8 replies to this topic

[Olof Persson]

Hello!

My first post here.

I'm currently working my way through actions points raised in a Hazop that I participated in. Several action points are concerning dimensions of tank vents, some to atmosphere and some to a scrubber system. The two worst case scenarios that have been identified are;

1) Leveling between communicating atmospheric storage tanks via a shared suction line (HDPE, DN100, PN16). The storage tanks are 188 m3 GRP tanks, (D 3,75 m, H 17,5 m) and in the worst case scenario levelling occurs between a full and an empty tank. Overflow pipe on both tanks is a DN100 HDPE PN16 pipe.

My first question is if it's reasonable to dimension the storage tanks and vent so they can handle levelling as described above? A simple solution is to interlock the block valves on the storage tanks to prevent leveling but this is done at the cost of "operational flexibility".

2) A 16 m3 CE reactor is emptied using compressed air at 3 bar(g) into a atmospheric tank, in case a control valve fails 18 m3 of compressed air will expand into the atmospheric tanks with the risk of rupture of, or damage to, the atmospheric tank.

Question two, what kind of software (or even better) spread sheet would be useful when performing the required dynamic pressure drop calculations required to size the vent and overflow pipe of the atmospheric tank?


I believe that I'm asking for a lot here but any help would be appreciated. I'm sure there are things that need clarifying so please fire away!

BR

Olof Persson


M.Sc. Chem.Eng

Consultant

[ankur2061]

Olof,

Norml venting of atmospheric or low pressure storage tanks is dealt in detail in API STD 2000. There have been many discussions on the forum on normal venting rquirement of atmospheric storage tanks. You can use the search feature of the forum using key words to get a lot of information on tank venting posts that have been discussed in the past on the forum. I will provide you a few links to start with:

http://www.cheresour...h__1#entry25536

http://www.cheresour...h__1#entry35912

http://www.cheresour...h__1#entry37821

http://www.cheresour...h__1#entry38996

Regards,
Ankur.

[Olof Persson]

Thank you Ankur.

ISO 28300 doesn't seem like it applies to my case mainly because the product kept in the tanks has very low vapour pressure and thermal venting is a non-issue (nor is it flammable), EN 14015 seems like a better standard to turn to for me. Thanks again!

BR

Olof Persson

M.Sc. Chem.Eng
Consultant

[ankur2061]

Olof,

One rule of thumb for low VP and non-flammable products is to size the vent equak to or one size higher than the largest of the inlet or outlet nozzle on the tank..

Thermal venting is due to the contraction or expamsion of the vapor in the vapor space caused by sudden reduction or sudden increase in ambient temperatures. If the vapor pressure of the stored liquid does not change over a fairly wide range of ambient temperatures then only the thermal venting case can be discounted otherwiswe it certainly needs to be considered in arriving at the total inbreathing and outbreathing rate.

Regards,
Ankur.

[kkala]

I'm currently working my way through actions points raised in a Hazop that I participated in. Several action points are concerning dimensions of tank vents, some to atmosphere and some to a scrubber system. The two worst case scenarios that have been identified are;
1) Leveling between communicating atmospheric storage tanks via a shared suction line (HDPE, DN100, PN16). The storage tanks are 188 m3 GRP tanks, (D 3,75 m, H 17,5 m) and in the worst case scenario levelling occurs between a full and an empty tank.Overflow pipe on both tanks is a DN100 HDPE PN16 pipe.
My first question is if it's reasonable to dimension the storage tanks and vent so they can handle levelling as described above? A simple solution is to interlock the block valves on the storage tanks to prevent leveling but this is done at the cost of "operational flexibility"..

