Release scenario query

[chemks2012]

Hello all,
I need your help.
We have a reactor with an internal coil [coil dia 2”, 40 schedule, coil design pressure is 15barg and vessel design pressure is 6barg] carrying saturated steam at 20barg [temperature 214.93 degC].
In fact 2” coil is supplied steam with a 1” line The steam inlet pipe has restricted orifice of size 15mm.
Reactor has a relief valve set pressure of 3barg [relieving pressure 3.3barg and relieving temperature of 146.38degC]

I want to calculate vapour release rate when reactor coil fails completely.

1) As per my understanding if I just calculate steam flow through 15mm orifice with differential pressure of 16.7barg [i.e. 20barg – 3.3barg], it covers the case of complete rupture of a coil. Is my understanding correct?
2) Also, for orifice calculation, I would consider temperature of 214.93degC and NOT the relieving temperature of 146.38degC. Please confirm/comment.

Would appreciate your help.
Regards,
KS

 

[chemks2012]

One more query please

3) Just want to understand what would be the relieving temperature in real world.

If the coil fails completely releasing steam at 214.93 degC but because the valve relieving pressure is 3.3barg it will have temperature of 146.38 deg C and NOT 214.93 degC i.e. steam will expand and temperature will be reduced. Is this correct?

Do you think for relief calculation purpose, I should consider relieving temperature as 214.93 deg C and not 146.38 degC?

 

[MortenA]

I think your reasoning is OK. You should be able to find the steam temperature using a steam table or a HX diagram following the isoenthalphy lines from 20 barg to 3 barg

 

[met11]

Is the reactor operating liquid full? For a full tube rupture, the instantaneous rate when a 2" coil ruptures (I'm assuming double-guillotine) will be a lot higher than the steady-state rate through the orifice. Can the reactor handle that initial "shock" of steam, as the volume of steam left in the coil is flashed into the reactor? If it's liquid-filled, the only way to do that may be by considering a rupture disc, or ruling out the guillotine failure scenario as it is unlikely to get a full guillotine. A pinhole leak may be more credible. If it can withstand that initial blast of steam (i.e. it's gas filled, or partially filled with liquid), then you could use the steam flow through the 15mm orifice to size the relief valve.

It's an isenthalpic flash across the RO. I didn't check the steam tables, but you will get superheated steam at a lower temperature. For relief, use the lower temperature because that's what it will actually be. Use the updated values as well for Z, Cp/Cv etc. with the higher quality stream.

Also, your initial conditions for relief may be different from just steam, depending on what's in the reactor. There would be a certain degree of mixing I assume. Composition would change over the relief event.

 

[chemks2012]

Hi MortenA

Thanks for your reply.
Those temperatures I stated are from steam tables.

With regards to query2, not sure what temperature should I take while calculating flow through an orifice. Is it equivalent to 20barg [i.e. 214.93degC] or 3barg [i.e. 146.38degC]?

With regards to query3, I believe I should base all my relief calcs on relieving temperature of 146.38degC though steam temperature is higher.?

Hi met11

Thanks for your reply. In fact I have two scenarios 1) coil failure above liquid level which would release only steam and relief valves will be sized for this scenario 2) coil failure below liquid which would result in to two phase release and bursting disc will be sized for this scenario.

My above query is for the coil failure above liquid level.