19 replies to this topic

[mkhooi]

When sizing PSV for boiling liquid (multi component) due o pool Fire, I’ve been told there are several methods to determine the latent heat of vaporization which are:

a) Rigorous method which are more tedious and represent the real scenario: http://webwormcpt.bl...nd-maximum.html (original from Joe Wong).

Using constant density method to bring the fluid from operating condition to relieving condition. After that, use a ‘Heater’ to vaporize 1% (mass) of the total inventory inside the vessel. The heater duty represents the sensible + latent heat. By using a cooler to down the liquid to the temperature before vaporization, we obtain the duty of the cooler which represents the sensible heat. Heater duty minus cooler duty is the Latent Heat of Vaporization.

c) From Hysys properties: Read directly from the stream ‘Mass Heat of Vap (kJ/kg). I have been told once that this property can only be applied for pure component and not valid for multicomponent. Is that so? Furthermore sometimes Hysys doesn’t show any value when we click on the liquid phase stream.

d) From Hysys properties: Using Mass Enthalpy (vapor) minus Mass Enthalpy (liquid). Seriously does this method apply for multi-component? I know it works for single component. However for multi component, the magnitude of the value doesn’t seem right. For eg. (Mass Enthalpy (vapor) = -4500 kJ/kg while Mass Enthalpy (liquid) =-150 kJ/kg). It would seem the liquid has more energy than the vapor phase which is not correct. Has anyone used this method or has any idea what this property represents for multi component?


Do feel free to drop your opinion as I have yet to determine which is the 'BEST' and Conservative method.

[mkhooi]

Sorry for being impatient, but was hoping whether Mr Joe Wong, or Mr Art and others could share their valuable opinions.


Thank you very much.

[Art Montemayor]

When I have to be conservative and don't have the Latent Heat of Vaporization of a liquid mixture - as in hydrocarbons - I frequently apply the the Latent Heat of Vaporization of Hexane.

This is an old-fashioned, but quick and conservative way to estimate a safe orifice size for a PSV in a pool fire scenario.

[sheiko]

d) From Hysys properties: Using Mass Enthalpy (vapor) minus Mass Enthalpy (liquid). Seriously does this method apply for multi-component?

Hi,

On item d)

Since specific enthalpy is an intensive property of the system, the difference between the vapor and liquid specific enthalpies has no physical meaning unless the liquid and vapor compositions upon which each specific enthalpy is based are the same. Because the vapor and liquid compositions of a pure-component system are in fact the same, it is valid to determine the latent heat by subtracting the specific enthalpy of the liquid phase.

In multicomponent systems, the composition of the liquid phase differs from that of the vapor phase at any vaporization point, except for special cases such as azeotropes...

[vikask8427]

mkhooi,

You can find out latent heat of vap in HYSYS using flash drum operation, specially for PSV sizing. Just define your stream, flash 10% (depending on your approach) of its mass by giving heat duty to the flash drum. Now subtract the M*cp* delta T (for vapor and liquid outlet from flash drum) from flash drum duty. Divide it by your overhead mass. Here is your heat of vaporization, which you can use to size PSV for fire case.

Also, if your fluid is two phase, you can conservatively take liquid part of it to size PSV for fire case.

[sheiko]

QUOTE (vikask8427 @ May 22 2009, 02:01 PM) mkhooi,

You can find out latent heat of vap in HYSYS using flash drum operation, specially for PSV sizing. Just define your stream, flash 10% (depending on your approach) of its mass by giving heat duty to the flash drum. Now subtract the M*cp* delta T (for vapor and liquid outlet from flash drum) from flash drum duty. Divide it by your overhead mass. Here is your heat of vaporization, which you can use to size PSV for fire case.

Also, if your fluid is two phase, you can conservatively take liquid part of it to size PSV for fire case.

vikask8427,

I did not checked it, but i think your procedure (which is similar to the method B from the OP) will give the latent heat of the light components (which have a higher mass-based latent heat than heavy components for hydrocarbons in general) in a wide-boiling-point-range mixture, which may indeed be desired in some cases.

However, in order to determine the total latent heat of a mixture, i would follow these steps:

Step 1: Designate a material stream, denoted as Stream 1, to represent the mixture in fully liquid phase at its bubble point; specify its composition and pressure, and assign a vapor fraction of 0 (representing bubble point)

Step 2: Designate a second stream, denoted as Stream 2, having the same composition and pressure as Stream 1 but assign a vapor fraction of 1, representing the dew point.

Step 3: Since the composition of Stream 1 is equal to that of Stream 2, the latent heat, is the difference between the specific enthalpies of the two streams.

Having said that, to ensure calculating relief loads that are neither excessive nor underestimated, the latent heat should be determined for incremental vaporization points of the system (until vaporization of the entire mixture). The volumetric relief load (or relief valve size) should then be calculated using the properties of the vapor generated at each of the increments. This approach gives a far more accurate indication of the required relief valve size.
You can refer to the article "Designing for pressure release during fires - Part 2" by S. Rahimi and al. released in Hydrocarbon Processing 2007.

Edited by sheiko, 30 October 2009 - 04:23 AM.

[maldini]

Dear all;
I want to ask about some methods I use it to get the latent heat
1) by hysis program I make reference stream to that I want to get its latent heat but I but I adjust the pressure to be the relieving pressure of the safety valve (as the pressure becomes high the latent heat decrease) and the temperature to be the bubble point

2) by the graph in the API 520 but this is only for single component but it sometimes becomes conservative

About point 1) is it right method and if is it right why specially we reach the bubble point as the fire heat is high we can reach the dew point or to be super heated

Can any one give me an advice?

