Don't underestimate the back of a napkin.
As engineers, we know that inspiration strikes when it decides the time is right—not on our timetable. Many ideas pop up when there are no convenient means to properly document or photograph the concept. Twenty-first-century engineers, even those of us with 30 years of experience, don’t use a notepad and pencil as much as we did. But even now, the odd exception to this rule is the back of a napkin.
Conversations help engineers build ideas as they expose a problem rolling around their brains to a second or third set of eyes. When those conversations, and possible beverages, start flowing the mind can generate some great ideas. There’s a wild story where Pixar animators came up with the concepts for four movies over lunch while doodling on a napkin. Another example is Robert Metcalfe, who famously sketched his first ethernet designs on a napkin. The concept of the napkin piquing inspiration is so universal in the tech space that the Digibarn Computer Museum has a few on display—just don’t let its 90s-era webpage deter you.
Documenting our design schemes on napkins is a time-honored tradition. But with current technology, it doesn’t take long to transfer those ink stains on wood pulp into a simple 1D simulation. By creating that simulation early in the design and development process, engineers can quickly find design issues, constraints and system boundaries. When we start simply and add to that simulation throughout the development cycle, we can also save time and money.
The Benefits of a Simple Simulation
When engineers study physics, a basic example that comes up is the elevator and counterweight. The earliest exposure a student gets is that a counterweight lessens the effort required by the lifting mechanism. The elevator and counterweight have a given mass, so we can find the acceleration and tension in the cable.
This is the perfect starting point for a simulation. A simple 1D simulation can model the mass of the elevator, the mass of the counterweight, the tension in the cables and the acceleration of the car. From this system-level view, engineers put several variables together. And with a few educated guesses, they can assess if the concept looks feasible.
The conceptual design of an elevator needs to know how many floors the elevator will travel, how many riders are expected per hour or day, and the laws and codes for the specific region. All this data can be added to the simulation, once they are discovered, to add fidelity to the model.
Adding Complexity as We Go
Using a model-based systems engineering approach, we can easily add constraints to the simulation. While researching elevator codes, regulations and laws, engineers will find values for some of these parameters. The simulation they’ve been constantly expanding becomes a great location to document their findings.
Rules for the size of the elevator car, its speed, acceleration and dampening forces can all be added to the model. This fleshes out the simple simulation and gives it more utility.
Moving from a 1D simulation to a fully 3D system gives us the ability to bring in information from other departments. For instance, during the concept and design phases, we can sniff out potential problems with the footprint of each object added to the simulation. Is there enough room for the doors to open and interface with the safety and braking systems? Does our control system have enough real estate between the passenger compartment and the walls? Do electrical cables and wires have enough length to go from the bottom of the building to the top? And how much of a safety factor do we build into these simulations along the way? All of these can be answered piecemeal as the simulation grows in complexity.
Once this simulation model moves from a simple proof-of-concept to a fleshed-out model, we can run several other studies. Finite element analysis can tell us how much stress is in the cables when all the required safety elements are connected to the model. Thermal analysis, impact tests and compressible flow dynamics can all tell us more about the system. And those studies are easier, faster and cheaper to get started as the once simple model gains more details.
Engineers should also consider how changes are inevitable for most projects. The customer is feeling competing constraints just like the simulation engineer, and these constraints can shift footprints, loads and aesthetics. Sometimes we even get lucky and find a specific grade of construction material that could save us a few weeks or drop the cost of the project. But when these changes happen, especially on a system as safety-focused as an elevator, the changes need to be thoroughly evaluated before moving forward. Another benefit of the already-existing simulation is the ability to quickly evaluate these proposed changes. This concept not only saves money from a budget point of view, but it can also save on the costs of late-stage redesigns, recalls and issues.
Starting Small to Build Big
Creating the digital twin of our simulation will help us understand the performance of the full assembly over time. But that twin can get a humble beginning from the back of a napkin. Engineers have industry knowledge that tells us which components will wear out first, and a reasonable assumption for wear patterns and possible malfunctions, but the digital twin can give us clues about how our specific system will behave one or five years into the future.
Our simple initial model that started as a 1D understanding of the system will help to build this twin and understand the 3D assembly and interaction between parts and services. In other words, beefing up that initial model can help maintenance and facilities crews working on the design, and its digital twin, years in the future.
Creating a small simulation early to prove a concept or discover all the constraints in the system has the potential to save time and money. Once the simple simulation is working to our satisfaction, adding complexity to that system ensures that our model is closer to reality. Digital manufacturing gives us the opportunity to build our simple models up and understand how control systems, inventory flow and humans will affect the big picture. But it can all start with a napkin.
