Jump to content

Featured Articles

Check out the latest featured articles.

File Library

Check out the latest downloads available in the File Library.

New Article

Product Viscosity vs. Shear

Featured File

Vertical Tank Selection

New Blog Entry

Low Flow in Pipes- posted in Ankur's blog

Safety Valve Capacity

This topic has been archived. This means that you cannot reply to this topic.
7 replies to this topic
Share this topic:
| More

#1 Guest_Guest_afdmello_*

  • guestGuests

Posted 17 July 2005 - 03:55 AM

We have safety valves on a initiator feed vessel blanketed with N2,they are designed for fire,control valve failure and overfilling:

Initiators are peroxides solutions with heptane and they are susceptible to decomposition.The valve flow rate is 18000kg for fire;780kg for N2 failure and much less for overfilling.The set pressure for the valve is 5bar.

This means that a Pressure Relief Valve can also act as thermal relief valve which in the case here is overfilling.

what does the fire flow rate mean? does it mean that the valve when full open can relieve 18000kg/hr in the event of a fire outside the vessel and heat from the fire increasing the pressure in the vessel.


#2 djack77494


    Gold Member

  • ChE Plus Subscriber
  • 1,282 posts

Posted 22 July 2005 - 04:50 PM

I read your description of the question to mean that three relief scenarios were considered in sizing the Initiator Feed Vessel's PSV. Those scenarios were:
1. External fire.
2. Failure (open) of N2 regulator.
3. Overfilling.
Each of these scenarios may result in overpressuring the tank, and therefore must be protected against through the use of one or more PSVs.

I interpret the 18000 kg/hr to mean that in the event of an external fire, enough heat will enter the tank to result in the vaporization of 18000 kg/hr of product at the relieving conditions. The relieving conditions are the setpoint (5 barg) + allowable overpressure (which depends on whether one or multiple PSVs are used). The temperature would be that where the fluid's vapor pressure was equal to the relieving pressure.

The fully open PSV would be sized to relieve a MINIMUM of 18000 kg/hr at these conditions. The actual PSV purchased would be capable of relieving more than 18000. In performing the sizing calculations, you would calculate both the relieving rate (kg/hr) and relieving pressure (barg) for each scenario. Usually, it is obvious which condition governs the PSV size. However, it is possible due to different overpressure allowances, that it is not completely obvious. In any ambiguous situations, I'd supply the potential vendors with all conditions.


#3 Art Montemayor

Art Montemayor

    Gold Member

  • Admin
  • 5,656 posts

Posted 22 July 2005 - 10:27 PM


First, allow me to reply to your direct question regarding the fire case. The fire flow rate is interpreted as per the description furnished in API RP 520 and 521: this is a scenario where an external "pool" fire occurs in the area of the vessel in question while the vessel contains a liquid charge. The external heat from the flames is considered to heat the contained liquid to the saturated state and then starts to vaporize it, causing an over-pressure in the vessel. As Doug has ably explained, the vapor produced by this Fire Case is at the temperature corresponding to the produced vapor pressure - which may be the relieving pressure . Please note that I stress the fact that the vessel contains liquid at the moment it is being relieved. If the vessel is devoid of liquids (only air or gas filled), the scenario does not fit. In that case, the vessel would probably fail due to structural failure and collapse due to the heat from the fire.

What would concern me is the phrase you employed: "Initiators are peroxides solutions with heptane and they are susceptible to decomposition". You don't mention if there is a credible "Decomp" scenario for the vessel. In other words, can the liquid charge decompose (with presumed violent gas generation) on its own, during storage, or upon being heated by an external fire? If so, does the stated vapor relieving capacity (18,000 kg/hr) represent the decomposition products? Also, is there a possibility of the relieving products ever being in a 2-phase mixture? If so, then the capacity of the relieving PSV should be designed with 2-phase flow in mind or else it will be undersized.

Your are probably on top of the problem and have considered this, but I just want to mention it in order to have peace of mind --- especially since you mention that over-filling of the vessel is a credible possibility.

Art Montemayor

#4 Guest_Guest_afdmello_*_*

  • guestGuests

Posted 23 July 2005 - 12:38 AM

Dear Art,

I hope you remember me.I am an plant operator in a polymer plant.I read an article of CSB investigation at http://www.csb.gov/c...msCaseStudy.pdf
and was thinking of our facility.

The vessel is having N2 blanketing and the inlet N2 and vent N2 PCV's are fail close.
The vessel temperature is maintained at 40 deg C with a small double pipe heater(2m length) on the inlet to the tank with steam at 150deg in the annulus and cooling water in the in pipe.This 40 deg water is sent to the jacket and then returns to the MW distribution.The steam valve is also fail close

The PSV information I found out in the data sheet in the archives.
The iniatiator SADT(self accelerated decomposition temp) is 40°C for 10kg at 98%conc. we are having a dilute solution of around 30% so the SADT is more than 40°C and I dont know how much it is.

So was just curious whether the fire case is also including a possible decomp case.
I am not competent to calculate whether a decomp is credible or not.

Just suppose the steam TCV starts leak when full close. The bypass valve of TCV starts leaking.No alarms are configured for the control loop or there is a TI to measure the tank content temp.And yes of course the liquid can decompose if heated by an external fire.

