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.

Having the legend "question" above the box where you want a title is .... not good.

2

Answers

  • glen_dewsburyglen_dewsbury Member Posts: 1,326 PRO

    Using the built in Wrap feature simplifies things like cam design. The graft length being Pi * Dia of course. This example would need clean up to remove sharp corners and the like. The graff makes timing and slope much simpler. There are other ways to accomplish this of course.

    https://cad.onshape.com/documents/ec9994607320087db3682d12/w/187cdb6ffb9d97c20a33dacf/e/99e818eef3ebeb13e7411e50

  • buk_browserukbuk_browseruk Member Posts: 28

    @eric_pestry "

    So when someone who claims to have as much experience as you do is
    complaining about how bad the software/UI/training is, citing they were
    unable to perform a certain simple task after two weeks of trying
    (including getting assistance in these forums), I'm afraid it tells more
    about yourself than the software…

    "

    Uh huh. The "two weeks" is due to the 3/4 days between posting a question, and it becoming visible. Remeber what I said about "documenting the new user experience here"?

    "

    so I really don't see why you weren't able to just do this if you have experience in Solidwork!

    "
    I never claimed any experience of Solidworks; because I do not have any. Please re-read what I wrote, and stop using strawmen to defend the indefensible,

  • eric_pestyeric_pesty Member, pcbaevp Posts: 2,753 PRO

    k, I misread the Solidworks part , I didn't realize you were quoting someone else. Side note: forums have this handy method for showing when you are quoting someone that can help avoid this sort of confusion, I would have thought a programmer/UI designer harping on clarity and usability would be actually use these, but I digress… In your own words, you claim to be competent in FreeCAD and Fusion (both parametric CAD software solutions that do have a lot of similarities when it comes to creating planes and general approach to modelling things), and have 40+ years experience in the world of 3D software and UI.

    I am also quite familiar with and quite competent with (despite my strong distaste for its horrible UI) FreeCAD. I'm also familiar with (have installed), Fusion, Blender, OpenSCAD, MoI and Plasticity. I trained (a long time ago) as a draughtsman; and subsequently made a 40+ year career as a programmer specialising in 3D graphics engines and UI design. I'm not a neophyte anywhere but here.

    I noticed that you seem to equate "having software installed", to being "familiar" with it, which might explain certain things… Speaking about "strawman", I didn't come up with the "two weeks" bit!

    …if it is this hard (over 2 weeks and counting) to work out/find out how to do something as trivial…

    You are claiming to have too much experience to need to go through the training while also claiming the software/interface is too complicated to figure out! In all this trying you don't seem to have picked up much, including some of the basic functions of the interface (selection boxes in feature dialogues), which is a bit embarrassing for someone with 40+ years of experience in UI design!

    Onshape is growing rapidly and has tons of new users asking questions here (and elsewhere). Most seem to be able to figure it out and learn to do things like you describe just fine after putting in a bit of effort, just saying…

    No one is forcing you to use Onshape but if you are, maybe you should consider taking it down a notch and learning a bit more about parametric CAD. It's not that it had to be complicated (it's not) but it does require a different way of thinking from the mesh/free-form software you mention being familiar with.
    Getting some help here is going to require a bit of an attitude adjustment but if you come back with some questions in good faith (and show that you actually attempted to learn something), we are always happy to do what we can to help!

    Does the 45s gif (that you asked for) seem like a complicated process to you (2 sketches and 3 features)? It's also parametric and fully adjustable on the fly:

    adjust.gif
  • ry_gbry_gb Member, csevp, pcbaevp Posts: 210 PRO
    edited May 20

    "FreeCAD, Fusion, Blender, OpenSCAD, MoI and Plasticity"

    Only two of those are are parametric and you haven't used Fusion (per your words "installed"). And FreeCAD isn't widely adopted.

    You just don't have any experience with parametric CAD and it's showing more and more. Nitpicking on @eric_pesty 's comment because he said Solidworks when you mentioned Fusion clearly shows your lack of knowledge. Any person claiming expertise would know they're just two sides of the same coin.

