Whenever a PHA or LOPA considers a low temperature hazard for carbon steel equipment, the team examines the P&IDs and possibly the mechanical data file to find the mechanical temperature rating.  The upper temperature limit is usually shown on P&Ds with the corresponding design pressure (MAWP).  Less frequently is the Minimum Design Metal Temperature (MDMT) also shown.  Engineers often expect carbon steel equipment to be, if not rated, at least suitable for service temperatures to  -20 F.  This is an often-assumed point below which many grades of carbon steel  can undergo a transition from a ductile to a brittle material.  Once a ASME code-rated pressure vessel becomes brittle, there is always a risk of brittle fracture.  The necessary & sufficient criteria for brittle fracture are:  brittle material, a flaw (e.g., crack), and sufficient stress to cause the flaw to propagate.   Of great concern is that stress component can often be satisfied by sufficient internal process pressure, much lower than MAWP.  Once carbon steel becomes brittle, its loses much of its ability to contain pressure without causing flaw propagation.  We should be cautious in assuming a carbon steel vessel will become brittle only at temperatures below -20F unless it is specifically rated by the mechanical designers with a MDMT.

“Fracture toughness” describes a material’s ability to resist brittle fracture by arresting crack (flaw) propagation.  A material with good fracture toughness is ductile; where cracks in the vessel exist, they will be stable.  A material with poor fracture toughness is brittle.   At ‘warm’ temperatures, carbon steels have good and relatively constant fracture toughness.  There is a temperature below which fracture toughness starts to decrease rapidly with decreasing temperature.  The material is not ductile, but not entirely brittle.  After this transition range and at very cold temperatures, fracture toughness becomes constant again, but the steel is now completely brittle.  A proper MDMT rating of -20 F for most carbon steels means that it will not enter this transition range until below -20 F.  The material remains fully ductile until -20 F or lower temperatures are reached.

Fracture toughness is measured using impact testing, such as Charpy Notch Test, which determines the amount of energy a specimen can absorb without propagating a flaw.  Current versions of the ASME code require impact testing in order to provide a MDMT rating to a pressure vessel.  There are a few testing exemption provisions based on thorough assessment of the steel grade and the range of expected service conditions.  Therefore, MDMT rating on the P&IDs for a relatively new carbon steel vessel can be ‘trusted’ for PHA / LOPA purposes.  However, the pre-1988 ASME code did not require any impact testing, and designers could assume the toughness of carbon steel down to -20 F.  Some grades of carbon steels in use pre-1988 do not have good fracture toughness down to -20 F.  I’m aware of operating companies who commissioned API RP 579 Fit-for-Service analysis of older equipment learn that the material has a Minimum Allowable Temperature (MAT) rating well above -20 F.   PV Newsletter. Vol 4, Issue 1. (June 20, 2011) : 1-9., Alejandro Vega, P.E. writes that some pre-1967 grades of carbon steel used in welded pressure vessel construction can have MDMT of approximately 100 F, which would not be suitable for service at/near ambient temperatures.  Although it was not a pressure vessel, remember the steel in the Titanic was analyzed and determined to be brittle at +32 F !  API Fit-for-Service calculations include assessment of the Critical Exposure Temperature (CET ), which is generally the coldest allowable metal temperature at the maximum allowable pressure.  CET and MAT are closely related because both can be describe as a loading vs. temperature envelope.  Developing these permissible operating limits should consider, in general:

  • Coldest one-day average ambient temperature
  • Lowest metal temperature under normal operating conditions
  • Lowest metal temperature associated with startup / shutdown, upset conditions, etc.
  • Potential for auto refrigeration due to depressurization

Therefore, during PHA / LOPA be cautious in making any assumptions about minimum temperature ratings for carbon steel equipment.  Consult the mechanical data files, ask a mechanical integrity specialist, and consider applying API RP 579 Fit-for-Service criteria before finalizing a determination of mechanical limits.  A quantitative risk assessment can be helpful in examining brittle fracture hazard scenarios, and determining suitability of continued operation of aging equipment based on risk.  Instrumentation & Controls can be used to reduce the risk using a definition of the safe operating envelope in terms of pressure vs temperature.  However, first we must have a good technical justification — not assumptions — regarding safe operating limits of low temperature and pressure.

For more information about an assessment of risk of brittle fracture hazards in carbon steel equipment, contact Kenexis.