8 replies to this topic

[jprocess]

Dear All,
Relief load calculation should be done for reception facility as a part of a gas gathering and NGL recovery project. The reception facility consists of a pig receiver, a finger type slug catcher and a condensate filter.
The concern here is "How fire case relief calculation for slug catcher should be done?"
This case study is a special one because:
1.Slug catcher is finger type pipe array, 8 fingers of 40'' nominal diameter, 200 m long. The liquids collection header is 40'' nominal diameter and estimated at 15m long.
A rough calculation results a wetted area of 4287 m2!!!!!! So as you know the absorbed heat during fire scenario is a function of wetted exposed area. If you consider ground as source of flame, according to API 521 a height of 25 ft above it should be considered for exposed wetted area. Consequently all the calculated wetted area should be considered as exposed area which results in high value of absorbed heat.
The first question is that in this special case, should the designer consider all the wetted exposed area?
Our approach to this question was according to part "5.20.2.2 Vapour from fire-heat input" of API 521 (2007 EDITION) that states:
"The extent of an assumed fire zone is a function of the design and installation features that permit confining a fire within a given area (see 4.3.14). Although the size of the assumed fire zone can vary, experience generally indicates that a fire that can be confined to approximately 232 m2 (2 500 ft2) of plot area will not affect the design of the main relief headers in processing areas where a depressuring flow discharges into the same relief header."
Also, a valuable paper published in chemical engineering journal (May 2000) titled" Fires, Vessels and the Pressure Relief Valve" by Wing Y.Wong states that:
"API RP 521 recommends that the probable maximum extent of fire incident is limited to a ground area ranging from 2500 to 5000 ft2.The equivalent circle has a diameter of about 57-80 ft. Most refineries use 5000 ft2"
So, we decided to use a value of 5000 ft2 for fire zone that equals to approximately only 25 m of slug catcher in comparison with its real length which as I stated earlier equals to 200 m.
2.Like any other fire relief calculation the designer needs to specify latent heat of vaporization for slug catcher contents.
The approach that usually used for latent heat calculation of a mixture is using simulators like HYSYS to flash 1% of contents at its bubble point and relieving pressure (Set pressure+Accumulation which is 21% for fire case+Atm.Pressure) in an imaginary heater with zero pressure drop and dividing heater duty by rate of vapor generated to obtain latent heat.
There are different ideas about the extent of flash that should be done: 1%, 5%, 5-30% and etc. All of them can be right but I believe that flashing 1% of liquid contents would obtain the most conservative value of latent heat.
By the way this approach can be followed until the liquid is below its critical conditions. At supercritical conditions HYSYS can not find a temperature for a stream compatible with its bubble point and relieving pressure!
Here the slug catcher design pressure is 47 Barg but its operating pressure is only 17 Barg .It seems that this high value of design pressure for slug catcher is specified according to shut off pressure of compressors that exist at production station.
So the slug catcher liquid contents will reach its critical conditions before the pressure safety valve reach its set pressure!
The same situation may occur for compressor suction drums which their design pressures should be specified with considering the settle out pressure of compression loop.
So the second question is that "how latent heat of vaporization should be calculated?!"
A. Part "5.15.3.2 Vapour" of API 521(2007 EDITION) states that:
"The recommended practice of finding a relief vapour flow rate from the heat input to the vessel and from the latent heat of liquid contained in the vessel becomes invalid near the critical point of the fluid, where the latent heat approaches zero and the sensible heat dominates.
If no accurate latent heat value is available for these hydrocarbons near the critical point, a minimum value of 115 kJ/kg (50 Btu/lb) is sometimes acceptable as an approximation."
B. The above mentioned paper states that:
"API RP 520 says that it is acceptable to use 50 Btu/lb as an approximation for latent heat. Some refineries specify slightly lower than 50 Btu/lb. The lowest value of latent heat of vaporization should never be less than 40 Btu/lb."
C. A recently published method in a paper in chemical engineering journal titled "Rigorously Size Relief Valves for Supercritical Fluids" contains valuable material about concerned subject. This paper can be downloaded freely from:
http://www.clarkson....gn/reliefv2.pdf
Although this paper is valuable and offer a new method but as you know with papers we can not talk to clients!
Our approach was to use a value of 50 Btu/lb for latent heat.
I should be grateful if you kindly would answer to these questions.
Warm Regards.
Mojtaba Habibi

[JoeWong]

Mojtaba,
I am suggesting you to revisit your earlier post...

http://www.cheresour...x...639&hl=slug

i believe it partly answer your question...

JoeWong

[jprocess]

Dear JoeWong,
Thanks a lot for your reply for this topic and also for my old topic.
My old and new topics are about one problem but as you see the new one is a review of our activitis and I send it to know your valuable comments about the path which we followed.
I read the old topic.
First of all I did not understand your mean by :
"The only different is the pool fire extent may be limited by the concrete slab below." specially the concrete slab part!
You and gvdlans speak about the likehood of pool fire.I agree that fire is a rare case in plant life but we always consider it for relief study.Do not we?!
You stated some points about considering the vertical portion of slug catcher:
" A slopping slugcatcher with 200m lenght may result the gas end high and potentially exceed 7.63m requirements. You may have the chance not to consider the gas end. Just check..."
I am willing to know your comment about 200m length and our approach,also about value of 50 for latent heat.
Thanks in advance.
Cheers.

