Modeling turbine housing

[yoshimitsuspeed]

I left the title vague because I suspect I will have ongoing questions on how to model this part properly.

The first question is based on properly modeling the scroll. My thought was to do a loft on rails however I can't figure out how to model the proper scroll curve.
The scroll is very simple based on the A/R.
http://www.turbobygarrett.com/turbobygarrett/turbine_housing_AR_and_housing_sizing

So the scroll starts at your determined AR where the inner curve of the scroll meets the base circle and the outer curve tapers from that measurement to zero where it meets the base circle.

I can kind of get something close to what I want using multiple conic curves but it's not perfect. I think I should be able to use a single conic curve to do this but can't figure out how. If a conic curve can't achieve this then how would you?

[Anonymous]

I haven't tried it yet, but it looks like you can import equation driven geometries in spreadsheet form, as shown here for a helix within a helix geometry:

https://screencast.autodesk.com/Main/Details/35f7beac-de8a-47c8-9ef7-0f927ab15b3d

Let me know if you try this, likely for a polar equation, and if you run into any snags, and I'll try it out too.

Jesse

[San_Escobar]

Funny enough! but I find harder to build the excel document (scripts and so on) rather than the 3D model itself. 

The reason is I am not sure whether I can add value on spreadsheet randonly like just drop in the value anywhere or if it most be well placed, like programming languanges, so that when you load the script (XL file) it happen what suppose happen.getting the maths right. 

In other hand, such shape I do use surface model for it so that I have high control to the model.  

do you know where I could find a great insight to improve my script skills?  

Cheers!

[Anonymous]

Hi, I made a video for you that illustrates making an equation driven table in s program like Excel, then importing the csv file to serve as a loft path or rails, to in this case just make a golden spiral 3d shape.  Hope you find it interesting 🙂

Jesse

[yoshimitsuspeed]

Thanks for the example as I really do need to learn how to do things like this but is this really the simplest and best way to do this inside F360?
It really seems like I should be able to sketch this in a couple clicks.

It doesn't really matter for something like a turbine housing but out of curiosity is this an actual golden spiral or are those just arc segments of an approximation of the golden spiral?
If it was actually a golden spiral then wouldn't it be possible to just have a start point and an end point with the formula that gives the path it should follow?

The way I am starting this I could do the spiral first but what if I couldn't or what if that doesn't work? What if I wanted to create this based off existing geometry?

[Anonymous]

Hi, you wrote:

"Thanks for the example as I really do need to learn how to do things like this but is this really the simplest and best way to do this inside F360?"

For equation driven geometry Fusion360 currently does not support this, except for driving one dimension with an equation of another dimension, but that's not going to give you geometries such as a polar curve with changing radius with respect to angle.

"It really seems like I should be able to sketch this in a couple clicks."

For most sketch geometries it is just a couple clicks, but for extremely powerful and essentially unlimited geometries that precisely follow various equations, it's really not that much work once you get the hang of it to make an equation driven spreadsheet and import it to get the equation driven spline for further 3d modeling.  If you're trying to design things like turbine housings, you could freehand splines to look approximately correct.  In fact that is a good way, you could insert an image of a turbine housing and use splines and/or arcs to get a very good approximation of the necessary curves.  But if you want it to be mathematically exact, having and using equations is the way to do that.  If you need help doing this let me know. 

"It doesn't really matter for something like a turbine housing but out of curiosity is this an actual golden spiral or are those just arc segments of an approximation of the golden spiral?
If it was actually a golden spiral then wouldn't it be possible to just have a start point and an end point with the formula that gives the path it should follow?"

Equations can be solved analytically, but for all graphical or physical representations of equations, this must be done numerically, i.e. as approximations via a dataset of 'test points'.  The spline approximation is quite good, and can be improved to whatever precision is desired by increasing the number/resolution of points.  

Good luck!

Jesse