External Design Pressure of API 650 Tank

[MOHAMMED89]

Dear All,

I need your advice on how to calculate the external design pressure for a API 650 tank, tank details are as follows:-

Tank Size - 30m (dia) x 25m (height)
Service - Aviation Fuel
Internal Design Pressure - 13 Kpag.
Remarks - The internal design pressure is finalized based on the nitrogen blanketing positive pressure.
Roof type - Fixed Cone Roof

what will be the external design pressure ?

Thanks in advance

 

[LittleInch]

It's absolutely minute.

Can't recall exactly but something like 4 or 6 inches water column. Edit: Actually one inch of water (see below).

Basically your blanketing system needs to be always positive or have a vacuum valve as a backup set at something very low.

It's in API 650 somewhere.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[goutam_freelance]

Please refer the following from API-650. But do check with latest code as mine is not up-to-date.

Engineers, think what we have done to the environment !

https://www.linkedin.com/in/goutam-das-59743b30/

 

[IFRs]

You should not use the API maximum design external pressure (vacuum) for your design.
You should use your actual vacuum based on operations.
Too many engineers look at API 650 and use the maximum allowed pressure or vaccuum that the standard allows.
This just makes for unnecessarily expensive tanks - all the way from design to material fabrication to erection, maintenance and operation.
Even your valves may be expensive - pilot operated style depending on your N2 flow rates.

Your blanketing system will have a valve that defines the worst case vacuum.
For the main body of the standard, the maximum vacuum is 0.25 kPa or 1 inch of water.

By the way, your 13 kPa pressure will require massive design alterations.
The roof plate only weighs abount 0.367 kPa so you will need to tie it down, the internal pressure will require a very large compression ring and a foundation and anchoring system capable of withstanding around 3 million pouunds of uplift.
All in all, my suggestion is to use a basic API 650 internal floating roof tank with circulation vents and zero internal pressure or vacuum.
Aviation fuel is very often stored in large atmospheric floating roof tanks - although is has low volatility the cleaner and drier you can keep it, the less expensive it is to maintain downstream quality.

API does have an annex to allow vacuum up to 6.9 kPa but for a tank this size the stiffening rings will be very expensive.

 

[MOHAMMED89]

Dear LittleInch, goutam & IFRs,

Thanks for your valuable advice.

 

[LittleInch]

IFRS offers great advice as always.

Blanketing pressure is 130 mbar.

Why?

That sounds very high to me for a 30 m diam tsnk.

So what are you going to do now? We like to know what happens next.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[IFRs]

I sometimes find that a blanket system is initially designed without consideration of the effect the pressure and vacuum have on the storage tank.

The fill and empty rates, environmental thermal effects, normal and emergency tank venting, purge gas availability and equipment costs all go into the initial design along with easily justifiable head rooom, safety factors, economical valves and piping, etc. It is easy to add a few kPa along the way at each stage of the N2 suppply and recovery system. Then, at the end someone goes "oh, what about the tank - we better check on that" and find out that for any substantially sized tank the effects are huge and the cost skyrockets.

On one set of tanks about the size of this one, they had to add many (many, many) tons of steel to each tank to counter the internal pressure (anchors were not possible due to foundation limitations) and this was after some hard conversations and tough decisions reducing all head rooms, safety factors etc to minimums and using multiple large pilot operated valves. In this case they had to blanket the tanks for several reasons.

In the OP's case blanketing is an easy option until the tank cost is examined. Hopefully the OP is early enough in the process to not get stuck with an economically unviable project.

 

[MOHAMMED89]

Thanks all for your valuable advices.

Actually let me clear my scenario to avoid confusions as raised above:-

The tank design pressure of 13Kpa is because the service product vapour pressure is about 10Kpa at 40degC, and to apply a positive pressure on the top of the product and by considering the set pressure of PVRV we arrived to a value of 13 kpa.

And also the tank is in FEED stage so no issues with the foundation, as we have room for design.

Actually we discussed with client to replace the nitrogen blanketing with internal floating room, But as per the latest industrial practice and code requirement we are forced to proceed with nitrogen blanketing.

 

[IFRs]

The vapor pressure is not "pressure" as it seems you may be thninking of it. The pressure inside the tank even if the tank were 100% sealed would not exceed atmospheric just because the aviation fuel has a vapor pressure of 10 kPa. Vapor pressure of a liquid is a partial pressure. If your produect is indeed standard AV Gas (which I suspect now since Jet Fuel has a much lower vapor pressure) then you have no need for any of the above discussion and a standard API 650 atmoshperic tank with conservation vents or an internal floating roof and circulation vents will do just fine.