Just food for thought Art. I will never forget your detailed replies to my propane tank query a year back.


#5 Art Montemayor

Art Montemayor

    Gold Member

  • Admin
  • 5,656 posts

Posted 26 July 2005 - 10:31 PM


It’s always nice to hear from you because of your obvious interest and dedication to doing a good and safe job. I wish more engineers would embark on their career with the same attitude.

Now that you’ve explained a little more of where your concern is coming from, I’m getting more concerned. There are parts of your reply I still don’t quite understand (and these could easily be resolved with an engineering sketch attached to the post, but I’ll only use that if it becomes necessary to clear up a lot of details) but the parts I do understand are:

1) you’re storing a fluid that is subject to decomposition;
2) You’re aware that the SADT (self accelerated decomposition temp) is 40 °C for 10kg at 98% concentration, but you are using a dilute solution of around 30%. However, you don’t know the SADT for this condition – although it probably is much higher.
3) You’re using a steam heater on the water used to keep the vessel at 40 oC; the steam is at 150 oC;
4) There is a by-pass valve around the steam TCV;
5) There are no alarms on the steam-heated water loop;
6) There is no TI on the vessel (& presumably no temperature recorder);
7) The vessel is subject to a pool fire taking place around its exterior.

I really don’t like what I’ve written and I would hope that you would clear me up if I’ve assumed and written wrong information. That would remove a lot of the concerns I have.

The Fire Case (as described in API RP 520 & 521 (which would be the PSV design basis) does NOT include a possible decomp case. A decomp case is a totally different and separate issue that must be specifically handled according to the specifics related to the application. I would never try to safe guard a vessel against a possible decomposition with a PSV. This is a very serious and dangerous possibility that must be defined according to the potential over pressurization rate. A decomposition normally generates over pressure much faster than a runaway reaction – and we should all be wary and stay away from runaway reactions because of the danger. I would classify a decomposition reaction as an explosion. And I can’t depend on a PSV to save me from an explosion.

The only good part that makes me feel better is your interest and consultation through this forum. It shows you are concerned and trying to do something. I’m elated that you are reading the CSB accident reports – an action that all engineers should be doing on a scheduled basis.

Another area of concern I have is that you don’t indicate that there is a documented file of calculations and explanations on the vessel, the PSV, the SADT at 30% concentration, and all instrumentation related to this part of your process. You should have this information available in detail in order for you and others to fully understand how it is expected that the personnel and the equipment are fully protected from all possible and credible hazardous situations.

Please keep us informed as to what this situation is and confirm if my concerns are valid. There must be ways or means to avoid such situations as I’ve described – regardless of where you’re located and for whom you work.

I wish you all the luck in the world.

#6 Guest_Guest_afdmello_*_*

  • guestGuests

Posted 27 July 2005 - 09:40 AM


All what you presumed are perfectly right.The scene is just as you have written and is going on for the last 8 yrs.I dont know what calculations have been done.

As I told you because of the dilution the SADT may be considerably high.
The CSB investigation report just caused some interest in me.I will tell my superiors about my concerns and maybe they will incorporate some alarms or interlocks.Thank you so much for the encouraging reply -as always!!!

your point no 7 that the vessel is subject to a pool fire on the exterior.

I dont know what is a pool fire?

I saw the PSV data on the data sheet but I dont have access to other data.
Where can I find the data related to SADT of a 30% solution?


#7 Art Montemayor

Art Montemayor

    Gold Member

  • Admin
  • 5,656 posts

Posted 27 July 2005 - 05:21 PM


Thanks for the timely reply. Although I feel concerned you (or anyone else) could be saddled with such a potential hazard, I’m relieved that you seem to be pro-active about resolving it and remaining on top of the situation until it is resolved with regards to safety.

A Pool Fire is defined as a fire originating from the ignition of accumulated liquid fuels that form a pool in, or near, the immediate area of the subject vessel. The impact of a pool fire is that it will burn and impart not only direct flame impingement on the vessel, but also contribute tremendous radiation effects – doing all this until it literally burns itself out (which could easily be well after the vessel fails due to over pressure or structural damage. One way to fight a potential pool fire is to build and construct efficient and adequate drainage for the combustible liquid to travel elsewhere and away from the immediate area.

For data regarding the characteristics (especially decomposition) of the Peroxide Solution, I would get in immediate contact with the manufacturer of the chemical. They should be able to tell you all the safety information you require in handling their chemical in a safe and predictable manner. Ask them direct and specific questions with regards to your application of their product. They surely can tell you the SADT at any % concentration.

I would not proceed in making any improvements on this installation without having at least 2 or 3 Hazops on the subject: One preliminary at the onset of the project; another immediately after all engineering is finished and the proposed improvements are approved for installation; and a final one prior to the start up of the modification.

Good Luck and keep up the good engineering thinking and action.

#8 Rammohan


    Brand New Member

  • Members
  • 3 posts

Posted 19 January 2006 - 07:25 PM

Dear Art,
My name is Rammohan and work as an Instrument Engineer for HHI in Ulsan, Korea on a FPSO projcet.
I would like to receive your excel workbook.
My E-MAIL is rammohan2000@gmail.com
With warm regards,

Similar Topics