7 replies to this topic

[bot]

Hi All,

We are in a process of having an AFC commissioned on a Flare Drum Pumpout Line ending up in a Crude Tank. This Crude Tank also receives general slops tank as a result of unit(s) upset. Don't mix these slops with oily sewer water (or storm drain).
What we have done so far is that we have picked up the worst (hottest & biggest) reliefs to made their way to respective unit flare drum then through the common flaredrum pumpout line into the Crude tank.
Hot reliefs thus picked are around 160 to 170 degC. If this hot load makes it way to the tank, causing the water in the tank to boil off - the roof of the tank will collapse with a tank out of service for more than 8 months.

I have done some quick heat balance to find out that for a mixing efficiency of 50%, the temperature of the water would increase from 25 to not more than 29.5 degC. We have also recommended a minimum tank heel of say X meter to protect the tank from potential failure.

Group of people supporting the commissioning of the cooler is saying that the pocket of vapors rising through the beds of Water and Crude if not back-absorbed would reach the roof disturbing its balance and thus resulting in catestrophic roof failure.

We have operated like this for last 12 years (without cooler) with all sort of emergencies successfully handled. Then why to have this commissioned now.

Could anyone share their experiences. or Advise accordingly .

[Qalander (Chem)]

Dear bot

Hello/Good Morning,

The Scenario you reffered reminds me some memories of my previous employer's days experiences Although of slightly different nature, as this sharing with you might be helpful in finding ways and means for safer conditions.

We had a blended F/Oil tank(Fixed cone Roof) recieving various incoming diversified nature componets i.e Kerosene~ VDU Slops tank's liquids, HVGO~PDA Resids,Vaccume bottoms and reduced crude from atmos bottoms.

The Tank was had hot insulated shell and Bottom MP steam Heating Coil Usual Temperature was kept very closed to 85~90 celcius.Occassionally settlled water was observed and regularly drained out to underground Oil Water sewer system. However we met with a Small Roof Rupture damage incident that we analyzed to occur due to some pocketed water(already at near boiling point 90~95 celcius) coming in contact with some much hotter incoming stream thus making sort of spontaneous volumetric expansion of water into steam
the tank's venting system capacity could not cater for this spontaneus volumetric outflow into tank vapour space
exerted huge pressure more(much higher than design of few inches of water column) and shell roof single fillet joint was torn around 15~20 feet.Thanks to God nothing more serious happenned.

This is indicative of possibilities of Catastrophe to take place although system remained operative for many years prior to the incident.

Moreover a possiblity of some propane(solvent)pocket could not be ruled out as the VDU slop tank might have been in-out with plant upsets could route pocketed streams into the tank with similar outcome (although this was not validated during analysis);Various corrective measure were followed strictly to yield safer outcomes thereafter.
Dear bot Hello/Good Morning,
The Scenario you referred reminds me some memories of my previous employer's days experiences Although of slightly different nature, as this sharing with you might be helpful in finding ways and means for safer conditions.
We had a blended F/Oil tank (Fixed cone Roof) receiving various incoming diversified nature components i.e Kerosene~ VDU Slops tank's liquids, HVGO~PDA Resids, Vacuum bottoms and reduced crude from atoms bottoms.
The Tank was had hot insulated shell and Bottom MP steam Heating Coil Usual Temperature was kept very closed to 85~90 Celsius. Occasionally settled water was observed and regularly drained out to underground Oily Water sewer system.
However we met with a Small Roof Rupture damage incident that we analyzed to occur
due to some pocketed water (already at near boiling point 90~95 Celsius)
coming in contact with some much hotter incoming stream
thus making sort of spontaneous volumetric expansion of water into steam
the tank's venting system capacity could not cater for this spontaneous volumetric outflow into tank vapor space
exerted huge pressure more(much higher than design of few inches of water column) and shell roof single fillet joint was torn around 15~20 feet. Thanks to God nothing more serious happened.
This is indicative of possibilities of Catastrophe to take place although system remained operative for many years prior to the incident.
Moreover a possibility of some propane(solvent)pocket could not be ruled out as the VDU slop tank might have been in-out with plant upsets material could route pocketed streams into the tank with similar outcome (although this was not validated during analysis);Various corrective measures were followed strictly to yield safer outcomes thereafter.

Although not directly related to Flare Drum liquids but may help.

Additionally in case of higher temperature volatile liquids containing streams inflow into even Floating Roof tanks disturbs the floating roof annular sealing system and occasionally liquids do find their ways even on top.
Hope above helps
Qalander

[bot]

Qalander Sb,
Thanks for your much valued response.
Referring to what you have just shared above, does this necessarily imply that we should avoid all forms of vapour pocket forming within a non-open roof tank?
Just thinking, if it can blow the chip off a cone roof tank then - a floating roof tank is more prone to these pockets.

Referring to a Potential LPG Leak through, we have already considered that bit, specifying tank minimum heel to cope with that sort of issues.

