Old Dogs and New Tricks... or I just can't figure out this new-fangeled software!

I’m an engineer. As such I am no stranger to CAD software. Also, whenever I interview one of my strong points is that I can learn just about any piece of software quickly… but I’m finding that to be less and less true as time goes on.

I started CAD using Bentley Microstation in school. As soon as I graduated I transitioned to AutoCAD and never looked back. I am strictly 2D, but when I got my 3D printer a couple years ago I started learning 3D design. I did ok at it. I am finding myself less successful when it comes to designing 3D objects for 2D printing however… so I thought I would try out some other pieces of software.

I decided to focus on OnShape. It seems like a relatively easy piece of software, and based on the demos and such it looks like everything is pretty much automatic. However, I’m having a hard time reconciling the “sketch” vs “shape”. I get that I need to draw a rectangle sketch and then extrude it to a shape. But I can’t figure out the interaction between sketches and existing shapes. Example, I have a shell (roughly a 3"x3"x4" box with 1/8" thick walls). I want to split that shell into two pieces, a lid and base to make a box. I don’t want a straight-line however, I want an angled edge to my box. Here’s the part in Onshape:

And here’s the part with the cutting plane that I would like drawn in…

(I drew this in BlueBeam).

I can’t figure out how to do this in Onshape for the life of me. Heck, I can’t even figure out how to snap the sketch for the cutting plane to the existing object. It is very frustrating. I suppose perseverance and devouring any and all videos that I can find are probably the key. I am relatively certain that, as limited as my 3D AutoCAD skills are, I could probably whip this out pretty quickly.

Have I finally become the “old guy” struggling with new software like this? So stuck in my ways that I can’t fathom another way of doing it? How much AutoCAD do I need to unlearn in order to be successful with Onshape? I had similar issues with a project that required Microstation a few years ago. I’m so ingrained in AutoCAD now that it took forever for me to get back into MS and figure out how to do things that are SO easy in ACAD.

What really gets me is that this 3d stuff is simply for the purposes of modelling and visualizing because the final product is a 2d sheet. So really I’m trying to learn all of this when it really isn’t strictly necessary.

Still, I’d like to be able to utilize this great tool, particularly being able to design in the Cloud so I can do it at home or work, where with ACAD I can design in both places, but I have to copy the file into Google Drive, etc… So are there any decent tutorials out there for Onshape that I haven’t found yet? Do I just need to plug along trying things out as I go?

Here’s the link to the Onshape file.


Can you publiclly share your Onshape document and post the link here? I, or another Onshape user here could then use that doc to create the parts you want so that you can see one or more ways to do it. You could also try posting on the Onshape user forum and someone there will probably do an animation showing how it is done.

If it makes you feel any better CAM software has the same effect on me.


Added the link to the original post.

If you’re used to AutoCAD I highly recommend Fushion360. It’s got cloud storage for all your models and it’ll be more intuitive to work with.


Thanks for the advice. I am not as fond of having to download software vs. Onshape which is 100% in browser. But my work doesn’t care if I install software, and at the same time, I was able to knock out my design in like 15 minutes without looking at a tutorial. I must be missing something with Onshape.

See this public document for a couple of ways to do what I think that you wanted.


This is just a copy of your public document and the 2nd two Part Studios each have a different way of make a box that is split into a bottom and a cover.

The first one creates a 2.75-in cube with a sketch and a subsequent extrude. This is just a solid block. A sketch is then created on the side cube to add a line that intersects the right top corner on the side face and a point 0.75" down from the top on the left hand side. This new line is then extruded into a surface that extends through the depth of the box. The new surface is then used to separate the block into a lid and bottom with Onshape’s Split command.

My second approach starts with a sketch that outlines the side of the box and includes the line that defines the boundary between lid and bottom as seen in profile. The lid and bottom are then created in separate extrudes, with one extrude using the triangular sketch face that defines the lid profile and the other extrude using the trapezoidal sketch face that defines the bottom part of the box. Note that the separate face portions of each sketch are used for the two extrudes. If you use the entire sketch you will just end up with a solid box.

You could also create lid and bottom as separate sketches and extrudes.

Hope that helps and doesn’t make things more confusing for you.

If you are still stuck, try asking for more help here:


Once again, I am foiled by 3D programs. I really got used to Sketchup and could do it there, but the parametric nature of OnShape and the work flow still intimidates/baffles me. I guess I’ll have to do some more tutorials. I do like the browser nature of this.

Thanks, but the document won’t load. It says that it doesn’t exist or I am not authorized to access it.

I had the same problems that you did. I spent years in SolidWorks in school, then got a job using AutoCAD for 2D things, and when I got my 3D printer I tried Onshape and for the life of me couldn’t get it to do the things that I wanted to. Switched to Fushion and it’s been smooth sailing since.

1 Like

Oops - forgot to make it public, but it should be fixed now.


CAD seems to be one those subjects where you just need to find one that thinks the same way you do. Once you do that things start to go a lot more smoothly.


You might not be thinking about the problem in the way a parametric software thinks you should. The way I would create that part is in two pieces - an upper and a lower. You can start from the same box, select one side and sketch your cut onto that side, then remove either the entire top or entire bottom of the box to create your two parts. This means you have one body per part file and is generally more accepted by the software I am experienced in than having two bodies in the same part file.

I can’t believe how much I’m going to sound like my old boss (whom I don’t really aspire to be anything like) right now…

The software should be designed to work the way that the user thinks. This becomes a problem when one user thinks differently than another, and ultimately this is why we have options for any given type of software. Someone didn’t think that the existing software worked properly, so they designed new software that was more in line with what they thought was the better method. We can learn any software given enough time, but if there is an option out there that is better for that individual, then why not use it? In this case, Fusion is far more familiar and intuitive to me than Onshape, so I’m going to move forward on that platform.


I agree and would have probably approached it the way that you suggested, but was trying to work out a solution as an extension of the OPs initial approach.

One of the great things about this sort of software is that there are so many different solutions to a problem,.

As someone deeply involved with software design I can’t agree.

One problem is that people think differently. Were you good in Algebra or Geometry? Most people have a definite leaning. How did you do with Trigonometry vs Calculus? How about Calc of the Unreal Variable?

The other problem we bump into regularly is that people (in aggregate) tend to get boxed in how they think, approach problems and create solutions - and they coalesce into some level of commonality (which conflicts with those who can’t conform because they’re Algebra people in a world of geometricians).

On top of all that is the frantic pace of change. People think differently and have fundamentally different ways of looking at things now vs 20 yrs ago - that computer in your pocket (& 3 billion other people’s pockets) radically changed the landscape of the possible. Most people lag the implications of the changes happening around them. That’s why really good design and elegant new approaches look “obvious” but only in retrospect. We all have those “why didn’t I think of that” moments.

Sometimes it takes someone who thinks radically differently to achieve progress.

My suggestion is that users are responsible for selecting software that comfortably fits the way they think. Neither the software (or designer) nor the user is responsible for conforming to the other. That’s why there is variety in solutions in the market. The market will determine if the solution fits enough people to be worthwhile.

< end rant :slightly_smiling_face: >