Details

Parent Category: Safety Info Posts

Category: Chemical Process Safety (PSM/RMP)

Published: 26 August 2023

This past week, we saw the 2nd catastrophic failure of an NH3 nurse tank in three months.  This is indeed a rare occurrence.  Both incidents appear (from news accounts and photos) to be circumferential weld failures on these vessels.  Both vessels failed while they contained NH3 (levels unknown), but neither was actively being filled at the time of their failure.  The incident this past week was said to be caused by the excessive heat the area was experiencing; let's all hope that is just nonsense and someone talking about matters they know very little about.   In both incidents, it appears neither tank was being moved at the time of their failure.  In the MN incident, the farmer stated he was walking away from his equipment when the tank failed.  In the IA incident, it seems (at least at the time of this post) that the tank was sitting in storage with other NH3 tanks when it failed.

The IA incident shows (see pics below) the power of a Liquified Gas under Pressure.  Notice in Pic #2 (aerial pic) where it appears to show where the tank was sitting when it failed (in storage with other nurse tanks), traveled several hundred feet, struck a Propane Bobtail truck, knocking it over, as well as struck other nurse tanks and knocking them over. 

The last two pics below show, just like the MN incident, that BOTH tanks had clean separation along a circumferential weld.  At this time, we do NOT know the history of these two PVs, but there have been several incidents in the past where a nurse tank was "repaired" and this work was NOT done to code.  But these PVs failing clean along a circumferential weld may be an indicator that investigators will surely be looking into. 

Then we have this "2015 study" after the 2012 catastrophic failure of a nurse tank in IA where the study declared that "9% of the 200,000 nurse tanks in the U.S. nurse tank fleet may contain leg-weld fatigue cracks", which was determined to be the cause of the 2012 failure.  But the failure mode in that 2012 event looked very different than these two most recent events.

 

SPECIAL NOTE: Also this month, we have a very similar type of tank,  a 500-gallon LPG tank, that failed somehow, killing the worker filling the tank at the time of the failure.

Membership Content