Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations SDETERS on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Design Pressure and Design Temperature 1

Status
Not open for further replies.

Leclerc

Chemical
Aug 22, 2002
73
[I have posted this in the API Codes Forum a few days ago, but no bites, yet]

As I understand, within the vessel Codes there is an allowable accumulation ( say 10%) which may be applied to design pressure/ MAWP. Also, where relief valve set pressure is set at this design pressure/ MAWP and there is overpressure of, say, 110% of set pressure, the valve relieving pressure is at the vessel allowable accumulation pressure.

Now to temperature:
For a system where vessel contents have a temperature coincident with pressure at the bubble point;

What should the vessel design temperature be? Should it be the bubble point at the design pressure/ MAWP, or should it be the bubble point at relieving pressure ?

 
Replies continue below

Recommended for you

Leclerc, I recommend that the design temperature be the LARGER of:

1) Operating temperature + 50F
2) Relieving temperature, but excluding fire case.

 
CJKruger, what you appear to imply, is that, within the codes, there is no such thing as "allowable over temperature".

thank you.
 
this question comes up all the time. The worse case scenerio is where the fire case has a vessel heating up to 1000 degrees F at blocked flow. The intial fressure is 50 psig and at fire case the pressure rises by (1460 R)/(520 R) or to about 175 psig and the relief is set at 300 psig. At this point the vessel will fail before it ever gets to 300 psig due to lowered yield strenght.

then add in a mixture, like gasoline, in a tank. As temp goes up, the boiling liquid raises pressure, the relief goes off, temp rises and rises and yet all the liquid doesn't boil away as the metal yield strenght drops till finally a failure. You could put 100 4J6 valves on a 100 gallon tank and you will get a failure eventually with gasoline in the tank.

The good news is that the remaining energy is vey small and the radius of expose is equally small. It's all risk management.
 

Dcasto is exact and correct. This question comes up all the time – and I detect an inability to recognize the most obvious and common-sense analysis that Dcasto has painted for us. Common sense tells us:

Mom always let us boil our own Easter Eggs on a gas range where the 1,000+ oF flames directly impinged on the pot holding the water where our eggs were being boiled at 212 oF. Why didn’t the aluminum pot simply melt (or collapse) when subjected to 1,000+ oF??? The answer is obvious: the water heat sink absorbed all the heat put to it through the pot’s walls – keeping the pot’s wall relatively cool compared to the flame temperature (and closer to the cooler 212 oF). As long as an inventory of water is maintained in the pot, the pot’s metal and mechanical integrity is protected by the heat sink. The moment all the water is boiled away …..goodbye pot.

This example of simple, direct-fired, heat transfer is illustrated every day in industrial boilers – especially the fire-tube (“scotch-marine”) variety. A flame is created and maintained 24-7 in a closed environment where the extremely hot flame and resultant gases are transferring heat to a pressure vessel filled with water and generating saturated steam as a result. There are very important liquid level controls maintained on such a steam generator for the simple reason spelled out above in the previous paragraph – should the liquid water disappear, the result would be a boiler rupture with a lot of pressure being released instantaneously – a boiler explosion. And this was the precise reason the US government got involved at the turn of the last century and mandated that something had to be done to prevent people getting killed in boiler explosions during the steam engine and locomotives era. The result was the ASME code that we all love so dearly.

Today, the rules are the same: the same exact thing happens in an industrial pressure vessel that is filled with liquid and is exposed to a pool fire. The vessel’s integrity is secure as long as the vessel has liquid inventory to serve as a heat sink. And it is this liquid inventory converted to saturated vapor that serves as the medium that furnishes the “over-pressure” that is relieved through a PSV. As long as the excess vapor is being relieved the vessel is protected – and so is everything around it. It is when the liquid is depleted that a definite hazard is created: as Dcasto explains, the collapse of the vessel is inevitable if the flames continues. It took the API a long time to accept this simple and straight-forward reality in their standards 520 and 521 - but at least they finally did. It hasn’t been until recently that people have now realistically accepted the fact that mere PSVs do not totally protect a liquid-filled vessel – and much less, a gas-filled vessel. That has now brought on the API procedures for depressurization – which is the smartest and most positive approach for personnel protection – especially combined with water sprays and heat-resistant insulation.

The point I want to add and make sure is taken into consideration in this thread is that an engineer – particularly a chemical engineer – should realize that you can’t “design” a pressure vessel pressure for the worst case relief scenario. For that purpose, you have safety devices, instruments, alarms, and procedures to implement. It is, however, vital that one know what is physically happening to the vessel during – for example – the pool fire case. Every engineer should know by now that the worse heat transfer film coefficient known is a gas coefficient. That is why gases – such as air and Freons – are used as insulators. In fact, Artic insulation works well not because of the solid material used, but because of the insulating air it contains. The point here is that a 100% gas filled pressure vessel exposed to a pool fire is destined to total failure – and simultaneous violent rupture – unless it is protected by a water spray or insulation. This is due to the insulating properties of the internal gas film that effectively retards efficient heat transfer from the external flame to the internal gas. There is NO HEAT SINK, and the vessel walls have to take the brunt of the heat accumulation because they can’t transfer it anywhere else. Depressurization is a mitigation but it will not ultimately save the vessel; however, it will protect the environment and personnel from the mechanical failure.

All of this is explained or inferred in API 520 and API 521 and these documents should be the basis for good, safe design when dealing with pressure vessel relief.
 
Here's a couple of more points:

1. The set pressure of the relief devices also depends on how many used but that gets into design.

2. Fire-sizing, though it often is the worst case (largest valve) and the easiest to calculate, it does not always yield the largest relief valve (diameter) or rupture disc area. It is best to consider all the other possibilities and make the selection.

3. Finding the temperature maximum should not be an issue. If you are heating with steam, the maximum system temperature is the saturation temperature. Even if a process operates at only 150 F with 35 psig steam (281 F),the maximum is not 150 F (the control SP) but 281 F because that's the assymptotic temperature the vessel will eventually see if the steam valve fails open.

4. The bigest valve isn't necessarily the best choice. Nor is installing several small-large valves together. The outlets for the small valves should be isolated from the larger valves to prevent backpressure.

Dirk Willard
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor