How 3D is becoming less relevant for me

[asid61]

A little while back, I was thinking about how I actually use CAD nowadays. As I get better and better at CAD, I found myself relying on sketches more and more and on features and parts less and less; now I'm able to refine my layouts to the point where I can take a layout sketch that I've been working on for 6 hours and turn it into a full gearbox in 2. They're not even particularly complex layouts; it's how I apply them that has changed.
The key to it is not just how I use sketches nowadays, but also how I think about them. When I first started using them (back in late 2014 for ABS-101 and stuff) I didn't think much about their eventual result; I just went step by step in making holes in plates and went back to move them around if they didn't work in 3D- a few relations defined how far apart gears were to each other, and that's about it. But as I got better and better at visualizing how everything fits together in the 3D sense, I'm now thinking about how the gearbox will look in the end as I make the layout.
Thinking with depth in mind has opened up new worlds with layout sketches and makes designs possible that previously I wouldn't have thought of, like cases where gears appear to hit spacers but actually end up in different planes. As many know, using layout sketches often makes designing faster because even the worst computer can deal with a single sketch very fast, so I didn't need to wait on large 3D assemblies to update themselves. In fact, after some time, the 3D aspect of CAD became less and less relevant; it's only a minor tool that I use as a final check of depth and fittings. It's ironic that as I use more and more 3D CAD, I get more and more accustomed to using 2D instead of 3D. Nowadays about half my projects never make it past layout stage because either a) they're unworkable or b) I already know what the 3D version will look like. It's like going for full completion in a video game after finishing the main storyline; sometimes I'm up to it, other times I'm not.
Maybe it's just a side effect of not having a powerful enough computer, but I thought that shift from 3D to 2D was interesting. Has anybody else experienced this sort of change as they've gone through using CAD?

[echin]

I have seen a similar transition in my CAD, though I typically use a large number of connected sketches and work planes rather than a single sketch. Sometimes, this gets a bit extreme. For example, last season, 3467's drive base template had over 20 individual sketches on different planes.
I think for me, this CAD style is mostly driven by wanting all of my parts to update themselves when I make a change, which ultimately saves time and makes it much easier to make minor adjustments.

[AdamHeard]

Yup, it's nice to be able to work this way for speed early on.

It's super useful for FRC particularly.

The downside is you need to make sure you actually fully follow through to 3d so it can be communicated to others.

[Michael Hill]

Originally Posted by echin

I have seen a similar transition in my CAD, though I typically use a large number of connected sketches and work planes rather than a single sketch. Sometimes, this gets a bit extreme. For example, last season, 3467's drive base template had over 20 individual sketches on different planes.
I think for me, this CAD style is mostly driven by wanting all of my parts to update themselves when I make a change, which ultimately saves time and makes it much easier to make minor adjustments.

Have you tried assembly layout sketches? They're powerful*

* - if used carefully

[InFlight]

I would suggest the design process starts even earlier.
What are the requirements?

What motors ( rpm and torque curve) can I use
What ratio is required to match the motor and output.

Spacing needed for gear, sprocket or pulley system.

Once you have requirements and math done, you can then look at different implementations that would most efficiently solve the problem.

[echin]

Originally Posted by Michael Hill

Have you tried assembly layout sketches? They're powerful*

* - if used carefully

I just looked them up, and they appear to be a Solidworks equivalent to what I'm currently doing in Inventor. I'm just starting with Solidworks at the internship I have over the summer, so thanks for the suggestion!

[asid61]

Originally Posted by InFlight

I would suggest the design process starts even earlier.
What are the requirements?

What motors ( rpm and torque curve) can I use
What ratio is required to match the motor and output.

Spacing needed for gear, sprocket or pulley system.

Once you have requirements and math done, you can then look at different implementations that would most efficiently solve the problem.

That's pretty fair.
For me it varies; sometimes I'll start with JVN calculator, other times I'll design the mechanism first and then look for how to move it around. I think there are uses for both methods. Fortunately, the existence of the Versaplanetary lets you cheat a little bit by sticking a 1 or 2 stage in as a placeholder, and let prototyping figure out what it should be exactly later on.

The 973 RAMP videos have also really opened up new horizons. For anybody looking to get into more layout sketching, the RAMP videos are great.