    You can't BOTH claim to be an expert while also hiding behind being a newcomer to Onshape.

    On the other hand, I keep coming back to this thread because I'm interested to see what new creative reply you come up with to justify your whining. I'm always up for some good drama.

    Ramon Yip | glassboard.com

  • jim_zamecnikjim_zamecnik Member Posts: 83 ✭✭

    Buk said:

    The "two weeks" is due to the 3/4 days between posting a question, and it becoming visible.

    A gross exaggeration I think. There is a moderation delay for new members for your first few posts, I'd guess that's to filter out the spammers, scammers, and miscellaneous jerks that are everywhere on the web. In my experience the delay was half a day at most, and only for my first few posts. Once the moderators trust you to behave yourself your posts will become instantaneous.

    No need to reply to this, I won't be back to read it. Arguing with the people trying to help seems counterproductive.

  • MDesignMDesign Member Posts: 1,402 PRO
    edited May 20

    @buk_browseruk Can you show me a gif of your system doing the same thing in 4 features? This took about 1 min per history log after a practice round as you can see by my suppressed features.

    animate6.gif

    The text based help system is right where most systems put it…. in the menu system.

    animate.gif

    Disclaimer: I haven't read any of the thread to know what's going on. I just saw a challenge and accepted it. LOL

  • buk_browserukbuk_browseruk Member Posts: 28

    It's hard to tell given the speed of the animation — mine tool "so long" because I deliberately paused at each step so it was clear — but I suspect that your groove is only 5.6568542494923801952067548968388 mm wide* where the Right plane splits it.

    ie. Not the required the constant width that my method produces.

    Note 1: that's sin(45°) * your 8mm starting width.
    Note 2: width is minimum width as measured orthogonal to and between the groove side faces. Ie where the cam follower needs to be a good fit to prevent slop.

  • buk_browserukbuk_browseruk Member Posts: 28

    Two further points:

    1) Fusion doesn't have "mate connectors". It does have Joints and As-built joints which are probably what mate connectors were before they got abused extended; hence my not thinking to look there for a solution.

    2) Your Split solution is wrong. It does not produce a constant width slot. Constant height but not constant width. Hence a follower that fits snugly at the top and bottom of the groove, cannot passes around the sides.

    "Side note: forums have this handy method for showing when you are quoting someone…"
    Yes. but unlike good forum software that if you use the quote button after selecting a subset of a post, it quote just the selected passage, here it quotes the entire post regardless of preselection. Hence, to have a purposeful interchange, I felt this was a better if tedious process.

    Despite your ad hominim attacks and spurious strawmen, I've attempted to stick to arguing the subject, not the man.

  • buk_browserukbuk_browseruk Member Posts: 28

    @ry_gb "You can't BOTH claim to be an expert while also hiding behind being a newcomer to Onshape. "

    Sorry, but politely, you are wrong. First, I didn't "claim expertise", I stated that I was not a neophyte. Big difference.
    But equally, I'm a perfectly adequate typist, but it doesn't mean I could right Ukrainian if you put a cyrillic keyboard on my system.



  • martin_kopplowmartin_kopplow Member Posts: 1,417 PRO

    Since the recent update, with Mate Connectors now being stackable, this has become even easier. Create one for the angle, cylinder reference and offset, and one on top of the first one to sketch upon.

  • MDesignMDesign Member Posts: 1,402 PRO
    Answer ✓

    You are correct. I might not have been cc'd on the precise dimensional requirements of your "one last attempt". But be that as it may, my question remains but revised to 5 features.

    image.png
  • eric_pestyeric_pesty Member, pcbaevp Posts: 2,753 PRO

    About "sticking to the subject", it's hard to tell from the thread title and added "requirements" along the way.

    As far as the "attacks" go, I'm just matching your energy… This was your response to my (accurate and addressing the issue from the first post, followed up with a detailed video containing multiple examples) first post in this thread:

    What would I be "mating" or "connecting" this plane to?

    I only want this plane so I can create a sketch on it.

