Set Pressure for a Liquid Thermal Relief Valve on a Butadiene Pipe 5

[lcms]

Dear Folks,

I have a 1,500 ft non insulated long pipe (sun exposed) which we use on butadiene liquid transfer between two vessels in our plant.

This pipe is a 3" seamless carbon steel ASTM A-106 B schedule #40 (wall thickness=0.216") alleged to be specified as 150# class. However, I have looked up on ANSI B31.1 Code for Pressure Piping that this pipe could bear pressure as high as 1,600 psi at 100F. I have to believe in it since I have myself testified that my line has already achieved pressures above 600 psig without any damage.

My question is: - What pressure shall I have to set on my Thermal Relief Valve? Our existing TRV was adjusted to 10 kgf/cm2 (~ 150 psig) considering our internal pipe specification (class 150#) but we had blocked it after realized that it opens every time we start the transfer (the transfer itself is performed near 15 kgf/cm2 ~ 225 psig)? My best figure by now would be 300 psig, considering some papers regarding piping design I read somewhere along my 20 years of Chemical Engineering practice (unfortunately, I couldn’t remember where).

Could any of you gentlemen help me on this issue?

Thanks in advance.

lcms

 

[StoneCold]

If you have been in this business for twenty years you should know that it is not the pipe that sets the pressure rating, it is the flanges. The pipe can take much higher pressure than the flanges and their bolts. Blocking in the thermal relief is asking for trouble, and on butadiene.??????


Goodluck, you are going to need it.

StoneCold

 

[EGT01]

lcms,

Seek the advice from some one at your facility that is experienced in this area of engineering. From your description, a more thorough review of your system is warranted.

First thing you need to do is determine to what code and/or specifications that your piping system has been constructed. ASME/ANSI B31.1 refers to Power Piping and I question whether that is appropriate as opposed to B31.3 which is Process Piping. The piping code will define the overpressure protection requirements but you should verfiy that any company piping specifications don't introduce any additional limitations.

Also, if your normal transfer pressure is about 225 psig you probably need to consider more than thermal expansion as an overpressure concern and make sure that the source of the butadiene does not present a concern (for example pump deadhead).

 

[Ashereng]

Flange rating of 150#, A105 CS at 100°F is good to 285 psi.

You should be okay since your line pressure is only 225 psi. If you set your thermal relief to 250 psi, would that work?

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[lcms]

Dear Sirs,

Thank you gentlemen for your valueable help.

Nearly to what Mr. Ashereng recomended, I will set my TRV on 230 psi, that is high enough to avoid leakeage during transfer and low enough to prevent damage on my line.

Regarding Mr. Stonecold and Mr.EGT01 recomendations, I do appretiate your concern, and that is exactly why I am working on this delicate issue. Unfortunately there is nobody else in my site experienced in this engineering field (piping) and I will have to take my chances to accomplish this task. As you seen, things are not going very weel so far (as Stonecold mentioned, I was very surprised myself when discovered that the TRV was blocked).

Sorry if I am not using wisely the forum, and I do appretiate if one of you could explain how to do it better (Did I make anything wrong ? Please clarify it better and I promise I will try not do this again)

Regards,

lcms.

 

[Latexman]

If the transfer occurs at 225 psi, a metal seated PSV set at 230 psi will start leaking at about 200 psi (90% of set pressure). If it gets opened fully by a pressure transient, like a valve closed too quickly, it will not reseat in liquid service until about 155 psi (70% of set pressure).

You need a higher set pressure still.

Check the MAWP of the rest of the components in your system first. Do you have 150 psi valves? The MAWP of the system is determined by the weakest link. I suspect you will find the flanges and bolts are the weakest link, but until you go thru the exercise, you do not know.

If the flanges and bolts do set the MAWP at 285 psi, I'd set the TRV at 285 psi. Transients may still be a problem, but that issue can be stopped with a block valve or stopping the pump.



Good luck,
Latexman

 

[Ashereng]

Latexman brings up a good point. Any system may be derated, hence, the caution to determine/ascertain the MAWP.

What I quoted you is from standards. What your system will bear may be different.

Latexman is also correct in that your pressure relief valve will start to lift at 90%. I suggested 250 psi. At 90%, it is 225 psi, which is what you gave. If this is too tight, you can go up more, to the MAWP.

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[lcms]

Dear Friends,

You are both right, but I had already took this on account when I chose the 230 psi set pressure. As a matter of fact, I have a weakest point in my line, not mentioned before, that is a DS-150 massflowmeter rated for 230 psig (and that was my main source for assigning 230 psig as the specified set pressure).

Another very important point I had ommited before (sorry, folks) was a rupture disk included upstream the TRV which is commonly used to avoid problems related to butadiene polymerization (popcorn). This device, as far as I know, prevent opening before set pressure as described by Latexman.

To sum up, I will remain on 230 psig, counting on the rupture disk to prevent opening before TRV set pressure and keep in mind my massflowmeter integrity (weakest pint).

Thank you very much for your kindly support on my thread, and I truly hope I could help you in a near future as much as you help me now.

Sincerely,

lcms.


can see, avoid the