Low Pressure PSV Calculation 2

[GeorgeGrande]

I am designing an organic fluid tank.

Tank size: 45 m3.
L: 27.9 feet
D:8.5 feet
H elevation: 8.5 feet

Horizontal cylindrical vessel.

I followed the "Crosby engineering handbook" recipe for calculating "Fire conditions sizing for vaporing liquids" and got all the importent values. Yet, i think the size dont fit the conditions.

This are my calculations. It led me to relieving area of 2" and wet area of 545 sq feet. Relief pressure is 0.5 Barg.

L feet 27.9 length
D feet 8.5 Diameter
V feet2 1590 Volume
f 0.3 Insulated vess'
LH Btu/lb 148.8 Latent heat
F 6.8 Liquid level
H feet 8.5 Above ground
K1 15.3 Effective total height of liquid surface, feet.
E 6.8 Effective liquid level,
B deg 126.9 Effective liquid level angle,
T oR 570 SAT temp at 0.5 barg
P psia 21
P1 psia 39.91 Relieving pressure, absolute. This is the set pressure (psig) + overpressure (psi) + atmospheric pressure.
C 325 table
K 0.975
Kb 1
M 72.15
Z 1

I think it should be around 3 or 4 inches.

If some one can find what am i doing wrong i would be
grateful.

Thanks in advance...

 

[MortenA]

First: Why do you think it should be larger?

To check: Did you use the formula for choked flow or non-choked flow? My guess is that at the low pressure the flow will not be critical?

Best regards

Morten

 

[GeorgeGrande]

I think my problem is with the flow out. I get about 7700 lb/h.
I think that while dealing with volatile organic fluid at low pressure and high temp, i sould get larger flow.

 

[SeanB]

What is the heat input you have calculated? What environmental factor did you use? If you have calculated your heat input correctly your relieving rate should just be your heat input divided by your latent heat at relieving conditions.

 

[GeorgeGrande]

I calculated it by :

Q=21000*environmental factor*wet area^0.82

I calculated the flow from the answer.

 

[MikeClay]

What is the design standard? API-2000 is for low pressure storage tanks (< 1.034 barg). The equations for Q give twice the value of API-521.

MortonA also makes a good point. Subsonic flow will also increase the size of the relief device.

--Mike--

 

[SeanB]

Using the wetted area you stated above of 545 ft² and an environmental factor of 1.0 I calculate the heat input using your equation as 3.68 MMBtu/hr. When I divide that by your latent heat of 148.8 Btu/lb I get a relieving rate of 24,744 lb/hr.

 

[GeorgeGrande]

The environmental factor is 0.3 (for this vessel). I also checked mortonA point, my factor is 150/100 gives 1.66. for sonic flow the factor is above 1.7. I guess it makes calculation even tougher. Isnt there a calculation way thats take all this factors. isnt there anybody out there who solved this problem?

 

[pleckner]

"MikeClay" is on the right track. Your tank is atypical for API 2000 (horizontal, cylindrical) but because of the desgin pressure, it indeed falls within API 2000 scope and not ASME Section VIII, Div. 1 scope (nor API RP520/521).

I quote from APR RP520, "Atmospheric and low pressure storage tanks covered in API Standard 2000 and pressure vessels used for the transportation of products in bulk or shipping containers are not within the scope of this recommended practice."

ASME and API RP520.521 only apply to vessels designed for pressures greater than 15 psig. Horizontal cylindrical tanks are usually designed for pressure service.

Since your vessel sits pretty low, I would take the entire surface area in the calcuation and use the appropriate equation in API 2000. The venting rate will be calculated in terms of SCFH of air. You would use a venting device more in line with a ProtectoSeal or Groth (weighted pallet)unit rather than a Crosby or Farris relief valve. You can use pilot operated relief valves intended for low pressure relief service as well.

 

[MortenA]

another thing - P1= 39.91 psia=1,75 barg - seems somewhat higher than the stated 05 barg? this btw higher than 15 psig.

Best regards

Morten

 

[pleckner]

The stated set pressure is about 7.25 psig (0.5 barg). If atmospheric pressure is 14.7 psig, then the stated set pressure in absolute terms is 21.95 psia; off from the stated 21 psia but OK. Now, the allowed over pressure for an ASME coded vessel for fire is 21% of gage set pressure, NOT absolute pressure so the over pressure would be about 1.52 psig (0.21 x 7.25), making the relieving pressure 7.25 + 1.52 or 8.77 psig. In absolute terms, relieving pressure would be 23.47 psia, not the 39.91 psia given. It appears 'GeorgeGrande' is working in psia to get the overpressure and then adding it again to atmospheric pressure to get the relieving pressure, and this is wrong.

