Transient Computational Fluid Dynamics

It is another day at the plant and as usual, your boss calls and says there is a meeting in the conference room that he wants you to attend regarding a compressor failure. That is all you know, but from experience you know it must be something major since all the “brass” will be in the conference room. The meeting starts with an overview from the production folks on what happened and their plan to put a spare into service that has not been refurbished and it had already failed once before. The bottom line is to get the plant back up and running safely, as the margins are good at this time. The vendor is on the plane and crossing the big pond as the conference goes on.

As it turns out the compressor has experienced another blade failure. However, this time it was clear the inlet guide vane also failed. Once again the vendor confirmed that this problem has not been experienced and it is unlikely that it is a problem with the compressor. At the meeting all kinds of plots from finite element analysis and computational fluid dynamics suggest no problems. Also, interference diagrams (refer to Figure 1 for an example) suggest there can be no interaction between the inlet guide vanes and the compressor blades. Surge is suggested, but the compressor was operating far away from surge. Yes, it is another case where everything shows no problems, but you are not about to tell this to your production manager as you don’t want to be booted out. So what happened?