Shortly after beginning to work in the electronics cooling field (1985), I heard about visions of a paperless engineering environment. The benefits were obvious, especially when faced with what to do with desktop-sized plots and drawings. For the most part, the vision has been realized. We store information electronically and routinely collaborate across geographic boundaries in real time as long as there is access to a common database. This engineering environment has many benefits for thermal engineers. Assuming common software, we can quickly send models to colleagues for assistance and review. We also can store substantially more information and details, especially when compared to folded paper drawings.
In spite of the ease of data storage, I frequently encounter thermal analyses and test documents that do not consist of much more than a set of presentation slides. A recent example was revisiting products that were designed and analyzed a few years ago. This product had gone through the typical design review process where thermal analysis presentations were tailored for peer reviews and then for customer reviews. Each tailoring removed more detail, and even though the reviews were judged complete, the summary type information in a presentation leaves out many details. In this particular example, it was possible to find the thermal model simulation files which revealed what was analyzed, but the background information of how the boundary conditions, heat loads, and material properties were obtained was lost. While there are many reasons why only a presentation document is generated, a common reason is that once the reviews are completed the project did not want to spend any more money generating reports.
We all are exposed to the pressure of reducing the time to perform thermal analyses and testing and may even generate metrics to show that we are faster now than in the past. The thermal analyst often works in a consultant type role where his time is bought by the pound (or Kg to stay consistent with our SI unit policy). There is a short term cost saving to stopping the effort as soon as the customers and managers are satisfied that the thermal analysis is sufficient. However, the costs saved by not documenting can easily be overwhelmed by the costs of recreating the information if more effort is needed in the future.
Part of the problem with presentation documents when viewed from an archival perspective is that the person generating them often needed to satisfy several constraints. There was probably a page count or time limit, and back-up material never receives the same amount of attention as the main body. There is also a marketing type pressure to show how this thermal design is similar to other products that have worked and this leads to an emphasis on past success rather than a more thorough engineering review. Adding color isotherm plots from a simulation can make the presentation look appealing but often do not do much more than show that a simulation model exists. An additional point is that adding details to a presentation can make them look cluttered. To compensate, these details are communicated orally and at best are only available in notes and memories.
I encourage all of our readers to examine their analysis presentations to improve them from an archival perspective, and make an extra effort to create separate thermal analysis and test reports. While the desire to be finished can be significant near the end of an effort, improved documentation and the associated cost avoidance offsets spending the additional time.
“Those who cannot remember the past are condemned to repeat it.” (George Santayana)
