A small amount of test data can go a long way to reducing the analysis effort. And it can be easy and inexpensive to acquire.
For example, this could be the temperature of electronic components within a circuit card, or wind loads on a structure. The methods by which electronic component manufacturers provide heat dissipation data can vary widely. These can be anywhere from mounting the component on a small circuit card and applying power, to suspending the component in midair and applying power to it. Quite obviously, the heat dissipation characteristics are going to vary considerably under these two scenarios. Neither one of these scenarios will accurately reflect the installation of the component on a custom board.
As an example, a circuit card component temperature can be acquired in several ways:
Using an infrared sensor makes this task quite simple, effective, and quick. Thermocouples can provide temperatures inside an assembly, while IR sensors require the assembly to be opened.
The best way forward is to acquire test data from a functioning prototype where acquiring this information by computer simulation would be difficult to do accurately. The general idea here is to extrapolate over the shortest “distance” possible i.e. using the most direct method available. This will undoubtedly shorten the time and therefore costs involved in generating an accurate simulation.
One area that comes to mind is combining fluid flow simulations with heat transfer interactions. Fluid flow simulations can be computationally expensive. Not only does the actual assembly model have to be simulated, but also the fluid flow throughout the assembly model must be simulated. If a fairly simple and inexpensive test can provide temperature information under real-world conditions, the simulation effort is greatly simplified.
Combining CFD and test data leads to much better Aerodynamic analysis. Having designed and conducted many wind tunnel and other aerodynamic-related tests over the years, I have come to appreciate having some real-world data available to calibrate a CFD analysis effort. Test results likely do not exist for the exact simulation that you may be after, but using test results as a way to get a “sanity check” on your CFD model can provide a real boost in confidence that you are on the right track.
Literally thousands of aerodynamic tests conducted by our military, universities, corporations and other government agencies are available at no charge. Often it is possible to combine the results of several tests along with a basic understanding of fluid mechanics to arrive at a pretty good estimate of the results you need. From there, running the final CFD analysis becomes much faster along with having the confidence to back up your FEA/CFD results with actual data.
Norman T. Neher, P.E.
Analytical Engineering Services, Inc.
Elko New Market, MN
www.aesmn.org
A small amount of test data can go a long way in reducing analysis effort—and it’s often inexpensive to collect. Whether measuring the temperature of electronic components or estimating wind loads on a structure, simple, real-world measurements can save both time and money.
Manufacturers often provide heat dissipation data for electronic components, but the methods vary. Some test a component mounted on a small circuit card, while others suspend it in midair. Unsurprisingly, these scenarios produce very different results—and neither accurately reflects the component’s performance on a custom board.
For circuit card components, engineers have several options:
Thermocouples — measure temperatures directly at the component.
Infrared (IR) scanning — capture temperatures across the component and surrounding areas.
IR sensors vs thermocouples — IR requires opening the assembly, while thermocouples can measure inside a closed assembly.
The best approach is to collect data from a functioning prototype. Direct measurements reduce the “distance” between testing and simulation, producing more accurate results in less time.
Some problems, like fluid flow coupled with heat transfer, can be computationally expensive. Not only must the assembly be modeled, but the fluid movement through it must be simulated as well. A quick and inexpensive test—like measuring real-world temperatures—can greatly simplify the simulation effort.
The same principle applies in aerodynamics. CFD provides powerful insights, but test data makes it better. After years of designing and conducting wind tunnel tests, I’ve learned the value of having real-world measurements to calibrate CFD models. Even if test data doesn’t perfectly match your case, it provides a much-needed “sanity check” and builds confidence in the results.
Thousands of aerodynamic tests conducted by NACA, NASA, universities, corporations, and government agencies are available at no cost. By combining data from multiple tests with a solid understanding of fluid mechanics, engineers can often estimate results with surprising accuracy. From there, CFD runs are faster, and the results are backed by real-world evidence.
Use data to guide simulations — even small tests make models more reliable.
Prototype when possible — direct measurements reduce complexity and error.
Don’t reinvent the wheel — leverage free, historical test data to calibrate your models.
Balance simulation with testing — the best results come from integrating both.
