Parametric Design and Workflow

[cyucelEJ2QP]

Hello, hope everyone is well,

Can I please get your opinion on one point? I started designing a model rocket cone, it is going well (thanks to great help from this forum). However I keep asking myself if I am doing it right.

Please see the screenshot (or the sketch called Cone Side), when I finalised this sketch I ended up typing 17.33 for that mid-height radius, then put in rounded but hard coded values for spline handles. I feel like after these changes my design is no longer parametric, the spline as is can handle few mm change up and down but if I reduce noseConeHeight and rocketOuterDiameter by 50% I end up with an ugly shape.

Is there a practical way to parameterise all settings of such a spline so that the design stays fully parametric ? Or is this too much to ask and there is a balance between parametric and practicality ?

Next stage is to create a small hole on top of the chute cover, please see the sketch called Cover Lips. The tiny square you see there is to align right on the cutout, roughly where it is now, but strictly speaking "wallTickness" away from the the edge of the cutout, not in X or Y direction but along the spline.

Again I found myself hardcoding distances, eyeballing from projections etc. It feels like every step I take after this point will make the design less parametric.

What is a best (good) practice here ? How do professionals parameterise their rocket cones ? 🙂

[TrippyLighting]

---

@cyucelEJ2QP wrote:

...

Is there a practical way to parameterise all settings of such a spline so that the design stays fully parametric ? Or is this too much to ask and there is a balance between parametric and practicality ?

🙂

---

By definition, in Fusion 360 when you are working with a timeline based design it is parametric, regardless whether or not you use user defined parameters. Just thought I'd clear that up!

For a fit point spline I'd say no. A fit point spline is a 5-degree multi span spline and as such quite a complex thing.

A 3-degree single span spline might be easier to define with user parameters and perhaps scaling formulae.

[cyucelEJ2QP]

So the workflow is: parameterise (user) as much as possible and hardcode the rest and when a size modification is needed fix it on the timeline and fix anything else that may be broken due to the size change ?

[TrippyLighting · Accepted solution]

---

@cyucelEJ2QP wrote:

So the workflow is: parameterise (user) as much as possible ....

---

What is it you think you are accomplishing by  making so many dimensions user parameters?

How frequently are you going to change dimensions?

In my designs I only give user parameters to a very few dimensions I know I will change more frequently.

I make only very simple sketches and name sketches and other objects appropriately, so I can find them quickly and then change dimensions/parameters in context.

Otherwise I'll have to come up with ever more elaborate descriptions and naming conventions for the user parameters. 

[laughingcreek · Accepted solution]

---

@cyucelEJ2QP wrote:

...Is there a practical way to parameterise all settings of such a spline so that the design stays fully parametric ? Or is this too much to ask and there is a balance between parametric and practicality ?...

---

I also rarely define user parameters.  I try to reference sketch articles directly whenever possible, such as using a sketch point for "extrude to".  then to change the extrude, instead of editing the extrude I just change a dimension in a sketch.  it becomes a visual process vs an exercise in finding the right extrude feature and/or digging thru the parameters table to find the right parameter to make a particular change.

on to your question of parametrizing a spline so it maintains a decent shape.  that will be hard (impossible?) to do with a fit point spline, but fairly easy to do with a control point spline.  I don't have time to create an example ATM, but here is a video from a guy I wish would do more tutorials that does a good job of explaining how to do it.-

https://www.youtube.com/watch?v=BFHaQGhxURs

[Bunga777 · Accepted solution]

First, it would be better to present the final goal of the target model.

Next, it would be better to present where the dimensions could change, and the range of those dimensions.

Otherwise, if all possibilities are considered, the design may never be finished.

[cyucelEJ2QP]

@TrippyLighting @laughingcreek @Bunga777 Thanks for your answers. I'll stop overthinking about parameterisation 🙂

[TrippyLighting]

---

@cyucelEJ2QP wrote:

... I'll stop overthinking about parameterisation 🙂

---

I don't think you've been overthinking it. Your questions are perfectly valid and there is not necessarily one way that fits all.