    I can't belive how stupidly difficult is is to do this.
    Nearly to hours of sessions with chatgpt and google's AI whilst waiting for my disappeared posts here to reappear, and nothing.

    Dozens of attempts to simple tilt a plane to 45°.

    In my current CAD, click "Move", Click the plane, click the direction handle. Type 45°. Done

    Who exactly is "on the attack" here? Also at the time you posted this, @jim_zamecnik had already provided a detailed (and valid) answer to your question so why are you still complaining about not being able to do it after dozens of attempts and hours or "research"?! The mate connector suggestion was just a slightly more efficient way to solve the problem as it bypasses the creation of the plane entirely.

    Also it's spelled ad hominem, if you are going to use big words, at least spell them correctly!

    @ry_gb summed it all up perfectly.

    2) Your Split solution is wrong. It does not produce a constant width slot. Constant height but not constant width. Hence a follower that fits snugly at the top and bottom of the groove, cannot passes around the sides.

    You are not helping your case making incorrect statements like this with such confidence!

    groove width.gif

    In case you are actually interested in learning something, setting the "profile control" option ensures the sweep profile remains aligned with both the sweep path and outer surface of the cylinder to produce a constant width groove.

    image.png

    Here's yet another different way to create this geometry (in addition to the multiple answers already provided by multiple users), I didn't even use mate connectors or splits so maybe that will be up to your standards?

    NoMCs.gif

    I named the public document with these examples after the title of the thread so it should be easy to find:

    https://cad.onshape.com/documents/baeecc255800d33b8c76a9db/w/20a80fef6cbb84fef0115739/e/fbe702cda4f9819797d60582?configuration=Angle%3D0.7853981633974483%2Bradian%3BDiameter%3D0.1%2Bmeter%3BGroove_Depth%3D0.005%2Bmeter&renderMode=0&tangentEdgeStyle=1&uiState=6a0f3fc9f0296bdc42fb5679

  • buk_browserukbuk_browseruk Member Posts: 28

    I believe I have recreated your method to produce a 10mm wide "geodesic" groove:

    image.png

    But when I measure that gap it is not 10mm?

    image.png
  • buk_browserukbuk_browseruk Member Posts: 28

    @eric_pesty :"Also it's spelled ad hominem, if you are going to use big words, at least spell them correctly!"

    Si “ad hominem” male scripsi, certe tamen ex meo ingenio protuli; tu vero primum quaerendum, deinde exscribendum habuisti.

    Color me disappointed. At least you quoted me correctly — unlike A.N.Other — in "What would I be "mating" or "connecting" this plane to? …", but nothing in that — or any post — was an "attack". Neither of you nor OS. Just confusion at the term "mate connector" for the functionality described. Little wonder:

    z

    CAD                  Terminology
    ANSYS SpaceClaim  	Origin / Coordinate System   
    Onshape           	Mate Connector               
    SolidWorks        	Coordinate System            
    Autodesk Inventor 	UCS / Work Features          
    Fusion 360        	Joint Origin                 
    PTC Creo          	CSYS                         
    Siemens NX        	Datum CSYS                   
    CATIA             	Axis System                  
    Autodesk AutoCAD  	UCS                          
    FreeCAD           	LCS (Local Coordinate System)
    Solid Edge        	Coordinate System            
    Rhino 3D          	CPlane                       
    Blender           	Empty / Transform Orientation
    MoI			Construction planes / CPlanes / Orient axes
    Plasticity		Gizmo pivot / construction grid / transform pivot
    
    Onshape’s mate connectors are unusual because they intentionally merge:
    
    * local coordinate system,
    * snap/alignment point,
    * mating reference,
    * and lightweight datum geometry
    
    into a single object. Most traditional CAD systems keep those as separate entities.
    
    

    So, no attack. Just confusion at the term, and a little frustration at: a) the implication that it should have been obvious to me, and b) the delays caused by my new user status. And the why behind my earlier "in defense of the indefensible"; you cannot rightly defend that which is not being attacked.

