Thermal/Vaporization Relief Sizing Help

[sofarfromhome]

Can someone help me determine what type of sizing I should use in this scenario?

I have a double pipe heat exchanger preheating liquids feeding a stabilizer feed tank. The exchanger is being put into service to pre-heat the liquids for better water/HC separation in the feed tank. The operator does not know the composition of the HC liquids but it is water and CH4+ at ~250psig. The heat medium is hot oil at a max of 600F.

The HC tube side is 3" steel pipe and was hydrotested to 1875psig. The shell side is 6" pipe and was tested to 435psig. The vessel is 20' long.

There is the potential for the HC side to get blocked in and heated by the hot oil. I am recommending a relief valve be installed. However I am uncertain as whether to size it as a thermal or vapor/fire relief. At 600F the water and some of the lighter components could vaporize. How did you suggest I calculate the heat of vaporization for a feed with an unknown composition? Use steam or the lightest component? What set pressure do I use?
Thanks!!

 

[ggordil]

First you need to determine your composition. You are going to have to ask for more information, maybe an MSDS or something. CH4+ is too vauge.

To determine if you are relieving vapor or liquid, you need to determine if the HC side is going to vaporize at your relieving pressure. If the bubble point is below 600 oF at 1875+10% psig (most likely, you need to check company standards), you will have vaporization in the heat exchanger. If it is above 600 oF, then thermal expansion will occur.

 

[sofarfromhome]

Is it bad practice to set the relief pressure low enough to use thermal relief instead of vapor?

 

[MortenA]

thermal relief=expansion if fluid?

When the liquid is boiling then the temperature is semi stable. Only semi if its a mix of varoius compounds (e.g. an oil) - since the boiling point will change as the lght components evaporates.

But for thermal relief the temperature will continue to increase untill sooner or later your liquid starts to boil.

Best regards

Morten

 

[ggordil]

Most likely, your relief pressure is based on the MAWP not the relief valve set pressure. That is not a standard, but I believe it is the approach most companies use.

By the way, the lower the relief pressure the lower the bobble point temperature will be. So, no, you can not lower your set pressure to ensure thermal expansion as opposed to vapor relief. For that you would have to INCREASE your set pressure.

 

[don1980]

It's important to know the hydrocarbon fraction but the composition of that fraction isn't too important. Most hydrocarbons have an Hvap that's about 15% that of water. So, the key question is the H2O-HC split. Ask you client if the stream is primarily oil or water. If it's mostly oil then just assume it's 100% oil for the relief calculation. If the client says it's mostly water, then ask for an estimate of the max HC fraction. The maximum HC fraction will be your worst-case sizing basis for the safety valve.

 

[Chance17]

The HC tube side is 3" pipe.
Why not invoke the 133% ANSI short term overpressure contingency ??
Install a thermal relief valve matching the heat input rate.
The set pressure would be based on exterior flange rating.

 

[MortenA]

Yitbos

You cant avoid the vaporisation - unless 1.33 gives a vapour pressure at 600 deg F thats lower than this pressure!

See my previous post for the argumentation.

BTW: Sofar: Re the Heat Of Vapourisation then API 521 gives you a curve for HOV - as far as i recall 50 BTU/lb is the lowest value quoted - you could use this value (double chack it first tough )

Best regards

Morten