Workflows improvements - 3D printing/ product development etc

[Mill3r]

Hi all,

I’d like to open a discussion about design workflows — whether that’s for project work, 3D printing, or general product development. As more people get involved in self-design and rapid prototyping, I think it’s an interesting area to share ideas and approaches.

I’ve been using SketchUp Pro for many years across construction, engineering, and 3D printing, so I’d say I’m fairly experienced — though not quite a wizard yet.

One thing I often notice is how different SketchUp feels compared to parametric modelling tools like Fusion 360. In Fusion, the ability to work with timelines and modify earlier design steps is incredibly useful. In SketchUp, once you’re too far along — especially if the model includes details like chamfers or complex edges — going back can be a real challenge. Sometimes, it’s easier to start again from scratch.

To manage this, I usually:

• Model as much as possible before adding fillets, chamfers, or other finishing details.

• Copy and paste key versions of the model at various milestones to create a kind of manual timeline within the file.

I’m curious — how do others approach this?
Do you have your own system for managing design evolution in SketchUp, or do you think Trimble could introduce features that better support iterative or parametric workflows in future updates?

I’m not looking for wizard-level shortcuts that rebuild a model in two seconds — just practical approaches that work for everyday users. I’d be really interested to hear how others organise their design and product-development process in SketchUp.

Thanks

[DaveR]

Generally I always model with the idea that there will be changes required down the road. If I’m doing CDs or shop drawings I know that the project isn’t final for a long time, if ever, so I model to make edits as simple as possible. That’s part of what drives me to use only components and no groups. (And that’ll get the knickers of some folks in a twist. ) I use a consistent workflow. I make all components report as solid even for models that aren’t destined for 3D printing. Solids are cleaner and easier to modify when needed.

When I model for 3D printing and I know the design is not final, I tend to leave things like filets as separate solid components. Here’s another example of that. The rim of the toothed pulley is a separate solid from the hub so the hub can be modified without touching the toothed rim. They don’t have to be combined before export to .stl.

In general with a clean, consistent, methodical workflow I don’t find the absence of parametric modeling features to be a problem.

[Mill3r]

Nice, similar then try to keep things separate as much as possible.

what do you mean by: I tend to leave things like filets as separate solid components.

[Mill3r]

What this 3D printed? what did you use it for?

[DaveR]

@Mill3r

what do you mean by:

I was thinking of something like a pipe tee with a filet where the central leg of the tee joins. If modeling for 3D printing I’d model the filet a sort ring that sits in place at the juncture.

Another way you could leverage this idea of breaking things into separate solids would be for something like the hex portion of the cap on this grease cup. It could be left as a separate component so its dimensions can be adjusted without changing the shape of the dome.

I modeled the toothed pulley for the local FIRST Robotics team. They printed it to use on a robot but I don’t have any photos of it.

[ateliernab]

@Mill3r

To manage this, I usually:

• Model as much as possible before adding fillets, chamfers, or other finishing details.
• Copy and paste key versions of the model at various milestones to create a kind of manual timeline within the file.

I’m curious — how do others approach this?

same. in my files there is a physical (spatial) timeline, with branches and variations. the finetuning comes quite late, and I’ll keep a rough version of the model next to it, so that if I need to change something, I can work with the pre-detail version.

@DaveR

The rim of the toothed pulley is a separate solid from the hub so the hub can be modified without touching the toothed rim. They don’t have to be combined before export to .stl.

that too. I too try to work with only solid groups / components, therefore most of what I model is technically 3dprint-ready. the slicer doesn’t care that in SU it’s separate entities, it just wants a clean SLT.

edit : this is 6 solid groups / components.

select all, export as one stl.
and this is the single stl sliced.

yet I can still work on the separate parts in SU.

[3DxJFD]

@ateliernab

in my files there is a physical (spatial) timeline, with branches and variations.

I like the sound of this. Will you show an example?

[ateliernab]

this is for a pen clip. first iteration at the origin, every work session / post prototype rework I move one slot to the right. up and down are just variations (or safekeeping) within a session.

same here. front left is all the parts as I worked on them. on the right is a test fit. in the back, the separate parts to be exported and printed.

I work my 3d models like I work my physical ones on my desk.
on a long project, I’ll duplicate my file every now and then, cleanup the copy and continue. that way I can always rollback to a previous iteration

[3DxJFD]

Thanks. That makes sense. I thought you might have meant something like ‘select all, move over X amount, new iteration at the model origin’. Likewise, shift everything forward or back. But the desk analogy makes it clear.

Then you just have a ‘feel’ for a ‘complete’ iteration that ought to be saved. Do you back those up to Trimble Connect? Or do you have a naming convention (e.g., V1, V1.1) or maybe dates are enough?