[gvdlans]

I consider a pool fire likely at locations where I have equipment or piping (partly) filled with flammable liquids and where there are possible leakage points (e.g. flanged connections, small bore piping). For the middle part of a finger-type slugcatcher I do not consider a pool fire likely provided you have large bore, all welded pipes, as well as a corrosion monitoring program in place.

[JoeWong]

"The only different is the pool fire extent may be limited by the concrete slab below." specially the concrete slab part!"

This is quite inline with gvdlans explaination. Slug catcher liquid end with has a lot flange connection, control valve, level gauges, etc. All these are leak source. Normally liquid containment e.g.concrete slab is provided underneath to capture any liquid. Those this is the ideal location for pool fire.

However, slug catcher body (long finger) is just a pipe. Hardly any liquid leak at this area and no concrete slab aroung undeneath area. Hence, there is unlikely to have any fire. This was based on the assumption where very unlikely tank truck passing this area and severe leakage around this area and proper drainage system provided to evacuate any HC away from slugcatcher area.

You stated some points about considering the vertical portion of slug catcher:
" A slopping slugcatcher with 200m lenght may result the gas end high and potentially exceed 7.63m requirements. You may have the chance not to consider the gas end. Just check..."

Sometime the liquid end is elevated, slopping of slugcatcher may lift the gas end higher than 25 ft, thus, you have some chance not to consider pool fire around this area.

about value of 50 for latent heat.

This is just a guideline given by API521. There are many approach on fluid latent heat. Please read this post...

http://www.cheresour...?showtopic=3742

Please also read the personal note send sometime ago.

Hope this help.
JoeWong

[markk]

If your slug catcher is a finger type is it designed to the piping codes or to the vessel code (ASME VIII)?

If it is designed to the piping code and not the vessel code you may not need a relief valve for fire case at all - after all you don't put fire relief valves on other pieces of pipe which could be blocked in - only vessels.

Since your slug catcher is designed for the same pressure as the upstream pipeline - I assume the upstream pipeline is good for the shut in pressure of the upstream system or has it's own relief valve - you wouldn't need a relief valve on the slug catcher at all.

If you do require a fire case relief valve then since the fluid is supercritical (i.e. in dense phase) the relief case will be gas expansion rather than boiling fluid.

You will need one on the pig receiver however.

[JoeWong]

markk,
Welcome...

Some questions from your statement, appreciate your advice.

You will need one on the pig receiver however.

I have seen pig receiver designed per piping code. Do you think relief valve is required in this case ?

If it is designed to the piping code and not the vessel code you may not need a relief valve for fire case at all - after all you don't put fire relief valves on other pieces of pipe which could be blocked in - only vessels.

Above statements implied that RV required for system with low inventory i.e. receiver BUT not required for a system with large inventory i.e slugcatcher ? Isn't the slugtacher with large inventory more dangerous ? Do you mind to share your the logic behind ?

JoeWong

[markk]

Sorry, I can see the statements are confusing. I will try to explain.

The pig receiver can trap in liquid. If the liquid is a lower temperature than the solar radiation temperature for the plant then a thermal relief case will exist even if the pig receiver is designed to piping codes.

Whether you put a fire relief valve on either piece of equipment is down to the risk since it is not required by the piping code itself.

Due to the operation mode of pig receivers the risk of leakage is much greater than for finger type slug catchers. However the propagation consequences of fire at a slug catcher are obviously higher than at a pig receiver.

So is one required at the slug catcher? I certainly wouldn't remove one that is already there unless I could make a good case for it being safe without one. This would probably have to include showing that a leak could not develope a pool fire under the slug catcher by arranging drainage to take the liquid away from the slug catcher.

In a strange coincidence the client has (today) questioned whether a fire relief valve is required for the slug catcher on the job I'm currently working on. Although I'm not working on the slug catcher area on this job, I'll keep an eye on the outcome.

[JoeWong]

markk,
Thanks for your quick response.

The pig receiver can trap in liquid. If the liquid is a lower temperature than the solar radiation temperature for the plant then a thermal relief case will exist even if the pig receiver is designed to piping codes.

Whether you put a fire relief valve on either piece of equipment is down to the risk since it is not required by the piping code itself.

Due to the operation mode of pig receivers the risk of leakage is much greater than for finger type slug catchers. However the propagation consequences of fire at a slug catcher are obviously higher than at a pig receiver.

Your statements implied that
i) If pressure containing part design to PIPING code i.e. B31.3, NO PSV for fire. If design to VESSEL code i.e. ASME Sect 8 Div II, PSV shall be provided.
ii) Higher inventory with higher risk

I hope i have correctly interpreted your statements.

You may see that INVENTORY(associates RISK) come into picture from (ii).

Your statement in earlier post...

If it is designed to the piping code and not the vessel code you may not need a relief valve for fire case at all...

I would say requirement of PSV should NOT be 100% judged from the design code itself. The RISK and CONSEQUENCE e.g. INVENTORY associates risk & consequences shall come into consideration to define if a overpressure protection i.e PSV is required. Conservative approach shall always in mind whenever engineer is dealing with SAFETY and something UNSURE...

Again thanks for sharing your idea...


JoeWong