I was planning to run a fluid dynamic model for a pocket of the steam travelling up a column of the water and crude (or slop). From there, we could atleast come to some concrete results. I personally don’t like approximation and/or quick fixes.
I would be on a lookout if anyone has something to add to the topic.
However, I guess I have to resign to Snr. Process Engineer's will to commission this cooler back to its purpose.

[Zauberberg]

Bot,

The fact that you didn't have any bad experience with pumping 160-170C hot hydrocarbon condensate back to the slop oil tank is only because you haven't encountered this situation in practice - and I sincerely hope it will remain that way.

Introducing hot hydrocarbons (or water) in a storage/slop oil tank would be quite dramatic, especially if amounts to be pumped are substantial - which is exactly the case in the event of full flare system load (power/CW failure etc.), so the necessity of downstream cooler is not anymore a question of engineering judgement.

But even in such case, the problem remains. If flowrate of condensed light hydrocarbons is such they can be flashed-off in the downstream tank, you'll have a serious safety issue. I remember a story back from my refinery days when light naphtha storage tank has been set on fire after only a few hours of pumping off-spec (high RVP) product. Please do a full assessment/scenario developement for all possible cases, and then you'll be in position to make good engineering decision. Do not think much about what seniors are saying, but always try to observe things from an engineering, common-sense perspective.

Good luck,

[Qalander (Chem)]

Bot,

The fact that you didn't have any bad experience with pumping 160-170C hot hydrocarbon condensate back to the slop oil tank is only because you haven't encountered this situation in practice - and I sincerely hope it will remain that way.

Introducing hot hydrocarbons (or water) in a storage/slop oil tank would be quite dramatic, especially if amounts to be pumped are substantial - which is exactly the case in the event of full flare system load (power/CW failure etc.), so the necessity of downstream cooler is not anymore a question of engineering judgement.

But even in such case, the problem remains. If flowrate of condensed light hydrocarbons is such they can be flashed-off in the downstream tank, you'll have a serious safety issue. I remember a story back from my refinery days when light naphtha storage tank has been set on fire after only a few hours of pumping off-spec (high RVP) product. Please do a full assessment/scenario developement for all possible cases, and then you'll be in position to make good engineering decision. Do not think much about what seniors are saying, but always try to observe things from an engineering, common-sense perspective.

Good luck,

Dear bot/Zuberberg Hello and Good Morning,
This was astonishhing /fine coincidence that my friend zauberberg said what I would have.
Merely I would supplement my previous post as regards floating Roof Storage Tank case;

that you have to Thoroughly analyze the each scenario for proper addressing in design;

as for extra volatility potential components there are measures in design

to provide Vapour emmission (vent) Valve(s) spread out uniformly and even equipped with flame arrestors as the locality dictates.
Hope this helps
Qalander

[bot]

But even in such case, the problem remains. If flowrate of condensed light hydrocarbons is such they can be flashed-off in the downstream tank, you'll have a serious safety issue.

Z & Q,
Much thanks for your valued response too. I would like to add something here about the above quoted para.
Yes it’s a major safety breach if you are planning to send a high volatile load, down to the slop system. I crunched few numbers to accommodate that.
I would require some feedback on my philosophy.

I have considered LPG – sent down to the slop system. LPG would have a tendency to get absorbed by raw slops within the tank. If so, based on that, I ran quick simulation to find out (on an hour basis) the volume of the Slop required to cool down the hot LPG along with considerable absorption. Using that value, I gave Ops min tank heel they should have the whole time (or atleast before commencing pumpout operation).

Though that would help cushion the affect but isn’t an absolute solution.

[Qalander (Chem)]

But even in such case, the problem remains. If flowrate of condensed light hydrocarbons is such they can be flashed-off in the downstream tank, you'll have a serious safety issue.

Z & Q,
Much thanks for your valued response too. I would like to add something here about the above quoted para.
Yes it’s a major safety breach if you are planning to send a high volatile load, down to the slop system. I crunched few numbers to accommodate that.
I would require some feedback on my philosophy.

I have considered LPG – sent down to the slop system. LPG would have a tendency to get absorbed by raw slops within the tank. If so, based on that, I ran quick simulation to find out (on an hour basis) the volume of the Slop required to cool down the hot LPG along with considerable absorption. Using that value, I gave Ops min tank heel they should have the whole time (or atleast before commencing pumpout operation).

Though that would help cushion the affect but isn’t an absolute solution.

Dear bot,
May be some mechanical solution should supplement
e.g. A slow diffusion into the tank heel material using numerous tiny perforations
Specially designed to provide better and almost complete intimacy for suggested absorption; if mangeable.
Pardon my wild thought again!
Best regards
Qalander

[Zauberberg]

Bot,

I haven't seen anywhere that anything lighter than C5 boiling range has been continuously pumped to the light slop oil tanks. Whether they are fixed or floating roof, it doesn't make too much difference. As I mentioned before, it is a serious safety issue.

I believe the best solution would be to have flare VRU (vapor recovery unit) for fractionation between C2-/C3C4/C5+, from which you'll be able to produce compressed fuel gas, LPG, and light naphtha cuts.