PSV case for Fire 2

[razook]

We have pipilines of sizes 8 to 10 inches containing hydrocarbon liquids which also contain valves at both ends. ( Can be isolated). Should we consider fire case for this isolated pipe and provide PSV or is it that we consider PSV for the connected vessel.

 

[pleckner]

Is this a long pipe line or a process line? The answer may have an impact on your decision.

Ask yourself:

1. Are there flanges between the valves that can leak?
2. You obviously can leak at the valves but can the leak produce a pool that can sustain a fire?
3. If there were a pool of liquid, is there an ignition source?
4. Is the pipe within 25' elevation of where this pool may develop?
5. The vessels, if ASME coded, will certainly have PSVs. These PSVs will protect the piping as well, assuming of course the pipe is open to the vessels and not isolated. But then you must ask if you should include the surface area of pipe in the vessel fire calculation. The answer is yes, especially for the larger sized pipes. Unfortunately, many engineers forget about the piping within the fire zone figuring that their surface are is insignificant compared to the vessels they are attached to. This is true for some but not all. The engineer needs to make that judgement call and should always err on the conservative side.

I am anxious to see how "BigInch" weighs in on this with his experience in pipe lines.

 

[BigInch]

BigInch weighing in,

A pipeline should NEVER depend on the vessel reliefs for several reasons,

[COLOR=white red]Pipeline PSVs will probably need a different pressure setting, as Msr Pleckner and myself discussed, was it just yesterday? [/color] Apparently there are max permitted overpressure differences between the BVP, B31.3 and B31.4 and B31.8. I haven't checked the paragraphs yet, but beware!

1.) Vessel PSVs will NOT usually have sufficient capacity to relieve the plant piping, never mind a long pipeline. I can't say for Msr. Pleckner, but I know of many vessels that have probably a 1-1/2" valve, where the connecting pipeline has 2 x 8 or a 10" or valves. Besides that really depends on the number of parallel vessels connecting to the pipeline and their configurations and such anyway.
2.) there is no guarantee that valves entering a vessels will be open, especially when YOU need them.
3.) Usually there are ESD valves between a pipeline and the station or plant that may be physically located well inside the plant where segments may be exposed to fire, but before the ESD valves, but that does depend on the plot layout. They will close if there is a fire and no connection to the vessel will be available.
3.) The pipeline operators need to know when the pipeline PSVs are open and will usually have a flow switch signal on the lines to the PSVs that mayalso be connected to a SCADA system (if SCADA is used or one will be installed at some time in the future, so control information will be lost.
3.) often there are two maintenance organizations within a company, (or the vessel might even belong to a different company) one handling inside plant and the other outside plant and responsibility for testing and maintenance is often confussed.
4.) Same comment for operators. Different operators may be responsible for in plant and pipeline valves. Check with the plant and pipeline supervisors.
5.) accurate testing and maintenance of PSVs need to be kept with a pipeline database completely different from a plant's records.
6.) Lastly, I don't like depending on someone else to guarantee the safety of my pipelines, and it may be a code issue as well. I don't know the technical answer to this, but I will propose the question. Does a vessel PSV located in a plant count for pipeline protection? I doubt it. By definition they are different facilities and may even be under different codes (see pressure setting note above). If its a compressor or pump station covered under the same design code as the pipeline, you stand a better chance with that argument, but if its a code break, good luck. Offshore, I believe MMS will not permit a pipeline without separate relief valves, but check the latest info on that.

So, to summarize, no.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[pleckner]

BigInch:

By the way, you can drop the "s", it's Phil Leckner.

For the most part, I think we agree.

My expertise is in plant design so I don't have a real good feel for long pipeline design. Perhaps the industry has a different approach for these and as this is your expertise, I yield to your wisdom and judgement.

But (there is always at least one "but", isn't there?) B31.3, 2004 edition, Chapter IX, High Pressure Piping, Part 6, Systems, Section K322.6, Page 118 - gives the same pressure relief requirments for pipe as does ASME Section VIII, Div 1 for vessels. And I have to add that in all my years, I've never recommended putting a PSV on process piping within the plant boundary (and perhaps I might have to re-think this in the future, see my comment at the end) unless it was liquid full and I had to protect it against thermal expansion, and that was usually due to possible solar heating.

As I pointed out above, and at the risk of repeating myself, during the fire calculation for vessels the piping surface area is seldom included in the wetted surface area for the vessel, but it should be; and I am guilty of this as well.

I always specify my piping with a design pressure and temperature equal to that of the source vessel (again, within the plant boundary). I don't know why it wouldn't be this way in every case? In every company I've been with (all E&Cs), this has pretty much been the rule.

Lastly (as I hinted above) B31.3, Appendix G, Safeguards does make one think about at least giving it some more thought!

 

[BigInch]

Msr.=Misure -Fr.

First let me add to the reasons above,
7.) Possible alteration or modification of plant equipment in future.

OK, had a look at B31.3 - but the 2002 edition. I give a summary of what I found,
*********************************************************
302.2.4 Self-limiting events +20% if < 50 hrs & 500 hr/yr

322.6.3 Liquid thermal expansion set pressure < test press and < 120% design pressure

322.6.3 Relieving apacity of any pressure relieving device shall include consideration of all piping systems which it protcts.

