station piping 31.3 tied in to the pipeline with higher MOP - reverse flow protection

[rob5377]

Dear All,

The gas plant is tied-in to the pipeline A. The gas plant design pressure is P1, and the pipeline A DP is P2. The pipeline after few kilometres will end with the metering station and will be tied to the another pipeline (B) than design pressure is P3 see attached sketch)

The Client suggested that pipeline A and everything downstream should have the design pressure P2=P3.
so P1 < P2 = P3
1st concern is how to prevent the gas plant from the gas flow-back scenario after Plant ESD action for example. I could not find any information in 31.8 for this case.

One of the solutions that comes to my mind could be to install pressure switch downstream the ESDV-3 and to close ESDV-3 once the pressure is close to P1. But then it is pointless to design both the pipeline A and the metering station for P3 design pressure.

The client does not want to hear about HIPP
I have not came across in the 31.8 how to prevent from back flow overpressure scenario. I will go through 31.3, but maybe someone is familiar with any requirements for that.
I would appreciate your comments and any advices.
Thank you.

 

[BigInch]

In general I think your client has a pretty high level of knowledge. HIPPS are NOT permitted as a pressure protection system by B31.8. If you read the code you will find that electronic protection systems are not a requirement for overpressure protection, however RELIEF VALVES ARE REQUIRED if overpressure is possible, thereby making a HIPPS totally redundant. You can waste the money on them if you want, but you will STILL need to have relief valves. It is a very good sign that your client knows this.

My remaining comments are as follows,

1.) It is not clear how your plant will experience high pressure from backflow after the ESDs close. BUT, if it is somehow possible, say ESD valves do not close and P3 is hi hi, or a compressor inside the plant goes to hi hi pressure, the only correct way to protect the plant is to install relief valves in the appropriate locations.

2.) ESDV-1 is not in the proper location. The code requires that the PIPELINE BE ISOLATED FROM THE PLANT, so that should be located at the dividing line between pipeline and plant, the code break line, or even closer to the plant, not way out there on the pipeline. Technically when this ESD closes you will isolate the pipeline from the launcher, but pipeline components, the launcher and the feed line, will still be connected to the plant. In many situations there may be a considerable amount of high pressure gas stored in the pipeline's feeder line which could backflow from there into the plant.

3.) Pressure fluctuations during pig launching could trip ESDV-1 just before, or just as the pig is running though the valve. Another reason that I always locate ESD valves off the pipeline,between pipeline and plant/metering/treating stations whenever possible.

4.) ESDV-2 should also be located between meter station and pipeline.

5.) ESDV-3 is pretty close to where it needs to be. It IS isolating the pipeline from the meter station. Perhaps it should be at moved a tiny bit to the code break line to fully comply word for word with the B31.8 code.

6.) It is extremely unusual not to see at least one check valve installed at the metering station. Most installations have at least one check valve upstream and many will have one downstream too. NO BACKFLOW should be permitted to go through a meter. If the system is bi-directional, then two separate flow paths, each one way only, each with check valves in the proper directions, should be installed. Lockable, double block and bleed valves should be installed at any common connecting points to prevent any possibility of reverse directional flows from going through the meters.

What would you be doing, if you knew that you could not fail? Ans. Bonds and derivative brokering.

 

[rob5377]

BigInch thanks for replay.

It is not clear how your plant will experience high pressure from backflow after the ESDs close. BUT, if it is somehow possible, say ESD valves do not close and P3 is hi hi, or a compressor inside the plant goes to hi hi pressure, the only correct way to protect the plant is to install relief valves in the appropriate locations.

The only source the plant may experience high pressure from is the pipeline.The plant design pressure is 80 barg, the pipeline A design pressure is 100 barg. Certainly there is no chance to "boost" the pipeline A above 80 barg, but the HH pressure may come from the pipeline B: say there is an ESD at the plant. ESD-1 is closed. In meantime due to backflow from the pipeline B pressure in pipeline reaches HH. Then there is a risk of opening the ESDV-1 and the gas with HH pressure backflows into the plant.

ESDV-1 is not in the proper location. The code requires that the PIPELINE BE ISOLATED FROM THE PLANT, so that should be located at the dividing line between pipeline and plant, the code break line, or even closer to the plant, not way out there on the pipeline. Technically when this ESD closes you will isolate the pipeline from the launcher, but pipeline components, the launcher and the feed line, will still be connected to the plant.

