19 replies to this topic

[shin29]

Hello all !
I want to size a safety valve for Air vessel ( design pressure 10 Barg) for fire service.
I am not clear how to calculate required capacity ( kg/hr)
and how to decide for what services safety valve should be designed for

i am expacting smbody's kind reply

shin.

[fallah]

shin,

Try to use API 521 and API 520 Part I.

Fallah

[TS1979]

PSV for fire case is for vessel in liquid service. If your case is for compressed air not liquified air, fire case is not applicable for your case.

[prabhakaran.nitt]

Calculation is based on fixed expansion of the vapor or gas due to increasing temperature from the fire exposure. To calculate worst case - use Higher pressure and lower operating temperature.

[shin29]

Thanks to all for your kind reply , i shall do some homework . than get back to you
Regards,
sahil

[fallah]

PSV for fire case is for vessel in liquid service. If your case is for compressed air not liquified air, fire case is not applicable for your case.

Hi,

Fire case is applicable for all pressure vessels can be exposed to fire regardless of their containing fluid state.

Fallah

[Lowflo]

For a compressed air vessel, sizing for fire exposure is a waste of time and money. Installing a PSV is necessary, but the user get to pick the sizing basis, and there's no code that requires you to do something senseless like sizing an air vessel PSV for fire exposure.

[shin29]

Greetings to all
sir, i have gone through API 520 part-1 and API 521, but while sizing , in one of the equation i need to mention the wall temperature which i am unable to find out .
kindly guide me how to calculate the vessel wall temperature in case of fire , while sizing PSV for fire cause for that vessel/

[fallah]

sir, i have gone through API 520 part-1 and API 521, but while sizing , in one of the equation i need to mention the wall temperature which i am unable to find out .
kindly guide me how to calculate the vessel wall temperature in case of fire , while sizing PSV for fire cause for that vessel/

shin29,

API 521 recommends maximum vessel wall temperature, T_w, for carbon steel materials equal to 593 C (1100 F). Then you can use this value in your calculations.

Fallah

[Robert Montoya]

My friend the fire case does NOT APPLY to this vessel, because it is assumed that the vessel is located in an area where a diameter around 20 meters there any source capable of generating fire around the vessel. Generally this type of vessels as well as the compressor, are located in a service area.

To protect the PSV vessel should be designed for blocked discharge.

[shin29]

Thank you sir for your kind reply,

I must mention that the vessel is located near Gas Turbine generator. How should i calculate relieving Temperature?

Is it = ( Relieving Pressure/Normal operating Pressure)* Operating Temperature ?

regards

[Robert Montoya]

You are right; the way for calculate the relief temperature is:

T1 = (P1/Pn) x Tn

T1 = relief temperature, °K;
P1 = pressure psia = Ps x 1.21; Ps is the set pressure.

Pn and Tn are the operating temperature and pressure; °K, psia.

[Olidin]

I am new to this topic but it's interesting so I'll chime in.

Section 5.15.1.2 in API 521 discusses external fire effect on unwetted vessel surface. For a vessel full of vapor, all the area is unwetted, and per section 5.15.1.2.2, because vapor conduct heat poorly, "Heat input ... can, in time, besufficient to heat the vessel wall to a temperature high enough to rupture the vessel"

I guess this is why we often don't size PSV for vessel contains only vapor since the rupture may have already fail the vessel before PSV relieves.

However, I did not find where API mentions that a PSV is not required against fire for vapor full vessels.

Section 5.15.2.2.2 provides detailed equations for calculating the relieve load.

However, the assumptions for the equations include: "... the vessel is uninsulated and has no mass, that the vessel wall temperature do not reach rupture-stress temperature, and that there is no change in fluid temperature. These assumptions shouldbe reviewed to ensure that they are appropriate for any particular situation..."

Though this give me another question: How do we know if the vessel will rupture because of vapor expansion first (then PSV is a mean to protect against fire) or because temp reaches beyond rupture temperature (which PSV protection is not appropriate).

Edited by Olidin, 27 September 2012 - 05:41 PM.

[fallah]

Olidin,

In most fire cases of vessels containing vapor, because using a PSV alone cannot be adequate for vessel protection it will rupture first before PSV opening and to avoid this situation provision of depressuring or special fire fighting facilities is needed.

Indeed, equations presented in Section 5.15.2.2.2 of API 521 will result in very conservative relief loads because: 1) the vessel is supposed to be uninsulated, 2) vapor is assumed to be an ideal gas, 3) temperature of the vapor inside vessel is assumed to be uniform,....
Hence normally for a vessel containing vapor which is subject to a pool fire a relatively small PSV is adequate mostly in order to code compliance...and therefore PSV sizing for such situation can be ignored.

Fallah

[pallavikhatri]

Dear All
I have a query related to PSV on an unwetted vessel. my normal operating temperature is 45 deg C and normal pressure is 0.3 kg/cm2g while the design pressure is 10.5 kg/cm2g. for fire case my relieving temperature is coming to be much higher (1122 deg C approx.) than the maximum wall temperature that can be sustained by carbon steel vessel. so in that case what should be the mitigating plan to avoid such higher temperatures?

Thanks in advance
Regards
Pallavi

[fallah]

pallavi,

As per API 521, to mitigate going to such high temperature in un-wetted wall vessels you can apply: water sprays, depressuring, fireproofing, ...

Fallah

[Tsquare09]

Hi Shin,

As you have confirmed that the compressed air vessel is within a fire zone, then you need to size the PSV for fire case. However there is no procedure in API 520/521 to determine the relief load for external fire for a vessel with 100% air content. You may use the following steps below:

1. Determine number of moles of air in the system at normal condition

2. Determine relieving temperature using ideal gas equation

3. Then determine the temperature rise due to fire using properties at relieving conditions

4. Assumed a reasonable time (say initial 20 secs) for the temperature to rise from initial condition and then work out the new number of moles of air generated due to fire (this will be greater than the initial moles at normal conditions)

5. Determine the difference in the number of moles and then calculate the flow rate.

Use the same step to work out the flow rate assuming the vessel is still exposed to fire for another 10 secs. Whichever rate is greater should be taken as your required relief rate.

Stephen

Edited by SolaX@PI, 27 November 2012 - 07:00 AM.

[shin29]

Dear stephen,
thanks for your kind response but i am unaable to understand what do you exactly mean by " number of moles of air generated " as the mass of air do not change with temperature rise, though raise in pressure occure. Do you suggest to calculate relieving capacity as per following:
1. We can calculate amount of heat absorbed by gas in given time (say 20 sec)
2. Than we can also calculate temperature rise in specified time and pressure increase , thus relieveing capacity required will be amount of gas ( may be in moles) to be relieved for keeping the pressure to its initial state

I dont know if it is a better way in case of PSV design in fire case . As i am logically covinced with the reason given my MR fallah in post #14

Warm regards,
Shin

[kkala]

I remember a local case of sizing PSVs for fire on vessels containing only gas or vapor (2007). Procedure complied with API RP521 (1997), which indicated sizing formula just after the known formulas on heat transmitted to the vessel during fire (Q=21000FA^0.82). This was probably similar to Stephen's (tsquare's) post above, although API does not give explanations. It must be what Olidin mentions.
However we believed that this was rather formal than effective, offering little safety in case of fire. Additional measures suggested by fallah in post No 9 had to be adopted.

[Subhendu]

Hi I am new to this site, I found this very helpful.. I need one small help.

I want to size a PRV located on a vessel for Fire case, can any one guide me how to calculate the relieving temperature and latent heat of vaporization using HYSYS software.

Hope to get a quick reply from you all !