Disaster vacuum

Lees (Loss Prevention in the Process Industries, 2d ed., Butter-worths, London, 1996), BP (Hazards of Trapped Pressure and Vacuum, 2003), and Kletz (What Went Wrong —Case Histories of Process Plant Disaster, Gulf Publishing Company, 1989) include additional case histories providing valuable lessons about how equipment failures and human errors can combine to inflict vacuum damage. 

Every year I run a chem lab and when someone is doing a vacuum filtration, suddenly I ll hear a scream and a moan of anguish, as water backs up into someone s filtration system. Usually there s not much damage, since the filtrate in the suction flask is generally thrown out. For vacuum distillations, however, this suck-back is disaster. It happens whenever there s a pressure drop on the water line big enough to cause the flow to decrease so that there is a greater vacuum in the system than in the aspirator. Water, being water, flows into the system. Disaster. 

This section lists fundamental potential pitfalls when working with a vacuum system. If you read no other section in this book, read this one. By following the rules and guidelines that you are directed to within this section, many hours, and perhaps weeks, of problems will be avoided. It is better to avoid the situation of not having enough time to do it right, but plenty of time to repair the wrong. This list is not meant to be comprehensive—it cannot be. It is, however, a collection of the more common disasters that occur on a laboratory vacuum system. 

Due to variations in equipment, controllers, and designs, what you see on your system will probably not be what you see in this book. You will have to accommodate and respond accordingly. In addition, re-read Sec. 7.1 and be sure you understand how accidents (disasters) can happen and how they can be avoided. That knowledge in itself will be the first major step to successful vacuum practice. 

Suggestion 3 has limited practicality. It can only be effective after work on the vacuum line is complete and you wish to shut down the system. This plan will not help vacuum lines left unattended and/or whose liquid nitrogen has boiled off. Regardless, if frozen oxygen boils off faster than can be released by the vent, the results would be the same as if no vent were available. Thus, the potential for disaster is still present. 

---