As the cliché goes: “timing is everything.” This often applies to forensic analysis of fuel gas leaks and their subsequent migration, whether it be natural gas or propane. Usual questions are, how long did the leak run before an explosive mixture was produced? What was the gas concentration at the time of the incident? Would the odorant have been detectible? The answers to these questions lead to an understanding of the incident.
Simple analysis using the volume of a structure and assumed parameters such as air changes can be done in an attempt to answer these questions. However, the simple analysis typically assumes a homogeneous mixture. In actuality, the mixture of air and gas is most often not homogeneous during a gas leak incident. For example, methane (the prime constituent of natural gas) has a density less than that of air and therefore tends to rise. However, the two gases readily mix. These are competing effects that are difficult to model using purely analytical (i.e. hand calculation) approaches. Empirical testing is an option, however it is not always practical. Simulation using computational fluid dynamics (CFD) is often the most accurate and practical method for fuel gas concentrations over time during a gas leak. CFD is based on centuries of scientific testing and research into the behavior of fluids (i.e. gases and liquids). CFD is also used to evaluate aerodynamic capabilities in aerospace and automotive industries. Example results from such simulation are shown below.
These results are taken ten minutes after the start of flow of natural gas simulating a leak. The gradation or change with respect to position is shown as different colors. This is a transient analysis, meaning that the effect of time is taken into account. The results at different times can be analyzed with respect to the incident. This could be anywhere from seconds to minutes to hours. Further analysis can be run such that the majority of the volume is at a concentration higher than the upper flammable limit (UFL or UEL), if required. This is difficult and potentially dangerous to empirically test. Simulation also produces demonstrative images and/or videos that more easily communicate the physics of the incident. Answering these questions about fuel gas migration is just one of many applications for this advanced simulation capability.
Please contact me for further discussion about fuel gas migration.