Monday, January 13, 2014

Reader Comment – 01-13-14 – Containment

There was a brief but interesting comment about containment posted to my Sunday blog post about the Freedom spill. W.V. Peter Hunt wrote:

“The block wall containment appears to be too low and too close to the tanks to prevent a high level leak from spraying over the wall, even though it may have the required capacity.”

This is a topic that I’ve never heard discussed with respect to containment, but it is almost certainly correct. Given the proper product viscosity (and I have no idea what the viscosity of Crude MCDM is; it is not listed in the Eastman MSDS) and size and height of the hole in the tank, it is possible that material could spray out over the containment wall.

The news descriptions of the containment failure in this particular instance do not seem to indicate that this was the culprit in this leak, but we will probably have to wait for the CSB investigation to be sure.

Most hazard discussions about potential tank failures that I have taken part in have dealt with the failure of lines coming out of the tank (particularly where those lines connect to the tank) or due to corrosion at the base of the tank. Random holes in the sides of tanks have not come up as standard failure modes in any of the discussions that I have been part of; probably a failure in imagination as much as anything else.

Deliberate holes in tank walls have certainly come up in discussions in this blog post about the use of firearms in and around chemical facilities, but I have never addressed those in conjunction with containment. This actually might be a very effective method of attacking any number of chemical storage facilities, particularly where those with tanks very close to facility boundaries.

With this in mind, can any of my ChemE readers point me at a calculation for the spray distance for a hole in a storage tank?

1 comment:

Jim said...

Assuming the hole is big enough that the liquid is flowing freely, the theoretical distance can be calculated as follows. Note that things get complicated unless we assume that the loss rate is small relative to the volume of the tank. Otherwise we have to use calculus to allow for the change in the height of the liquid as the liquid escapes. This is based on Torricelli's Theorem.

Procedure :- 


ha = height of the hole above tank's bottom; 
hb = height of water in the tank; 
h = height of water column above the hole-level; 
v = speed at which the stream gushes out of the tank's hole horizontally;
t = time taken by the stream gushing out of the hole to fall to the ground; 
g = acceleration due to gravity; 
s = distance traveled by the gushing horizontal stream before it falls to the ground; 

The tank is assumed to be placed on the ground; 



We have to find out the value of "s";

h = hb-ha = 15ft 

v = sqrt(2*g*h) = 30.98ft/sec

t = sqrt(2*ha/g) = 0.97 sec; 

s = v*t = 30ft

/* Use this with templates/template-twocol.html */