11 replies to this topic

[Evgeny]

Dear all,

There is a case I would like to hear your solution on.

Column. Demethanizer (cold section of ethylene unit)
Top: T=-26, P=31.6 C1+H2
Bottom: T=25, P=36.9 C2+

There is need to implement ESD so that all liquid is drained first and column is depressurized.
Contractor suggested to add automatic drain valve on bottom and deP valve on top.

But I think that deP valve can add more risk since if it opens before all liquid drained, the bottom T will be extremely droped (below design T).

Can we go without deP valve but drain valve only? Or how reliable this dual-valve system would be?
What would you suggest?

Thanks

[Homayun]

Evgeny,

It all depends where you want to drain the liquid and how you want to depressure your demethanizer. If the drain system is a lowe pressure system, then your bottoms are gonna evaporate and I don't recommend that. I think the best way is to first depressue the tower and then drain any liquid that is left there.

Depressuring should be automatic.

[Evgeny]

Hi Homayun,

Drain system is cold blow down (dry flare) and of course all C2+ will evaporate. Its not problem
Depressurizing first is not possible as it cools down bottom too low (its not stainless steel)

So, now I am wondering is it possible to drain&depressurize via drain valve. Although I have never seen such colution.


Evgeny

[Zauberberg]

As indicated in the previous post, normally there is no automatic (liquid) blowdown valve but rather a depressurization valve somewhere in the column overheads. This may be questioned in case that the tower is fed by stream(s) containing significant quantities of C5+ components which are still in liquid phase when blowdown/depressurizing is completed and the system remains at the final pressure. In ordinary circumstances, the available system volume should be more than sufficient for full liquid containment.

You will clearly identify the need for liquid BDV once when you see which section of the unit is isolated by ESD valves in case of emergency shutdown. I would expect ESD valves on all feed lines to be closed, which essentially eliminates the need for liquid blowdown facilities. If there are any specific items in your case that we are not aware of, then please upload the P&IDs of the system under consideration and we will be able to give you more accurate answers.

[Homayun]

Yevgeny,

What actually causes the bottom of tower to become too cold is vaporizing Liquid; but you already knew that. So if your drain system is connected to cold blow down, then you could drain the tower first - before depressuring - if your cold blow down system is good for the final temperautre.

What would happen is that immediately downstream of the drain valve, your liquid starts to evaporate and gets too cold. Make sure that you have 2 drain vavles in series. There have been major incidents with similar configurations draining LPG tanks. A single valve could freeze and fail leading to potential leakage. Without a second valve in series upstream, one can not isolate the system. So be aware of that.

So the next question is: is your drain system capable of handling 2-phase flow?

good luck..

[fra.telli]

Ciao Evgeny

may I ask a couple of dumb questions?

a - Normally ESD is actuated in order to rapidly depressurize reactants that could led to dangerous reactions if "leaved alone" inside the vessel (as for PSD/LSD). Normally 'drainage' is performed after depressurization, so: is the C2 volume so valued to risk liquid removal before gas ESD in presence of ESD signalling? Is it worth to risk for a drainage line freezing and a gas bypas towards drainage system?

b - Are you sure that depressurizing from 36 barg to 7 barg will cause metal cooling down to -10°C or lower? do you have such a huge gas volume and so little metal mass?

I think you can have a delayed gas ESD deP in favor of partial liquid drainage (using automatic valve), but back to question (a), wath's healthy?

Sorry, I've got a third question:

c - who "sold you" a process vessel that won't stand ESD temperature?

[gvdlans]

Hello Evgeny,

You wrote

There is need to implement ESD so that all liquid is drained first and column is depressurized.

Can you describe what that "need" is? So what scenario are you thinking of? Is it an external fire?

In that case, why don't you just block in your tower and leave it, so without draining/depressurizing?

What seems to be missing here is a structured, scenario/risk based approach...

[Evgeny]

Hello gvdlans,

Actually, according to standards we were required to design in such a way that during an emergency situation (e.g., fire) you can empty all inventories and depressurize.

When a cryogenic service is involved, you have to first drain, and then depressurize. Otherwise, you cool down your vessels to a very low T.

That's why I was worried when I saw depressurization and drain valves on one column (see above). To me, depressurization could develop into a more hazardous situation to the operation than without it.

These are on/off valves connected to a single ESD.

I know that leaving the column under pressure could, in some cases, be more logical. But you always need the possibility to empty the vessels and do it fast.

[gvdlans]

In case of an external fire, the liquid inside the column provides cooling to the vessel wall, preventing the column from failing. Draining away the liquid makes things worse. In this case of a Demethanizer column, depressurizing is essentially the same as draining, since you will boil off the liquid.

There are better ways to protect against this scenario (such as drainage systems that drain away the spilled flammable liquid, active and passive fire protection).

Having a standard does not mean you have to stop thinking.

Edit: added "In this case of a Demethanizer column". In other situations, depressurization can be a good option to prevent vessel collapse as a result of an external pool fire.

Edited by gvdlans, 29 October 2010 - 10:02 AM.

[Art Montemayor]

Evgeny:

If the emergency scenario is an external pool fire around the demethanizer, then Guido has asked the right question and responded with the proven, safety-related response. The immediate remedy to a fire case is the Pressure Safety Relief Valves (PSVs) on the column and the unit. As has been mentioned, under this scenario, the Emergency Shut Down valves would be actuated and the unit would be “blocked in” – with a normal, operational amount of liquids within the unit exposed to the effects of the external fire.

