There is a misconception about CAD and design where some will quickly fire up their 3D modelling software of their choice when given a project. This has been a major grievance of mine when working in a team and having to resolve issues later down the line that could have been avoided if team members realised CAD just everything else is just a tool. A tool that allows us to develop complex machines but just like any other bit of equipment things can go wrong. I always compare CAD is the engineering equivalent of a spade to gardening. Sure you could just dive in a start shovelling head 1st to landscape your garden but without proper planning ahead you just end up with a sad mess.
Here is a short summary of how you should tackle a design problem and when you should start using CAD during the design process.
Step 1: Gather requirements
At the beginning of a project you should gather your customer’s (external and internal) requirements as this will give you the goals you need to achieve.
Step 2: Brainstorm/research
Come up with ideas with colleagues and sketch out ideas on what the sub-systems of your satellite should be. Include members of your team who aren’t involved with your system as both of your work can have an impact on each other. You may only be doing the mechanisms for a solar array but that could have an impact of the centre of gravity or interfere with a radio, ADCS or electronics board. Plus having an outsider asking the “obvious” questions will help provide concepts.
During this step you can research components in the market to see if you can just purchase something than make a custom. Sometimes it is cheaper and easier to implement a space qualified component than something bespoke.
Step 3: Sketch ideas
We are at step 3 and not even started doing CAD. You should out ideas on a sheet of paper or whiteboard and just brain dump your ideas. As you are sketching think of the things you will need to calculate, how you are going to manufacture these parts. Manufacturing is something that can be missed out the last thing want to do is spend 100 hours modelling a component or subassembly and realise you want to make them as a sheet metal part. Even though most CAD software can allow you to just convert parts into sheet metal in my experience this has never worked as intended. In addition every manufacturing technique available has specific design requirements therefore your model might not be manufacturable. Always consider how your part will be getting made or how your satellite will be getting assembled to avoid costly and time consuming rework.
Step 4: Bring out the calculator (or Excel)
You finally got some ideas you want to develop into a functional design, it is time to get some. Before you can design your ADCS system you will need to define how big your magnetorquers and reactions wheels need to be to orientate your satellite. Carrying out some initial calculations will help you in your design as you can define the dimensions of your parts without the guesswork. Your sketches and calculations will be your rough guide of what you are designing.
Step 5: CAD
Now it’s time to fire up your CAD software of choice and start your initial designs, such as initial concepts, volume claims of each subsystem, subassemblies of your satellite etc.
Step 6: Refine your design
As the project and design progresses you will acquire more information, knowledge and experience problems which will require you to make adjustments and refine your design. You might need to review or make more calculations which will feed into your design. You can use other modelling tools like FEA to help you develop a design that is fit for purpose. This can only be done if you know how your design will interact within its environment so that you can create a simulation that would be representative of its real life counterpart.
Once the design is ready it’s time for manufacturing, testing and release of your design.
Summary
CAD is nothing but a tool and you should drive the software and not be driven by it. CAD is like every other computer software where garbage in will result in garbage out. Also you need to understand the limitations of your CAD software as each vendor has its own quirks that will require you to find a work around to overcome them.
Also it is very easy to make a design that looks good on the computer screen but will completely fail in the physical world. For example your solar array is going to sit perfectly in the 3D model but you picked the incorrect tolerance, manufacturing process or didn’t take into account the load paths or stresses that would occur which could result it in breaking apart or jamming. By jumping straight into CAD you could end up making costly mistakes further down in your design process. This also includes FEA, without a proper grasp and understanding of your products environment, you will have a simulation that could provide a false positive of the design being fit for purpose. Obviously not everything can be accounted for and an unknown problem that no one could foreseen could cause a failure during testing, but a structured approach should help you avoid the obvious and unavoidable problems. Spending more time in understanding the customer requirements, problems and possible solutions before jumping to CAD can save you time and money in the long run.
Follow me on Instagram:
https://www.instagram.com/wyrm_engineering_store/
For those of you engineers or designers who use 3D modelling why not take a break with our stylish Design Make Test coffee mug.