 

[LittleInch]

I agree - you're not thinking correctly here.

13kPa is actually quite low in terms of vapour pressure - I assume this at some sort of elevated storage temperature?

So yes, in a freely vented fixed roof tank, some of the product will vapourise and mix with the air to create a potentially explosive / flammable mixture and over time some it will escape via the vents, especially during filling. This and potential issues to do with product quality mean that inert gas blanketing is called for. But not at 130mbar.

You need to go and revise that to something much lower.

The vacuum / negative pressure issue remains though. supply of gas needs to exceed max vapour changes either because the vapour suddenly cools if it rains on the tank (really) and also the max outflow rate.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[e43u8]

@IFR

"...The pressure inside the tank even if the tank were 100% sealed would not exceed atmospheric just because the aviation fuel has a vapor pressure of 10 kPa."

Suppose the tank has been filled with aviation fuel at atmospheric pressure and a temperature corresponding to a vapor pressure of 5 kpa (partial pressure of aviation fuel in mixing with air) then the tank was 100% sealed and the temperature goes up to 40degC means the vapor pressure would be 10 kpa. At this situation the partial pressure of the aviation fuel increased by 5 kpa hence the pressure of the vapor space inside the tank will be increased above the previous atmospheric pressure means the pressure inside the tank will be almost 1 atm.+5 kpa.

 

[IFRs]

I'm not really good at this, my understanding of it is incomplete, I'm happy to be wrong or to have misunderstood the question.

The example of a closed container was perhaps not a good one, especially for temperature changes. On the other hand, remember to use absolute temperatures when trying PV=nRT type evaluations. From 60C to 100C (about 333C to 373C) the pressure should increase about 12% assuming a bunch of things - it will not double.

What I was trying to get at was that the partial vapor pressure of fuel does not change the absolute pressure of a system at equilibrium.

If you partially fill a container with fuel and then put on the cover, the pressure inside won't change unless temperature changes. You can fill a glass with fuel, put a sheet of paper on the top and it won't be lifted off by the 5kPa vapor pressure even though the paper weighs much less than 5kPa. If you had a perfect seal it would not even smell. For this example it does not matter if the room tempeature is 40C or 80C as long as it is not 100C (or whatever the boiing point of the liquid is) and the temperature is constant during your experiment.

From britannica.com: Vapour pressure is a measure of the tendency of a material to change into the gaseous or vapour state, and it increases with temperature. The temperature at which the vapour pressure at the surface of a liquid becomes equal to the pressure exerted by the surroundings is called the boiling point of the liquid.

If you have a tank with a piston floating on the liquid surface, the piston will remain in contact with the liquid surface until the vapor pressure of the liquid exceeds both the atmospheric pressure above the piston and the weight of the piston. If the piston is weightless it will still not rise above the liquid surface until the vapor pressure exceeds atmospheric.

Gasoline can have a vapor pressure of around 70 kPa (10 psi) yet you don't see rectangular gas cans changing shape or floating roofs lifting off the liquid surface inside tanks.

My whole point (I think) is that the TVP of the fuel does not suggest that the tank needs to be designed for pressure, and the blanketing gas does not need to overcome the fuel's vapor pressure in order to enter the tank, because it is a partial vapor pressure. It is very common to use internal floating roof tanks for volatile liquid storage, and the US EPA allows such emission controls for liquids whose TVP is up to 11.1 psi [76.6 kPa] (see 40 CFR part 60 subpart kb paragraph 60.112b).

 

[LittleInch]

IFRS,

Being a bit pedantic here but this "(about 333C to 373C) " needs to be 333K to 373 K, not C

And gasoline vapour pressure is 70kPa absolute or 10 psia. In this case the "a" makes a difference...

You do get some fuel cans expanding when it gets warm and then a hissss when you remove the cap, but after that initial expansion of vapour it doesn't start to boil or emit a stream of vapour. Any gas under a floating roof tends to find its way to a vent. Even floating roofs have vapour emission, it's just a lot less than an open surface of liquid.

but you main point is correct AFAIK, i.e the vapour pressure is not the same thing as the blanketing pressure which the OP has massively miscalculated.



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[e43u8]

Please think about the thermal out-breathing in a sealed fixed roof tanks and suppose the relevant facility for pressure relief, say pressure breather valve haven't been considered. Doesn't the fluid temperature increase due to ambient temperature change, hence the vapor pressure increase, pressurize the tank above the design pressure mostly a little bit higher than atmospheric pressure?

 

[LittleInch]

Yes, but that's what your PVRV valve is for.