The engineering safety logic applied to this scenario is basically the following:

• The internal liquid inventory within the unit is a temporary “saving grace” for the unit: the liquids represent a heat sink equal to their latent heat. In other words, while the fire is raging outside the unit, the liquids inside are “boiling” and creating excessive vapor that must be relieved or vented. That is the purpose of the PSVs – to relieve all the internal vapors generated by the effects of the heat transfer from the fire. While there is liquid within the unit (within the height of the external flames), the unit can withstand the effects of the flames by simply generating internal vapors – which help to keep it relatively cool (compared to the fire). This is what happens in the normal operation of a direct-fired boiler: you have flames on one side of a steel wall and boiling water on the other.
• HOWEVER, once the liquid is depleted, drained, or evaporated, you have a very serious problem: the unit and all its vessels are exposed to reaching temperatures that reduce the safe, allowable stress of its material of construction. Mechanical collapse and failure is the next stage that the unit is exposed to unless other factors are used to save it. Some of the other factors are: insulation, water sprays, and finally depressurization. The latter, depressurization, does not insulate or remove heat from the vessels. All depressurization does is relieve the internal pressures within the unit and allow the material of construction to last longer due to less imposed stress from internal pressure. If the fire continues to persist, there is no further hope for the vessels and they will ultimately fail mechanically. All depressurization does is alleviate the stresses on the unit – temporarily. It does not assure nor promise salvation or redemption for the unit.
• During an external fire, if you have a gas-filled vessel without any hope of relying on latent heat within it, then you are at the mercy of any insulation or water spraying that you may have on it. If you have no insulation, water spray, or internal liquids, then the time that the vessel is exposed to the fire will tell whether it will fail or not. It is not practical to rely on insulation for protection from an external fire because it has been found that few insulation applications can resist the damage from fire. Water sprays can be effective – but they must be available and dependable for 100% of the time. Usually that is not the case. Internal liquid inventory continues to be the most dependable heat sink that can help the vessel during a fire. I know this sounds simplistic – but it has been proven to be the most practical and dependable field solution on fighting a fire – at least while the internal liquid exists and before it all evaporates.
• As simple and as practical as this logic is, it wasn’t until the last 20 or so years that the API started to give importance to the fact that gas-filled vessels present a potentially bad situation during an external pool fire. I’ve been an engineer for 50 years and for the first half of my career I can remember that the practice and importance of depressurization was not even mentioned in such documentation as API 520 and 521. Today, we have a much better knowledge and established practices of dealing with hazardous scenarios. Depressurization is important. It helps to buy time and is an attempt to stave off imminent vessel mechanical failure. However, internal liquid vaporization also helps to save the vessel by keeping it cool. It is important to understand and take into consideration that once all internal liquid is evaporated within a vessel, the need for depressurization is the next immediate step in order to buy time – should the fire still continue.

The above explains why I sponsor what Guido is stating: just because a plant standard procedure exists doesn’t mean that we should stop applying engineering logic and purpose to addressing a potential hazard. Draining the internal liquids of a unit during a fire scenario actually takes away a degree of inherent fire protection for that unit. While I am not against the need to drain liquids from a demethanizer, I feel that the importance that those liquids could pose during a fire is such that it should be discussed and evaluated.

[kkala]

Probably following may be of interest, even though the fire case discussion has been complete(indicating depressuring only after depletion of liquid).
1. Depressuring valve & drain (and PRV) can coexist at same tower, as seen in a local naphtha desulfurization colun. The depressuring valve is remote operated, so operator undertakes the task to open it on LLLA (no link to the valve).
2. As previous post advises, all liquid (CH4, C2H4, etc) of the column should be boiled off during fire, before depressuring is started (to avoid low temperature and extend protection period).
3. Another case can be a sudden need to empty the column for maintenance. The column should be drained from bottom (see post 21 Oct) without depressuring until all liquid goes out. There is still a drop in temperature (remaining liquid evaporates, so that its vapor occupies the empty space), but this can be estimated and assumingly results in higher temperature (compared to depressuring). Of corse depressuring valve and any available vent / manhole will be opened at the final stage.
4. Just for the above case, introduction of an inert gas (e.g. nitrogen) into the column would help to raise the boiling point of the contained cryogenic liquid (depending on the pressure developped), thus avoiding too low temperatures. This seems logical (heard once in a company lecture), even though never seen it applied.

Edited by kkala, 03 November 2010 - 12:33 PM.

[fallah]

There is need to implement ESD so that all liquid is drained first and column is depressurized.
Contractor suggested to add automatic drain valve on bottom and deP valve on top.

But I think that deP valve can add more risk since if it opens before all liquid drained, the bottom T will be extremely droped (below design T).

Can we go without deP valve but drain valve only? Or how reliable this dual-valve system would be?
What would you suggest?

You need to refer to Drainage Philosophy of your project,as the first priority.

Normally,Two types of drainage have been considered
− maintenance drainage
− operational drainage

♦ maintenance drainage
Before starting the maintenance operations for a vessel, the liquid inventory is minimized inside the vessel by transferring liquid to other vessels by means of process lines. In general the remaining quantity of liquid corresponds to the volume of liquid below the LLS.
Before being drained to the process closed drain, the vessel is isolated and depressurised. Drainage will be performed by opening adequate valves ; liquid under slight pressure conditions is sent by gravity flow to the dedicated sump.
♦ operational drainage
Operational drainage is the pressurized drainage of minor inventory such as :
− liquid volume contained in level instruments and drained manually

Low temperature condition that would be resulted due to depressurizing,should be considered in vessel's min design temperature determination.