    What with you jumping on a typo and A.N.Other crowing "He said: 'My CAD'", He he.", as if it was some thunderous faux par to abbreviate "The CAD system I use most, and am most familiar with and prefer…" to "My CAD…".

    You two should form a double act and go on the road. You'd go down a storm at frat house parties everwhere after about 11pm.

    Oh wait. Its already been done


    Which of them was your insiiration?

  • buk_browserukbuk_browseruk Member Posts: 28

    "I like how you keep moving the goalposts by adding additional requirements so that only you can be right. "

    The requirements didn't change; you simply waded in mouth-deep and inferred them incorrectly.

    "just go use Fusion already. It's already on your computer. "

    That's the official verdict of PTC for whom you've elected to speak is it?

    "post Title has NOTHING to do with the ACTUAL question"

    Deliberately so.

    Did you read that title? Of course not. Probably could not see it past those chips on your shoulders.

  • ry_gbry_gb Member, csevp, pcbaevp Posts: 210 PRO

    I'm just glad you're hiding under an internet username because this would be really embarrassing for you.

    Ramon Yip | glassboard.com

  • Derek_Van_Allen_BDDerek_Van_Allen_BD Member Posts: 970 PRO

    My preferred solution until I get around to getting a better body sweep feature cooked up is to leverage ruled surfaces and thicken operations to get the thickness of my slots to be consistent. The deficiency you see in the previous solutions is notable because body sweep algorithms are an open ended area of research for NURBS based modeling kernels. Even the Parasolid developers at Siemens themselves haven't really implemented a good generalized algorithm that handles convex geometry robustly and they'll tell you in their docs that it should only really be used for visualization of machining toolpaths instead of generating them. And then you peek under the hood of those wonderful Moduleworks CAM algorithms and find a million line / arc segments and mesh geometry jammed into the trenchcoat of modern machining softwares so in effect nobody is solving the body sweep problem analytically at all in the present sense.

    So here's my own batch of useful lies.

    image.png

    Loose steps:

    • Draw your toolpath curve
    • Use a ruled surface to generate the mid-surface of where your tool is gonna cut into your part and make the surface normal to the cutting axis
    • Thicken that surface to the diameter of the cutting tool
    • Boolean the tool geometry away from the part geometry

    To explain a little about why the other methods don't achieve the correct geometry, some of them fail to take into account the local orientation of a cylinder sweeping across a volume, or the cross sectional thickness of that cylinder sweeping across a volume, or both. The sweep methods will maintain one of these dimensions but not the other. Same with the wrap method.

    And a caveat about the ruled surface method as well: technically ruled surfaces never lie precisely on the input curves used to generate them as an optimization method for drawing an approximate surface, so this isn't geometrically perfect either from a purely literal topological perspective. But also literally nothing in industry from the front engine to the back is either, so you can ignore this fact and enjoy your slots within an acceptable E-6 tolerance value and move on to other fun aspects of machining.

  • MDesignMDesign Member Posts: 1,402 PRO
    edited May 22

    You sure are picky for just wanting to know how to create a 45 deg plane….LOL repeat question now 6 features. LOL for some reason I have to create use move boundary to get the ruled surface outside the cylinder… otherwise it'd still be 5 features. but now the depth is not maintained at 4mm.. Can you send me a STEP file from your CAD software?

    image.png
  • Derek_Van_Allen_BDDerek_Van_Allen_BD Member Posts: 970 PRO

    Essentially the same solution I've got going but instead of move boundary I used thicken new, then move face, then boolean. Potato potato.

    The reason we need to adjust that boundary or move that face is because ruled surfaces necessarily run an approximation of the input curves when they are generated. There are cases like this cylinder or @EvanReese's submission to my CAD battle where you'd expect that perfect analytic surfaces can be generated and wonder why they're being approximated, but it's because ruled surfaces are doing some best fitting magic to not implode your model with a billion control points for the non-trivial surface types where you involve twist or lock face orientations or change angle of inclination on some nodes but not others.

    At least, that's what I gather by reading between the lines of the way that it is and how it outputs geometry.

