-glass-and-polycarbonate-canopies .... These are designed around a system of standard profiles, glazing, fittings etc. each is adapted and made bespoke to site. Currently these are done in SolidWorks using in context assembly model driven by 2 sketches, multiple patterns and equations. Edit the sketches, edit the patterns, model updates, linked drawings update etc etc...pretty standard stuff.

So..the problem here is that 20% of the time we need to do extra customisation that requires edits that go beyond what the standard model variation allows. We use assembly cuts etc for this, but to be frank this is a workaround to having to make several in context assembly level new part files, which is tedious. Remember these structures are static. They do not move. This is a very very common use case.

So in OnShape, you need to be able to...

1. Model "assembly" using standard parts, patterns, mate parts in context etc inside a single part mode.
2. You need to be able to cut and chop up those standard parts
3. You need to have onshape then recognise the parts so you can save out a parts list and create drawings
4. You need to be able to select parts and say 'make these a group/sub assembly' whilst still maintaining the top down context

In general SolidWorks does these jobs well, but it is the 80 20 thing. 80% of the time the standard stuff flies through. In our latest analysis of the last 50 projects, 40 were standardish. The 10 that were not actually amounted to 45% of the time. That was simply down to dealing with having to figure out ways to cut up parts, then recreate said parts for drawing production...and trust me, when you have customers screaming at you for output that is the stage where errors creep in.

I would really like to see a demo of a static assembly using multiple common parts and others defined in context.


  • Options
    Mark_SandersMark_Sanders Member Posts: 17 ✭✭
    Hi Kevin, I think I just asked a similar question...in a slightly different way.
    Agreed importing/modifying is an area, I for one would like some help with.

    Good ... Init ;-) .... Well done guys !
  • Options
    billy2billy2 Member, OS Professional, Mentor, Developers, User Group Leader Posts: 2,040 PRO
    Kevin do you really think there is a model that can be built to handle all the changes that come along?

    I'm currently working on a medical device that has sizes: small, medium & large. The model won't adapt no matter what I do.

    The good news is that with a little help, I can tweak it and make the changes happen.

    My concerns are how to pass this to the next guy so he can tweak the model and get good results.

    How hard would it be for you to tell me how to make your model work. Can anyone change your model and get it to produce good results or does it take a model master like yourself to make it work?

    Back in the day when we were showing off parametrics in a demo the models were so fickle that any unwarranted change would have sent the whole demo into the garbage. The perception in the market place was that we could capture any/all design intent and build extremely robust models. The reality is we shook in our boots every time we made the big change and practiced day & night to make it happen. Obliviously we got lucky.

    I think parametric modeling has been over sold and I'm not sure there is a solution to your problem. It's non-deterministic.

    Instead of making parametric modeling more robust, I'm more focused on building models so others understand them. Maybe the solution to your problem is to hire more people that can control your model.

  • Options
    kevin_quigleykevin_quigley Member Posts: 306 ✭✭✭
    @bill not sure I follow your argument?

    The scenario I am describing is extremely common and probably 70% of the products we design come into this 'static assembly ' category. It is not about editing. It is about designing and tweaking. The fact is, for standard stuff parametric modelling works extremely well. Applications like DriveWorks very effectively semi automate repetitive modelling tasks. I'm talking about repetitive jobs here not one off tasks. The difficulty we face is when the standard systems have to adapted beyond the scope of the "automation".

    Usually this requires adapting the standard system, chopping and changing standard parts etc. we have tried pure direct modelling tools like Spaceclaim and they are good for specials but not so hot for the 80% standard jobs. These products are not complex. The parts are simple, but the way they go together can be tricky.

    We have also tried Synch Tech but to be frank, that was useless. The interface is dire and even the sales guy got confused trying to remodel one of these, then apply some edits.

    This particular business churns out 10+ proposal packages from 1 designer every day. Average is 30 mins a package...so that is update master model, update drawing package, save out DwGs and PDFs and issue. Automation of the standard systems is critical. Reworking the standards into specials needs to be slick.

    Bottom line is I don't care if the model is parametric or direct. As long as it can be quickly created and updated.
  • Options
    billy2billy2 Member, OS Professional, Mentor, Developers, User Group Leader Posts: 2,040 PRO
    Sounds like a great sales tool.
  • Options
    billy2billy2 Member, OS Professional, Mentor, Developers, User Group Leader Posts: 2,040 PRO

    I guess in a broad since you are looking for better parametric capabilities to help with your sales tool.

    My point, in addition to better parametrics, give us better tools to understand the existing/better parametrics. Have you ever worked on someone else's model? Look at the one you are building, how easy would it be for anyone to understand all that you've put into to it. I realize you're interested in getting it to work and being more powerful. I just want parametrics easier to understand and follow.

    SW is so frustrating for the 2nd guy who has to work on something the 1st guy built. Complicated or not, there just needs to be better tools to understand how a parametric model was built. Not sure but it sounds like your tricked out model would be very difficult for someone else to follow especially if you dumped it on someone with no instructions on how to use it.

    As a consultant, the majority of my work comes from other sources other than my own and it's made me bitter over the years. Why can't people build better models that are easily understood and make predictable changes?

    In general, instead of more parametric twist & turns, I'd rather have something easier for the 2nd guy to understand what the 1st guy did.

    Maybe I'm missing your point totally,

Sign In or Register to comment.