In many situations there may be a considerable amount of high pressure gas stored in the pipeline's feeder line which could backflow from there into the plant.

The code requirements appeals to me, however I do not see the scenario of high pressure gas from the feeder backflows into the plant. The pressure in the plant's piping and in the feeder is the same. So no matter the ESDV-1 valve is located you have the same problem. Do you specify the ESDV located on the feeder underground or aboveground?
For the fire case and ESD with blowndown I would locate the ESDV-1 valve underground and then where to have the code break on the valve?
On the code break there is a flanged isolation valve not shown on the sketch and this is the place where the code break is located.

My understanding of the check valve is to prevent from backflow through the metering station only not as a overpressure protection. My experience is that HAZOP would not approve the check valve to protect from backflow from the higher to lower pressure system. What is your experience?


thanks again for fruitful discussion.

 

[BigInch]

The plant is functioning with full discharge pressure into the pipelines and ESD1 closes. ESD1 closed due to a general shutdown because someone saw a fire and pushed the shutdown button. The fire at the compressor discharge weakens the discharge pipe and it ruptures. The pipeline compressor discharge - feed line will then backflow to the compressor and feed the fire.

What would you be doing, if you knew that you could not fail? Ans. Bonds and derivative brokering.

 

[BigInch]

Above ground or underground has no significance in deciding which code you use, or where the code break needs to be made.
B31.3 can be below ground, B31.8 above, or v/v.

Check valves at the meter only prevent metered gas from backflowing and either being lost forever, or from being metered again when flow returns to normal. Nobody likes to pay for gas two times.

Check valves are always considered to leak, never, never a positive seal.

What would you be doing, if you knew that you could not fail? Ans. Bonds and derivative brokering.

 

[rob5377]

The plant is functioning with full discharge pressure into the pipelines and ESD1 closes. ESD1 closed due to a general shutdown because someone saw a fire and pushed the shutdown button. The fire at the compressor discharge weakens the discharge pipe and it ruptures. The pipeline compressor discharge - feed line will then backflow to the compressor and feed the fire.

BigInch, now I see your point.
Apology that I have not put the whole story on the table which makes difficult to discuss.
There is an ESD level one which in case of fire blowdowns the plant piping (upstream the ESDV-1). So the fire can be fed with the gas only if the ESDV-1 leaks no matter if it was on the feeder or on the pipleine.

Does the code requires that the PIPELINE BE ISOLATED FROM THE PLANT by ESD valve? On the updated sketch I marked-up the isolation valve on which the code break is.
I could not find in 31.8 that ESD valve shall not be covered by 31.8.

 

[BigInch]

843.3.3
Compressor station shall be provided with ESD system by means of which the gas can be blocked out of the station and the station gas piping blown down.

The extent of the shutdown required, compressors, fired heaters, blowdown activation, etc. leaves no question that it must be an automatic shut-down system, therefore manual valves are not counting.

If you have an ESD Level-1 shutdown valve inside the plant, why do you need the one out on the pipeline?

If a P1 Plant Level-1 ESD activates ESDV-1 to supposedly block the pipeline from the station, ESDV-1 does not block what might be a large amount of "pipeline pipe" and what could be a considerable amount of gas contained inside that pipeline in the feed line and the launchers. All that gas could backflow into the station piping to feed a fire or wait for an ignition source. Redrawing the code break line to some manual valve while you leave the real ESD valve way out there on the pipeline still leaves gas free to come into the station.

What would you be doing, if you knew that you could not fail? Ans. Bonds and derivative brokering.

 

[rob5377]

BigInch
What about the following scenario:
The plant is functioning with full discharge pressure into the pipelines and ESDV located at the pipeline feeder closes. ESDV closed due to a general shutdown because someone saw a fire and pushed the shutdown button. The fire is in the vicinity of the pipeline feeder (say close to the above ground flange of the ESDV) and weakens the feeder and it ruptures. In that case you feed the fire from the pipeline section few km in length (depends how far away is the LBV).