I think the contingency of flow from one tank to another, as described, is credible; it can be a result of inadverted manipulation of valves at the suction line. So tank vents have to be sized accordingly.
Local work practices for refineries consider max flows in & out for vent sizing of much bigger API 650 tanks. These max flows are usually due to pumps, being higher than natural flow from tank to tank (neglected).
Note: Check valves on suction line would reduce flow from tank to tank, see post of daryon in http://www.cheresour...h__1#entry39173. Nevertheless this is not usual in suction lines due to additional ΔP required by check valves, which need special care to be reliable. I would try to avoid it.
2) A 16 m3 CE reactor is emptied using compressed air at 3 bar(g) into a atmospheric tank, in case a control valve fails 18 m3 of compressed air will expand into the atmospheric tanks with the risk of rupture of, or damage to, the atmospheric tank.
Question two, what kind of software (or even better) spread sheet would be useful when performing the required dynamic pressure drop calculations required to size the vent and overflow pipe of the atmospheric tank?.
To my understanding approximately 3x18=54 std m3 of gas shall have to pass from vent(s) without causing overpressure. The question is to specify max gas flow release (to be varying) during emptying. It sounds like depressuring, with gas flow release specified by max Cv of mentioned control valve (quite open, max Cv to be used according to valve supplier). Max flow would then occur at the beginning of the release, when pressure is 3 Barg. As fas as I know from forum, a module of Hysys can be used (I have not experience on it).
It is assumed that reactor does not contain solid or liquid to result in polyphase flow. An orifice (suitably sized) is sometimes located on the line to tanks to control excess flow.
Welcome to the forum, generally more data and a sketch would help for more precise answers.

Edited by kkala, 10 June 2011 - 02:10 PM.

[Olof Persson]

Hi

Storage tanks
Ankur, vapour pressure of the medium is about 70 % that of water and the tanks are made from GRP with fairly low heat conductivity. With that in mind I believe that I can disregard the thermal in and out breathing and simply focus on the pressure caused by the hydrostatic pressure in case of levelling. Size of largest outlet (suction line) is DN150 so a DN200 vent line combined with the DN100 overflow line "feels" like it should be sufficient, will check of course.

kkala, lining up several pumps to the storage tanks is not possible and the pumps on the suction line from the storage tanks are protected against back flow with check valves. Still levelling of the tanks can occur as a result of manipulation as you said. Will try Chemcad to get a feel the size of the vent line.

Raector + atmospheric tank
You've understood me correct, 18 x 3 bar(g) = 54 std m3 of air is what's in the reactor when it has been emptied, this air will then pass through a DN100 CS/PTFE pipeline (no less than 25 meters long with a fair number of tees and valves on it) pushing the product in front of it before it expands into the atmospheric tank. The atmospheric tank will have about 25 m3 free volume when the air enters it. I have Chemcad but nothing else, wish I had FluidFlow or Aspen but I'll have to make do with what I've got.

Thank you both for your answers!

/Olof

[kkala]

If my understanding is right, mentioned 54 std m3 of air will pass from a control valve 100% opened, being at failure (so this control valve is F.O.). This offers the main restriction to flow, and venting capacity of tanks should consider this flow. Max Cv of this valve has to be used, taken from its supplier (or its catalogs).
As a simplified example, attached conval.xls tries to estimate this max air flow based on arbitrary data, e.g. max Cv=50. In the case of conval.xls, max air flow to pass from vents (when air is cut from control valve) is about 0.84 std m3 /s.
An orifice on the 4" discharge pipe could be tried, if judged appropriate and flow through control valve is too high.
Note: I am practicing in refineries, where tank flows due to pumping are usually much higher, so I have not seen leveling flows investigated.

Attached Files

•  convalve.xls   15KB   108 downloads

Edited by kkala, 15 June 2011 - 03:06 PM.

[Root]

Olof,

2nd part of your question need lot of calculation and not simple to calculate vent size. It is better for you to consult "Chemical Process Safety Fundamentals with Applications" Chapter 8, 9.

In this book lot of equations are given with different scenarios.
Good luck

Mr Alof,

Not sure if you have got a solution to your question on software. I do have a suggestion here. There one called 'Enardo Sizing program' which does sizing of PVRV and EPRV (assuming you know what those terms are). If it's a emergency venting you will first need a wetted surface area in ft2 and set points on the EPRV and MAWP of the tank for the program to determine the vent size. Of course we have to always double check with API 2000 or your favourite spreadsheet if the numbers are jiving.

Another way to use this program which I believe is for validation is to give a vent size and ask for the SCFH(a) number that will keep the tanks just below its MAWP.

Sorry I am not very good in expressing - But hope the above helps.

Maditha