Wednesday, August 19, 2020

Lighter Stronger Faster

As the goals of FS teams keep evolving, the fundamental principle is to build a fast and performing car that travels between two points in the fastest time possible. Turning to the age-old relation a=F/m, one can broadly split the overall investment of a team into two categories:
  1. Investments which increase F
  2. Investments which reduce m
Over the years (considering combustion cars), the fraction of investment over the second category has been steadily increasing. Today, weight-reduction is done on a grass-root level where teams are ready to invest the extra amount and time to reduce as less as 100 grams of weight. This may be attributed to two factors. Firstly, the upper cap in the engine capacity limits improvements in the powertrain subsystem. So teams tend to explore other areas (i.e, weight-reduction) and are willing to invest the extra resources for the overall betterment of the car. The other factor being the availability of materials which have exponentially less weight compared to the conventionally used materials (Al, Steel, etc.).

As the investment for weight reduction increased, it set the era for carbon fiber development (or the other way round too). Carbon fibre perfectly aligns with the goal of increasing the strength to weight ratio. Since carbon fibre offers a lot of flexibility and compatibility to the needs of engineers albeit being on the expensive side, they have been used extensively by teams. The following are the key features of our Composites R&D.

Having a proper setup for Manufacturing

One of the features by which composites differ from conventional materials like steel and aluminium is that the components are completely manufactured in-house and involve a lot of labour. It becomes really messy while handling the resin also when performing layup. So it is necessary to have a clean, spacious and comfortable setup to house all the moulds, fibre cloth, resin and other miscellaneous stuff required for layup with proper disposal (as there are a lot of single-time-use products involved) and cleaning procedure so that it doesn't get messy for the people doing the layup. Small setup hacks like having the resin container away from the unused fibre cloth helps because if any accidental spill happens and affects the unused fibre cloth, it will be a waste of money (as carbon fibre is really expensive).

Material Selection

Bi-Directional Woven Carbon Fibre Cloth

 For manufacturing one single component, there are a lot of materials involved and hence choosing materials might be tricky. For instance, the notable materials that have to be selected are the grade of carbon fiber, resin type, core material (if any), mould material and other miscellaneous materials like release agent and the finishing agent. The usual factors like budget, availability and properties must be considered for all these components carefully to narrow down the list of materials.

Obtaining the properties of the Laminate

Since the actual manufacturing is done by us, the properties are highly dependent on the manufacturing process and values are bound to change each time. Obviously these errors are unavoidable but we can improve on the simulations by testing and finding the properties of a laminate that was manufactured in-house and using that in the simulations rather than externally given values since it might have a different manufacturing background and may not be bankable. The tests include (but not limited to) static tests to determine required mechanical properties and tests to determine the fiber fraction in the laminate. Note that more than one set of tests might have to be done to check the consistency of the setup.

Structural Design and Simulation

An iteration from the composite Rear Bell Crank simulations

Most of the simulations are done using ANSYS ACP with the properties obtained from the above tests as the material properties. Most of the boundary conditions are given by the respective subsystems/rules. A manual estimation is carried out on the basis of the boundary conditions to determine the direction of the plies. Using that as an initial case, successive iterations are performed with changes in number/direction of the plies to get the most optimum combination. In the case of sandwich panels, the thickness is also iterated (if possible based on availability) and simulated to improve bending stiffness.

Determining the Manufacturing Process

Cutting Carbon fibre cloth before the Layup process

This is one of the key steps for the composites subsystem. Start from scratch. If at all, there is a proposal for an improvement in the manufacturing process, care has to be taken if the extra investment in the process actually yields better properties. This can be simply done by comparing the properties of samples with and without the new method and deciding if the improvement of the properties is significant enough for the team to put the extra effort (both money and time). To summarize, just check if the investment is worth it. For example, we found that vacuum bagging a sample after wet layup increases fiber content by a significant number thereby giving a green signal for the process. On the other hand, we still use 2D hotwire cutting to manufacture our moulds rather than machining. Note that the manufacturing process does not only involve the layup process, but also includes manufacturing the mould.
The other aspect to keep in mind is to consult the respective subsystems about the manufacturing method. For example, the aero components will have a priority on the surface finish on one side. These points have to be taken care of while designing the mould and manufacturing the component.

Our Progress as a Team

We started off our journey in composites by manufacturing the seat and now we manufacture the entire aerodynamic package, steering wheel and other miscellaneous components like switch panels and bodywork. With the development of composites, we look forward to expanding the use of composites in subsystems other than aerodynamics. The primary goal is to replace components on the car with composites as much as possible if it aids with weight reduction without compromise in other properties. A preliminary list of components are selected for design. These components will be tested and validated properly before giving the green light to be on the car.

This is how our team handles the Composites R&D for making our car. If you feel that we have missed something, please do let us know in the comments.

No comments:

Post a Comment

Lighter Stronger Faster

As the goals of FS teams keep evolving, the fundamental principle is to build a fast and performing car that travels between two points in...