Just for a blanketed tank it is set a little bit ( few mbar) higher than atmospheric. Jus not 130 mbar higher than atmospheric....

I think we need to forget about temperature change in all of this. That's a different issue and subject.

The OP was initially asking about pressure in the tank below atmospheric pressure.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[IFRs]

Use API 2000 for sizing normal and emergency venting.
The tank design comes first.
A normal gasoline, AvGas or Jet fuel tank would be atmospheric design with either conservation (PV) vents, free vents or a floating roof and free vents.
If you want to nitrogen blanket the tank, the pressure is independent of the vapor pressure of the fuel, should not exceed the weight of the fixed roof plates and the vacuum should be 1 inch of water max - both to give you an economical tank design.

 

[HTURKAK]

I suspect that there is a mistake in OP's data... I am looking this thread and posts for a few days... and looked my past documents at my laptop and found one of the data sheets applicable for aviation fuel tank.

In past we have involved with a similar tank with dimensions - 40m (dia) x 25m (height),
Service - Aviation Fuel , Roof type - Aluminium Geodesic Dome Roof, Similar nitrogen blanketing concept and without IFR.
Design Pressure +10/- 5 mbarg. Operating temperature amb to 50°C..



Dear MOHAMMED89 (Mechanical)(OP), pls check your data again..The internal Design Pressure in original post could be 1.3 Kpag. or 13 mbarg. and provide more info. ( minimum ambient temp, filling and emptying rates,risk of rapid cooling....) to get more helpful responds..

I attached the following snap showing the vapor pressure of acetone to refresh the knowledges..

 

[e43u8]

@HTURKAK
Thanks for sharing a good example for clarification...
This example approves what i was saying about the pressure variation inside the tank due to temperature variation...

In fact, in third and fourth stages of your example the pressure inside the flask is higher than atmospheric due to fluid vapor pressure and, obviously, if the ambient temperature will increase; the pressure inside the flask will be increased accordingly due to higher vapor pressure...

 

[HTURKAK]

Dear e43u8 ; what ever you said is true from the material thermodynamics point of view. I want to remind also the evaporation is surface property. If you can avoid the contact of liquid surface with air / nitrogen ,( in this case ,with internal floating roof) evaporation will not occur so the vapour pressure will not develop or at least will be minimized....

The design and construction of storage tanks with this size and with design pressures 13 kPa or to full vapor pressure 40-50 kPa is not a viable option . Although API 650 appendix F allows design pressure up to 18 kPa , these are typically appendix F tanks , small sized shop assembled tanks.

The tank with D=30m, and with the internal pressure 13 kPa will experience 9200 kN uplift force.. That is, tank roof shall be designed for 13 kN/ M2 which is higher than highway bridge loading. 9200 kN uplift force will require 92000 mm2 total anchorage area, foundation having wt 15000 kN or tension piles..Moreover, The 13 kPa internal pressure implies 7 kPa external or vacuum pressure ( if we consider rapid cooling of the tank when it rains) , the bottom plates will not be typically 6 mm but a bottom frame with 15-20 mm thk. bottom plates will be necessary to resist vacuum pressure which i did not see, hear....

I have experience for the design of mega storage tanks but not an expert for the internal pressure and venting design.

I have similar opinion with IFR.. and IMO ,the subject tank;

- THE VENTING SHALL BE DESIGNED AS PER API STANDARD 2000,
- THE DESIGN PRESSURE SHOULD BE IN THE RANGE OF +10/- 5 kPa
- IN ORDER TO DICTATE THESE MAX/MIN PRESSURE, MINIMUM VENTING REQUIREMENTS SHALL BE ADOPTED WITH NECESSARY AMOUNT AND SIZE OF
VACUUM/VENT VALVES..
- SUITABLE SIZED VAPOR RECOVERY UNIT AND NITROGEN SUPPLY SYSTEM SHALL BE PROVIDED,
- THE USE OF IFR STILL A VIABLE OPTION ON THE TABLE TO MINIMIZE THE EVAPORATION .

 

[e43u8]

@HTURKAK,
Ok, i am involving the process design of the storage tanks for many years and agree with what you mentioned about the rules for storage tanks sizing based on API 650...
But the matter around which i did open a discussions, might different than what the OP initiated, was a basic concept claimed by IFR as follows:

"...The pressure inside the tank even if the tank were 100% sealed would not exceed atmospheric just because the aviation fuel has a vapor pressure of 10 kPa."

I did want just to say the pressure inside a sealed tank can exceed atmospheric pressure due to the vapor pressure increment of the fluid inside the tank...same as what is shown in the pictures provided in your previous post.