What are the most effective methods for testing and validating mechanical system designs and prototypes?

Some questions to start off with include: - Where will system be used geographically. This will determine certification required for piping or electrical and materials. - Cost / budget and timeline - System pressure, flow, and medium (air, gas venting, etc.) requirements.

Defining testing objectives and criteria is much easier when the product has good product requirements and subsystem functional requirements. Depending on how they are written, these requirements can act as a north star for effective testing. It is also important to use your engineering intuition to identify risk areas in a design and ensure there is adequate analysis and testing coverage for these risks.

To begin, you need to set priorities - what are the low TRL (technology readiness level) systems? Efforts should be made to test these first as they have the highest risk. Systems and components which are off-the-shelf or have design heritage will have lower priority. Secondly, you can define objectives and criteria for those specific systems that need early testing. While there are obvious KPI's, don't forget reliability and robustness - it is critical to demonstrate how the system's performance may change over time or when subjected to variations in the operating environment like temperature, humidity, ambient pressure, ground vibration, etc.

The voice of the customer is very important in this phase. Try to grasp what are the specific experiences that the users are hoping to get out of the design. Break it down into smaller bite-size objectives that work together to deliver those experiences.

In the aerospace world, "Test-Like-You-Fly" (TLYF) is used as a grounding principle for test method and setup. It is imperative that a system be grounded with the same boundary conditions and operating environment in test as it will be in operation. If using simulation to predict performance, it is important that the model be subjected to realistic boundary conditions and not bad approximations. Then, you must know the answer before you take the test. That is to say, you must already understand your system, know what you are looking to achieve in the test, and have an expectation for what will happen. That is critical in defining where to place sensors to acquire data that can actionably be used to validate and improve the system.

For certain environmental tests like UV, temperature, humidity and water ingress testing (reliability), there are established standards that allow quicker testing that is somewhat equivalent to longer environmental stresses. Eg., in water ingress testing, often devices are tested to a certain pressure for a very short period of time to find leaky units at the end of a production line. In this vein, testing during product development (to derisk designs) is very different from testing on the production line (to guarantee ongoing quality). Both have different goals and need different methods.

If in-house testing is not possible, integrating BIQ quality gates and IPVs can be done by quality lead or teammates. Creating subassembly modules can reduce field installation time and be tested for functional testing before shipped to field or customer installations.

Before getting into the instrumentation of the device, it is important to make sure where sensors need to be placed. For marine renewable energy devices, previous structural simulation studies coupled with CFD analyses make possible identify areas of maximum strain, stress and deformation that in turn are points of interest when physically testing the machine.

When using simulation for testing, it is important to realize the complexity of your specific test. Sometimes a very simple test like bouncing a ball can get very complex due to non-linear material and dynamic impacts. In the absence of access to resources, try to simplify the simulation into smaller samples, gain enough confidence in certain designs, and pass it on to contractors. Many companies out there can run simulations with higher accuracy and less time compared to you learning the entire setup.

Reverse engineering existing products will give a lot of insights into how some companies have resolved similar issues that you might be facing.