Figure 1 Design process
When developing the Claymore deployer or designing subsystems for satellites I follow the design process. The design process is a structured approach and covers from the initial stages of capturing requirements to release of the design. The design process helps identify where you are in the project and how to turn an idea into a functioning design.
As shown above (Figure 1) the design process isn’t completely linear as there is a feedback loop reverting back to previous steps. When carrying out a design/development project you will have to revert back to a previous step in order to progress. This is because new knowledge, changes in scope,requirements and problems will arise and you will have to make changes to your requirements or design.
Requirements
Before work should ever be carried out you should always record the requirements you must meet. This can be from the customer, mission needs and standards that you will need to adhere to from launch vehicles and organisations (e.g., ESA, NASA, ISRO, JAXA, Roscosmos). Keep in mind that your requirements may be subject to change due to project constraints, new developments as you progress and understanding the problems more. Therefore, always ensure your requirements are kept up to date.
Mapping out the processes of your mission, (including testing, launch and mission) will help you define requirements for each subsystem of your satellite. Also, as you are mapping out your processes it is important to begin the initial phases of a FMEA to identify potential issues that could be avoided.
This is also when you will begin to identify all the various subsystems in your satellite which will have their own specific requirements. For example, thermal, electrical interferences etc.
Sometimes it can be easy to go overboard with requirements therefore focus on the minimum requirements that allow your satellite to perform its mission and keep the nice to have ideas for future iterations.
Recommended tools to be used:
Research
This is when you research potential solutions that will enable the team to develop a satellite that will meet the mission requirements. It’s good to investigate if a solution to your problems has already been solved or if there is a solution to similar problem has been developed.
In addition, you shouldn’t just restrict yourself to space in your research as you will be surprised how you can take inspiration from other fields as well such as automotive, aerospace, naval, oil & gas etc. In previous pocketqube projects I took inspiration from the farming industry as they had a release mechanism I was looking for. After contacting their suppliers, I realised it wouldn’t be able to withstand the loads of launch, but I did find a space rated alternative that was used in cubesat missions.
Here are a list of recommended sources you should use during your research.
- Patents
- Academic papers
- Standards
- Industrial magazines
- Experts in your field
- Suppliers
- Networking events.
Recommended tools to be used:
- PESTEL/SWOT & TOWS
- Reverse planning
Initial concepts
Once the requirements and project scope have been identified and agreed it is now time to begin the initial concepts of your satellite and sub-systems. But before you jump to 3D modelling you should sketch out your ideas 1st. Always remember CAD software is like any other tool like a spade, CNC mill, drill in that you need to know what you are planning to do. Break down the satellite into its subsystems and how they will interact with each other. While sketching out your designs of your components and assembly you should consider how you will manufacture these components. For example, will the reaction wheels be turned, milled or 3D printed? Will the chassis of the satellite be a box extrusion, bolted panels, injection moulded, forged or bent sheet metal? It is good to consider these options early on as you want to avoid redoing your 3D models if you realise the part can’t be milled on a CNC and must be changed into sheet metal components. Also in some cases trying to convert a 3D model into a sheet metal part in CAD can potentially lead to errors in manufacturing as there are different design rules for each manufacturing process.
Brainstorm with your team to bounce ideas as the more concepts you have the easier it will be covering all your requirements. Tools such as 9 Windows and 3-6-5 are useful in planning out your satellite will interact in its environment and help you generate large amounts of concepts in a short space of time. While you are sketching and making the initial concepts on CAD you should review, add, or modify your requirements.
When you begin 3D modelling your satellite you should create volume claims for each subassembly (e.g., payload, radio, ADCS etc) of your satellite as this will help identify potential volume clashes before they arise. When you start to mature these initial ideas, you will come across a lot of issues which is where TRIZ can come in and help provide inspiration in solving your design contradictions.
Once you have developed some concepts and reviewed them with the team it is time to select the most promising concepts to progress further. I recommend advancing the top 3 designs as in my experience sometimes I had to revert to an old concept that was previous rejected as new information, change in scope and requirements made the “best” design unsuitable. A Pugh matrix is good to use to select these concepts as you can be able to quantify some of their attributes against the main requirements.
Recommended tools to be used:
- TRIZ
- 9 Windows
- 3-6-5
- Pugh matrix
- FMEA
- FEA
Detailed design
Once you have identified the most promising designs it is time to add more detail to your concepts enough that can carry out a second down selection where you will select the most viable design of your satellite. Once you have selected your concept it is now to finalise that design so you can carry out prototyping and testing. Depending on your design method such as using the V-model you might manufacture and test each subsystem before you do a final design. This means that you will be doing the detailed design, prototyping, and testing process concurrently.
For detailed designs FMEA will help you identify potential pitfalls in your design before they happen. In addition, this is where you map out the manufacturing and assembly processes for each component and assembly of your satellite.
Recommended tools to be used:
- Pugh matrix
- FMEA
- FEA
- Tolerance Stack
- Poke Yoke
- Manufacturing mapping
Prototyping
Figure 2 Prototyping process
In the prototyping stage (Figure 2) you will be carrying out various functional tests of each component and subassembly to prove out each working concepts. But before you do any testing you should carry of a design for experiments (DoE) for each test you want to carry out. This can help with scheduling in your project and ensure you are testing these components that would replicate their actual application.
By testing out each subsystem individually helps you spot potential failures that can be fixed before you put your satellite together. This will help debug your system when you put it all together and something fails as you had tested everything beforehand. During the prototyping stage it is useful to record every action, outcome and lessons learned in a report so you can refer to it when you need to make changes to your design. This is especially important if you made minor changes during the prototyping stage such as using a drill to enlarge a hole as you made an interference fit instead of a clearance fit by accident. The Ishikawa analysis and 5 whys tools can help you identify the root cause for each issue you face during the prototyping phase.
Once the testing and prototyping phase is finished you can finalise the design and manufacture the flight model for testing.
Recommended tools to be used:
- Lessons learned
- Design for experiments
- 3D printers
- Metal working machines (CNC, lathe, drills)
- PPE
- Ishikawa analysis
- 5 Whys
- FMEA
Testing
Once your satellite prototype is working and meets you or your customer’s requirements it is now time to qualify it for launch. For environmental testing you should refer to the relevant standards from your space agency (NASA, ESA, JAXA, ISRO, Roscosmos or China National Space Administration etc) or launch vehicle requirements.
A design for experiments for these environmental tests should be created as your testing centre will require the testing levels and your satellites attributes. This will ensure you test to the required standards.
Recommended tools to be used:
- Design for experiments
Release
Your satellite has qualified its testing campaign and is now ready for launch. Follow your launch provider’s requirements and the various legal requirements.
Once your mission begins it’s good to reflect on the project and keep the lessons learned up to date and ensure everything is recorded. You may have to repeat the same mission or develop out a slightly different payload therefore having this info available will avoid you having to relearn everything.
Recommended tools to be used:
- Lessons learned
There are many variations of the design process and the variation I use is explained here. As long as you have a structured approach in how you approach your project you should avoid the most costly mistakes and meet your mission needs.
If you need assistance in your pocketqube or cubesat mission please don’t hesitate to contact me at Andrew.dunn@wyrmengineering.com