• We have a heavy coker gas oil hydrotreating unit and its reactor MOC is 2.25 Cr, 1 Mo, 0.25 V. As per the licensor manual, the MPT (minimum pressurisation temperature) is 93C. However, as per the fabricator, it is 38C. The query is which MPT value will be governing and be used during pressurisation of the unit at startup.



  • Richard Rhead, Refining & Aromatics Consulting, rick.rhead@gmail.com

    The licensor wouldn't typically have access to the actual mill specifications for the steel composition used in the reactor. They would assume that a typical vintage steel (based on composition and manufacturing techniques) to have a typical MPT. Whereas the fabricator would know this and can determine a more accurate MPT.

    The licensor's MPT is more conservative. One thing to keep in mind that this should be truly for the reactor circuit. Meaning not just the reactor but other 2.25 Cr/1 Mo metallurgy, your combined feed exchangers? My experience is that the CFE's don't have this information readily available so I would go with the more conservative 93 C.



  • Jake Gotham, InSite Technical Services, jake.gotham@insitetechnical.com

    It is well understood that the MPT is the temperature below which the steel is assumed to be brittle.  What is less well understood is that this temperature can change with time, depending on the metallurgy and service conditions. Temper embrittlement is linked to impurities in the steel accumulating at grain boundaries over time. With heavy wall reactors formed from old steels, temper embrittlement can increase the MPT by up to 100°C, but modern steelmaking limits the impurities and hence the affect on MPT is usually quite small for a modern reactor.

    Some fabricators may anticipate in-service degradation in their calculation of MDMT (minimum design metal temperature), but usually they are quoting a figure for the reactor on the day it leaves the fabrication shop.  The licensor’s figure includes allowances for temper embrittlement and other mechanisms that might affect toughness during operation over the design life of the reactor. I suspect this is the reason for the difference between the two figures.

    You shouldn’t have much difficulty reaching an MPT of 93°C, particularly if your design allows you to establish liquid circulation at low pressure.  This carries the heat into the reactor much more effectively than gas circulation.



  • Eric Lin, Becht, elin@becht.com

    The fabricator's MPT number is the most accurate as they know best the impurities and material makeup of your hydrotreating reactor and should already have some conservatism built into the MPT calculation (this temperature should be considered the bare minimum to respect). There would be nothing wrong with following the higher 93C number as it is just more conservative but may present a challenge in colder regions of the world (like Russia or Canada).

    You can find out more about how MPT is calculated by the fabricators in my most recent PTQ submission here: https://www.digitalrefining.com/article/1002634/protecting-your-hydroprocessing-reactor



  • Nelmo Fernandes, PETROBRAS SA, nelmoff@petrobras.com.br

    We pressurize to no more them 25% of the design pressure until the wall temperature achieve 60°C. Being conservative, and to protect your reactor, I suggest you use the higher temperature.



  • Responsive image Petrochemical Market Trends Report
  • Responsive image Shut-off valves
  • Responsive image Tri block, double block and bleed valve from ZWICK
  • Responsive image Nasty stuff
  • Responsive image OpX® Digital
  • Responsive image FluegasExact 2700 effective combustion analysis
  • Responsive image Haldor Topsoe is now Topsoe
  • Responsive image Proven Vacuum & Jet Pump Technologies
  • Responsive image Monthly technical bulletin
  • Responsive image Flash point testing