Vessel Overpressure Radius (Facility Siting)?

[TiCl4]

I have been tasked with performing a facilities siting analysis of an existing facility that has a variety of pressure vessels on site. This task involves calculating the overpressure radius of vapor-filled vessels rupturing. Most vessels are 50 psig MAWP and under. I have no direct experience in this and my problem is two-fold:

First, what blast wave pressure is generally considered dangerous? I've found both CDC and FAA recommendations (Calculation of Safety Clear Zones for Experimental Permits) around 1 psi as the minimum standard for possible injuries.

Secondly, what is the industry standard on this analysis? Back when I was interning for an EPC I remember seeing an aerial plant layout with various overlapping circles representing blast radius of concern for various process equipment. This data was used for facilities siting for a new plant construction.

Is there a company that provides this kind of analysis? Or, perhaps, is there a reference for calculations of such blast radii?

 

[pierreick]

Hi,
Let you get a copy Of: explosion hazard in the process industries by Rolf K Eckhoff (ELSEVIER) .
Chapter 9 is about "destructive and harmful effects by pressure, solid fragments and heat from accidental explosion"

Good luck
Pierre

 

[shvet]

It is not obvious what you are talking about - evaluation of a consequence of what:
- an explosion of a vapor cloud
- a vessel/piping rupture without explosion
- explosive/flammable vapor leakage and spreading
- personal risk of death/injuries for personnel or residental area

For vessel/piping rupture you can try to use para. III-2 ASME PCC-2-2015.
For explosive vapor spreading you should use a dedicated standard, e.g. ATEX.

Equipment layout and zoning depends on many factors. Which one do you want to evaluate?

 

[Latexman]

In my experience, BakerRisk is one of the best in this area.

Good Luck,
Latexman

 

[TiCl4]

Shvet,

The PHA recommendation is to evaluate the location of the administrative/QA areas to determine if they are protected from overpressure blast radius/fragments in the case of pressure vessel rupture due to internal overpressure.

Thanks others for the resources - I’ll check them out.

 

[MortenA]

@OP: "irst, what blast wave pressure is generally considered dangerous? I've found both CDC and FAA recommendations (Calculation of Safety Clear Zones for Experimental Permits) around 1 psi as the minimum standard for possible injuries." Its rarely the pressure wave that kills people but debris shrapnel etc.

--- Best regards, Morten Andersen

 

[LittleInch]

Try searching for
"FERA risk analysis consultants"

Then see what crops up.

What country are you in?
Most safety regulators will have guidance.

Is this just overpressure? I.e. not fire or are the contents flammable?

At 50 psi are the vessels able to rupture or do they leak only? You might not have a problem if they can't break.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[shvet]

A breaf calculation from ASME PCC-2-2015:
Input - rupture of vessel 1m3 containing air at 3.5barg
Output:
- eardrum rupture distance is 0.66m
- glass window break distance is 1.13m
- min safe fragments throw distance is 50m

 

[TiCl4]

Thanks again all. MortenA, the FAA paper safe zone calculation had some direction on distance thrown of harmful debris.

LI, the location is in the US. We have many pressure vessels on site, with some smaller air receiver tanks at 150 psig MAWP to large reactors at 50 psig. The facilities siting recommendation is to evaluate the rupture case of each, determine the overlapping threat zones created by each, and determine how the main populated areas in the plant fall in relation to these threat zones. The case is simple overpressure. One of the issues is that I don't know what the actual pressure would be upon vessel rupture due to pressure vessel design margins. There are other issues as well - is pressure vessel failure due to overpressure typically a main body or head weld failure, or will it start popping off nozzles first (or, like most things, is it a "it depends" answer?)?

LI, the released contents could either be flammable or not. One of the reasons I want to analyze the consequence of failure of the vessel only and not the following release scenario two-fold: First, this recommendation is simply for performing the facilities siting analysis for failure of the vessel itself, regardless of either the event probability or initiating event (the vessel could overpressure due to high inert gas feed pressure). Second, the presence of a flammable vapor cloud can occur independently of vessel rupture and can be addressed in a separate siting analysis.

Lastly, anything that could cause vessel rupture has been evaluated via LOPA and is being mitigated to 1E-5. This analysis for facilities siting is above and beyond that mitigation concern.

Svet,

ASME PCC-2-15 is Repair of Pressure Equipment and Piping. This has vessel rupture calculations in it?

In summary, all, I am going to have an outside firm do this analysis. There is too much I do not know about failure modes and mechanisms of pressure vessels to be able to confidently perform this siting analysis myself, even after some of the provided written resources.

 

[LittleInch]

I think your worst case issue is the dished end weld failing. Hence why they tend to locate long horizontal vessels with respect to that and not point them at anything which could be at risk of failure, or jet fire. Then you also need to consider the end head going at speed in one direction whilst the vessel goes in the other.

Vertical tanks can launch like a rocket.

The worst case would be some sort of BLEVE, which is why I asked about contents. then you get a shock wave from the gas release then something like 10 times that from the explosion of the flammable liquid which has turned into gas. Propane tanks at ambient temperature are like that.

You would have to say that overpressure of a vessel would have to be 150 or 200% of design pressure as it is tested at 150% no?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[TiCl4]

You mean like this?

https://www.youtube.com/watch?v=6qcrwNM74sg&ab_channel=DavidDrummond

ASME VIII only dictates hydrotesting at 1.3x MAWP currently. I believe older vessels were 1.5x due to greater variability in material quality or calculation accuracy in days of yore. However, safety margins in design (you don't design right below the yield strength of the material) mean it will rupture at some higher, but unknown pressure.

FYI our flammables have flame propagation speeds below the sonic velocity. Any ignition of vapor clouds will result in deflagration rather than detonation. I have no idea if that substantially decreases lethality of a vapor cloud explosion.

There are too many unknowns here for doing it myself. I'll get experts involved.

 

[Compositepro]

Fuel/air mixtures can detonate, but propagation velocity will be limited to sonic velocities. In high explosives the detonation velocity may be over 8,000 m/sec, but the shock wave will only propagate at sonic velocity.

For methane/air:

"Using a direct method to initiate detonation in test mixtures, detonation is sustainable in
methane in air mixtures over a range from 5.3 to about 15.5%. The range for the lower
detonation limit is between 4.9 to 5.3% methane in air, and the range for the upper detonation

limit is between 15.5 and 15.7% methane in air. This new range almost encompasses the

entire normal combustion limits of 5 to 16% methane in air as reported by Cashdollar et al.

(2000). This range is larger than that observed by Wolanski et al. (1981), who reported a
range of 7 to 13% for methane in air. The methane used in these experiments is natural gas

containing about 97.5% methane and from 1.5 to 1.7% ethane; however, the presence of

ethane is considered negligible. The detonation velocity varied from 1,512 to 1,863 m/s over

this composition range, and the measured detonation velocities agreed with computed DCJ
values to within 2.5%. The measured pressure behind the peak shock pressure ranged from

1.2 to 1.7 MPa (250 psi), and agreed with computed CJ pressure to within experimental error

associated with this pressure measurement."

https://www.cdc.gov/niosh/mining\/UserFiles/works/pdfs/madea.pdf

 

[shvet]

ASME PCC-2-15 is Repair of Pressure Equipment and Piping. This has vessel rupture calculations in it?

Yes, it as. See appendix III.