BUT, there is a more fundamental problem here. As I and others pointed out, this is not an ASME coded vessel but an API 2000 vessel and as such, does not warrant the allowable over pressure. The relieving pressure must be 0.5 barg without any over pressure allowed. 'GeorgeGrande' needs to calculate the vent size based on API 2000 requirements, not ASME.

 

[GeorgeGrande]

Thank you all for Guiding and helping. It is good to know you are out there.Each one of you....

 

[GeorgeGrande]

Well, using your help and readind API 2000 i applied two companies.
The first one gave me again the 2" proposition (For vent and fire). still i thought it is impossible using 2" as i am using the same size at design pressure of 10 Bar.
The seconed company gave me: 2" for venting and 12" for fire sizing. That seems like a train more than a tank but is is also sound more reliable.
Can i have your opinion in this subject.

 

[pleckner]

1. Which vendors did you talk to?
2. Which models of their vents did they quote you?
3. What was the venting rate they based their calculations on?
4. What is the flash point of your liquid?
5. What is the pump-in rate they used for venting in their calculations?
6. What set point (pressure) did they use in their calculations?
7. Can you give us some of the vapor properties for the fire case (API 2000 puts everything in terms of air)?

The size of a vent is very much dependent on the set pressure you gave the vendors and the models of the devices they are choosing for you. For instance, I tried running your numbers through ProtectoSeal's sizing program and it couldn't even pick a device with the set pressure you stated in your original post. As I manipulated information, I was getting a whole host of vent sizes. This is why I am asking you my question number 6.

One way for you to check this yourself is to calculate the venting rate for both out-breathing and fire per API 2000 as we discussed above. Then do a hydraulics calcuation using these venting rates. Choose a line size (nozzle will be the same size as the line). Start at the final vent pressure (atmosphere if going to atmoshere or some system pressure if venting into a flare system for example). Work backwards and calculate the pressure in the vessel. This pressure cannot exceed your vessel design pressure of 0.5 barg. If it does, then you started with too small of a vent size. If the pressure is very much less than 0.5 barg, thenyou chose too large of a vent size.

The problem with your system is that you are designed for an intermediate pressure, somewhere between what even API 2000 is meant to cover and ASME.

To summarize, let us know the details of what these vendors used in their calculations and the devices they are proposing and perhaps we can evaluate these. Try doing the calculations base strictly on hydraulics and see what you come up with. We can then try to address what type of device to use.

 

[mechprocess]

I have done similar relief flow requirement calculations and I found that the APA 520 section 3.6.3 Sizing for subcritical flow: Gas or Vapor equations produce the same results as those in API 2000 using the coefficient of discharge method.

SCFH is determined from heat and the relief capacity is based on the area sizing equations.

 

[pleckner]

These equations are all derived from the same source and are only presented differently, which is why they produce the same results when plugging in the same system information.

My problem with using API 520 in this particular application is that it still discusses the issue in terms of ASME and allowable over pressures. API 2000 does not allow over pressure and as pointed out above, the vessel in question does not fall within ASME Section VIII, Div. 1 scope. Therefore, P1 in the equations has to be the set pressure only and not include over pressure. Note that you can still have a subcritical system and thus need these API equations for an ASME scope vessel, and this is why the over pressure is included in the API 520 definition for the inlet pressure, P1.

Secondly, I don't like any of API equations except in terms of obtaining theoretical flows and size, precisely because of the use of coefficient of discharge values. For API, these are generic in nature and will change depending on the actual device used. This is why ProtectoSeal can't give me a device with the set pressure I used in the sizing software. Their devices are limited in some aspects as will be devices from other vendors.

So, bottom line, calculate the vent rates (normal venting and emergency/fire) and go to the vendors to get the proper hardware. Vendors will be glad to assist, they don't bite.

 

[GeorgeGrande]

1. I talked with ProtectoSeal and Groth .
2. a. 9812P for fire sizing, 19540 for normal breath.
b.2301A
3. 3600 SCMH for the first and 566 NCMH for the seconed one.
4.-18 deg C
5. 200 m3/h
6. 0.45 bar
7. n-pentane at 1.5 bara 538 kg/m3 (l) 4.3 kg/m3 (vap)

The tank is insulated and has a NFPA fire system.
Corrected offer stand on 6" for insulated tank (ProtectoSeal). Seems right as DP on 6" is 0.043 bar (by my claculations like you advised).

 

[GeorgeGrande]

Sorry..

for number 2:

a. 9806P instead of 9812P