322.6.3 (2) Liquid thermal expansion relief device which protects only a blocked-in portion of a piping system, the set pressure shall not exceed 120%
******
Chapter VII Nonmetallic Piping and Piping LIned with Nonmetals

A302.2.4 (a) nonmetallic - variations above P & T not permitted.
A302.2.4 (b) metallic lined - permitted if lining sutiable
******
Chapter VIII Category M Fluid Service
M302.2.4 Allowances in para. 302.2.4 is not permitted.
M322.6.3 Metallic piping - 110% of design pressure during operation of a pressure relieving system.
*****
Chapter IX High Pressure Piping
K302.2.4 variations in pressure above the design pressure not permitted, except during relieving
K322.6.3 (a) capacity of ? relieving devices must prevent pressure rise above 10% of design pressure AT operating Temp for 1 relieving device. No more than 16% above with >1 relief device. IF 1 device is set to D.P. and none set higher than 105% of design pressure.

(c) Supplementary pressure relieving devices provided
for protection against overpressure due to fire
or other unexpected sources of external heat shall be
set to operate at a pressure not greater than 110% of
the design pressure of the piping system and shall be
capable of limiting the maximum pressure during relief
to no more than 121% of the design pressure.
*********************************************************
B31.4 Transportation Systems for Liquid Hydrocarbons and other Liquids

402.2.4 Ratings - Allowance for Variations From
Normal Operations. Surge pressures in a liquid pipeline
are produced by a change in the velocity of the moving
stream that results from shutting down of a pump
station or pumping unit, closing of a valve, or blockage
of the moving stream.
Surge pressure attenuates (decreases in intensity) as
it moves away from its point of origin.
Surge calculations shall be made, and adequate controls
and protective equipment shall be provided, so
that the level of pressure rise due to surges and other
variations from normal operations shall not exceed the
internal design pressure at any point in the piping
system and equipment by more than 10%.

************************************************
B31.8

805.217 Overpressure protection is provided by a device or equipment installed for preventing the pressure in a pressure vessel, a pipeline, or a distribution system from exceeding a predetermined value. This protection may be obtained by installing a pressure relief station or a pressure limiting station.

843.411 Pressure relief or other suitable protective devices of sufficient capacity and sensitivity shall be installed and maintained to ensure that the masximum allowable operating pressure of the station piping and equipment is not exceeded by more than 10%.

843.442 Relief is required immediately after +displacement compressor or on pipeline.

=============== my conclusion is ===========================

Given the above posts, The B31.3 code appears to agree with BVP VIII in this respect.

B31.3 does not agree with B31.4 nor B31.8

Both B31.4 and 8 limit maximum pressures to 10% higher than design pressure.

B31.3 permits variations well in excess of 10% overpressure under some circumstances. Of note are the many possible allowances well above 10%, "with permission of the owner". The pipeline codes have no provision to exceed 10% overage under any circumstance. And in the USA, CFR 49, Parts 192 & 195 would legally prohibit a regulated pipeline from that practice.
************************************************
IMO all the abov tend to justify separate relief installations in between any facilities that are not immediately associated with a pipeline or pipeline pump or compressor station. That IMO could not possibly include any facility designed to B31.3, given the many exceptions to the 10% pipeline overpressure limit sumarized above.

I have always ordered PSVs on vessels associated with pipeline work to be fitted with PSVs set to a max of 10% over the design pressure and have always protected pipelines and associated facilities independently of liquid plant and refinery systems. I don't believe it is safe to do otherwise, esp. in the long term where changes and modifications can often occur without any exchange of information between pipeline company and refinery or liquid plant, power plant operators, etc.

I would like to hear other's opinions about this.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[pleckner]

I took spanish!

B31.4 appears to cover a different scope, sort of the same type of difference you might get between ASME Section VIII and Section I.

The quote you show for B31.8, Section 843.411 interests me.

I agree that more opinions would be nice...especially from lawyers, insurance people and municipal code enforcers!

 

[BigInch]

That's why I proposed the question about coverage across codes. It could definately be a spanner in the works.

B31.8/843.411 The 10% gives the same limit for gas as what is given for liquid pipelines in B31.4

I actually found para. 843.442 more interesting in that it apparently requires PSVs to be installed "ON THE PIPELINE", which would/could essentially prohibit them from being considered as pipeline protection, if they were to be installed on a vessel.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[razook]

Thank you BigInch and pleckner for your valuable suggestions and reference from codes. Though we could not fully conclude, I think I shall infer that pipe line area definetly has to be taken for fire wetted area if below 25 feet.
Regrding PSV on pipeline, although there is refernce about having one in pipeline, I suppose I can take a decision based on case to case.
I think I am concluding right. please comment

Razook

 

[BigInch]

Since you ask for more comments,

For the specific thermal relief in a fire case, and if there are no block valves between the protected pipe and the relief valve, I'd put it on the vessel. If there is a block valve between, I'd say no, put one on the piping and the vessel.

Non fire cases (overpressure protection), I would tend to put the PSVs directy where needed, one on the vessel, one on the pipeline, considering all reasons above, however if the above logic isn't relavent to your case, I'd see nothing wrong with only one on the vessel, but only if there are no block valves inbetween.

As you can tell, I do not like block valves in front of relief valves. Although they are permitted, I've seen too many of them forgotton and locked closed while a system is operating, so I avoid them whenever possible. Its much easier to keep an operation safe, if Safety is in the design.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com