  • MDesignMDesign Member Posts: 1,402 PRO

    Ah didn't see your post before I hit send @Derek_Van_Allen_BD yes that was my next method before the last one. move face after the fact felt clunky and I wanted to eliminate the boolean feature.

  • eric_pestyeric_pesty Member, pcbaevp Posts: 2,753 PRO

    Right, so instead of wondering why mate connectors might have been suggested and spending 5min figuring out what they are, (which would have made it clear why they were suggested in the first place), you jump to a smart ass comment of "What would I be "mating" or "connecting" this plane to?"…

    Notice how @jim_zamecnik noted in his post that it wasn't something he was familiar with and how I responded by providing more information in the form of an excellent video going over all their uses? By the way, creating a sketch directly on a mate connector, which you claimed wasn't covered in the video, is narrated and demonstrated no less than four times, 1:44, 3:42, 6:54, 13:30 in that video!

    You seem to say you're British, which rules out a language barrier hampering your comprehension abilities so it's hard to conclude anything else than you are just being obtuse and contrary on purpose!

    You then proceed to explain condescendingly that my proposed solution is wrong because it doesn't produce a constant width groove and it's not going to work for your (never expressly stated) purpose, except that it did in fact produce the correct geometry but you couldn't be bother to check that…

    I'm guessing admitting when you are wrong about something isn't in your habits but are you going to acknowledge that?

    I don't know if it's actually your intent but your interactions here look a lot like those of a troll!

    Anyway, at least this has been entertaining!

  • Derek_Van_Allen_BDDerek_Van_Allen_BD Member Posts: 970 PRO
    Answer ✓

    I tend to lean this direction because there are usually more steps to navigate like the cylinder insertion, some filleting, and maybe some other sweep or delete face operations that can be necessary if you go all multi-axial with your slotting operations. Things you need to consider when it's not cylinder on cylinder action. I usually err on the side of solving the most generalized version of a problem than optimizing for minimum feature count.

    Though in this case that would mean sliding straight back down the slippery slope of whitepaper research and doing Siemens' work for them, and unless they're gonna provide me with a developers' license I'm not going quite that far out of my way yet. So I'm cool with drawing the line at cylinders.

  • buk_browserukbuk_browseruk Member Posts: 28

    Thanks. I got that to work.

    image.png

    It doesn't quite survive the export/import to my CAD via STEP (for faster subsequent modifications; working in OS is like stirring molasses (for me!) yet.

    image.png

    Whether the discrepancy results from I haven't yet worked out.

  • buk_browserukbuk_browseruk Member Posts: 28

    Don't sweat it son; your youthful indiscrensions don't embarass me at all.

  • buk_browserukbuk_browseruk Member Posts: 28

    Yaddah, yaddah, yaddah.

    The only trolls here are Beavis and Butthead. (I worked out which inspired you!)

  • buk_browserukbuk_browseruk Member Posts: 28

    @MDesign "Can you send me a STEP file from your CAD software?"

    I persisted here because my own attempts in my cad also suffer the same problems. It seems to be a "nurbs thing" as Derek_Van_Allen_BD explained above.

    BTW. My asking about the tilting the plane was my first problem, the rest as shown in my gif was to head off the question "Why do you want to tilt a plane." Simples; cos that they way I do it in my current CAD.

  • Derek_Van_Allen_BDDerek_Van_Allen_BD Member Posts: 970 PRO

    If you're using Designspark Mechanical to finish your edits you should export and import as a parasolid file to avoid conversion issues from .step files. Step is meant to be an interchange format that isn't always a perfect representation of the source geometry if other software packages don't have the same capability for calculating or rendering surfaces, but it seems both Onshape and DSM are based on the same kernel underlying so you should be able to pass a parasolid from one to the other.

    If you still see discrepancy between the distances in both programs that would point to the measure tool using different logic in either software.

  • MDesignMDesign Member Posts: 1,402 PRO
Sign In or Register to comment.

Howdy, Stranger!

It looks like you're new here. Sign in or register to get started.