Tankage Photo by Navintar - Designed by Freepik |
The Rest of the Story - The Exploding Tank Paradox
Michael A Eastman
Senior Technologist - System Integration Design Development, Analysis, Configuration Data Management, Application Development
Usually the HR interviewer will ask what was a problem that needed to be solved and how did you apply your skills to that end. This is another article in the continuing series 'The Rest of the Story' that goes further back into my history as a young contract engineer and deals with how I solved a problem with a chemical tank farm that exploded.
In my reading of other publishers on interviews and what the interviewer is looking for, I came across an article by Lou Adler dated February 20, 2014: How to Blow an Interview. In it he indicated that "How You Answer and Ask Questions Will Determine if You Get the Job." He continues with the real process, "...candidates aren’t judged on how well they do their jobs; they’re judged on how well they describe how they do their jobs." Furthermore he adds - "there were a few points to note that all interviewers are attempting to evaluate the following:"
- How skilled you are and how you applied these skills on the job
- If what you've accomplished is comparable to what needs to be accomplished
- How you’d fit with the team, work well with the hiring manager, and fit with the company “culture”
- Your level of drive, initiative, and motivation
- Your upside "potential"
For more of his article and many more please go to Lou Adler's LinkedIn profile. To answer each of the points that he outlined above I started my process for telling the stories of my achievements - Dragon Slaying Stories (per Liz Ryan) or how I accomplished that achievement - The Rest of the Story (per Paul Harvey). I prefer to show a few steps of that process to give people more of an idea of what it took to get a solution.
With that in mind I reach back to my early years when I was working at a Corporate Engineering Division - Chemical Intermediates (CED-CI) in St. Louis about 1974 - 1979.
The titles that this was accomplished under: Consultant Associate conducting Chemical Engineering Research, Chemical Plant Design with drafting of Architectural, Piping, Electrical, Structural and Tankage Farms.
As listed under the accomplishments section: "There was an incident at a chemical intermediates plant and several tanks blew up. My supervisor gave me the specifications on the tanks and the types of chemicals that they were mixing at the time. His instructions were simple, find out what happened and report back to him. I analyzed the mixture ratios and the tank structures and found that they were mixing the solutions too early, creating hot spots on the glass-lined steel tanks. This in turn increased the pressure inside the tanks and at the same time was weakening the structure. I redesigned the tanks and fixed the mixture ratios so that the temperatures did not create a hazardous situation."
The Process:
The first is to verify the basic information that I had. A quick trip to the research library to check out the chemicals being used involved other details that I knew would be needed for the tank design. For the three chemicals I was able to verify the specific gravity, viscosity (roughly defined as a liquid's resistance to flow), surface tension, capillary rise along with pressure/temperature charts. As noted in chemistry, the specific gravity and viscosity changes as the pressure/temperature changes. An example is shown in the following chart:
Next would be the combinations of the three chemical intermediates (Here I will call them A, B and C) that would produce the desired chemical product. There are six total combinations, but only one combination that will provide the results that we are looking for. For this the research library came in useful again. The different interactions between the three were analyzed to find the amount and timing needed for optimum chemical interaction. Luckily, I had access to the Gould Mainframe Computer, a healthy dose of Fortran (Level V) as well as certificates in Disspla (a graphical display language for Fortran). Ternary graphing was used to find the mix needed.
The basic Ternary graphic would look like this:
How to read it:
The other variables in the graphing process involved temperature, percentage of chemical interaction and time. Pressure is held constant where all three have optimum interaction.
Next was the tank design to accomplish what the graphs show. I had a choice between two common diameters of tanks, 6.5 foot or 7.5 foot. I produced a flat of each tank to determine which would give me the room and timing needed. I also produced a flat of what the original design looked like to give me a reference. The 'flats' are produced by taking the geometry of a cylinder and laying it out using the 2πr equation. The 6.5 foot diameter did not give me enough timing for the chemical interaction, so I layed out the flat for the 7.5 foot diameter.
This is the design from the original engineer. He was using a 6.5 foot diameter tank and all the inlets were in a straight line. In this arrangement the chemicals would reach critical temperature within one quarter of a tank circumference. This caused a hot spot that melted thru the glass lining of the tank, then through the steel casing. The pressure would spike which caused a rupture. With the heat, chemicals and oxygen together we have a nice explosion.
The next drawing shows a overhead of the three main inlet piping, 20 inch, 14 inch and 10 inch. I also drew up the dimensions of the tank, the most important would be the interior.
The next drawing is a mashup of the top view and side with the piping to get an idea of the orientation of the structure. Basically a one-off to check my dimensions.
For reading of the original article and others you can go to my LinkedIn Site.