[maldini]

sorry and also how can i calculate the exact relieving temperature because most of the times it becomes very very high (as per API Pn/Pr * T operating = T rel. which is very high)
and if the design temp. of the equipment is about 200 c and the relieving temp. reaches 300 or 350 c is it ok

[mkhooi]

When I have to be conservative and don't have the Latent Heat of Vaporization of a liquid mixture - as in hydrocarbons - I frequently apply the the Latent Heat of Vaporization of Hexane.

This is an old-fashioned, but quick and conservative way to estimate a safe orifice size for a PSV in a pool fire scenario.

Mr Art,

I suppose The latent heat of vap of Hexane is the same as figure quoted under 5.15.3.2 of API 521 where it states that

" If no accurate latent heat value is available for these hydrocarbons near the critical point, a minimum value of 115 kJ/kg(50Btu/lb) is sometimes acceptable as an approximation."

Please correct me if this is not the case.

[sachindhopade]

hope this helps u
http://www.processca...atent_Heat.aspx

[ankur2061]

hope this helps u
http://www.processca...atent_Heat.aspx

Sachindhopade,

Are you an agent of the above mentioned website or just a plain idiot who doesn't know what he is talking about.

[vikask8427]

mkhooi,

You can find out latent heat of vap in HYSYS using flash drum operation, specially for PSV sizing. Just define your stream, flash 10% (depending on your approach) of its mass by giving heat duty to the flash drum. Now subtract the M*cp* delta T (for vapor and liquid outlet from flash drum) from flash drum duty. Divide it by your overhead mass. Here is your heat of vaporization, which you can use to size PSV for fire case.

Also, if your fluid is two phase, you can conservatively take liquid part of it to size PSV for fire case.

vikask8427,

I did not checked it, but i think your procedure (which is similar to the method B from the OP) will give the latent heat of the light components (which have a higher mass-based latent heat than heavy components for hydrocarbons in general) in a wide-boiling-point-range mixture, which may indeed be desired in some cases.

However, in order to determine the total latent heat of a mixture, i would follow these steps:

Step 1: Designate a material stream, denoted as Stream 1, to represent the mixture in fully liquid phase at its bubble point; specify its composition and pressure, and assign a vapor fraction of 0 (representing bubble point)

Step 2: Designate a second stream, denoted as Stream 2, having the same composition and pressure as Stream 1 but assign a vapor fraction of 1, representing the dew point.

Step 3: Since the composition of Stream 1 is equal to that of Stream 2, the latent heat, is the difference between the enthalpies of the two streams.

Having said that, to ensure calculating relief loads that are neither excessive nor underestimated, the latent heat should be determined for incremental vaporization
points of the system (until vaporization of the entire mixture). The volumetric relief load (or relief valve size) should then be calculated using the properties of the vapor generated at each of the increments. This approach gives a far more accurate indication of the required relief valve size.
You can refer to the article "Designing for pressure release during fires - Part 1" by S. Rahimi and al. released in Hydrocarbon Processing, November 2007.

Sheiko,

You are right. Procedure that I described considers latent heat of lighters only. While sizing PSV for fire case, we should consider latent heat of lighters only. First lighters will evaporate wherein relief rate will be maximum (considering latent heat). Later on there will not be so high rate. Also, we should neglect sensible heat for fire case.

[jerald04]

From GPSA Section 5 'Relief Systems', it says that

"The latent heat will approach a minimum value near the critical. When no better information is available, a conservative minimum value of 110kJ/kg is typically used".

Comparing the heat of vapourisation values of other components, 110kJ/kg is considered very conservative.

[thakur]

dear,
u can easily calculate latent heat of vaporisation through following formula.dont need to go for any complicated method.

L= sumation of(Li*Mi*Y)/M

where
L= latent heat of vaporisation for mixture
Li= latent heat of vaporisation for individual component
Y= vapor fraction of individula component
Mi= molecular weight of individual component
M= molecular weight of mixture of vapor

U can easily obtain vapor fraction from aspen or hysis by flashing.

[demank]

how about if the vapor fraction very low near zero? the above equation still valid to that condition?

[thakur]

how about if the vapor fraction very low near zero? the above equation still valid to that condition?

dear,
yes the above equaltion holds valid for such condition also.
u can cross check it if you want.

[demank]

how about if the vapor fraction very low near zero? the above equation still valid to that condition?

dear,
yes the above equaltion holds valid for such condition also.
u can cross check it if you want.

If vapor fraction near zero, then latent of vaporization of mixture also zero, rigth?
I think there is an important parameter about this which is temperatur.
So relief temperature must be defined first for calculating latent heat.
Because of different temperature, different vapor fraction when the fluid still not yet 100% vaporize.

[thakur]

yes u r absolutely right.
relieving temprature ultimately gives implication on the vapor fraction.vapor fraction changes according to temprature and hence the latent heat.

[bobbobs]

I am a junior engineer and very new to relief valves. I have been trying to gather information on how to determine relieving temperatures.

For the fire case, shouldn't the relieving temperature be the boiling point of the mixture?

[JoeWong]

bobbobs,
Welcome...
This thread typically for Latent heat discussion. Suggest you open a new thread...