The Structural Aspects of Car Safety
Timeline
Before the project
I have always been interested in cars and how they work but also what happens when they crash. So I would watch crash test videos and notice the different ways each car deforms differently not to mention disassembly and destruction was quite fascinating to me as well.
Beginning project goals
My goal is to determine the safest body style of a car and what makes a specific body style the safest but there were a few things to consider before I could do my testing:
Which variables to change and which to have stay the same
In order to make the results as accurate as possible I have to keep the amount of variables to an absolute minimum and keep the variables that I do change in my control so if something goes wrong or it doesn't have the desired result, I know why.
The variables in my project before even thinking about the tests themselves were about the design of the frames and what I should make different about each one. The most noticeable change is of course the general shape of each frame but I can also change the different components of each frame. I can change the sizes and thicknesses of things such as the A pillars or the firewall.
How to fabricate the models
The ultimate goal is to make multiple metal versions of each frame with differing variables
How to do the crash tests consistently
There are a few ways to consistently run each crash test. The most consistent way that I can think of is with a ramp, a wall at the end and letting gravity do the work. Seeing as the frames are going to have different masses though for maximum consistency I think I will make a pulley system in which I have a weight attached to a rope which is attached to the frame. I drop the weight from a determined height and that in turn pulls the frames down into the wall at the same speeds.
What makes a car safe?
There are many aspects of car safety from the active systems such as airbags to the passive systems like crumple zones and now systems to prevent crashes altogether. My project focuses more on the passive systems and the structural safety of cars, specifically the frame of a car. The goal of a safe frame is obviously to protect the driver and any passengers. It will prevent deformation in the cabin and make sure that nothing enters the cabin in the event of a crash. It will also be able to transfer energy effectively so that during a crash the acceleration of the car, from whatever speed it was going to zero, is as slow as possible. This is to reduce the amount of force that the driver and passengers experience and the goal of the crumple zone which is specifically designed to crush and deform so the area around the cabin remains the most rigid.
Designing the frames
I looked at many different cars of the same body type and found the similarities between the components of each frame and created a sort of generic version for each body type.
Using a program called Fusion 360, I started out making the frame with a reference picture to be used for proportions mostly and then tweaked it to follow the generic system I have. When I had the general shape of what the end product would look like I then spent most of the designing time with tweaking the shape and size of each component until the frame was exactly the way I wanted it. This process got faster after the first one as I already had a sense of proportions from designing the first one so I didnt have to spend as much time on tweaking.
Computer crash testing
I wanted to do some crash testing with a program called BeamNG Drive. It is a soft body physics simulator that uses something called a jbeam structure which is applied to a 3d object and is what tells the program the strengths in different parts of the 3d object so that it can accurately simulate deformation of the object. The jbeam structure is made up of a bunch of lines basically that create a mesh in the shape of the object and I set the strength values for the lines. the problem is that the program updated and the jbeams that I made no longer work.
Getting real
When I 3D printed my designs I was able to get a real sense of what my frames would look like in real life and I could see any changes that I would have to make to the design.
I had a few ideas for how i would fabricate metal versions of my frames. I thought about forging or casting the frames but after talking with a blacksmith he had shown me why I couldn't do either of those. He showed me some cast pieces and explained that with casting the metal is too rigid for what I want to do. So I determined that I should use sheet metal to form my frames.
To form the sheet metal I have a workbench set up with tools such as vices and metal break used for bending the sheet metal. For now I'm using the guillotine cutter to make straight cuts for the floor pan and roof but for the sides which have curves I'm using shears until I figure out a more consistent and faster way hopefully with an electric shears or nibbler.
In the future
Due to things going wrong and delays, mostly with finding an outside mentor and facilities, I haven't yet been able to do any crash tests yet or have a metal frame fully built yet so here is what I plan to do in the weeks ahead.
Get the metal frames built and figure out how to build them faster and more efficiently.
I am still in the process of building a frame out of sheet metal which is proving to be harder than previously anticipated. I am trying different ways to form the sheet metal and through this trial and error I will also figure out the best way to make consistent frames quickly and hopefully make four of each body type to crash test.
Figure out how exactly I want to crash test them and then do the crash tests
What I have determined so far is that I definitely want to use gravity to move the frames that way they are as consistent as possible. I still need to go through the trial and error phase for that as there will undoubtedly be problems along the way in building the ramp and wall system.
The Final Product
With the metal pieces all that is left is to put them together. i riveted together the pieces using the tabs that i had put in to hold them together as i soldered them. I used a blow torch and long sticks of solder for my soldering job. i learned how the aluminum and the solder would react and adjusted my technique accordingly throughout the job with mixed results. it was harder than i thought as i had to heat the aluminum to get the solder to bond to it but i couldnt heat it up too much as i could melt the aluminum if it go too hot. because aluminum cools down very quickly it was hard to keep it at the right temperature. the final result worked out about as well as i expected with my soldering when i first started looking a lot worse than the end of my solding job. something that i didnt take into account but realized very fast was that the aluminum would deform as it heated up so my aluminum sheets were no longer flat but that is also because of the thickness of this particular trial aluminum i used for this first model. if i had time to make a car out of the thicker aluminum then the deformation wouldnt be as much of an issue.