The chemical leak that occurred in Charleston, WV is a prime example on how the upfront costs of installing an automated safety system can pay for itself many times over.
On January 9th it was discovered that the chemical 4-methylcyclohexane methanol was leaking into the Elk river in Charleston, WV. The chemical was being stored in tanks at an old Pennzoil oil terminal now operated by Freedom Industries. The chemical is used in the process of coal preparation. Both the tank and the secondary containment had failed. The chemical was flowing through cracks in the secondary containment dike (which was built from cinder blocks) and flowed over the bank and into the Elk river. A short distance downstream happens to be the largest water treatment plant in the state, serving the largest population concentration in the state. The water treatment plant was contaminated and compromised, along with the all of the distribution systems it serves. This incident has affected the life’s of 300,00 or so people and many businesses, as the tap water could only be used for flushing toilets and for fire fighting purposes.
I have read that the normal procedure at the location was to gauge the tanks with a gauge stick.
If the secondary containment had been up to par, and if there had been a safety system in place, the results of this incident could have been much less impactful.
I am not going to get into the politics of the situation, rather just make my case for an automated safety system.
Imagine if the tanks were being monitored with liquid level monitoring technology and this data was being brought back to a central point such as an RTU or PLC. With this data you could immediately determine tank level on a specified update interval of your choosing and implement rate of change alarming. Rate of change alarming simply alarms if the tank level changes faster than a preset value. For example, you would not expect the tank level to change when no product is being introduced or removed from the tank, so if the level were to change 3 inches in 20 minutes, you could trip an alarm and also calculate the amount of product lost using the tank dimensions, and know immediately the severity of the situation. When the tank(s) are being serviced you could disable the rate of change alarm by simply installing a switch on a pedestal that would be engaged while the product is being transferred to or from the tank. For safety reasons the alarm would only be disabled for a preset amount of time after the switch or button is engaged in case the operator forgets to disengage the switch/button.
The alarming on this system could be set to “cry out” notifying plant employees via text message and email, etc…
The secondary containment could contain a float switch and this float switch could trip an alarm both notifying the operators of the abnormal condition and engage a pump to transfer material to a predetermined tank.
Nowadays with the technology available this can be done quickly and rather cost effectively, which could have possibly saved this company from the financial problems to come (and I am going to go ahead and speculate they are facing bankruptcy, and possibly criminal charges), the impact to the people of the Kanawha Valley, and the impact to the environment.
(As I write this the impact is still reverberating throughout the communities affected)