Skip to main content

Reverse Engineered Objects: Progress II and Final Renders

 





 
        In this project we are tasked with reverse engineering an existing object to study methods in modelling, and to develop a competent level of understanding CAD software such as RHINO 7. I am not gonna lie but the beginning stages of this project was rather hellish. The idea of translating an existing object into digital software, with correct dimensions was not only tedious but it also reveals flaws and imperfections on the object of study. 

I found that measuring each individual facet or extrusions in the gadget showed me that even if the object was designed in CAD with mathematical precision, the final physical object is still bound by "Physics" and would never be perfect to scale. My object of choice was a low-cost and mass produced video-game controller. We were tasked with measuring each part with a digital calliper in order to make the models precise. But the problem was visually even though my object seemed to have straight lines or perfect cylinders when viewed with your own eyes. Not a single part of the object, such as the screw posts which gave me a lot of problems, and measure similarly with an offset sometimes of almost 2mm. This made me realize that the physical object had some considerable warping which did not look good when recreating it digitally.




In a brief meeting with the instructor they informed me of how this material was developed and it was through a process of injection molding. In which plastics or rubbers are heated and pressed into a metallic mold, during the cooling process materials such as plastics may shrink up to 25%. This may have been the reason why some screw posts in the object don't all have the same measurement. To solve this issue i've just decided to find the average between similarly looking parts and used it as the basis for duplicates.



My object had four different materials on it. Mainly Plastic(Shell, Buttons & Triggers), Metal(Screws, Solder), Wood(Bottom of PCB), and Rubber, which gave me a diverse amounts of options to experiment with in rhino. 

I began modelling the top shell of the controller which was flat, this made it easer for me to start the project in 2d making measurements and placements easier.


Then I proceeded with the buttons and the two triggers because I needed to use them to cut the surface of outer shell creating the holes we see above and below. 


This section is the other half of the shell which had an angle cut into it, this was difficult at first but i just decided to  elongate the screw posts from the top shell. This was to determine the placements of the larger posts we see below. I just subtracted the difference in heights between both sides to determine the true angle of the cut-out.



Originally, I planned on creating the PCB's textures by hand by using the hatch function on rhino and implementing a different material on each one. But unfortunately I was incredibly short on time since i've had already invested a lot on modelling and just decided to use photo textures. (I have a life too!) 


Here is the top part of the PCB with the metallic etchings. On original iterations of the rendering the metallic textures appeared flat. So i've decided to play with the luminosity and gamma(emission of light) levels in photoshop and experimented until I was satisfied with the effect. 

Fun fact! Rhino accesses the material live so this means if you edit the PNG or JPG in realtime on Photoshop/Lightroom and saved it with the exact naming convention. It would automatically update the texture in Render Mode, this function made it easier to adjust the light levels on the source image(Yay!).




Exploded View on Rhino. 









Below are the final HD renders of the project, I did utilize blender for rendering because Rhino 7 is not yet fully optimized for it on MAC computers and takes hours to complete. 












Comments

Post a Comment

Popular posts from this blog

OBDF310 Milestone Project 1: Stacked Slices

         My current sliced model for this weeks blogpost. I wanted to test modelling an object in Rhino first and using the mesh as the starting point for the contour slices in grass hopper. I ended up experimenting on different extrusion depths and the vector direction for the slices.  Modularity of grasshopper allows for rapid iteration and gives me a couple of choices to pick from. I decided on the vector slices along the y-axis which gave me a very simplified version of the bird but still recognizable.  Rough model doesn't need to be detailed because the contour slices simplifies everything else.
Post a Comment
Read more

Algorithmic Surface Modelling

  This week surface morph animation and some extra stuff.  Starting out with a divided curve I was able to use it determine points for circles and then using said circles to loft a structure. I then utilized the random numbers to offset the radius for each section in the loft and by utilizing "replace item" I was able to move a circle around the geometry. To show this variating animation without changing the seed values in the random numbers node. Bonus: Sending Grasshopper data via OSC/UPD to other programs I really wanted to use the program in conjunction with my main creative tools just to give me more reasoning to use grass hopper. You can utilize geometry in grass hopper and use them on other programs like MAX msp and Touch Designer for live visual effects or any other use cases. But as a caution try to re-mesh the models or simplify,  because grasshopper uses a lot of data and sending them live can cause it to crash if the models are too complicated.  Geometry used to C
Post a Comment
Read more

OBDF 310 Milestone: Stacked Slices Laser cut Final

  This week in my Object design class, we were able to finalize out designs for laser cutting slices. My penguin model has gone through a couple of design adjustments mostly to have a cleaner and readable structure. simplifying the model makes it easier to distinguish, spherical curves on the original brep reference doesn't seem to provide appealing slices so I decided to go with a more geometric form. Once processed through grass hopper I didn't have a need to include wings because the tapered form still translates to something that appears like a wing.   On the left are the initial slices, the right is the cleaner side decided to mirror to create a more readable model. Above is the layout for laser cutting, I ended up layering the cut-outs manually because the grid method shown in last weeks class was quite troublesome. I just couldn't figure out how to pull of the grid placement on vector direction and didn't want to spend too much time confusing myself. (The penguin
Post a Comment
Read more