Regards

 

[BigInch]

Not a creditable scenario, since the chance of a fire IN the plant piping located inside the compressor station is many more times greater than a fire farther away where the code break and ESD valve should be. Furthermore the object is to protect people that may be working at the station and the high value equipment located within, not some piece of pipe on the pipeline. Thus the optimum location is as near to the compressor units as possible, but still outside the high fire risk zone. The high fire risk zone is where all the piping flanged connections to equipment overlap with the radaii of possible electrical sources of ignition. If you do an electrical area classification study, like you should, these high risk zones will become apparent and the ESD should be located safely outside those hazardous zones.

What would you be doing, if you knew that you could not fail? Ans. Bonds and derivative brokering.

 

[rob5377]

BigInch, the plant is the brownfield and the scrapers are located inside the plant fence, not outside as it should be done.

The pig launcher is located 8 meters away from the existing metering skid.

 

[BigInch]

That's petty close, but probably still outside the hazardous plant areas.

So are the launchers "pipeline" or "plant" piping. The ESD should divide them.

What would you be doing, if you knew that you could not fail? Ans. Bonds and derivative brokering.

 

[rob5377]

The launchers are pipeline piping.
BigInch thanks for your comments and advices, but the discussion is a bit offtopic. I could not find any code requirements for the situation that lower design pressure station is tied-up to the pipeline with higher design pressure. If the pressure spec break was at the ESDV-3 my proposal would be to install checkvalve, PS and ESDV-3. The PS would triger esd at the metering plant. The pipeline A would be protected against hi pressure. There would be no risk of the back flow to the station upstream ESDV-1 since pressure on the both sided of the ESDV-1 would remain the same.

 

[BigInch]

[/b]HINT[/b] Please start reading the code on page 1 and continue until you have finished. You will never learn anything by making word searches in Adobe Acrobat. I see many engineers in training having English as a second language that try searching codes, but without even knowing the right words to search for, they don't often find what they are looking for. Please don't be one of them. Sorry if you don't think it applies to your case, but this next paragraph is in bold black type.

845 CONTROL AND LIMITING OF GAS PRESSURE
845.1 Basic Requirement for Protection Against Accidental Overpressuring

Every pipeline, main, distribution system, customer’s meter and ..., if
connected to a compressor or to a gas source where the failure of pressure control or other causes might result in a pressure that would exceed the maximum allowable operating pressure of the facility (refer to para. 805.2.1), shall be equipped with suitable pressure-relieving or pressure-limiting devices. ...[/quote]

For the requirements for PRs, see
ss 845.3 Requirements for Design of Pressure Relief and Pressure-Limiting Installations
and
ss 845.4 Capacity of Pressure-Relieving and Pressure-Limiting Stations.

A check valve by itself doesn't qualify in meeting the above criteria.

What would you be doing, if you knew that you could not fail? Ans. Gov lobbyist.

 

[rob5377]

BigInch,

Thanks for advice, but I usually read the text first. That is true not from the page 1 until the end but sometimes I skip not relevant sections.
The point is I treated the text literally and could find any example of overpressure protection case due to back-flow. "Gas source" definition was in my case too narrow.

By accident I came across Shell DEP which shows the ESD valve location on the pipeline instead of on the feeder, so as usual more than one option is acceptable.

Regards
Rob

 

[BigInch]

You're welcome. IMO, you are not correct in blindly accepting Shell's design as compliant to B31.x and more than one option may not be acceptable. Shell doesn't necessarily design to B31 code in all their facilities around the world. You see Shell, no pun intended, is the client and they can decide what criteria they follow when there is no specific overriding legal design code in effect, which is true in may of their locations around the world. In the US, for example, the law is CFR 192 & 195, which is pretty much the same, but it can vary, so they must at least follow the most stringent requirement, the CFRs actually controlling in any case of doubt. So, in your case it depends what your design contract specifies, assuming you are not working on a Shell project. If you are and your contract specifies Shell DPs, you're OK. If your contract specifies B31.x then you are incorrect in accepting a Shell standard without at least confirming first with your client that such a deviation will be an acceptable and you are not legally bound to a B31.8-2010 edition code. I have already pointed out in a similar recent thread that even a Shell DP may not be acceptable to Shell under all circumstances. Even they allow for cases where their standards are not applicable.

What's your definition of a gas source. Only a gas well, or what? A source in this case is any place in the system that high pressure gas can arrive into your piping system. I can't imagine what other definition could apply.

OK, it's your ice. Skate out as far as you want.

If it ain't broke, don't fix it. If it's not safe ... make it that way.