I’m getting the sense that the OP’s ‘manual timeline’ is like this (but maybe more often in separate files). Interesting topic anyway… I’m going to keep following along

[ateliernab]

@3DxJFD

Then you just have a ‘feel’ for a ‘complete’ iteration that ought to be saved. Do you back those up to Trimble Connect? Or do you have a naming convention (e.g., V1, V1.1) or maybe dates are enough?

I’m 100% local, mainly because I use the same account for pro, personal projects and teaching, I model on my mac but I teach (mostly) on my PC + web. that way I don’t have my models all over the place
also, as a student, I did 3 years abroad, moving from country to country in couple of month sessions, so most of my workflows relies on not having a stable connexion. even today. (and I have a daily backup of my mac on a server, so I’m relaxed).

firstly, if it’s a small project, I’ll contain it in one SU file like in my previous screenshots. most of the 3d printing I’ve made myself fits in this category. no need for 10 files when making a table, just make one file, and if you need to cleanup a bit, make 10 tags in it, one per iteration, and hide them.

this is one way to deal with it, a folder for every major step, because it’s not just SU, it’s also photos, photoshop files, scans… on these dates I had a meeting with the client, took lots of notes, and moved on to the next version.

this is more frequent for me, not by dates but by versions. it’s nothing precise, sometimes it’s a feel, I’ll go 4 steps in a day, and sometimes I’ll stagnate on a step for a week.
plus, since all of these are duplicates over time, it’s very simple to make a layout file showing all the evolutions. just duplicate the page, update the content to the new version, BAM.

(each Mk bears the same number as the file. simple)

[3DxJFD]

First, I like the Layout versions idea. Very clear.

100% local as in ‘external drive’ for switching between machines? Or projects stay local on each machine (personal on Mac, teaching on PC) so models aren’t all over the place because they are on the right machine for the job? Either way, I too tend to have a “Projects” folder and everything (pics, survey, estimate, point cloud, info, etc.) goes in it on my machine. I back that up to an external drive. I like TC and have tried to switch too it, but I haven’t made a complete jump over.

So I see the versions. Makes sense. Would you use an extension that saves components as versions? Let’s say you have a “Widget 1” component in your ACME working model and it’s a great Widget but then you develop the really -really- good Widget 2 and you want to use it. You don’t want to lose Widget 1 (so you move over in your model space and continue with Widget 2) and/or you save Widget 1 (ACME model version 1 with Widget 1) in your project folder. But if you could select Widget 1 and save as ‘Widget version 1’ to your project folder, and then later save Widget 1 as ‘Widget version 2’, and so on, would you if you could later select your component and swap (like paste in place) between component versions? This would be similar to a ‘timeline’ (if displayed in order) or a component versioning system (name it the version you want).

[Mill3r]

thank you, that’s a useful insite. I wasnt too far off the mark myself on how i do it.

Do you have any suggestions on what could be done in future updates if any that is?

[ateliernab]

@3DxJFD

100% local as in ‘external drive’ for switching between machines? Or projects stay local on each machine (personal on Mac, teaching on PC) so models aren’t all over the place because they are on the right machine for the job?

so, the mac mini is my day to day machine. so it’s all on it. it has a backup so I’m fine.
on the pc, by default, you’ll find the files I require to teach, plus a couple of personal projects that I can use to show them more complex models.
the only time when I need to transfer some other files on the PC is when I’m travelling for some time. Then, I’ll either use an usb stick, or trimble connect, depending on what I need to transfer.

@3DxJFD

Would you use an extension that saves components as versions?

probably not. what you describe is quite simple to do with vanilla tools, right click on a component, save it locally, with a smart name, poof.
and I’m a visual guy, having components in folders is not the same as having them on screen. when it comes to managing my library (and I’m digressing here), I know some people have folders of neatly named components, ready to be imported. I prefer having a couple of SU files that contain several components, like a virtual warehouse, where I can go in, compare, explore… and ultimately, copy-paste the right one to my project.

@Mill3r

Do you have any suggestions on what could be done in future updates if any that is?

not much, really. since su21, the file size is almost halved, so it’s not an issue. on a mini m4 with metal, performances are also not an issue… I’m a partisan of “if it’s not broken, don’t fix it”.
The way I work is version-agnostic, it doesn’t really rely on any exstension, it also works on ipad, on web. in fact it also works fine in affinity, archicad or even word.

[3DxJFD]

The EW file you linked has a bit of a Neo in the training matrix when he’s gearing up, or trip to the big box store vibe.

@ateliernab

simple to do with vanilla tools

Yes, right click, Save Out. Easy. A little related gripe is the Components Panel isn’t too user friendly IMO. But that’s not part of the workflow you’re describing.

I see what you’re saying about being visual. I like that too. A problem I sometimes have is forgetting in which file something is in to open to see it. Or where I put it…

All in all I think we work in a couple of similar ways.

Design Scribbles

[ateliernab]

@3DxJFD

A little related gripe is the Components Panel isn’t too user friendly IMO. But that’s not part of the workflow you’re describing.

indeed. I rarely use it, it’s not even on my screen by default.