Relief Pipework : Thermal Stress temperatures - and also plastic pipes 1

[jamesbanda]

Basic Question: We have vessels that only have a credible case of fire for overpressure on highly volatile fluids. E.g. with a boiling point at 7 barg of 250 C. So therefore the relief temperature is 250C.

Is it correct for pipework stress anaylsis o use 250C, and 7 barg as the relief conditions for the linw which exceeds standard 150lb pipework and requries 300 lb pipework, and for the vessel or should one assume in a fire this is not likely..

equally i've seen the case with Plastic pipes.
vessel design P, 7 barg 150C, , pipework design 7 barg, T 150C but relief pressure is 7 barg, but 300 C due to organics.. is this acceptable, again only case is fire..clearly two considerations most plastic pipes have metal exterior which wont melt but some fibre class pipes would stand the vessel design P but simply melt and fall off in a fire.. how does one normally consider this ?

i get different answers from enginers..

 

[don1980]

The PSV inlet piping, and the piping within the protected system, must be as mechanically strong as the vessel, or stronger. Otherwise, the piping is a weak link that limits the set pressure and relieving pressure.

One of the basic fundamentals of good process design is that you select a pipe spec that's as strong, or stronger, than the vessel. Never design trap for someone. Engineers will intuitively assume that the vessel is the weak link. In special cases where the piping is the weak link, then put a clear note on the P&IDs to alert everyone to that fact. Generally that'll only happen when you use a surplus vessel (e.g. 300 psig) in a lower pressure system.

So, to answer your question, if you're exceeding the allowable pressure limits of the pipe (refer to ASME B31.3) then you should lower the relieving pressure, or implement HIPS, or upgrade the piping.

Note that you're allowed to exceed design temperature limits during fire exposure. In such cases, it's the owner's responsibility to assess that risk and take appropriate protective measures, where necessary, to ensure that the system is safe.

 

[jamesbanda]

Thanks is it documented that you are allowed to exceed the documented temperature in a standard..

i know you cannot exceed the design pressure..

 

[don1980]

ASME, and other codes too, are mute on this issue of temperature. That's not because it's not a significant risk, it's just too complicated to write prescriptive requirements that are universally practical. The best thing to do is to remain silent and leave this to a case-by-case risk assessment by the user. That's what ASME and other pressure relief codes do.

If you look in ASME Sec VIII UG-125 - UG-140 all you'll find pressure limitation requirements, but you won't find similar requirements for temperature.

 

[moltenmetal]

You run into such problems when you try to address thermal relief problems by means of pressure relief devices. This is a common mistake arising from a tendency on the part of many engineers to think of every vessel in the first instance as if it were a water boiler. The ASME vessel codes do seem to have that bias built into them in my opinion, arising from ASME's history.

If you have non-firesafe pipework connected to a vessel containing a fuel, or being fed a fuel, you should consider firesafe automatic isolation valves in an attempt to keep the inventory of the vessel or of the fuel source system from dumping through the melted/ruptured pipe during a fire. Such valves need fusible links in their actuation packages (plastic tubing in their air supply lines if pneumatically actuated) that will cause them to go to their closed position before the fire gets hot enough to prevent them from closing (i.e. because they jam when hot or because the actuator melts off).

Similarly, if fire is a credible case and your vessel contains only a gas or high-boiling liquid, and it is lined with nonmetallic material or fitted with soft non-firesafe gaskets etc., it's important to consider thermal relief separate from pressure relief- otherwise the gaskets or lined flange faces may become your thermal relief and may let go before the pressure relief valve does, with the leakage going to an uncontrolled or undesired location.

Unless I've missed something, B31.3 does not require piping to be separately relieved, nor does it require piping to be designed to survive fire intact.